text
stringlengths 87
880k
| pmid
stringlengths 1
8
| accession_id
stringlengths 9
10
| license
stringclasses 2
values | last_updated
stringlengths 19
19
| retracted
stringclasses 2
values | citation
stringlengths 22
94
| decoded_as
stringclasses 2
values | journal
stringlengths 3
48
| year
int32 1.95k
2.02k
| doi
stringlengths 3
61
| oa_subset
stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|
==== Front
J Infect
J Infect
The Journal of Infection
0163-4453
1532-2742
The British Infection Association. Published by Elsevier Ltd.
S0163-4453(22)00684-3
10.1016/j.jinf.2022.11.024
Letter to the Editor
Persistent olfactory dysfunction 2 years after onset of COVID-19
Deng Yi-Ke 123
Shi Ke-Tai 123
Liu Zheng 123†⁎
Zeng Ming 123†⁎⁎
1 Department of Otolaryngology-Head and Neck Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, P.R. China
2 Insititue of Allergy and Clinical Immunology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, P.R. China
3 Hubei Clinical Research Center for Nasal Inflammatory Diseases, Wuhan, P.R. China
⁎ Corresponding author: Zheng Liu, Department of Otolaryngology-Head and Neck Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No. 1095 Jiefang Avenue, Wuhan 430030, P.R. China.
⁎⁎ Corresponding author: Ming Zeng, Department of Otolaryngology-Head and Neck Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No. 1095 Jiefang Avenue, Wuhan 430030, P.R. China.
† Joint corresponding authors.
29 11 2022
29 11 2022
25 11 2022
© 2022 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
2022
The British Infection Association
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Keywords
Coronavirus disease 2019
Olfactory dysfunction
Psychophysical test
==== Body
pmcTo the editor
Growing evidence support that olfactory dysfunction (OD) is one of the major sequalae of long COVID-19.1 Long-term unrecovered OD may significantly reduce the quality of life of patients.2 However, previous studies were mostly based on self-reported symptoms and lacked control groups, making it difficult to evaluate the real association between severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and persistent OD. Therefore, we read with great interest the recently published study in the Journal of Infection by M S Otte et al. on the topic of persistent OD in patients with COVID-19. Using psychophysical test, they found that 50% of patients still suffered from OD at 7 weeks after the start of COVID-19.3 Nevertheless, no study reported recovery rates of OD more than 2 years after infection. Long-term careful evaluation of OD recovery after SARS-CoV-2 infection is critical to understand the chronic symptoms of COVID-19 and to guide treatments.
We followed up our previously reported cohort of 1,172 hospitalized COVID-19 patients, who were infected by alpha variant of SARS-CoV-2 during the first pandemic.4 The severity of OD was scored by patients on a numerical rating scale (NRS) of 0–10, with 0 being “sense of smell is completely normal” and 10 being “complete loss of sense of smell”. In this cohort, 134 patients reported OD during the onset of illness4, and 116 of them completed 2-year follow-up (Fig. 1 A). Patients whose NRS scores dropped to 0 were defined as fully recovered, while those with declined scores (>0) were defined as partially recovered, and the patients not having declined scores were defined as not recovered. Twenty-one of the 116 patients with self-reported OD (SROD group) finished the Sniffin’ Sticks test (Burghart Messtechnik, Holm, Germany). Among 1,038 patients without self-reported OD, 187 finished 2-year follow-up and completed the Sniffin’ Sticks test. As a disease control group, 21 of the 187 patients (NSROD group) were sex and age-matched to SROD group (Fig. 1A). In addition, 21 of 1,862 healthy subjects, who underwent Sniffin’ Sticks test and denied history of COVID-19, rhinitis, sinusitis and OD, were 1:1 sex and age-matched to SROD group as a normal control group (NC) (Fig. 1A). More information is provided in the Online Repository.Fig. 1 The flow chart and evaluation of OD.
A, The flow chart. B, Self-reported recovery rates of OD in SROD group. C, OD evaluated by Sniffin` Sticks test in patients with self-reported totally recovered. D, OD evaluated by Sniffin` Sticks test in patients self-reporting not recovered. Abbreviations: OD, olfaction dysfunction; NRS, numerical rating scale; SROD, self-reported olfactory dysfunction; NSROD, no self-reported olfactory dysfunction; NC, normal control.
Fig. 1
The NRS scores of the patients in both NSROD and NC group were zero. At 24- month after symptom onset, we found that 91.4% (106 of 116) and 5.2% (6 of 116) patients in SROD group reported complete and partial recovery, respectively, while 3.4% (4 of 116) reported not recovered (Fig. 1B). No significant association was found between the self-reported OD 24 months after disease onset and the initial severity of COVID-19 and comorbidities (Table S1).
Both patients in SROD and NSROD group had significantly lower TDI scores compared to the subjects in NC group. Consistently, we found that only 28.6% (6/21) and 14.3% (3/21) patients were normosmia in SROD and NSROD groups, respectively, which were significantly lower than that in NC group (14/21; 66.7%) (Table 1 ). In contrast, the ratios of heterosmia were significantly higher in SROD and NSROD groups (71.4% and 85.7%, respectively) than that in NC group (33.3%) (Table 1). No difference was found between SROD and NSROD groups (Table 1).Table 1 The TDI score and age-adjusted olfactory function grading of subjects 24 months after onset symptoms evaluated by the Sniffin’ Sticks test.
Table 1Characteristics COVID-19 patients Normal subjects
With self-reported olfactory dysfunction Without self-reported olfactory dysfunction
Subject, N 21 21 21
Gender, male, N (%) 11 (52.4%) 14 (66.7%) 10 (47.6%)
Age, years, median (IQR) 63 (42, 66) 51 (40, 59) 54 (41, 58)
TDI score, median (IQR) 23.5 (20.1, 29.9) * 24 (21.5, 26.3) * 29 (24.6, 30.5)
Threshold, median (IQR) 4 (1, 5.8) 1 (1, 4) * 5.25 (2.5, 6.5)
Discrimination, median (IQR) 9 (8, 13) 11 (9, 12) 11 (10, 13.5)
Identification, median (IQR) 11 (9, 12) * 11 (10, 12) 12 (11, 13)
Normosmia, N (%) 6 (28.6%) * 3 (14.3%) ** 14 (66.7%)
Heterosmia, N (%) 15 (71.4%) * 18 (85.7%) ** 7 (33.3%)
Hyposmia, N (%) 13 (61.9%) 18 (85.7%) 7 (33.3%)
Anosmia, N (%) 2 (9.5%) 0 0
Abbreviations: TDI, threshold, discrimination, and identification; COVID-19, coronavirus disease 2019; IQR, interquartile range; Data are presented as medians and interquartile ranges for continuous variables and numbers with percentage for categorical variables. P values were calculated from Mann-Whitney U 2-tailed test, or χ2 test or Fisher's exact test. *P<0.05 vs. Normal control, **P<0.01 vs. Normal control.
Among 21 patients finishing psychophysical test in SROD group, 18 of them reported complete recovery of OD and 3 reported not recovered. Surprisingly, the Sniffin’ Sticks test revealed that only 33.3% (6/18) of the self-reported completely recovered patients reached to normal sense of smell, while 61.1% (11/18) and 5.6% (1/18) of them were still hyposmia and anosmia, respectively (Fig. 1C). In addition, 2 of the 3 subjects who complained persistent OD were hyposmia (66.7%), and one (33.3%) was anosmia (Fig. 1D).
The persistence of OD in long COVID-19 over 2 years remains unclear. In this study, we found that although 96.6% of the patients reported that their sense of smell was recovered after 2 years, only one-third of them actually recovered according to the psychophysical test. This result is in line with another study, which reported that the prevalence of OD evaluated by psychophysical tests was higher than that estimated based on self-reported information at 6-month from diagnosis of SARS-CoV-2 infection.5 Collectively, it suggests that the self-reported evaluation is inadequate to evaluate OD in long COVID-19. Olfactory function varies with age and gender.6 To adjust these confounding factors, 21 individuals 1:1 matched to patients infected with SARS-CoV-2 were enrolled. The ratios of normosmia and heterosmia in NC group in this study were in line with those reported in general population.6 We found that the prevalence and severity of OD in SROD group were significantly higher than those in NC group. It indicates that SARS-CoV-2 infection is the cause of persistent OD which may last up to 24 months. High prevalence of persistent OD in long COVID-19 argues for more attentions with development of target therapies.7
A very recent study based on self-reported data showed that 88.2% of patients reporting COVID-19–related smell or taste dysfunction completely recovered within 2 years.8 However, no study, to the best of our knowledge, has investigated the olfactory function in COVID-19 patients after two years using psychophysical tests. Surprisingly, we found that 85.7% patients in NSROD groups also had heterosmia when evaluated by psychophysical test, further indicating that the prevalence of OD in long COVID-19 might be much higher than those self-reported, even for those without complain of OD at the onset of illness and 2-year follow up.
In conclusion, our current findings suggest that there is still a significant proportion of patients having persistent OD 2 years after SARS-CoV-2 infection. However, additional studies are required to clarify the long-term olfaction condition in COVID-19 patients infected by the other variants, since distinct SARS-CoV-2 variants may have different abilities to cause OD.9 Our study is limited to the small sample size of subjects completing the psychophysical test. Nevertheless, our study confirms the great discrepancy between psychophysical tests and self-reported function, which highlights the importance of psychophysical tests in the evaluation of OD in the long-term follow-up of COVID-19 patients. Last but not least, the high prevalence of OD in long COVID-19, even 2 years post-infection, argues for more attentions with the development of target therapies.
Author contribution
Yi-Ke Deng collected clinical data, performed data analysis and prepared manuscript; Ke-Tai Shi collected clinical data; Ming Zeng and Zheng Liu designed the study, interpreted data, and prepared manuscript.
Declaration of Competing Interest
The authors have no conflicts of interest to disclose.
Appendix Supplementary materials
Image, application 1
Image, application 2
Funding sources
This study was supported by the 10.13039/501100001809 National Natural Science Foundation of China (NSFC) grants 82071025 (M.Z.), 82201260 (Y.K.D.), 82130030 (Z.L.) and 81920108011(Z.L.), and the Key Research and Development Program of Hubei Province 2021BCA119 (Z.L.).
Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jinf.2022.11.024.
==== Refs
References
1 Groff D Sun A Ssentongo AE Short-term and Long-term Rates of Postacute Sequelae of SARS-CoV-2 Infection: A Systematic Review JAMA Netw Open 4 2021 e2128568
2 Pinto JM Wroblewski KE Kern DW Schumm LP McClintock MK. Olfactory dysfunction predicts 5-year mortality in older adults PLoS One 9 2014 e107541
3 Otte MS Klussmann JP Luers JC. Persisting olfactory dysfunction in patients after recovering from COVID-19 J Infect 81 2020 e58 32592702
4 Song J Deng YK Wang Het Self-reported Taste and Smell Disorders in Patients with COVID-19: Distinct Features in China Curr Med Sci 41 2021 14 23 33582900
5 Boscolo-Rizzo P Menegaldo A Fabbris Cet Six-Month Psychophysical Evaluation of Olfactory Dysfunction in Patients with COVID-19 Chem Senses 46 2021
6 Hummel T Kobal G Gudziol H Mackay-Sim A. Normative data for the "Sniffin' Sticks" including tests of odor identification, odor discrimination, and olfactory thresholds: an upgrade based on a group of more than 3,000 subjects Eur Arch Otorhinolaryngol 264 2007 237 243 17021776
7 Denis F Septans AL Periers Let al. Olfactory Training and Visual Stimulation Assisted by a Web Application for Patients With Persistent Olfactory Dysfunction After SARS-CoV-2 Infection: Observational Study J Med Internet Res 23 2021 e29583 34003765
8 Boscolo-Rizzo P Fabbris C Polesel Jet Two-Year Prevalence and Recovery Rate of Altered Sense of Smell or Taste in Patients With Mildly Symptomatic COVID-19 JAMA Otolaryngol Head Neck Surg 2022
9 Menni C Valdes AM Polidori Let Symptom prevalence, duration, and risk of hospital admission in individuals infected with SARS-CoV-2 during periods of omicron and delta variant dominance: a prospective observational study from the ZOE COVID Study Lancet 399 2022 1618 1624 35397851
| 36460170 | PMC9707023 | NO-CC CODE | 2022-12-07 23:16:16 | no | J Infect. 2022 Nov 29; doi: 10.1016/j.jinf.2022.11.024 | utf-8 | J Infect | 2,022 | 10.1016/j.jinf.2022.11.024 | oa_other |
==== Front
Comput Ind Eng
Comput Ind Eng
Computers & Industrial Engineering
0360-8352
1879-0550
Elsevier Ltd.
S0360-8352(22)00749-5
10.1016/j.cie.2022.108761
108761
Article
How, When, & Where temporary hospitals fit in turbulent times: A hybrid MADM optimization in the middle east
Karbassi Yazdi Amir a
Muneeb Farhan Muhammad b
Wanke Peter Fernandes c⁎
Hanne Thomas d
Ali Adnan b
a School of Engineering, Universidad Católica del Norte, Larrondo 1281, 1781421, Coquimbo, Chile
b School of Management, Xi’an Jiaotong University, Shaanxi, Xi’an, PR China
c Center for Logistics Studies, COPPEAD – The Graduate School of Business Administration Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
d Institute for Information Systems, University of Applied Sciences and Arts Northwestern Switzerland, Switzerland
⁎ Corresponding author.
29 11 2022
1 2023
29 11 2022
175 108761108761
© 2022 Elsevier Ltd. All rights reserved.
2022
Elsevier Ltd
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Governments have been challenged to provide temporary hospitals and other types of facilities to face the COVID-19 pandemic. This research proposes a novel multi-attribute decision-making (MADM) model to help determine how, when, and where these temporary facilities should be installed based on a set of critical success factors (CSFs) mapped in an uncertain environment. We portray the available facilities for temporary hospitals based on the CSFs that must be considered to make critical decisions regarding the optimal position based on the government’s strategic decision-making process, thus indirectly providing better services and maximizing resources. In relation to earlier work, this research builds upon hybrid Pythagorean fuzzy numbers to find weights in Best-Worst Methods and rank temporary facilities based on evaluation by an area-based method for ranking. Policy implications and future directions are derived.
Keywords
Temporary hospitals
Best worst method
Evaluation by an area-based method of ranking
Pythagorean fuzzy numbers
Delphi method
COVID-19
==== Body
pmc1 Introduction
The healthcare industry provides vital services to modern societies to integrate advanced and optimized operations (Mazloumian et al., 2022). However, healthcare has recently faced challenging circumstances in terms of operating robust operations vs services (Samadi et al., 2021). The far-reaching COVID-19 pandemic forces global industries to coordinate their efforts in supplying and transporting goods and services to help control the pandemic, including the hospital care industry. For instance, blockchain technology has been assessed (Samadi et al., 2021); new trends have been recognized (Hsu et al., 2021); business to business alliances have been formed (Madanaguli et al., 2021); six sigma projection and prioritization has been enumerated (Can et al., 2021); and sustainability-oriented innovation frameworks have been implemented (Elabed et al., 2021).
Admittedly, the responsiveness and propitiousness of healthcare operations significantly contribute to patient survival in global rural and urban areas (Can et al., 2021, Liu et al., 2021). Indeed, Hejazi (2021) called for “requirements to provide fast and quality services to patients and creation of value for health care centers.” Accordingly, we argued that the significance of hospitals in times the COVID-19 pandemic (a) leverages consumer rights for medical care and instituting temporary hospitals and (b) emerges as humanitarian and economic strategic pathways to serve and thrive. Therefore, this study aims to identify the strategic critical success factors (CSFs) to bolster temporary hospitals and significantly control operational and medical costs.
Strategic and operations developments in temporary hospitals comprise complex interfaces between various health care clusters that can overwhelm limited human resources, operations, and optimizations, as well as strategic decision-making (Hsu et al., 2021, Mazloumian et al., 2022). Such profound practices become more convoluted by the inherent variability of the various CSFs. In particular, the unplanned landscape of the CSFs and the uncertainty about patient recovery mean the temporary hospitals must have vibrant CSFs in place that can strongly stimulate instantaneous operations and strategic decision-making to alleviate poverty and economic vulnerability (Brandt et al., 2021, Lu et al., 2021). Therefore, this study intends to optimize and rank the strategic CSFs related to the site selection of temporary hospitals during the turbulent waters of the COVID-19 pandemic.
Congestion of temporary hospitals is a matter of global import, and operations research techniques have been extensively utilized to conceal the dearth of relevant research. Therefore, this study proposed a novel combination of Pythagorean fuzzy numbers of the best-worst method (PFNBWM) with the Pythagorean fuzzy number evaluation area-based method ranking (PFNEAMR) method for selecting suitable hospitals to accommodate the unprecedented COVID-19 diseases. The key advantage of such a hybrid approach is that it can map the CSFs related to where such temporary hospitals might be set up and operated in a multi-dimensional fashion. Indeed, it is evident that providing the best type of services in temporary hospitals and managing referrals to medical personnel will gratify patients, especially in rural and urban communities (Yang et al., 2021, Zhang et al., 2020).
This study explores the optimal 16 potential CSFs for temporary hospitals to reduce waiting time and increase amenities according to diverse conditions, thus enhancing the productivity of temporary hospitals. Moreover, we pursue the strategic action plan to answer where such temporary hospitals should be located. This action plan strategy decreases the waiting time for triage, intensifies the utility of temporary hospitals, and reduces the patient uncertainty vis-à-vis the COVID-19 pandemic. As Sarkar et al. (2021) noted, the decision must to be taken swiftly because hospital capacity is in constant flux due to the increasing number of infected people. Therefore, we have answered this call by providing different scenarios to improve the potential CSFs of temporary hospitals and put in place a strategic action plan for instant service delivery. In doing so, we used a hybrid multi-attribute decision-making (MADM) model to rank the selected CSFs by the PFNBWM and PFNEAMR methods and established a model to find an action plan strategy for temporary hospitals.
The rest of the article comprises the following sections. The literature review is presented in Section 2. Section 3 revisits the MADM and PFN methods. The research methodology is described in Section 4. Section 5 focuses on the data analysis while the final section sets out some managerial implications and conclusions of the study.
2 Literature review
2.1 Transitioning of temporary hospitals during the COVID-19 pandemic
The impact of the COVID-19 pandemic goes beyond the spread of the disease and other transitioning related to acute respiratory syndrome (Sarkar et al., 2021). The impact on the supply chain has been particularly significant in the Middle East, in countries such as Iran, where it has economically disrupted and debilitated particularly impoverished urban and rural communities (Moosavi & Hosseini, 2021). However, what is important is that controlling the disruptions requires maintaining temporary hospitals management and rapid responsiveness to healthcare and public welfare (Govindan et al., 2021). The economic downturn triggered by the COVID-19 pandemic in Iran coincides with the economic sanctions against the nation (Lotfi et al., 2022). Although various of these sanctions have been in place for the last four decades, since May 2019 (Danaei et al., 2019), the unilateral sanctions imposed have increased dramatically to an almost total economic lockdown, thus posing severe barriers to designing and deploying health resources in the fight against COVID-19 (Han et al., 2020).
As a result, all prevention, diagnosis, and treatment sectors are directly and indirectly jeopardized, and the country’s temporary hospitals management are falling short in combating the crisis (Lotfi et al., 2022). In fact, the lack of medical, pharmaceutical, and laboratory equipment to establish new temporary hospitals and even to reequip the old ones has scaled up the COVID-19 burden (Han et al., 2020). Thus, the Iranian government has established various strategies and rules to control the spread and boost public welfare. The first case of COVID-19 was recorded in the province of Qom on February 19, 2020 (Abdi, 2020). The disease is still increasing and spreading day by day, thus exacerbating the economic recession (Han et al., 2020, Lotfi et al., 2022).
2.2 Site selection and optimization of temporary hospitals
Numerous factors should be considered jointly when selecting the site for temporary hospitals (Lotfi et al., 2022 ). Such factors should be prioritized for optimum placement (Sarkar et al., 2021). Moreover, the literature is prolific in describing alternative MADM methods for selecting from among location alternatives, using applications based on single approaches and relying on a single MADM model (Yang et al., 2021, Hejazi, 2021). For instance,
Kmail et al. (2017), who inserted geographic information system data into analytic hierarchy process (AHP) methods for hospital site selection, used the following criteria or CSFs: land use, distance to hospitals, intersection with the main road, distance to dumping sites, industrial area distance, and elevation. Khaksefidi and Miri (2015) identified and prioritized CSFs that impacted hospital location in Iranian provinces by using hybrid MADM methods: AHP, technique for order of preference by similarity to ideal solution (TOPSIS), ELimination Et Choix Traduisant la REalité (ELECTRE), and simple additive weighting (SAW). Accidents, population-density, quality of the road, distance from the center of the region, distance from two other main cities, and local climate were the criteria considered.
Şahin et al. (2019) studied the application of AHP for hospital location selection based on competitors, demand factors, environmental conditions, accessibility, related industry, and government factors. Another example in terms of more elaborate hybrid MADM approaches is found in Adalı and Tuş, which structured a decision framework involving criteria importance through inter criteria correlation (CRITIC), TOPSIS, evaluation based on distance from average solution (EDAS), and combinative distance-based assessment (CODAS) to select a hospital site. Cost, land, strategy, market condition, transportation, environment, demography, and geology were the CSFs selected.
Chauhan and Singh (2016) pointed out how hybrid approaches using fuzzy logic or grey analysis could help sustain healthcare operations. In this regard, Sen (2017) applied grey additive ratio assessment (ARAS-G) for hospital location selection. Site condition and surroundings, accessibility and traffic, patients, cost, future considerations, climate conditions, and noise were factors for hospital site selection. In turn, Kahraman et al. (2019) used Fuzzy TOPSIS for hospital location selection, considering cost of land, land topography, building cost, population density, education level, economic conditions, proximity to transport, availability of infrastructure, and proximity to markets. Similarly, Miç and Antmen (2019) performed fuzzy TOPSIS (FTOPSIS) for healthcare location selection based on demographic structure, investment costs, environmental factors, travel time, travel costs, and infrastructure. Senvar et al. (2016) also presented a similar study using FTOPSIS.
Çelikbilek (2018) showed how hospital location can be selected using Fuzzy VIseKriterijumska Optimizacija I Kompromisno Resenje (FVIKOR). Building, cost, population density, prospective population, distance to the social center, distance to medical suppliers, distance to other institutions, easy access for ambulances, easy access to transportation, hospital demand at the location, and parking lot availability were the CSFs considered.
In contrast to previous authors, Kutlu Gündoğdu et al. (2018) presented hesitant fuzzy sets (HFS) and evaluation based on distance from average solution (EDAS) for hospital location selection. They pointed to the applicability of the SERVQUAL method for selecting hospital location. Hashemkhani Zolfani et al. (2020) showed how to select a hospital site by a grey MADM method during the COVID-19 pandemic. They considered 10 criteria for evaluation to obtain 5 alternatives (hospital sites in Istanbul). Methods used were criteria importance through intercriteria correlation (CRITICT) and grey combined compromise solution (CoCoSo). The result indicates which alternative received the highest and lowest priority. Table 1 summarizes the previous studies on hospital location selection and state-of-the-art methods.Table 1 State-of-the-art methods.
Pythagorean Fuzzy Number *
EAMR * *
Best-Worst Method * *
CoCoSo *
Hesitant Fuzzy Sets *
VIKOR *
ARAS *
Grey Number * *
Fuzzy Sets * * * * *
CODAS *
EDAS * *
CRITIC * *
SAW *
ELECTRE *
TOPSIS * * * * *
AHP * * *
Method Proposed (Hashemkhani Zolfani et al., 2020) Kutlu Gündoğdu et al. (2018) Çelikbilek (2018) Senvar et al. (2016) Miç and Antmen, (2019) Kahraman et al. (2019) Sen (2017) Chauhan and Singh (2016) Şahin et al. (2019) Khaksefidi and Miri (2015) Kmail et al. (2017) Method Proposed
Incorporating MADM methods for site selection projects is prevalent among operations researchers (Yang et al., 2021, Ye et al., 2021). We studied prior research on how MADM methods can be used for sustainable supplier evaluation (Lo et al., 2021), development of an integrated programming approach (D. Liu et al., 2020), and supply–demand resilience (Mohammed et al., 2021). Such potential MADM methods can be categorized into decision matrix and pairwise comparison methods. In contrast with previous studies, we applied new paradigms of MADAM to speculate on how, when, and where the temporary hospitals sites were established with robust proficiency and efficacy. Therefore this study used the best-worst method (BWM) (Muneeb et al., 2020) and evaluation area-based method for ranking (EAMR) to aid decision makers (DMs) in strategic decision making. With these tools, DMs can formulate strategic pathways to rapidly to select, identify, and optimize temporary hospital sites (Karbassi Yazdi et al., 2018, Rezaei, 2015, Yazdi et al., 2020, Karbassi Yazdi et al., 2021).
3 Multi-attribute decision-making methods
3.1 Best-Worst method
MADM methods incorporate decision-making processes that are more accurate than statistical methods. Indeed, some of them rely on pairwise alternative comparison approaches, while others are based on decision-matrix (D. Liu et al., 2020, Lo et al., 2021). Pairwise comparison approaches are used to obtain weights for alternatives, and decision matrix approaches are used to rank alternatives using multiple criteria directly (Muneeb et al., 2020, Ye et al., 2021). The best-worst method has several advantages compared to other methods and this is, discussed in Rezaei (2015), who first introduced this model. Two vectors are compared, one of which is the best vector (benefits) and the other the worst vector (costs). The nonlinear minimax model is applied for computing the maximal absolute weighted ratio difference while simultaneously minimizing it. The following steps yield the weights of the BWM alternatives.Step 1. The criteria of the model must be determined. Criteria are shown as C={c1,c2,⋯.,cn}.
Step 2. The kind of criteria is determined. These criteria can be the best (maximization criteria) or the worst (minimization criteria).
Step 3. Allocate preferences to best criteria based on a 1–9 scale with the preferences of best criteria of B being indicated asAB=(aB1,aB2,...,aBn). It is obvious that aBB=1.
Step 4. The method for allocating preferences to the worst criterion is the same as for the best criterion. Preferences of the worst criterion of W are indicated asAw=(aw1,aw2,...,awn).
Again, it is obvious that aww=1.Step 5. Final weights are extracted based on the following model. These weights are shown as (w1∗,w2∗,...,wn∗)
Maximum absolute differences wBwj-aBj and wjww-awj will be minimized for all j. The following equation shows this computation.minmaxj{wbwj-aBj,wjww-awj
s.t.(1) ∑jwj=1wj⩾0,forallj
This model can be rewritten as follows.(2) minζs.t.wbwj-aBj⩽ζwj⩾0for all j
After solving this model, the result shows the final weights of the BWM.
3.2 Evaluation by an area-based method of ranking (EAMR)
EAMR is a kind of decision matrix approach introduced by Keshavarz Ghorabaee et al. (2016). This model uses beneficial and non-beneficial criteria for ranking alternatives. Their steps are detailed as follows. Step 1. A decision matrix Md is created as:
(3) Md=Mijd=x11d…x1md⋮⋱⋮xn1d⋯xnmd,1⩽i⩽n,1⩽j⩽m,1⩽d⩽kxij=(xij1+xij2+...+xijk)kY¯=[xij¯]
The number of decision-makers is represented as k; d is the index for the d-th decision-maker, and thecriterionvalueofalternativeiforcriterionjofdisMij. n is the number of alternatives and m the number of criteria.Step 2. Eqs. (4), (5) point out an average of the decision matrix
(4) xij=(xij1+xij2+...+xijk)k
(5) Y¯=[xij¯]
where xij¯ indicates average value performance (criterion value) of alternative i and criterion j and Y¯ is the average decision matrix, such that 1 ≤i≤n,1≤j≤m. Step 3. The weighting matrix (weighting vector) Wp is designed as follows:
wp=[wjp]m×1=[w1pw2p⋮wmp]1⩽j⩽m,1⩽p⩽k,
(6) wp=[wjp]m×1=[w1pw2p⋮wmp]
p is the index of the pth decision-maker and wjp is the respective weight of criterion j for1⩽j⩽m,1⩽p⩽k,.Step 4. The average weighting matrix (weighting vector) W¯ is calculated as follows:
(7) wj=(wj1+wj2+...+wjk)k
(8) w¯=[wj]m×1
Step 5. The normalized average decision matrix from Y¯ denoted as N is calculated by:
(9) nij=xijej
(10) ej=maxi∈{1,2,...,n}(xij)
(11) N=[nij]n×m
where 1⩽i⩽n,1⩽j⩽m.Step 6. The normalized weights of the decision matrix v are:
(12) vij=nij×wj
(13) V=[vij]n×m
Step 7. The sum of normalized scores for beneficial criteria (G+i) and non-beneficial criteria (G-i) are calculated as
(14) G+i=(v+i1+v+i2+...+v+in)
(15) G-i=(v-i1+v-i2+...+v-in)
v+ij and v-ij are normalized weighted values for beneficial and non-beneficial criteria.Step 8 (RV) is the rank of value based on G+i and G-i: (1≤i≤n). They are ranked according to scores of alternatives that are shown by G+i and G-i.
Step 9. Calculate the appraisal score (Si) based on the rank values:
(16) Si=RV(G+i)RV(G-i)
Si is the score of the highest alternative (Amiri and Antucheviciene, 2016, Karbassi Yazdi et al., 2018).
3.3 Pythagorean fuzzy sets (PFS)
Yager (2013) first introduced PFS, which is a method that describes the membership and non-membership degrees that satisfy the condition that the sum of the squares of the membership and non-membership degrees is equal to or less than one. In some cases, membership and non-membership degrees assigned by a decision-maker to a given alternative satisfy this inequality.
Let us first define the PFS P in the fixed non-empty domain X, which can be denoted as:(17) P={x,μp(x),υp(x)}
where μPx represents the membership degree function, such that X →[0,1] and υP(x) represents the non-membership degree function, such that X →[0,1] for all × ∈ X of set P. The limiting conditions for the membership degree values are given next:(18) 0⩽(μp(x))2+(υp(x))2⩽1∀x∈x
The respective hesitant degree is denoted as πP(x) and computed as:(19) πp(x)=1-(μp(x))2+(υp(x))2∀x∈x
Let us also consider two PFS numbers, such that P1=(μp1(x),υp1(x)) andP2=(μp2(x),υp2(x)), also assuming that λ>0. PFS operations are defined as in Peng and Yang, 2015, Yager, 2016, Yager, 2013 such that:(20) P1⊕P2=(μp1(x))2+(μp2(x))2-(μp1(x))2,(μp2(x))2,(νp1(x),νp2(x))
(21) P1⊕P2=(μp1(x)+μp2(x),(νp1(x))2+(νp2(x))2-(υp1(x))2,(υp2(x))2
P1⊖P2=(μp1(x))2+(μp2(x))21-(μp2(x))2,υp1(x)υp2(x) if(22) μp1(x)⩾μp2(x),υp1(x)⩽min{υp2(x),υp2(x),πp1(x)πp2(x),νp1(x)⩾νp2(x)
P1P2=(μp1(x)μp2(x),(μp1(x))2+(μp2(x))21-(μp2(x))2 if(23) μp1(x)⩽min{μp2(x),μp2(x),πp1(x)πp2(x),νp1(x)⩾νp2(x)
(24) λp1=(1-(1-μp1(x))2)λ,νp1(x))2)λ
(25) P1λ=(μp1(x))2)λ,1-(1-υp1(x))2)λ
(26) P1c=(νp1(x),μp1(x))
4 Research methodology
4.1 Problem description
Despite the limitations of higher hospitalization during the pandemic, attempts were made by the Iranian government to make the best use of the scarce resources available during the COVID-19 pandemic (Takian et al., 2020, Lotfi et al., 2022). The primary complications for all countries in providing facilities for treating patients are related to the deployment of physicians, nurses, medicines, and intensive care unit (ICU) equipment (Hejazi, 2021). Therefore, the current study maps the relative importance of the CSFs that influence principled site selection of locations for temporary hospitals.
4.2 Model procedure
Step 1.Identification of CSFs: The CSFs for selecting the best location for building temporary hospitals are extracted by studying previous research and conducting interviews with experts.
Step 2.Filtration of CSFs: A Delphi questionnaire was applied to experts to filter CSFs for selecting temporary hospital sites. Delphi methods were used for screening and customizing factors (reaching consensus). In this study, since many CSFs are extracted from previous studies, these factors must be adapted to the circumstances of this paper by using the Delphi method. A preliminary questionnaire for evaluating factors using a Likert scale was created and distributed among the experts. The minimum number of DMs needed to fill out this questionnaire is controversial among researchers. Some believe the minimum is 100, while others believe 5 to 15 is optimal (Brady, 2015, Dunham, 1998). In the questionnaire, DMs specify their preferences of the factors based on a Likert scale with values from 1 to 5, 7, or 9. The CSF is rejected if the average of DM preferences for it is less than an acceptance score; otherwise, it is accepted. In this research, a 5-point Likert scale was used, and if the average score of a CSF was more than four, it was considered acceptable; otherwise, it was rejected. Readers may refer to Gordon, 1994, Linstone and Turoff, 1975.
Step 3.Allocate decision maker (DM) preferences by Pythagorean fuzzy numbers (PFN) for each criterion and transfer them to crisp data: Once PFN are defined, they are transferred to a crisp number after allocating them to each criterion.
Step 4.Test of independency of factors: When we want to use a decision-making method for ranking alternatives in the first step, we must take into account that these factors may depend on each other. Whenever factors present codependency, the result will not be reliable. Therefore, grey relational analysis is applied for handling this issue. The data do not present dependency when the degree of greyness is less than 0.5 and in this case, subsequent steps can be carried out.
Step 5.Finding weights by Pythagorean fuzzy numbers of the best-worst method (PFNBWM): Since EAMR uses a decision matrix, it needs weights for the criteria. PFNBWM was the approach chosen to compute these weights.
Step 6.Ranking possible temporary hospital building sites by Pythagorean fuzzy numbers evaluation area-based method for ranking (PFNEAMR): In this step, the possible hospital site is selected by ranking the DM preferences.
Step 7.Sensitivity Analysis: This helps DMs compare the results of this hybrid approach with other alternative approaches for finding the best solution.
Fig. 1 summarizes the research framework adopted in this research.Fig. 1 Research framework.
4.3 Sample of decision makers
Six DMs were selected for this research from civil engineering, physicians, and transportation experts. The information on the profiles of these DMs is shown in Table 2 .Table 2 DMs’ information.
Expert No. Work Experience (Years) Education Level
1 28 MD
2 26 PhD
3 27 MSc
4 28 PhD
5 29 MSc
6 32 MD
4.4 CSFs for selecting temporary hospital sites
The list of these CSFs is given in Table 3 .Table 3 State-of-the-art CSFs.
Source Abbreviation Component
Cost CO Sen, 2017, Kahraman et al., 2019, Miç and Antmen, 2019, Çelikbilek, 2018
Land Strategy LS Adalı & Tuş (2019)
Market condition MC Adalı & Tuş (2019)
Transportation TR Adalı and Tuş, 2019, Çelikbilek, 2018
Environmental EN Şahin et al., 2019, Miç and Antmen, 2019
Demographics DE Miç and Antmen (2019);
Geological factors GE Adalı and Tuş, 2019, Kahraman et al., 2019, Miç and Antmen, 2019
Financial support FS Kahraman et al. (2019)
Competitors CM Şahin et al. (2019)
Demand DM Şahin et al., 2019, Çelikbilek, 2018
Accessibility ACC Şahin et al., 2019, Sen, 2017
Government GO Şahin et al. (2019);
Building structure BS Kahraman et al., 2019, Çelikbilek, 2018;
Workers WO Adalı & Tuş (2019)
Emergency access considerations EAC Şahin et al., 2019, Sen, 2017, Çelikbilek, 2018
Traffic TA Sen (2017)
Waste disposal site WDS Kahraman et al. (2019)
Distance from center of province DCP Senvar et al., 2016, Çelikbilek, 2018
Distance from two other cities DTC Senvar et al., 2016, Çelikbilek, 2018
Easy Access for Ambulances EAA Çelikbilek (2018)
Availability of Parking Lot APL Çelikbilek (2018)
4.5 Delphi method and results
Table 4 reports the DM preferences for each CSF based on a 5-point Likert scale. Sixteen factors out of 21 factors were filtered based on DMs’ preferences.Table 4 Delphi results.
Factors DM1 DM2 DM3 DM4 DM5 DM6 Average Accept/Reject
1 Cost 5 4 5 4 3 5 4.333333333 Accept
2 Land Strategy 5 4 5 3 3 4 4 Accept
3 Market condition 3 4 3 5 3 3 3.5 Reject
4 Transportation 5 5 4 5 4 5 4.666666667 Accept
5 Environmental 4 5 5 4 3 5 4.333333333 Accept
6 Demographics 5 5 4 3 4 4 4.166666667 Accept
7 Geological factors 4 3 3 4 3 5 3.666666667 Reject
8 Financial support 5 5 4 5 3 5 4.5 Accept
9 Competitors 3 3 3 4 3 4 3.333333333 Reject
10 Demand 5 5 5 4 5 3 4.5 Accept
11 Accessibility 5 5 4 4 5 3 4.333333333 Accept
12 Government 4 5 5 4 3 4 4.166666667 Accept
13 Building structure 3 3 3 4 3 5 3.5 Reject
14 Workers 3 3 4 3 4 5 3.666666667 Reject
15 Emergency access considerations 5 5 5 4 4 5 4.666666667 Accept
16 Traffic 5 5 5 4 5 3 4.5 Accept
17 Waste disposal site 4 4 5 5 5 3 4.333333333 Accept
18 Distance from center of province 5 5 5 4 4 3 4.333333333 Accept
19 Distance from two other cities 5 5 4 4 4 3 4.166666667 Accept
20 Easy Access for Ambulances 5 4 5 3 5 5 4.5 Accept
21 Availability of Parking Lot 5 5 5 5 4 3 4.5 Accept
4.6 Advantages of the proposed hybrid approach
4.6.1 Advantages of best worst method
The BWM is one of the most widespread pairwise comparison methods used in research (Muneeb et al., 2020). Compared with traditional pairwise comparison methods such as AHP, REMBRANDT, and MACBETH, BWM needs fewer pairwise comparisons. However, recent advances of AHP, such as Parsimonious AHP, Cybernetic AHP, and Express AHP, have overcome the original disadvantage and require fewer pairwise comparisons than BWM. One of the merits of BWM compared to these AHP improvements is that it is less complex and more user-friendly in a big-data context. High consistency and compatibility when using big data are other BWM advantages (Aboutorab et al., 2018).
The advantageous positioning of BWM and its comprehensive fuzzy concepts are universally applicable to numerous fields, ranging from manufacturing and supply chains (Ahmadi et al., 2017), sustainable outsourcing, and partner selection (Garg and Sharma, 2020) to sustainable architecture. Nevertheless, the theoretical contributions remain salient for underpinning the COVID-19 context when contemplating the best locations for temporary hospitals. The primary purpose of the BWM method is to identify the best and worst weight vectors (Rezaei, 2015, Aboutorab et al., 2018). Furthermore, pairwise comparisons concerning the best for other weight vectors and the worst are employed. The final weights of alternatives can be obtained from these weighted vectors by through linear programming. The consistency of fuzzy preference relations is patterned and confers additive persistence and multiplicative consistency. Finally, an algorithm is deployed seeking to amplify the consistency of the fuzzy preference relationships when an anticipated consistency level is not accomplished. As argued by Rezaei (2016, p. 26), “BWM provides more information about the optimal solution.” Thus, this study posits that the BWM method might be fruitful for selecting and identifying the sites of temporary hospitals and confers advantages for implementing future managerial and practical strategies to combat the turbulent waters of the COVID-19 pandemic.
4.6.2 Advantages of Pythagorean fuzzy numbers
Because the PFN method uses distance measurement, its accuracy is better than that of other methods. As stated by Fei and Deng (2020), “Pythagorean fuzzy set, initially extended by Yager [2013] from intuitionistic fuzzy set, is capable of modeling information with more uncertainties in the process of MCDM, thus can be used on [a] wider range of conditions” (p. 1). Therefore, the current study used PFN to optimize the site selection of temporary hospitals. PFN is functional in several real-world complications involving more uncertainty than other approaches and extensions of the fuzzy set theory, such as intuitionistic fuzzy sets, hesitant fuzzy sets, and interval type-2 fuzzy sets (Alcantud, 2016, Torra, 2010). The intuitionistic fuzzy set was first put into practice by Atanassov (1999) and proved highly efficient in unpacking uncertain information. The intuitionistic fuzzy set is described by a membership grade and non-membership grade, given the situation that is less than or equal to 1. However, in other applications and under real-life conditions, the decision-makers may convey their preferences using concepts suitable to situations under uncertainty, such as fuzzy numbers including the specification of membership functions, which may pose multiple difficulties. Thus, to overcome this drawback, Yager (2013) planned the advantageous PFN to handle problems involving uncertainty.
4.6.3 Advantages of the EAMR method
In MADM methods based on a decision matrix, there is no way to identify whether the result is reliable or not, so data statistics methods must be employed to gauge their reliability. EAMR is a more reliable method than similar methods such as TOPSIS or VIKOR (Amiri and Antucheviciene, 2016, Karbassi Yazdi et al., 2018).
4.6.4 Advantages of using hybrid methods
The rationale for using such hybrid methods is that EAMR, as mentioned in Section 4.6.3, is one of most reliable MADM decision matrix methods. However, this method requires primary weight. To obtain primary weight, MADM methods based on pairwise comparison methods must be used. Many such pairwise comparison methods exist, and one of them is BWM, which is more friendly and less-complex than other similar methods (as noted in Section 4.6.1). Today, DMs must face many complex and vague environments, making decision-making all the more onerous, especially when they must make recommendations in terms of by exact numbers. Pythagorean fuzzy numbers is a fuzzy number type that has more accuracy because of the distance method. The combination of EAMR, BWM, and Pythagorean fuzzy numbers enable a strong approach with high computation accuracy for ranking alternatives in uncertainty environments.
4.7 The DM preferences scale
See Table 5 .Table 5 Presented the PFNs.
Meaning PFNs(μ,ν)
Very Very Low {0.1,0.99}
Very Low {0.1,0.97}
Low {0.25,0.92}
Moderate {0.4,0.87}
Fair-Moderate {0.5,0.8}
Fair-Good {0.6,0.71}
Good {0.7,0.6}
Very Good {0.8,0.44}
Very Very Good {1,0}
5 Data analysis
5.1 Dependency analysis
Typically, numerous factors are affected when choosing the best alternatives. In other words, researchers do consider all relevant factors when analyzing alternatives, but this does not mean that these factors necessarily affect each other (Fargnoli & Haber, 2019). If factors have a dependent relationship, the result is not reliable. The only relationship allowed is that these factors affect alternatives. Grey relational analysis (GRA) is used to determine whether the factors in this research depend on each other. Before analyzing the data, the dependency rate of the CSFs is computed by GRA (Wu et al., 2020). Hence, in the first step, we have m alternatives with n criteria for Yi.(27) Yi=(yi1,yi2,⋯..,yij,⋯,yin)
Yi is the degree of importance of alternative i based on criterion j. For normalization, Yi must be transferred to Xi.(28) Xi=(xi1,xi2,⋯..,xij,⋯,xin)
One of these formulae (29)–(31) is used for this transfer:(29) xij=yij-min(yij)maxyij-min(yij)thebiggestisbest
(30) xij=maxyij-yijmaxyij-min(yij)thesmallestisbest
(31) xij=yij-y∗maxmaxyij-y∗,y∗-min(yij)whenthevalueclosesttoy∗isbest
The normalized dimensionless decision matrix is denoted by N, and each entry of the matrix is indicated by Gij∗.
For a benefit criterion, Gij∗ is calculated as follows:(32) Gij∗=[Gij_Gjmax,Gij¯Gjmax]
(33) Gjmax=maxGij¯
For a cost criterion, Gij∗ is calculated according to (32), (33):(34) Gij∗=[GjminGij¯,GjminGij_]
(35) Gjmin=min1≤i≤mGij_
In the last step, the GRA coefficients are computed based on the following formula:(36) γxoj,xij=Δmin-rΔmaxΔij-rΔmax
The Δij must be determined before calculating these coefficients.(37) Δij=xoj-xij
Hence, Δmin is the smallest amount of Δij and Δmax is the largest amount of Δij, while r is a distinguished coefficient (Julong, 1989, Kuo et al., 2008). Table 6 presents the decision matrix after using Eqs. (27), (28), (29), (30), (31), (32), (33), (34), (35), (36), (37). A normalized matrix based on this is calculated.Table 6 Presenting the decision matrix.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
A1 0.0773 0.0078 0.0773 0.3069 0.3069 0.1043 0.0115 0.0078 0.0773 0.1043 0.0078 0.0078 0.3069 0.3069 0.0115
A2 0.0773 0.0773 0.0773 0.3069 0.3069 0.0115 0.0078 0.0773 0.1043 0.0115 0.0078 0.0773 0.3069 0.0115 0.0078
A3 0.0115 0.0115 0.0078 0.0773 0.0078 0.0773 0.3069 0.3069 0.3069 0.0115 0.0078 0.0773 0.1043 0.3069 0.0773
A4 0.1043 0.1043 0.0115 0.0078 0.0773 0.3069 0.0115 0.0078 0.0773 0.3069 0.0078 0.0773 0.3069 0.0773 0.3069
A5 0.0115 0.0115 0.0115 0.0078 0.0078 0.0773 0.0773 0.3069 0.0078 0.0773 0.3069 0.0115 0.0078 0.0773 0.3069
A6 0.0078 0.0773 0.3069 0.0078 0.0773 0.3069 0.0115 0.0078 0.0773 0.3069 0.3069 0.0078 0.0773 0.0773 0.0078
A7 0.3069 0.3069 0.3069 0.3069 0.0078 0.0773 0.0078 0.0115 0.0078 0.0773 0.1043 0.0078 0.1043 0.0115 0.0773
A8 0.3069 0.3069 0.3069 0.0078 0.0773 0.0773 0.0773 0.0115 0.0078 0.0115 0.0115 0.0078 0.0773 0.0115 0.0078
A9 0.3069 0.0773 0.0078 0.3069 0.3069 0.0773 0.0115 0.0078 0.0773 0.3069 0.1043 0.0115 0.0078 0.0773 0.3069
Min 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Max 0.31 0.31 0.31 0.31 0.31 0.31 0.31 0.31 0.31 0.31 0.31 0.08 0.31 0.31 0.31
Max-min 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.07 0.3 0.3 0.3
Star 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99
A grey relational analysis was then conducted. Table 7 shows the GRA coefficients. The GRA result indicates that there is not any dependency among the CSFs.Table 7 Presenting the gray coefficients.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
A1 0.07 0 0.07 0.3 0.3 0.09 0 0 0.07 0.09 0 0 0.3 0.3 0
A2 0.07 0.07 0.07 0.3 0.3 0 0 0.07 0.1 0 0 0.07 0.3 0 0
A3 0 0 0 0.07 0 0.07 0.3 0.3 0.3 0 0 0.07 0.1 0.3 0.07
A4 0.1 0.1 0 0 0.07 0.3 0 0 0.07 0.3 0 0.07 0.3 0.07 0.3
A5 0 0 0 0 0 0.07 0.07 0.3 0 0.07 0.3 0 0 0.07 0.3
A6 0 0.07 0.3 0 0.07 0.3 0 0 0.07 0.3 0.3 0 0.07 0.07 0
A7 0.3 0.3 0.3 0.3 0 0.07 0 0 0 0.07 0.1 0 0.1 0 0.07
A8 0.3 0.3 0.3 0 0.07 0.07 0.07 0 0 0 0 0 0.07 0 0
A9 0.3 0.07 0 0.3 0.3 0.07 0 0 0.07 0.3 0.1 0 0 0.07 0.3
Δ min 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Δ max 0.301 0.301 0.301 0.301 0.301 0.301 0.301 0.301 0.301 0.299 0.301 0.07 0.301 0.299 0.301
rΔ max 0.151 0.151 0.151 0.151 0.151 0.149 0.151 0.151 0.151 0.149 0.151 0.149 0.151 0.149 0.151
Δ min + r Δ max 0.151 0.151 0.151 0.151 0.151 0.149 0.151 0.151 0.151 0.149 0.151 0.149 0.151 0.149 0.151
5.2 Pythagorean fuzzy numbers best-worst method (PFNBWM)
Table 8 reports on the best criterion found using Eqs. (1), (2), and (17), (18), (19), (20), (21), (22), (23), (24), (25), (26). Then, based on that, the resulting criterion is compared to other criteria, and the DM preferences are assigned to them in terms of PFN (Hendiani et al., 2021) (see Table 9, Table 10, Table 11, Table 12 ).Table 8 Presenting the DMs’ preferences.
LS TR EN DE FS DM ACC GO EAC TA WDS DCP EAA APL
DTC {1,0} {0.4,0.87} {0.5,0.8} {0.4,0.87} {0.4,0.87} {0.5,0.8} {0.5,0.8} {0.8,0.44} {{0.4,0.87} {0.5,0.8} {0.5,0.8} {0.4,0.87} {0.5,0.8} {0.5,0.8}
Table 9 Computed analogously for the worst criterion.
LS
TR {1,0}
EN {0.6,0.71}
DE {0.7,0.6}
FS {0.7,0.6}
DM {0.8,0.44}
ACC {0.8,0.44}
GO {0.5,0.8}
EAC {1,0}
TA {1,0}
WDS {0.8,0.44}
DCP {1,0}
DTC {1,0}
EAA {1,0}
APL {0.5,0.8}
Table 10 Presented the transfer of PFN numbers to crisp numbers regarding the best criterion.
LS TR EN DE FS DM ACC GO EAC TA WDS DCP EAA APL
DTC 0.30685 0.31167 0.10430 0.31167 0.31167 0.10430 0.10430 0.07732 0.31167 0.10430 0.10430 0.31167 0.10430 0.10430
Table 11 Computed analogously as the worst criterion.
LS
TR 0.30685
EN 0.0115
DE 0.00778
FS 0.00778
DM 0.07732
ACC 0.07732
GO 0.1043
EAC 0.30685
TA 0.30685
WDS 0.07732
DCP 0.30685
DTC 0.30685
EAA 0.30685
APL 0.1043
Table 12 Presented the final weights of the criteria.
Factors LS TR EN DE FS DM ACC GO EAC TA WDS DCP DTC EAA APL
Weight 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666
A pairwise comparison method must also be used for the required weights along with the PFNEAMR (Hendiani et al., 2021). One of them is BWM, and in this research, BWM is combined with PFN (Zhou & Chen, 2020). The linear programming method based on Eq (2) was used for finding the weights using PFNBWM. The results show the weights of each alternative after solving this model with LINGO version 19. The criteria in this model are CSFs used for hospital site selection. This model consists of four parts. The first section is a set of constraints related to the best criterion, while the second section points to the constraints for the worst criterion. Section three deals with the constraint regarding the sum of the weights, which must be one. The final section shows a set of constraints that satisfies all positive weights. Eq. (38) shows the PFNBWM linear model solved using LINGO version 24 software. To solve the model, it is converted into a standard linear model for formulating in LINGO. The solution of the model shows the weights of factors.
Min k*(38) -k∗m1≤m13-0.30685∗m1≤k∗m1-k∗m2≤m13-0.31167∗m2≤k∗m2-k∗m3≤m13-0.1043∗m3≤k∗m3-k∗m4≤m13-0.31167∗m4≤k∗m4-k∗m5≤m13-0.31167∗m5≤k∗m5-k∗m6≤m13-0.1043∗m6≤k∗m6-k∗m7≤m13-0.1043∗m7≤k∗m7-k∗m8≤m13-0.07732∗m8≤k∗m8-k∗m9≤m13-0.31167∗m9≤k∗m9-k∗m10≤m13-0.1043∗m10≤k∗m10-k∗m11≤m13-0.1043∗m11≤k∗m11-k∗m12≤m13-0.31167∗m12≤k∗m12-k∗m14≤m13-0.1043∗m14≤k∗m14-k∗m15≤m13-0.1043∗m15≤k∗m15-k∗m1≤m2-0.30685∗m1≤k∗m1-k∗m1≤m3-0.0115∗m1≤k∗m1-k∗m1≤m4-0.00778∗m1≤k∗m1-k∗m1≤m5-0.00778∗m1≤k∗u1-k∗m1≤m6-0.07732∗m1≤k∗m1-k∗m1≤m7-0.07732∗m1≤k∗m1-k∗m1≤m8-0.1043∗m1≤k∗m1-k∗m1≤m9-0.30685∗m1≤k∗m1-k∗m1≤m10-0.30685∗m1≤k∗m1-k∗m1≤m11-0.07732∗m1≤k∗m1-k∗m1≤m12-0.30685∗m1≤k∗m1-k∗m1≤m13-0.30685∗m1≤k∗m1-k∗m1≤m14-0.30685∗m1≤k∗m1-k∗m1≤m15-0.1043∗m17≤k∗m17m1+m2+m3+m4+m5+m6+m7+m8+m9+m10+m11+m12+m13+m14+m15=1m1≤m2m2≤m3m3≤m4m4≤m5m5≤m6m6≤m7m7≤m8m8≤m9m9≤m10
m10≤m11m11≤m12m12≤m13m13≤m14m14≤m15m1>0m2>0m3>0m4>0m5>0m6>0m7>0m8>0m9>0m10>0m11>0m12>0m13>0m14>0m15>0k≥0
All resulting weights of criteria equal 0.0666. The next step is ranking the temporary hospital sites by EAMR. Equation sets (3), (4), (5), (6), (7), (8), (9), (10), (11), (12), (13), (14), (15) and (11), (12), (13), (14), (15), (16), (17), (18) were used to obtain the final ranking as reported in the next section.
5.3 Pythagorean fuzzy number evaluation area-based method for ranking (PFNEAMR)
First, the decision matrix is created, as presented in Table 13 .Table 13 Decision Matrix of PFNEAMR.
LS TR EN DE FS DM ACC GO EAC TA WDS DCP DTC EAA APL
0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666
Site → 1 {0.8,0.44} {0.7,0.6} {0.8,0.44} {1,0} {1,0} {0.5,0.8} {0.6,0.71} {0.7,0.6} {0.44,0.8} {0.5,0.8} {0.7,0.6} {0.7,0.6} {1,0} {1,0} {0.6,0.71}
Site → 2 {0.8,0.44} {0.8,0.44} {0.8,0.44} {1,0} {1,0} {0.6,0.71} {0.7,0.6} {0.8,0.44} {0.5,0.8} {0.6,0.71} {0.7,0.6} {0.8,0.44} {1,0} {0.6,0.71} {0.7,0.6}
Site → 3 {0.6,0.71} {0.6,0.71} {0.7,0.6} {0.8,0.44} {0.7,0.6} {0.8,0.44} {1,0} {1,0} {1,0} {0.6,0.71} {0.7,0.6} {0.8,0.44} {0.5,0.8} {1,0} {0.8,0.44}
Site → 4 {0.5,0.8} {0.5,0.8} {0.6,0.71} {0.7,0.6} {0.8,0.44} {1,0} {0.6,0.71} {0.7,0.6} {0.8,0.44} {1,0} {0.7,0.6} {0.8,0.44} {1,0} {0.8,0.44} {1,0}
Site → 5 {0.6,0.71} {0.6,0.71} {0.6,0.71} {0.7,0.6} {0.7,0.6} {0.8,0.44} {0.8,0.44} {1,0} {0.7,0.6} {0.8,0.44} {1,0} {0.6,0.71} {0.7,0.6} {0.8,0.44} {1,0}
Site → 6 {0.7,0.6} {0.8,0.44} {1,0} {0.7,0.6} {0.8,0.44} {1,0} {0.6,0.71} {0.7,0.6} {0.8,0.44} {1,0} {1,0} {0.7,0.6} {0.8,0.44} {0.8,0.44} {0.7,0.6}
Site → 7 {1,0} {1,0} {1,0} {1,0} {0.7,0.6} {0.8,0.44} {0.7,0.6} {0.6,0.71} {0.7,0.6} {0.8,0.44} {0.5,0.8} {0.7,0.6} {0.5,0.8} {0.6,0.71} {0.8,0.44}
Site → 8 {1,0} {1,0} {1,0} {0.7,0.6} {0.8,0.44} {0.8,0.44} {0.8,0.44} {0.6,0.71} {0.7,0.6} {0.6,0.71} {0.6,0.71} {0.7,0.6} {0.8,0.44} {0.6,0.71} {0.7,0.6}
Site → 9 {1,0} {0.8,0.44} {0.7,0.6} {1,0} {1,0} {0.8,0.44} {0.6,0.71} {0.7,0.6} {0.8,0.44} {1,0} {0.5,0.8} {0.8,0.44} {0.6,0.71} {0.8,0.44} {1,0}
The transferred score from the Pythagorean fuzzy sets matrix is shown in Table 14 .Table 14 Presented the Pythagorean fuzzy sets matrix.
LS TR EN DE FS DM ACC GO EAC TA WDS DCP DTC EAA APL
Weight 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666 0.0666
Site → 1 0.0773 0.0078 0.0773 0.3069 0.3069 0.1043 0.0115 0.0078 0.0773 0.1043 0.0078 0.0078 0.3069 0.3069 0.0115
Site → 2 0.0773 0.0773 0.0773 0.3069 0.3069 0.0115 0.0078 0.0773 0.1043 0.0115 0.0078 0.0773 0.3069 0.0115 0.0078
Site → 3 0.0115 0.0115 0.0078 0.0773 0.0078 0.0773 0.3069 0.3069 0.3069 0.0115 0.0078 0.0773 0.1043 0.3069 0.0773
Site → 4 0.1043 0.1043 0.0115 0.0078 0.0773 0.3069 0.0115 0.0078 0.0773 0.3069 0.0078 0.0773 0.3069 0.0773 0.3069
Site → 5 0.0115 0.0115 0.0115 0.0078 0.0078 0.0773 0.0773 0.3069 0.0078 0.0773 0.3069 0.0115 0.0078 0.0773 0.3069
Site → 6 0.0078 0.0773 0.3069 0.0078 0.0773 0.3069 0.0115 0.0078 0.0773 0.3069 0.3069 0.0078 0.0773 0.0773 0.0078
Site → 7 0.3069 0.3069 0.3069 0.3069 0.0078 0.0773 0.0078 0.0115 0.0078 0.0773 0.1043 0.0078 0.1043 0.0115 0.0773
Site → 8 0.3069 0.3069 0.3069 0.0078 0.0773 0.0773 0.0773 0.0115 0.0078 0.0115 0.0115 0.0078 0.0773 0.0115 0.0078
Site → 9 0.3069 0.0773 0.0078 0.3069 0.3069 0.0773 0.0115 0.0078 0.0773 0.3069 0.1043 0.0115 0.0078 0.0773 0.3069
The average decision matrix is presented in Table 15 .Table 15 Presented the average decision matrix.
LS TR EN DE FS DM ACC GO EAC TA WDS DCP DTC EAA APL
Site → 1 0.0639 0.0079 0.0694 0.2610 0.2610 0.0934 0.0220 0.0104 0.1040 0.0090 0.0090 0.0272 0.2362 0.3205 0.0104
Site → 2 0.0639 0.0788 0.0694 0.2610 0.2610 0.0103 0.0149 0.1038 0.1402 0.0090 0.0090 0.2703 0.2362 0.0120 0.0070
Site → 3 0.0095 0.0117 0.0070 0.0066 0.0066 0.0693 0.5866 0.4118 0.4126 0.0090 0.0090 0.2703 0.0803 0.3205 0.0697
Site → 4 0.0862 0.1063 0.0103 0.0658 0.0658 0.2749 0.0220 0.0104 0.1040 0.0090 0.0090 0.2703 0.2362 0.0808 0.2764
Site → 5 0.0095 0.0117 0.0103 0.0066 0.0066 0.0693 0.1478 0.4118 0.0105 0.3548 0.3548 0.0402 0.0060 0.0808 0.2764
Site → 6 0.0064 0.0788 0.2755 0.0658 0.0658 0.2749 0.0220 0.0104 0.1040 0.3548 0.3548 0.0272 0.0595 0.0808 0.0070
Site → 7 0.2535 0.3129 0.2755 0.0066 0.0066 0.0693 0.0149 0.0154 0.0105 0.1206 0.1206 0.0272 0.0803 0.0120 0.0697
Site → 8 0.2535 0.3129 0.2755 0.0658 0.0658 0.0693 0.1478 0.0154 0.0105 0.0133 0.0133 0.0272 0.0595 0.0120 0.0070
Site → 9 0.2535 0.0788 0.0070 0.2610 0.2610 0.0693 0.0220 0.0104 0.1040 0.1206 0.1206 0.0402 0.0060 0.0808 0.2764
Table 16 describes the average weighted matrix.Table 16 Average weighted matrix.
LS TR EN DE FS DM ACC GO EAC TA WDS DCP DTC EAA APL
Site → 1 0.1089 0.0135 0.1183 0.4448 0.4448 0.1593 0.0375 0.0178 0.1772 0.0153 0.0153 0.0464 0.4026 0.5463 0.0177
Site → 2 0.1089 0.1344 0.1183 0.4448 0.4448 0.0176 0.0254 0.1769 0.2390 0.0153 0.0153 0.4607 0.4026 0.0205 0.0120
Site → 3 0.0162 0.0200 0.0119 0.0113 0.0113 0.1181 1.0000 0.7020 0.7033 0.0153 0.0153 0.4607 0.1368 0.5463 0.1187
Site → 4 0.1469 0.1813 0.0176 0.1121 0.1121 0.4687 0.0375 0.0178 0.1772 0.0153 0.0153 0.4607 0.4026 0.1377 0.4712
Site → 5 0.0162 0.0200 0.0176 0.0113 0.0113 0.1181 0.2520 0.7020 0.0178 0.6048 0.6048 0.0685 0.0102 0.1377 0.4712
Site → 6 0.0110 0.1344 0.4696 0.1121 0.1121 0.4687 0.0375 0.0178 0.1772 0.6048 0.6048 0.0464 0.1014 0.1377 0.0120
Site → 7 0.4322 0.5333 0.4696 0.0113 0.0113 0.1181 0.0254 0.0263 0.0178 0.2055 0.2055 0.0464 0.1368 0.0205 0.1187
Site → 8 0.4322 0.5333 0.4696 0.1121 0.1121 0.1181 0.2520 0.0263 0.0178 0.0227 0.0227 0.0464 0.1014 0.0205 0.0120
Site → 9 0.4322 0.1344 0.0119 0.4448 0.4448 0.1181 0.0375 0.0178 0.1772 0.2055 0.2055 0.0685 0.0102 0.1377 0.4712
The normalized matrix and the weighted normalized matrix are then calculated and the final ranking obtained, as shown in Table 17 .Table 17 Final ranking.
Site1 1.223832554 0.281353779 4.349799599 4
Site → 2 1.022262321 0.524415542 1.949336432 7
Site → 3 1.911851209 0.368526155 5.187830445 2
Site → 4 1.102864895 0.524415542 2.103036252 6
Site → 5 1.041339314 0.755735332 1.37791535 8
Site → 6 0.991377604 0.796253675 1.24505247 9
Site → 7 1.046879976 0.348636238 3.00278589 5
Site → 8 1.235458296 0.113305312 10.90379855 1
Site → 9 1.42413383 0.287358426 4.955949437 3
This study aims to propose a novel MADM model to help determine where temporary hospitals should be installed. For instance, we utilize BWM, EAMR, and PFN methods to help distinguish and analyze the optimal location temporary hospitals. Our results suggest the fundamental factors for calculating where such temporary hospitals should be set up. In doing so, we identified the 16 most significant CSFs from a pool of 21 relative CSFs. For example, a couple of CSFs selected from our findings (ambulance accessibility and availability of parking) are crucial factors and have major implications for official government and healthcare departments in relation to installing temporary hospitals. The current study deployed PFNBWM and PFNEAMR to find the 9 most appropriate sites for temporary hospitals, as presented in Table 17. Moreover, by expanding and utilizing the current study and its findings, these sites can prompt the Iranian government and healthcare ministries in optimal location of temporary hospitals. This approach will not only help to combat COVID-19 but also have a significant indirect effect on rural communities and increased public welfare.
5.4 Sensitivity analysis
In this section on evaluating the reliability of our hybrid model, the results obtained are compared to those computed using alternative MADM methods resembling a similar computational structure. These methods are TOPSIS, VIKOR, and WASPAS. The final results are compared in terms of the Pearson coefficient correlation. When the significance level (denoted as α) is less than 0.05, these correlation results can be considered correlated and thus similar in the case of a positive correlation coefficient. Our sensitivity analysis shows that the result of our model, representing a compromise solution among all these individual methods, has the highest reliability because it has a linear relationship with all TOPSIS, VIKOR, and WASPAS methods. Table 18 and Fig. 2 show these sensitivity analysis results.Table 18 Reliability and validity of the results.
TOPSIS VIKOR WASPAS
Pearson Coefficient 0.867 0.667 0.883
Significance 0.002 0.05 0.002
Fig. 2 Presented the sensitivity analysis.
6 Conclusions
The healthcare industries in the 21st century era have faced severe challenges such as the COVID-19 pandemic. As such, temporary hospitals have been at the forefront of the fight against these challenges (Han et al., 2020, Govindan et al., 2021). Thus, rapidly providing temporary hospitals/institutions, equipment, and physicians to combat such challenges is of the utmost importance (Weiner et al., 2020). In response to such situations, the current study addresses the crucial questions as to how, when, and where to set up such temporary hospitals by utilizing the EAMR method in a combination of PFN with hybrid MADM methods.
The hybrid approach adopted in this research allowed the ranking of alternative locations for temporary hospitals by starting with the initial mapping of cost and benefit criteria in BWM. Although, GRA revealed independence among CSFs, alternatives and criteria weights are still endogenously related. This happens due to the perceptions of the decision-makers, since the trade-offs among location alternatives expressed by means of criterion weights cannot be clearly ascertained in terms of cause-effect relationships, but rather in terms of feedback relationships. Hence, a number of intermediate steps performed using EAMR were deemed necessary to remove this inherent bias. The fuzzyfication of EAMR steps via PFN helped in finding a better balance between membership and non-membership functions of a given alternative in light of positive and negative criteria based on the averaged perceptions, mitigating spurious feedbacks among alternatives and the respective weights assigned.
Our findings implicate “distance from the center of the province,” and “distance from two other cities” to be the top criteria for the Iranian government in installing temporary hospitals. Therefore, we suggest that these CSFs help support and assist in protecting the capital city from COVID-19 and guide in providing a better distribution of healthcare services. Additionally, this study suggested 16 potential CSFs for optimal location of temporary hospitals. These specific attributes include cost, land strategy, transport system, financial support, demand & accessibility, consideration of government within the emergency circumstances, distance from the capital, vehicle strategies, and parking lots.
In addition, such CSFs may provide support in combating crises and increasing economic integrity, both of which have significant indirect effects on poor communities and public welfare. As mentioned above, 20 % of the overall Iranian budget for this year, approximately 10 trillion rials (US$ 237.5 million), has been allocated to combating the COVID-19 pandemic (Guler, 2020). Along with these efforts, the current research contributed to selecting sites for temporary hospitals, seeking to enhance healthcare services, public welfare. The study also contributes indirectly in combating future disasters and supporting the poor economically to maintain and improve their situation. The final CSFs presented in this study can improve performance in a context where resources are scarce and induce implementation synergies. For example, the current study finds that the most prominent factor is “waste disposal site,” a factor that has been neglected in prior studies.
6.1 Managerial implications
Despite the spread of COVID-19, decision-making plays a vital role when setting up temporary hospitals since the wrong decision could affect rural communities and the entire country. As argued by Sharfuddin (2020, p. 248), “it is ironic that this pandemic has attacked one thing most precious to modern civilization, which is human liberty.” Temporary hospitals have a critical role in fostering the comeback of human liberty and emergency readiness (Auener et al., 2020). Some advantages of temporary hospitals include the ability to rapidly triage large numbers of patients, provide emergency clinical services, support local clinics and laboratories, and decrease the risk of a virus. Policymakers and governments must keep these factors in mind to perform actions/controls for optimal decision-making in times of crises such as COVID-19 and to establish a strategic roadmap for temporary hospitals. This study, using MADM methods, provides such a road map for policymakers to use CSFs for effectively locating temporary hospitals.
The government and health authorities play a key role in the decision-making process and, therefore, should hire experienced healthcare managers to retain dynamic communication with several local and international hospitals, such as the Chinese hospital built in Wuhan. Healthcare managers should use a participatory policy approach and be acquainted with CSFs and their importance and implementation. Finally, the managers should also recognize when the time is right to set up temporary hospitals and where to offer that space.
By now, it is well stablished that distinct variants of COVID-19 (e.g., alpha, delta, and omicron) display different levels of infection severity and symptoms among the population. As the pandemic evolves into its third and distinct waves continue to emerge, decision-makers could revisit the hybrid approach depicted in this research for a better management of existing resources. While the availability and proximity of large and well-equipped ICUs were critical in the initial first phases of the pandemic, smaller and widely distributed family health-care clinics run by paramedics may now be desirable for purposes of vaccination and rapid testing. The experience brought by the Spanish flu pandemic in 1918 shows not only that a pandemic can last for several years, but also that the profile of victims in terms of gender and age and the severity of symptoms may vary substantially and even include a hemorrhagic component, something not thus far seen.
6.2 Future directions
Although our study already involves several DMs, future research could focus on diverse perspectives to compare the most relevant stakeholder groups, formal institutions, non-government organizations, and their supportive actions to establish temporary hospitals in times of crisis. In addition, future research could also involve a broader range of MADM methods and different approaches. For instance, the site selection problem could be formulated as a facility location problem usually done in logistics (Hanne & Dornberger, 2017), which may allow for an efficient selection from a nonfinite set of possible locations.
Another future research stream, while broader than the temporary facility location issues, relates to the systematic treatment of the endogeneity issues among alternative criteria and weights, which are inherent in MADM applications. Hybrid approaches and exhaustive testing have proven to be critical for achieving robust results, not only in terms of ranking of alternatives, but also for mitigating bias that may emerge when decision-makers assign weights. The best combination of approaches, however, is still a matter of future debate and research.
CRediT authorship contribution statement
Amir Karbassi Yazdi: Conceptualization, Formal analysis, Investigation. Farhan Muhammad Muneeb: Formal analysis, Writing – original draft. Peter Fernandes Wanke: Writing – original draft, Supervision. Thomas Hanne: Writing – review & editing, Project administration. Adnan Ali: Formal analysis.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
==== Refs
References
Abdi M. Coronavirus disease 2019 (COVID-19) outbreak in Iran: Actions and problems Infection Control and Hospital Epidemiology 41 6 2020 754 755 10.1017/ice.2020.86 32192541
Aboutorab H. Saberi M. Asadabadi M.R. Hussain O. Chang E. ZBWM: The Z-number extension of Best Worst Method and its application for supplier development Expert Systems with Applications 107 2018 115 125 10.1016/j.eswa.2018.04.015
Adalı E.A. Tuş A. Hospital site selection with distance-based multi-criteria decision-making methods International Journal of Healthcare Management 2019 1 11
Ahmadi H.B. Kusi-Sarpong S. Rezaei J. Assessing the social sustainability of supply chains using Best Worst Method Resources, Conservation and Recycling 126 2017 99 106 10.1016/j.resconrec.2017.07.020
Alcantud J.C.R. Some formal relationships among soft sets, fuzzy sets, and their extensions International Journal of Approximate Reasoning 68 2016 45 53 10.1016/j.ijar.2015.10.004
Amiri M. Antucheviciene J. Evaluation by an area-based method of ranking interval type-2 fuzzy sets (EAMRIT-2F) for multi-criteria group decision-making Transformations in Business and Economics 15 3 2016 76 95
Auener S. Kroon D. Wackers E. Van Dulmen S. Jeurissen P. COVID-19: A window of opportunity for positive healthcare reforms International Journal of Health Policy and Management 2020
Atanassov, K. T. (1999). Intuitionistic fuzzy sets, in: Intuitionistic Fuzzy Sets. Studies in Fuzziness and Soft Computing, Physica, Heidelberg, pp. 1-137.
Brady S.R. Utilizing and adapting the Delphi method for use in qualitative research International Journal of Qualitative Methods 14 5 2015 10.1177/1609406915621381
Brandt T. Wagner S. Neumann D. Prescriptive analytics in public-sector decision-making: A framework and insights from charging infrastructure planning European Journal of Operational Research 291 1 2021 379 393 33041471
Can G.F. Toktaş P. Pakdil F. Six Sigma Project Prioritization and Selection Using AHP–CODAS Integration: A Case Study in Healthcare Industry IEEE Transactions on Engineering Management. 2021
Çelikbilek Y. Group Decision Making for Hospital Location Selection Using VIKOR under Fuzzy Environment Igusabder 5 2018 435 450 10.38079/igusabder.425439
Chauhan A. Singh A. A hybrid multi-criteria decision making method approach for selecting a sustainable location of health-care waste disposal facility Journal of Cleaner Production 139 2016 1001 1010 10.1016/j.jclepro.2016.08.098
Danaei G. Farzadfar F. Kelishadi R. Rashidian A. Rouhani O.M. Ahmadnia S. Ahmadvand A. Arabi M. Ardalan A. Arhami M. Iran in transition The Lancet 393 10184 2019 1984 2005
Dunham R.B. The DELPHI technique 1998 Organizational Behavior
Elabed S. Shamayleh A. Daghfous A. Sustainability-oriented innovation in the health care supply chain Computers & Industrial Engineering 160 2021 107564
Fargnoli M. Haber N. A practical ANP-QFD methodology for dealing with requirements’ inner dependency in PSS development Computers & Industrial Engineering 127 2019 536 548
Fei L. Deng Y. Multi-criteria decision making in Pythagorean fuzzy environment Applied Intelligence 50 2 2020 537 561 10.1007/s10489-019-01532-2
Garg C.P. Sharma A. Sustainable outsourcing partner selection and evaluation using an integrated BWM–VIKOR framework Environment, Development and Sustainability 22 2 2020 1529 1557 10.1007/s10668-018-0261-5
Gordon T.J. The delphi method Futures Research Methodology 2 3 1994 1 30
Govindan K. Nasr A.K. Mostafazadeh P. Mina H. Medical waste management during coronavirus disease 2019 (COVID-19) outbreak: A mathematical programming model Computers & Industrial Engineering 162 2021 107668
Guler, B., 2020, Iran to cover 90% of COVID-19 patient’s expenses. https://www.aa.com.tr/en/corporate-news/iran-to-cover-90-of-COVID-19-patients-expenses-/1783168.
Han R.H. Schmidt M.N. Waits W.M. Bell A.K.C. Miller T.L. Planning for mental health needs during COVID-19 Current Psychiatry Reports 22 12 2020 1 10 31912372
Hanne T. Dornberger R. Computational Intelligence in Logistics and Supply Chain Management 2017 Springer Cham
Hashemkhani Zolfani S. Yazdani M. Ebadi Torkayesh A. Derakhti A. Application of a gray-based decision support framework for location selection of a temporary hospital during COVID-19 pandemic Symmetry 12 6 2020 886
Hejazi T.-H. State-dependent resource reallocation plan for health care systems: A simulation optimization approach Computers & Industrial Engineering 159 2021 107502
Hendiani S. Lev B. Gharehbaghi A. Diagnosing social failures in sustainable supply chains using a modified Pythagorean fuzzy distance to ideal solution Computers & Industrial Engineering 154 2021 107156
Hsu W.-C.-J. Liou J.J.H. Lo H.-W. A group decision-making approach for exploring trends in the development of the healthcare industry in Taiwan Decision Support Systems 141 2021 113447
Julong D. Introduction to grey system theory The J. Grey Syst. 1 1989 1 24
Kahraman, C., Gundogdu, F. K., Onar, S. C., Oztaysi, B., 2019, Hospital Location Selection Using Spherical Fuzzy TOPSIS. In: The 11th Conference of the European Society for Fuzzy Logic and Technology (EUSFLAT 2019). https://doi.org/10.2991/eusflat-19.2019.12.
Karbassi Yazdi A. Hanne T. Osorio Gómez J.C. García Alcaraz J.L. Finding the best third-party logistics in the automobile industry: A hybrid approach Mathematical Problems in Engineering 2018 2018 10.1155/2018/5251261
Karbassi Yazdi, A., Muneeb, F. M., Wanke, P. F., Figueiredo, O., & Mushtaq, I. (2021). Critical Success Factors for Competitive Advantage in Iranian Pharmaceutical Companies: A Comprehensive MCDM Approach. Mathematical Problems in Engineering, 2021.
Keshavarz Ghorabaee M. Zavadskas E.K. Amiri M. Antucheviciene J. A new method of assessment based on fuzzy ranking and aggregated weights (AFRAW) for MCDM problems under type-2 fuzzy environment Economic Computation and Economic Cybernetics Studies and Research 50 1 2016 39 68
Khaksefidi M. Miri M. Hospital Location in the Southern Fars Province by Using Multi Criteria Decision Making Techniques European Online Journal of Natural and Social Sciences 4 3 2015 638 645
Kmail A. Jubran J. Sabbah W. Jenin P. Coupling GIS-based MCA and AHP techniques for Hospital Site Selection International Journal of Computer Science and Information Security (IJCSIS) 15 12 2017
Kuo Y. Yang T. Huang G.-W. The use of grey relational analysis in solving multiple attribute decision-making problems Computers and Industrial Engineering 55 1 2008 80 93 10.1016/j.cie.2007.12.002
Kutlu Gündoğdu F. Kahraman C. Civan H.N. A novel hesitant fuzzy EDAS method and its application to hospital selection Journal of Intelligent Fuzzy Systems 35 6 2018 6353 6365 10.3233/JIFS-181172
Linstone H.A. Turoff M. The delphi method 1975 Addison-Wesley Reading, MA
Liu D. Li T. Liang D. An integrated approach towards modeling ranked weights Computers & Industrial Engineering 147 2020 106629
Liu W. Dridi M. Fei H. El Hassani A.H. Solving a multi-period home health care routing and scheduling problem using an efficient matheuristic Computers & Industrial Engineering 162 2021 107721
Lo H.-W. Liaw C.-F. Gul M. Lin K.-Y. Sustainable supplier evaluation and transportation planning in multi-level supply chain networks using multi-attribute-and multi-objective decision making Computers & Industrial Engineering 162 2021 107756
Lotfi R. Kheiri K. Sadeghi A. Babaee Tirkolaee E. An extended robust mathematical model to project the course of COVID-19 epidemic in Iran Annals of Operations Research 2022 1 25
Lu J. Zhang M. Zhang J. Xu C. Cheng B. Can health poverty alleviation project reduce the economic vulnerability of poor households? Evidence from Chifeng City, China Computers & Industrial Engineering 162 2021 107762
Madanaguli A.T. Dhir A. Talwar S. Singh G. Escobar O. Business to business (B2B) alliances in the healthcare industry: A review of research trends and pertinent issues Journal of Business & Industrial Marketing 2021
Mazloumian M. Baki M.F. Ahmadi M. A robust multiobjective integrated master surgery schedule and surgical case assignment model at a publicly funded hospital Computers & Industrial Engineering 163 2022 107826
Miç P. Antmen Z.F. A healthcare facility location selection problem with Fuzzy TOPSIS method for a regional hospital Avrupa Bilim ve Teknoloji Dergisi 16 2019 750 757 10.31590/ejosat.584217
Mohammed A. Naghshineh B. Spiegler V. Carvalho H. Conceptualising a supply and demand resilience methodology: A hybrid DEMATEL-TOPSIS-possibilistic multi-objective optimization approach Computers & Industrial Engineering 160 2021 107589
Moosavi J. Hosseini S. Simulation-based assessment of supply chain resilience with consideration of recovery strategies in the COVID-19 pandemic context Computers & Industrial Engineering 160 2021 107593
Muneeb F.M. Yazdi A.K. Wanke P. Yiyin C. Chughtai M. Critical success factors for sustainable entrepreneurship in Pakistani Telecommunications industry: A hybrid grey systems theory/best-worst method approach Management Decision 2020
Peng X. Yang Y. Some results for Pythagorean fuzzy sets International Journal of Intelligence Systems 30 11 2015 1133 1160 10.1002/int.21738
Rezaei J. Best-worst multi-criteria decision-making method Omega 53 2015 49 57 10.1016/j.omega.2014.11.009
Rezaei J. Best-worst multi-criteria decision-making method: Some properties and a linear model Omega 64 2016 126 130 10.1016/j.omega.2015.12.001
Şahin T. Ocak S. Top M. Analytic hierarchy process for hospital site selection Health Policy Technology 8 1 2019 42 50 10.1016/j.hlpt.2019.02.005
Samadi A.H. Owjimehr S. Halafi Z.N. The cross-impact between financial markets, Covid-19 pandemic, and economic sanctions: The case of Iran Journal of Policy Modeling 43 1 2021 34 55 32994651
Sarkar S. Pramanik A. Maiti J. Reniers G. COVID-19 outbreak: A data-driven optimization model for allocation of patients Computers & Industrial Engineering 161 2021 107675
Sen H. Hospital location selection with ARAS-G. The Eurasia Proc Science, Technology, Engineering, and Mathematics 1 2017 359 365
Sharfuddin S. The world after Covid-19 The Round Table 109 3 2020 247 257
Senvar O. Otay I. Bolturk E. Hospital site selection via hesitant fuzzy TOPSIS IFAC-PapersOnLine 49 12 2016 1140 1145 10.1016/j.ifacol.2016.07.656
Torra V. Hesitant fuzzy sets International Journal of Intelligence Systems 25 6 2010 529 539 10.1002/int.20418
Takian A. Raoofi A. Kazempour-Ardebili S. COVID-19 battle during the toughest sanctions against Iran Lancet (London, England) 395 10229 2020 1035 32199073
Weiner H.L. Adelson P.D. Brockmeyer D.L. Maher C.O. Gupta N. Smyth M.D. …Ahn E.S. Pediatric neurosurgery along with Children’s Hospitals’ innovations are rapid and uniform in response to the COVID-19 pandemic. Journal of Neurosurgery Pediatrics 1 aop 2020 1 3
Wu Y. Zhou F. Kong J. Innovative design approach for product design based on TRIZ, AD, fuzzy and Grey relational analysis Computers & Industrial Engineering 140 2020 106276
Yager, R. R., 2013, June, Pythagorean fuzzy subsets. In: 2013 joint IFSA world congress and NAFIPS annual meeting (IFSA/NAFIPS), IEEE, pp. 57-61.
Yager R.R. Properties and applications of Pythagorean fuzzy sets Angelov P. Sotirov S. Imprecision and uncertainty in information representation and processing 2016 Springer Cham 119 136
Yang M. Ni Y. Yang L. A multi-objective consistent home healthcare routing and scheduling problem in an uncertain environment Computers & Industrial Engineering 160 2021 107560
Yazdi A.K. Komijan A.R. Wanke P.F. Sardar S. Oil project selection in Iran: A hybrid MADM approach in an uncertain environment Applied Soft Computing 88 2020 106066 10.1016/j.asoc.2020.106066
Ye J. Zhan J. Xu Z. A novel multi-attribute decision-making method based on fuzzy rough sets Computers & Industrial Engineering 155 2021 107136
Zhang L. Zhu F. Xie L. Wang C. Wang J. Chen R. Jia P. Guan H.Q. Peng L. Chen Y. Peng P. Clinical characteristics of COVID-19-infected cancer patients: A retrospective case study in three hospitals within Wuhan, China 2020 Annals of Oncology
Zhou F. Chen T.-Y. Multiple criteria group decision analysis using a Pythagorean fuzzy programming model for multidimensional analysis of preference based on novel distance measures Computers & Industrial Engineering 148 2020 106670
| 36466770 | PMC9707025 | NO-CC CODE | 2022-12-07 23:16:13 | no | Comput Ind Eng. 2023 Jan 29; 175:108761 | utf-8 | Comput Ind Eng | 2,022 | 10.1016/j.cie.2022.108761 | oa_other |
==== Front
Physica A
Physica A
Physica a
0378-4371
0378-4371
Elsevier B.V.
S0378-4371(22)00899-8
10.1016/j.physa.2022.128341
128341
Article
Inner composition alignment networks reveal financial impacts of COVID-19
Upadhyay Shashankaditya a
Mukherjee Indranil b⁎
Panigrahi Prasanta K. c
a Department of Electrical Engineering, Indian Institure of Technology, New Delhi, 110016, India
b School of Management Sciences, Maulana Abul Kalam Azad University of Technology, Haringhata, Nadia, West Bengal, 741249, India
c Department of Physical Sciences, Indian Institure of Science Education and Research, Kolkata, Mohanpur, West Bengal, 741246, India
⁎ Corresponding author.
29 11 2022
29 11 2022
1283419 1 2021
16 11 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Elsevier B.V.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
We show that inner composition alignment networks derived for financial time-series data, studied in response to worldwide lockdown imposed in response to COVID-19 situation, show distinct patterns before, during and after lockdown phase. It is observed that significant couplings between companies as captured by inner composition alignment between time series, reduced considerably across the globe during lockdown and post-lockdown recovery period. The study of global community structure of the networks show that factions of companies emerge during recovery phase, with strong coupling within the members of the faction group, a trend which was absent before lockdown period. The study of strongly connected components of the networks further show that market has fragmented in response to COVID-19 situation. We find that most central firms as characterized by in-degree, out-degree and betweenness centralities belong to Chinese and Japanese economies, indicating a role played by these organizations in financial information propagation across the globe. We further observe that recovery phase of the lockdown period is strongly influenced by financial sector, which is one of the main result of this study. It is also observed that two different group of companies, which may not be co-moving, emerge across economies during COVID-19. We further notice that many companies in US and European economy tend to shield themselves from local influences.
Keywords
Financial networks
Inner composition alignment networks
Coupled time series
==== Body
pmc1 Introduction
Of the many modern constructs that define human existence in present age, markets form a special niche that facilitate exponentially larger proportions of resource allocation that could ever be achieved by an individual or a central command. Essential to sustain present day life, markets affect each individual in the society, often in subtle invisible ways. Governed by basic principles of diversity, interaction among components (companies and organizations etc.), and regulation, markets form a prime example of complex adaptive systems (CASs) [1], [2], [3]. A complex adaptive system viewpoint makes it possible to apply modern scientific tools to study such vast and diverse system as a market, studying which otherwise is a very challenging task. The use of modern scientific methods to study finance is a relatively new approach [4], [5], [6], [7], [8], [9], [10], [11] and several such studies have successfully characterized many different aspects of markets at different time scales [12], [13], [14], [15], [16].
COVID-19 and its aftermath has been encountered as one of the most unprecedented situation the world has ever witnessed. It is unclear whether such large scale disruptions in market activities in the form of forced lock-down all across the globe has ever been observed before in times of peace. However catastrophic and painful the COVID-19 situation has been for the average person, it presents researches with a unique paradigm to study finance at a macro-scale [17], [18], [19]. While the complete repercussions of the COVID-19 situation may only be clear with passage of time, we wish to present an indicative study of the COVID-19 situation as studied from a complex adaptive systems point of view that may help broaden our understanding in this respect.
Network theory, which can be seen as application of graph theory to real world problems, is one of the most common and widely accepted methodology to model and study markets as complex adaptive systems [20], [21], [22], [23], [24], [25], [26]. Networks are empirical representations of a complex system, where the underlying topology is a graph such that the various interactive components of the system can be seen as vertices and the interaction of interest between these components is modeled as edges or arcs (if the direction of interaction is also important). A formal definition for networks can be found in [27]. In the current study we wish to present a network representation of the global, Indian and US market where the vertices represent public companies that trade stocks across various stock exchanges and are connected by directed edges or arcs if the closing stock prices of these companies are co-evolving or significantly coupled as captured by their inner composition alignment [28].
We collected financial time-series data (closing indices of stock prices) of 743 large global corporations which comprises of 286 Global Fortune 500 (other than Indian and US) companies, 408 Fortune 500 US companies and 49 large-scale Indian companies for a total of 150 days, roughly representing a period from 1 January to 15 August, 2020. The 286 Global Fortune 500 companies can be further classified as companies belonging to the European region (including UK) which are 132 in number and 137 companies belonging to Asian economies and rest are from other regions. The collected data was divided into three equal parts of 50 indices each, representing pre-lockdown, lockdown and recovery period for COVID-19. The broad objectives of the current studies are to, (a) identify the factions or groups of companies that have co-evolved over the different periods and (b) identify how coupling between companies has changed during different periods. We further wish to get an insight into factors that help mitigate financial difficulties that arise due to COVID-19 situation.
2 Material and methods
The study of the financial impact of COVID-19 on the market presents a unique challenge from a data science point of view, as the size of time series data available to study financial impacts of worldwide lockdowns due to COVID-19 is very small (as low as 50 indices in our case). Many of the common methods that are traditionally used to convey or establish some type of relational information such as causality or information flow or indicate coupling relationship within and among time series data fail to be effective in case of time series of short lengths. In the following section we present results corresponding to very simple methods based on correlation matrices and establish how these methods are either inapplicable or present a limited understanding of the otherwise complex market.
2.1 Covariance and cross-correlation matrix based methods
Methods based on application of random matrix theory to the covariance matrices of stock movements is known as one of the benchmark methods to study financial time series data [29], [30]. In order for the random matrix methods to be applicable on the time series data, it is firstly necessary that we restructure the data such that the results are independent of the scale of measurement. To achieve the same, for a given time series Xi(t) for the stock corresponding to company i, we calculate the logarithmic price return which is defined as Ri(t,δt)=ln(Xi(t+δt))−ln(Xi(t)) where δt correspond to one day as daily closing indices of stock data is chosen for the study. The resulting logarithmic returns are normalized by subtracting the mean value of returns and dividing the result by the standard deviation of the returns as given by Ni(t,δt)=Ri−Ri¯σi, where Ri¯ is the mean values of logarithmic returns and σi is the standard deviation. The time series that has undefined entries (one such time series was found in both pre-lockdown and lockdown period) due to zero standard deviation were removed from the study. Once the normalized returns are calculated hence, we arrange the data as a matrix WN×(T−1) (N−1 in case when time series was removed as described above), such that T is the length of time series (corresponding to 50 days in each of the pre-lockdown, lockdown and post-lockdown cases) and N is the number of components (number of stocks equal to 743 in our case) and the resultant matrix has T−1 columns because of using log returns for the data. The (N×N) covariance matrix can now be calculated as A=WW∗, where W∗ denotes the transpose/Hermitean conjugate of W. It must be noted that the orientation of data is important in computing the covariance matrix. If we interchange the rows and columns and represent the data as a (T×N) matrix, then the covariance matrix would be calculated as A=W∗W, which is again a (N×N) matrix.
There are two cases of interest here. The Wishart and anti-Wishart cases describe the situations where length of the time series T and the number of time series N either obey the relation T≥N (Wishart) case or the relation N>T (anti-Wishart) case [31], [32], [33]. For uncorrelated time series, the resultant random covariance matrix, with N→∞, T→∞, such that Q=T/N≥1, i.e., the Wishart case has eigenvalue distributed according to the Marchenko–Pastur distribution. P(λ)=Q2π(λmax−λ)(λ−λmin)λ.
This distribution is bounded by the maximum eigenvalue λmax given by λmax=[1+1Q]2 and the minimum eigenvalue λmin given by λmin=[1−1Q]2 [34]. Typically, the Wishart case refers to the truly random (N×N) matrices where T≥N and all the N eigenvalues are positive. In the anti-Wishart case (T<N), the matrices (again of dimension N×N) carry some redundant information. To apply the Marchenko–Pastur distribution of eigenvalues when T<N, it is necessary to replace Q by Q′ with respect to the original Marchenko–Pastur definition, such that Q′=N/T. In this situation, the covariance matrix spectrum comprises of T positive eigenvalues and the remaining (N−T) eigenvalues are identically zero.
It may be mentioned here that there exist some ambiguity in the literature about the naming convention of the Wishart and anti-Wishart cases. For example, the anti-Wishart nomenclature is sometimes attributed to the (T×T) matrices when T<N [32], [33]. However in the present work, the anti-Wishart format refers to the (N×N) matrices when T<N.
In the present study, the closing stock prices corresponding to 150 days for 743 companies is divided into three equal parts and subsequently log-returns of data are calculated and then normalized. Thus effectively for calculating Q′ we use the value T−1 (which in our case becomes equal to 49) while N remains 743. However as discussed earlier, for pre-lockdown and lockdown period data for companies that contained undefined values upon normalization as a result of zero standard deviation in the data following the calculation of logarithmic returns (one such company found in both pre-lockdown and lockdown period) was removed from the study and thus N effectively becomes 742 in these two cases. Using Q′=N/(T−1), we obtain Q′=15.16 (a slightly different value of 15.14 in other two cases). Thus the largest and the smallest eigenvalue as predicted by random matrix theory (RMT) in our case becomes equal to 1.5796 and 0.5523 respectively (slightly different values equal 1.5800 and 0.5521 in the pre-lockdown and lockdown cases). For proper application of random matrix theory formalism, bulk of the eigenvalues of the covariance matrix should lie within the upper and the lower bounds as predicted by the Marchenko–Pastur distribution i.e. between λmax and λmin. It is observed that the non-trivial eigenvalues for the three periods viz pre-lockdown, lockdown and post-lockdown period, mostly lie beyond the range predicted by RMT. The few deviations from the bulk behavior provides important insights about the underlying system. For the present dataset considering the anti-Wishart case, it is observed that while the (N−T+1) eigenvalues are typically zero (N−T−2 in other cases), the non-trivial eigenvalues still showed deviation in large numbers from the bulk predicted by the random matrix theory, thereby indicating high correlation in the data. It is thus not possible to apply RMT framework in the present case due to the time series being very small and the resultant data thus being highly correlated.
We can further use a distance metric based on the cross-correlation matrix C given by dij=2(1−Cij) to construct minimum spanning tree corresponding to the cross-correlation matrix such that for the n−1 edges (where n is the order of the network i.e., number of stocks used in the study) in the tree, the sum of distance Σijdij is minimum for any choice of edges possible in the tree. Minimum spanning tree method is known to present a good graphical representation of co-moving stocks that cluster together [29]. The same is observed in our case where we observe clusters of co-moving stocks forming for the pre-lockdown, lockdown and post-lockdown phase with the help of minimum spanning trees for these periods. These results can be accessed as figures presented in the supplementary material.
It must however be noted that the minimum spanning tree method presents a very limited view of the whole wide market system where an array of possible interactions (as represented by a large number of edges or directed arcs in a network as against n−1 edges) add to the complexity of the system. Thus in order to thoroughly analyze the market we need to implement a different approach that is applicable on time series of small size.
In a typical study based on financial time series data such as the one undertaken here, we try to unravel the fundamental driving mechanisms that produce the stochastic processes, which in turn give rise to the time series that we intend to analyze. It is important to note that we usually have just one sequentially observed data set, i.e., a time series and try to infer the properties of the generating process from this single trajectory. However, as the time series data available is very small, other traditional methods used to infer causality or dependence in time series such as transfer entropy, Granger causality, method of mutual information, symbolic dynamics, Hurst exponent etc. [35], [36], [37], [38], [39], [40] may not yield useful results and are thus avoided in the current study.
A few methods to infer causality or dependence in data with small time series have been proposed in literature [41], [42], [43]. We however, choose to implement inner composition alignment between given time series as not only this method which can be implemented on short time-series identifies directionality of coupling, infers auto-regulation and does not depend on time, but at the same time we found it easier to implement as compared to other methods. Inner composition alignment (IOTA) as a method has been successfully implemented to derive functional brain networks using short electroencephalogram (EEG) data [44] and to infer causality in gene regulatory networks in case when the time-series data related to regulatory processes is very small [28].
2.2 Inner composition alignment networks
Inner composition alignment (IOTA) is a directionality preserving, permutation based measure between two time series which can be applied to time series of same size [28]. Consider two time series X and Y of equal length n. Now consider a permutation π which when applied to the time series X returns a non-decreasing time series π(X) i.e., π:(π(X))i≤(π(X))i+1,∀i:1≤i≤n−1. Let Y˜ be the series which is returned when the permutation π is applied to the time series Y i.e., Y˜=π(Y). Inner composition alignment from time series X to Y, represented ιX→Y, in effect measures the number of times the series Y˜ intersects (criss-crosses) the series π(X) i.e., if Y˜ becomes a monotonic sequence (non-decreasing or non-increasing) on application of permutation π on time series Y, then X and Y are highly coupled, otherwise the more fluctuations are found in series Y˜, the lesser the coupling.
The inner composition alignment from time series X to Y under the permutation π can be calculated as (1) ιX→Y=1−∑i=1n−2∑j=i+1n−1wijΘ[(Y˜j+1−Y˜i)(Y˜i−Y˜j)]Δ,
where Δ=(n−1)(n−2)2 is a normalization constant, wij is a weight and Θ is Heaviside step function i.e., Θ[x] is 1 for positive values of x and zero otherwise. For the purpose of our analysis we choose weight wij=1.
It must be noted that inner composition analysis will rate coupling between two time series higher or close to one if they are monotonic in the above mentioned sense i.e., despite if one series is increasing and the other is decreasing. For financial data however this property is undesirable, since it is important if the stock price of a company is increasing or decreasing as the change in values of stock prices makes all the difference between profit and loss. To overcome this limitation, we device a method that allows IOTA to differentiate between increasing and decreasing sequences.
We call a time series as uptrending if the mean of first half of the series is less than the mean of the series which in turn is less than the mean of the latter half of the series i.e., for a time series T of length n,n≥5, if Ti¯<T¯<Tj¯,1≤i≤⌊n2⌋,⌈n2⌉≤j≤n. Similarly, a time series T is downtrending if Ti¯>T¯>Tj¯,1≤i≤⌊n2⌋,⌈n2⌉≤j≤n. We modified inner composition alignment, ιX→Y, to take a positive value as its outcome if both series X and Y are either uptrending or downtrending. If one of the series X or Y is uptrending and the other is downtrending, we take the value of IOTA as negative, and we take this value as zero if no such trend could be established.
We further wish to assess if for a given time series T of a stock, there are no points in these time series such that due to presence of these few points the overall mean of either the first half or the second half of the time series change significantly enough to render the ordering of Ti¯, T¯ and Tj¯,1≤i≤⌊n2⌋,⌈n2⌉≤j≤n. That is to say for example if there are very small number of points p1,…,pk, where k is typically small such that there is a clear trend Ti−{p1,…,pk}¯<T¯<Tj¯ (where Ti−{p1,…,pk}¯ is mean of first half of the time series calculated without including the points {p1,…,pk}) while no such trend can be established when these points are included in Ti,1≤i≤⌊n2⌋. Thus we wish to establish that there are no such points in the time series studied here the values of which jump so much that by including these points we can no longer call the time series as uptrending or downtrending while the same can be done when these points are not included. In order to establish the same we take difference of time series Di(t)=Ti(t+1)−Ti(t) and calculate the Z-score Zi=Dimax−Di¯σ(Di) where Dimax is the maximum value of the differenced series Di, Di¯ is the mean and σ(Di) is the standard deviation of the series Di. The idea here is that for a jump point the difference from previous (or latter) value will be very large where the differences otherwise shall follow a normal distribution. For the three periods considered in this study i.e., the pre-lockdown, the lockdown and the post-lockdown period, we found the maximum Z-score across time series for all 743 stocks to be 5.9718, 6.7095 and 6.8571 respectively. Thus we conclude that there are likely no jump points in the data considered in this study. However, we present a detailed framework to qualify such jump points in time series data as supplimentay information and rigrously establish absence of jump points in the data used in the study. The added framework provides a detailed justification of the validity of proposed definitions of uptrending and downtrending time series and provides an explanation as to how the proposed definitions can be employed in the alasysis of time series data in a generalized setting.
To derive a network from time series which show significant coupling strength with respect to IOTA, we first calculate a coupling matrix (C)ij, where i and j represent the time series data corresponding to some specific companies incorporated in the current study. In order to find a given entry cij in the coupling matric C, we first permute or sort the time series i such that it becomes a strictly increasing time series and register such a permutation π. Next we apply this permutation π, which returns a strictly increasing or sorted time series i, to rearrange the entries in time series j into another time series j¯. Thereafter we apply the formulation used in (1) on j¯ to calculate the value of IOTA.
Once the values of IOTA are derived, we proceed to find if the time series i and j are uptrending or downtrending as defined earlier by comparing the value of mean of time series with means of first half and latter half of the time series and thus compute the modified IOTA values by comparing the trends in the timeseries. We likewise find the modified coupling matrix C¯, in which every entry (C¯)ij is modified IOTA value between time series i and j. In order to keep only significant coupling strengths in the directed network as arcs, we derive the directed network from the modified coupling matrix C¯ by keeping only the values above a given threshold and discarding the rest of the values as zero.
We took 150 closing indices of 286 Global Fortune 500 (other than Indian and US) companies, 408 Fortune 500 US companies and 49 large-scale Indian companies, and divided the data into three equal parts of 50 indices each, representing pre-lockdown, lockdown and recovery phase. We then calculated the pairwise inner composition alignment for each of the series for the three mentioned periods, keeping in mind if the series are uptrending and downtrending, and obtained three [743 x 743] matrices for each period, whose values represent the strength of coupling as captured by modified IOTA. We obtain directed networks for the three periods from the coupling matrices by considering vertices X and Y (companies represented by their time series) as connected by a directed edge if the modified value of IOTA, ιX→Y, is greater than a threshold value (say θ) of the modified IOTA. For the data corresponding to pre-lockdown period, we arrive at a threshold value of 0.9 by trial and error, considering that for thresholds below this value, the obtained networks become very dense and for values above 0.9, the networks become very sparse. We further keep the threshold value as 0.9 for the other two periods in order to make comparisons possible between the three considered periods. We have presented these results with the help of Fig. 1, where we show the change in the values of arc density and the number of strongly connected components of the networks generated hence, with change in the threshold value θ. The data related to Fig. 1 is tabulated in supplementary information.
Fig. 1 Change in the value of (a) arc density and (b) number of strongly connected components with change in the threshold value θ of the modified IOTA.
2.3 Analysis of inner composition alignment networks
We study the directed networks obtained for three periods to get an insight into the effect of COVID-19 situation on the markets. We first study the local properties of these networks, to quantify how individual companies are fairing in the situation. Network centrality is a basic idea in the study of networks, which proposes that some vertices in a network are more important than the others. Network centrality can be thought of as a real valued function on the set of vertices or edges in a network, that induces a total order on the set [45]. In a directed network, vertices with higher in-degree and out-degree can be considered as central vertices [46].
We consider a vertex to be more central if it lies on a large number of shortest directed paths between other vertices in the network [47]. Let δstv denotes the fraction of directed shortest paths between vertices s and t that contain v, such that, δstv=σst(v)σst, where σst is the number of directed shortest paths between s and t. The betweenness centrality of a vertex is given by, CB(v)=Σs≠t≠vδst(v) such that, the sum is being calculated for all the pairs of distinct vertices in the networks.
Furthermore, we wish to quantify the global structural properties of inner composition alignment networks which may reveal how the market is fairing as a whole during the different periods. A network is said to show assortative mixing if vertices of higher degrees are connected to other vertices with higher degrees, and the vertices with lower degree tend to be connected to other vertices with lower degree [48]. In a network, vertices with high degree tend to be adjacent to vertices with low degrees, then such a network is said to be disassortative. In a directed network, if we consider the end vertices of an edge as source and sink vertices (such that the direction is oriented from the source vertex to the sink vertex), then we can find the assortativitive mixing for in-degree to in-degree, out-degree to out-degree, out-degree to in-degree and in-degree to out-degree cases using an index called assortativity [49], [50].
We find the community structure for the networks representing three different periods to understand which set of companies are forming closely-knit groups during this time and hence may be affecting each other more than companies outside of these groups or communities.. A community in a directed network is a partition of the set of vertices into smaller sets with relatively larger number of edges between the vertices in the sub-network [51]. An arbitrary partition of the vertex set of a graph is said to form a good community structure if the value of modularity associated with the defined communities is close to one. We use an algorithm proposed by Louvain et al. to find the community structure [52]. We further find the strongly connected components in each network. A strongly connected component in a directed network is a set of vertices, such that, a directed path exists between any two vertices of the set [53]. The findings of this study are presented and discussed in the following section.
3 Results and discussion
We obtain networks with 743 vertices for three periods under study i.e., pre-lockdown, lockdown and recovery phase, using a threshold coupling strength of 0.9. By considering a threshold value of 0.9, we choose to consider only significant coupling between time series under study and thus companies under investigation. The choice of threshold is further justified using an example as shown in Fig. 2, where we visualize three time series with mutual different values of inner composition alignment. As can be seen in the figure, the time series with higher IOTA coupling strength appears similar to each other as compared to time series with lower IOTA coupling strength.
We find that the number of directed edges or arcs in the pre-lockdown phase for the threshold coupling strength of 0.9, is equal to 77 743. The arc density (number of arcs present divided by total number of arcs possible, which is n(n−1), where n is the order of the graph) for the pre-lockdown period is 0.1410 (number of vertices being 743). For the same value of threshold, the number of arcs found in the lockdown period is 12 835, which is a very steep decrease in the number of directed edges for a network of same order i.e., 743 vertices. The arc density for lockdown period reduces to 0.0233, making the network a sparse network. For the recovery phase the number of arcs further reduces to 7899 with an arc density of 0.0143, making the connections in the network even rarer for the same choice of threshold.Fig. 2 The time series ONGC.NS and TATAMOTORS.NS seems to be trending more similar to each other with an IOTA coupling strength from ONGC to TATAMOTORS being 0.9413 as compared to IOTA coupling strength from TATAMOTORS to GNZUF being 0.7007.
This trend in the reduction of number of edges clearly shows that as a result of COVID-19 situation, companies throughout the world show a propensity to de-couple from their global counterparts and a flourishing network of coupled interactions between companies all over the globe vanished to a very sparse network of coupling during lockdown period. Thus companies during and immediately after the COVID-19 situation tend to operate independent of each other shielding any influence other companies may be inducing on them.
A clear trend is observed when we look at best-ranked centrality indices for betweenness centrality, in-degree and out-degree centrality measures over the three periods considered for this study. It is observed that all the best-ranked centrality indices for the aforementioned measures over the three periods are dominated heavily by Chinese and Japanese companies. This trend is softly diluted for the lockdown phase, where companies from other countries (mainly US) appears in the best-ranked list along with Chinese and Japanese companies. However, for both pre-lockdown and recovery phase, the best ranked vertices are dominantly Chinese or Japanese organizations irrespective of centrality measure chosen for study. Since a high value for betweenness for a vertex indicate the greater degree of control the vertex has over disseminating information across the network, it seems only natural to conclude that these Chinese and Japanese companies that are frequently ranked high as per betweenness centrality, play a key role in channeling the COVID-19 situation across the network. The set of companies ranked higher by betweenness centrality are also ranked higher by in-degree and out-degree centralities, thus further supporting our inference.
It is further observed that the first five best-ranked vertices (or companies) as per betweenness centrality in the recovery phase belong to financial sector, indicating a key role played by the financial sector in mitigating the COVID-19 crisis situation. This observation is in contrast with our study of 2008 financial crisis [26], where we found that financial institutions dropped out of network in order to shield themselves from the effect of global recession. In COVID-19 situation however, banks are playing a role in recovery and thus the situation is fundamentally different as compared to global slowdown. The ten best-ranked companies as given by betweenness centrality for recovery period is given here as Table 1. The remaining tables for betweenness centrality, in-degree and out-degree centrality measures over the three periods are shared as supplementary information.
The values of assortativity calulated for all networks over the three periods are close to zero with exception of in-degree to in-degree assortativity value and out-degree to in-degree assortativity value for the recovery phase being −0.4374 and −0.4319, indicating that in vertices with both high in-degree and high out-degree tend not to be connected to vertices with high in-degree to some extent. For other degree combinations in the different periods of study however, no such trend could be established.Table 1 Best ranked companies as per betweeness centrality for the inner composition alignment network for the recovery period.
S. No. Company Sector Company abbreviation Betweenness
1 People’s Insurance, China Financial PINXY 5.3209E＋04
2 Sompo Holdings, Japan Financial NHOLF 2.8849E＋04
3 China Vanke Co., China Financial 000002.SZ 1.6211E＋04
4 Japan Post Holdings, Japan Financial JPHLF 1.1795E＋04
5 Chubu Electric Power, Japan Utilities CHJHF 7.4806E＋03
6 China Minsheng Banking Corp, China Financial CGMBF 7.4193E＋03
7 Shangdong Weiqiao Pioneering Group, China Consumer Discretionary WQTEF 6.7725E＋03
8 JFE Holdings, Japan Materials JFEEF 6.2523E＋03
9 China Shenhua Energy Co. Ltd., China Energy CUAEF 4.9838E＋03
10 Ford Motors, USA Consumer Discretionary F 4.1345E＋03
The global structure of inner composition alignment networks over the three periods show interesting results upon observation. It is seen that for pre-lockdown phase their are 20 communities with modularity of 0.1388, the largest of which are four communities of size 207, 166, 165, 103 and 87 vertices and the rest are singleton vertices. The very low value of modularity indicate that the community is poorly formed i.e., the number of arcs between the members of the communities are similar to number of arcs outside the members of the communities. It is further observed that most Indian companies belong to the largest and second largest community in pre-lockdown period, while US companies are evenly distributed across communities.
The value of modularity increases slightly to 0.3609 for the lockdown period, forming 14 communities, the largest of which are communities of size 356, 143, 141 and 93 while the rest are singleton vertices. The Indian companies belong mostly to the largest two communities. For the recovery period, the value of modularity increases to a further of 0.4956, indicating that for recovery period the network show the best community structure out of the three periods of study. It means that the number of arcs within the members of the community are considerably more than number of arcs between vertices outside the members of the community. The community structure for recovery period consists of 58 communities, three of which are large communities with 268, 289 and 131 members, while the others are singleton vertices.
Thus we infer that during the recovery phase, factions of companies are forming that are coupled strongly within a group to which the company belongs rather than to companies outside of the group, while no such trend existed before global lockdown. It also indicates that a large number of companies have isolated (or completely uncoupled) themselves from other companies in the recovery period for COVID-19 situation, as 55 companies in this phase does not belong to any community.
The evidence for global isolation of companies in response to COVID-19 situation is further supported by the study of strongly connected components in the networks for the three periods, where it is observed that the for the pre-lockdown phase there were a total of 56 strongly connected components, of which the largest strongly connected components were of order 628 and 61, while the rest were singleton vertices i.e., belonging to trivial strongly connected component of size one. For the lockdown phase, the number of strongly connected components become 293, while the size of largest strongly connected components decrease to 407 and 43, indicating that the rest of the vertices are singleton vertices. For the recovery phase, two large strongly connected components of size 305 and 163 are observed and the rest of 275 observed strongly connected components are singleton vertices or strongly connected components of size two.
Strongly connected component in a network is a set of vertices which can be reached by any other vertex from the set through a directed path. In terms of financial networks, it means that the a company belonging to a strongly connected components can influence another company from the set and can be influenced by another company from the set in return. Thus there is a greater propensity of bidirectional financial information flow within the members of a strongly connected components. A decrease in the size of strongly connected components and increase in the number of trivial strongly connected components show that in response to COVID-19 situation, companies globally are trying to shield the influence of extra players on their organization and operation.
We further wish to assimilate the effect COVID 19 had on different economies across the globe and in the current study we have chosen to look at Indian, Asian (other than India), European (including Britain) and US economies as standalone economies and in relationship to each other as give by the induced subgraphs of the global network of aforementioned companies. It should be noted that an induced subgraph for a given economy show the vertices i.e companies in relationship to other companies from the same economy. Thus a singleton vertex in an induced subgraph would mean that the company representing that vertex is not influencing any other company in the same economy, nor getting influenced by other companies from the local economy. However, it is possible that such a company may be experiencing global influences (from companies other than from its own economy).
We find that initially in the pre-lockdown phase, the network structure of all the four economies mentioned above show presence of a giant component, which in financial terms indicate that the companies in each economy are coupled together or co-moving with each other as one large group with exception of a few companies in each market. However, a completely different picture begin to emerge during lockdown and post-lockdown phase where it is found that generally two large components emerge in each economy, showing that the market may be getting dismantled into two groups, such that companies in each group may be coupled or co-moving with companies from within the group and not with companies within the other group. That is to say all economies show a divide in companies, possibly facing different market pressures.
It is further noticed that a lot of companies in the US and European markets have isolated themselves from the local economy (as is indicated by presence of a large number of singleton vertices in these economies), thus indicating that a lot of these companies are not co-moving with either of the groups emerging in these economies and thus the nature of forces influencing them (if any) are global. These results are presented as Fig. 3, Fig. 4, Fig. 5. Thus the current study presents a comprehensive view of how COVID 19 affected different economies and how the financial impacts of global lock-down were mediated and eventually begin to get mitigated across the globe by organizations raking high on different centrality indices. The role of network based study, especially in light of inner composition alignment capturing coupling between different organizations is thus acknowledged as a viable method in the study of complex adaptive systems.
Fig. 3 The four different economies in the global network of companies for the pre-lockdown period are shown here as (a) India in blue, (b) Asia in red, (c) Europe in yellow and (d) US in green.
Fig. 4 The four different economies in the global network of companies for the lockdown period are shown here as (a) India in blue, (b) Asia in red, (c) Europe in yellow and (d) US in green.
Fig. 5 The four different economies in the global network of companies for the post-lockdown period are shown here as (a) India in blue, (b) Asia in red, (c) Europe in yellow and (d) US in green.
4 Conclusion
We analyzed financial time series data (closing indices of stock price) of 743 companies from across the globe for a period ranging roughly between 1 January, 2020 to 15 August, 2020 to study the impact of lockdown imposed in response to COVID-19 situation on markets. We used inner composition alignment, a method developed to find coupling between short time series data, to come up with a network structure of the aforementioned 743 companies over three time periods i.e., pre-lockdown, lockdown and recovery phase.
We conclude that companies all across the globe have shown a tendency to decouple themselves from other organizations in response to COVID-19 situation. We find evidence in form of arc density, community structure in networks studied and strongly connected components present in networks in different periods to support our hypothesis. We further observe that Chinese and Japanese companies are ranked higher by different centrality indices over the studied time periods and thus some of these companies play a key role in channeling financial information flow before, during and after the lockdowns imposed due to COVID-19 situation. We further conclude that financial sector play a key role in mitigating the COVID-19 situation, as indicated by high values of betweenness centrality attained by the sector during the recovery phase.
We further conclude that as a result of COVID-19 situation, companies across different economies seems to be getting divided onto two major co-moving or coupled groups. We also notice that a lot of companies belonging to the US and European market seems to be shielding local influences or couplings.
We end by emphasizing that complete effects of COVID-19 situation on entire global economy may only be clear with adequate passage of time, which could be as long as a decade if not more, yet our study characterizes an initial indicative picture of crisis situation as excellently captured by inner composition alignment networks, which as a method has shown great promise to study short time series.
CRediT authorship contribution statement
Shashankaditya Upadhyay: Conceived the idea upon discussion and devised the methodology, Collected and curated the data, Performed the analysis and wrote the manuscript. Indranil Mukherjee: Conceived the idea upon discussion and devised the methodology, Collected and curated the data, Revision. Prasanta K. Panigrahi: Conceived the idea upon discussion and devised the methodology, Revision.
Uncited References
[54]
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary data
The following is the Supplementary material related to this article. MMC S1
.
Data availability
Data will be made available on request.
Acknowledgment
We would like to thank Ms Chaitali Chanda for her help with collecting the data. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Appendix A Supplementary material related to this article can be found online at https://doi.org/10.1016/j.physa.2022.128341.
==== Refs
References
1 Tesfatsion L. Economic agents and markets as emergent phenomena Proc. Natl. Acad. Sci. 99 Suppl 3 2002 7191 7192 12011395
2 Markose S.M. Computability and evolutionary complexity: Markets as complex adaptive systems (CAS) Econ. J. 115 504 2005 F159 F192
3 Levin S.A. Lo A.W. Opinion: A new approach to financial regulation Proc. Natl. Acad. Sci. 112 41 2015 12543 12544 26464480
4 Battiston S. Caldarelli G. Georg C.P. May R. Stiglitz J. Complex derivatives Nat. Phys. 9 3 2013 123
5 Lux T. Westerhoff F. Economics crisis Nat. Phys. 5 1 2009 2
6 Rua A. Nunes L.C. International comovement of stock market returns: A wavelet analysis J. Empir. Financ. 16 4 2009 632 639
7 Ghosh S. Manimaran P. Panigrahi P.K. Characterizing multi-scale self-similar behavior and non-statistical properties of fluctuations in financial time series Physica A 390 23–24 2011 4304 4316
8 Kulkarni V. Deo N. Correlation and volatility in an Indian stock market: A random matrix approach Eur. Phys. J. B 60 1 2007 101 109
9 Neeraj Panigrahi P.K. Causality and correlations between BSE and NYSE indexes: A Janus faced relationshi Physica A 481 2017 284 313
10 Pharasi H.K. Sharma K. Chatterjee R. Chakraborti A. Leyvraz F. Seligman T.H. Identifying long-term precursors of financial market crashes using correlation patterns New J. Phys. 20 10 2018 103041
11 Chakraborti A. Toke I.M. Patriarca M. Abergel F. Econophysics review: II. Agent-based models Quant. Finance 11 7 2011 1013 1041
12 Mantegna R.N. Stanley H.E. Introduction to Econophysics: Correlations and Complexity in Finance 1999 Cambridge university press.
13 Chakrabarti B.K. Chakraborti A. Chatterjee A. Econophysics and Sociophysics: Trends and Perspectives 2006 John Wiley & Sons
14 Gabaix X. Gopikrishnan P. Plerou V. Stanley H.E. A theory of power-law distributions in financial market fluctuations Nature 423 6937 2003 267 12748636
15 Sinha S. Chatterjee A. Chakraborti A. Chakrabarti B.K. Econophysics: An Introduction 2010 John Wiley & Sons
16 Mukherjee I. Chatterjee S. Giri A. Barat P. Understanding the pattern of the BSE sensex Physica A 482 2017 262 275
17 Goodell J.W. COVID-19 and finance: Agendas for future research Finance Res. Lett. 2020 101512
18 Zhang D. Hu M. Ji Q. Financial markets under the global pandemic of COVID-19 Finance Res. Lett. 2020 101528
19 Bonaccorsi G. Pierri F. Cinelli M. Flori A. Galeazzi A. Porcelli F. Schmidt A.L. Valensise C.M. Scala A. Quattrociocchi W. Pammolli F. Economic and social consequences of human mobility restrictions under COVID-19 Proc. Natl. Acad. Sci. 117 27 2020 15530 15535 32554604
20 Catanzaro M. Buchanan M. Network opportunity Nat. Phys. 9 3 2013 121 32226462
21 Atalay E. Hortacsu A. Roberts J. Syverson C. Network structure of production Proc. Natl. Acad. Sci. 108 13 2011 5199 5202 21402924
22 Battiston S. Caldarelli G. May R.M. Roukny T. Stiglitz J.E. The price of complexity in financial networks Proc. Natl. Acad. Sci. 113 36 2016 10031 10036 27555583
23 Onnela J.P. Chakraborti A. Kaski K. Kertesz J. Kanto A. Dynamics of market correlations: Taxonomy and portfolio analysis Phys. Rev. E 68 5 2003 056110
24 Sharma K. Chakrabarti A.S. Chakraborti A. Multi-layered network structure: Relationship between financial and macroeconomic dynamics New Perspectives and Challenges in Econophysics and Sociophysics 2019 Springer Cham 117 131
25 Kuyyamudi C. Chakrabarti A.S. Sinha S. Emergence of frustration signals systemic risk Phys. Rev. E 99 5 2019 052306
26 Upadhyay S. Banerjee A. Panigrahi P.K. Causal evolution of global crisis in financial networks Physica A 554 2020 124690
27 Upadhyay S. Mondal T. Pathak P.A. Roy A. Agrawal G. Bhattacharya S. A network theoretic study of potential invasion and spread of Lantana camara in a part of Rajaji Tiger Reserve, India Ecol. Complex. 40 2019 100793
28 Hempel S. Koseska A. Kurths J. Nikoloski Z. Inner composition alignment for inferring directed networks from short time series Phys. Rev. Lett. 107 5 2011 054101
29 Sinha S. Pan R.K. Uncovering the internal structure of the Indian financial market: Large cross-correlation behavior in the NSE Econophysics of Markets and Business Networks 2007 Springer Milano 3 19
30 Plerou V. Gopikrishnan P. Rosenow B. Amaral L.A. Guhr T. Stanley H.E. Random matrix approach to cross correlations in financial data Phys. Rev. E 65 6 2002 066126
31 Tao T. Vu V. Random covariance matrices: Universality of local statistics of eigenvalues Ann. Probab. 40 3 2012 1285 1315
32 Yu Y.K. Zhang Y.C. On the anti-Wishart distribution Physica A 312 1–2 2002 1 22
33 Vivo P. Majumdar S.N. Bohigas O. Large deviations of the maximum eigenvalue in wishart random matrices J. Phys. A 40 16 2007 4317
34 Sengupta A.M. Mitra P.P. Distributions of singular values for some random matrices Phys. Rev. E 60 3 1999 3389
35 Granger C.W. Investigating causal relations by econometric models and cross-spectral methods Econometrica 42 1969 4 438
36 Eichler M. Granger causality and path diagrams for multivariate time series J. Econometrics 137 2 2007 334 353
37 Wen X. Rangarajan G. Ding M. Multivariate granger causality: An estimation framework based on factorization of the spectral density matrix Phil. Trans. R. Soc. A 371 1997 2013 20110610
38 Vicente R. Wibral M. Lindner M. Pipa G. Transfer entropy—A model-free measure of effective connectivity for the neurosciences J. Comput. Neurosci. 30 1 2011 45 67 20706781
39 Adler R. Coppersmith D. Hassner M. Algorithms for sliding block codes-an application of symbolic dynamics to information theory IEEE Trans. Inform. Theory 29 1 1983 5 22
40 Schroeder M. Fractals, Chaos, Power Laws: Minutes from an Infinite Paradise 2009 Courier Corporation
41 Lake D.E. Moorman J.R. Accurate estimation of entropy in very short physiological time series: The problem of atrial fibrillation detection in implanted ventricular devices Am. J. Physiol.-Heart Circul. Physiol. 300 1 2011 H319 H325
42 Ma H. Aihara K. Chen L. Detecting causality from nonlinear dynamics with short-term time series Sci. Rep. 4 1 2014 1 10
43 Aste T. Matteo T.Di. Sparse causality network retrieval from short time series Complexity 2017 2017
44 Shi Z. Wang S. Wang J. Dai J. Hou F. Li J. Analyzing occupational stress based on inner composition alignment algorithm 2016 9th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) 2016 IEEE 1462 1466
45 Upadhyay S. Roy A. Ramprakash M. Idiculla J. Kumar A.S. Bhattacharya S. A network theoretic study of ecological connectivity in Western Himalayas Ecol. Model. 359 2017 246 257
46 Newman M.E. The structure and function of complex networks SIAM Rev. 45 2 2003 167 256
47 Freeman L.C. A set of measures of centrality based on betweenness Sociometry 3 1977 5 41
48 Newman M.E. Assortative mixing in networks Phys. Rev. Lett. 89 20 2002 208701
49 Newman M.E. Mixing patterns in networks Phys. Rev. E 67 2 2003 026126
50 M. Piraveenan, K.S.K. Chung, S. Uddin, Assortativity of links in directed networks, in: Proceedings of Fundamentals of Computer Science, 2012.
51 Pantula S.G. Gonzalez-Farias G. Fuller W.A. A comparison of unit-root test criteria J. Bus. Econom. Statist. 12 4 1994 449 459
52 Blondel V.D. Guillaume J.L. Lambiotte R. Lefebvre E. Fast unfolding of communities in large networks J. Stat. Mech. Theory Exp. 2008 10 2008 P10008
53 West D.B. Introduction to Graph Theory, Vol. 2 1996 Prentice hall Upper Saddle River, NJ
54 Gençağa D. Transfer entropy Entropy 20 4 2018 288 33265379
| 36465189 | PMC9707026 | NO-CC CODE | 2022-12-09 23:15:03 | no | Physica A. 2023 Jan 1; 609:128341 | utf-8 | Physica A | 2,022 | 10.1016/j.physa.2022.128341 | oa_other |
==== Front
J Neurol Sci
J Neurol Sci
Journal of the Neurological Sciences
0022-510X
1878-5883
Elsevier B.V.
S0022-510X(22)00373-2
10.1016/j.jns.2022.120511
120511
Article
Monitoring cognitive and psychological alterations in COVID-19 patients: A longitudinal neuropsychological study
Diana Lorenzo ab
Regazzoni Rossana a
Sozzi Matteo a
Piconi Stefania c
Borghesi Luca c
Lazzaroni Elisa d
Basilico Paola a
Aliprandi Angelo a
Bolognini Nadia be
Bonardi Daniela R. f
Colombo Daniele f
Salmaggi Andrea a⁎
a Neurology Unit, Ospedale “A. Manzoni”, Lecco, Italy
b Neuropsychology Laboratory, IRCCS Istituto Auxologico Italiano, Milan, Italy
c Infectious Diseases Unit, Ospedale “A. Manzoni”, Lecco, Italy
d Department of Mental Health, Ospedale “A. Manzoni”, Lecco, Italy
e Department of Psychology and NeuroMI, University of Milano-Bicocca, Milan, Italy
f Respiratory Unit, IRCCS INRCA (Italian National Research Centre On Aging), Casatenovo, Italy
⁎ Corresponding author.
29 11 2022
15 1 2023
29 11 2022
444 120511120511
21 9 2022
19 11 2022
23 11 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Elsevier B.V.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
SARS-COV-2 infection has been associated to long-lasting neuropsychiatric sequelae, including cognitive deficits, that persist after one year. However, longitudinal monitoring has been scarcely performed. Here, in a sample of COVID-19 patients, we monitor cognitive, psychological and quality of life-related profiles up to 22 months from resolution of respiratory disease.
Methods
Out of 657 COVID-19 patients screened at Manzoni Hospital (Lecco, Italy), 22 underwent neuropsychological testing because of subjective cognitive disturbances at 6 months, 16 months, and 22 months. Tests of memory, attention, and executive functions were administered, along with questionnaires for depressive and Post-traumatic stress disorder (PTSD) symptoms, psychological well-being and quality of life. Cross-sectional descriptives, correlational, as well as longitudinal analyses considering COVID19-severity were carried out. A preliminary comparison with a sample of obstructive sleep apneas patients was also performed.
Results
Around 50% of COVID-19 patients presented with cognitive deficits at t0. The most affected domain was verbal memory. Pathological scores diminished over time, but a high rate of borderline scores was still observable. Longitudinal analyses highlighted improvements in verbal and non-verbal long term memory, as well as attention, and executive functioning. Depression and PTSD-related symptoms were present in 30% of patients. The latter decreased over time and were associated to attentional-executive performance.
Conclusions
Cognitive dysfunctions in COVID-19 patients may extend over 1 year, yet showing a significant recovery in several cases. Cognitive alterations are accompanied by a significant psychological distress. Many patients displaying borderline scores, especially those at higher risk of dementia, deserve clinical monitoring.
Keywords
COVID-19
Long-COVID
Cognitive functions
Neuropsychology
Longitudinal assessment
==== Body
pmc1 Introduction
Severe Acute Respiratory Syndrome due to Coronavirus 2 (SARS-COV-2) has been shown to impact multiple organs beyond the respiratory system, including the brain [1]. In fact, along the first studies highlighting neurological [[2], [3], [4]], as well as psychiatric symptoms that could persist up to six months after the respiratory syndrome [5,6], it became more and more evident that also fatigue and cognitive impairments (often described as “brain fog”) could persist well beyond the acute infection, leading to the so-called “Long COVID” [[7], [8]]. Such sequelae could pose psychological, occupational, and social problems, also affecting the caregivers [9]. Moreover, concerns were raised as COVID-19 patients who recovered from respiratory disease could be more likely to develop cognitive decline and Alzheimer's disease [10]. Thus, it is not surprising that a growing number of studies have been characterizing COVID-19-related cognitive alterations and their neural correlates, such as [11,12].
The pathophysiological mechanisms underlying de novo onset of cognitive decline, or the worsening of pre-existing cognitive impairments, in patients with COVID-19 are likely multiple, ranging from the noxious effects of neural cells hypo‑oxygenation to toxic effects by cytokines released in the context of widespread inflammation [13].
Overall, despite methodological differences - e.g., telephone vs. in-person assessment, screening vs. second-level assessment [14,15] – converging evidence shows that cognitive deficits in COVID-19 patients can be appreciated from the early sub-acute stage [16] and few days after hospital discharge [[17], [18], [19]], up to 5–7 months [14,[20], [21], [22]] and even 1 year after hospital discharge [23,24]. A recent review by Crivelli and colleagues [14] on 27 studies including 2049 individuals, highlights a broad spectrum of cognitive impairments mostly involving executive functions, attention and long-term memory (i.e., the ability to learn new information and/or recall it at a later time).
However, to date, longitudinal monitoring of cognitive performance has been scarcely performed [16,22,25,26], with the vast majority of studies presenting only cross-sectional evidence [14]. To our knowledge, only three studies explored longitudinal, long-term, neuropsychological changes up to one year follow-up [23,24,26]. For instance, Ferrucci and colleagues [23] assessed a sample of COVID-19 patients after five months and one year from hospital discharge, with an extended neuropsychological battery. The authors found that around 60% of patients had a deficit in at least one cognitive function at five months, mostly in speed of processing (41%), long-term verbal (around 20%) and visuospatial memory (18%). Interestingly, at one year follow up, around 50% of the sample still showed a pathological performance in at least one cognitive domain, mainly in the same cognitive functions of the first timepoint. Similar results were found by Miskowiak et al. [24] and Mendez et al. [26]. Nonetheless, Ferrucci and colleagues [23] observed that speed of processing, attention, and verbal memory significantly improved after one year. Moreover, worse oxygenation in the acute phase was associated to worse verbal memory performance five months from discharge.
In the present study, we aim to enrich longitudinal evidence of cognitive dysfunctions by investigating neuropsychological changes in a sample of COVID-19 patients tested about 6, 16, and up to 22 months after the recovery of acute respiratory symptoms. Importantly, we complement the characterization of impaired cognitive domains by also considering sub-clinical (i.e., borderline), cognitive performance, almost neglected so far. We further assessed the impact of disease severity in the acute phase (i.e., the need for oxygen therapy) on the evolution of neuropsychological symptoms, as previous works suggested a pivotal role of respiratory impairments in explaining the pattern and the severity of cognitive deficits [20,23]. Moreover, as suggested by [21], the association between cognitive performance and mood disorders was also tested, along with a characterization of psychological well-being, and quality of life.
Finally, to assess the role of hypoxia alone as opposed to other mechanisms active in neurocovid patients, we compare cognitive and emotional profiles of COVID-19 patients to those of a sample of patients suffering from a common respiratory disease, i.e., Obstructive Sleep Apnea Syndrome (OSAS).
2 Materials and methods
2.1 Participants
From April 2020 to March 2021, 657 patients infected by SARS-COV-2 who received care for acute infection at the Manzoni Hospital (Lecco, Italy) were screened by a team of infectious disease specialists, in order to monitor long-term symptoms of SARS-CoV2 infection, with a particular focus on persistent respiratory, neurological and psychological sequelae (see Fig. 1 ). Specifically, neurological and neuropsychiatric changes (i.e., memory impairments, mood changes, neurological symptoms) were evaluated by means a check-list provided by the Neurology Unit (for details, see the Supplementary Material). In light of new-onset neurological symptoms, 74 patients (29 F; mean age = 60 ± 15.87, range 19–94) were referred to neurological examination; of those, 34 reported subjective cognitive alterations following COVID-19 (i.e., 5.5% of the whole population study), and 21 of them agreed to undergo a neuropsychological assessment around 6 months (timepoint t0) after resolution of acute respiratory problems. A sub-group of patients was further evaluated at two supplementary timepoints i.e., on average, at 16 months (t1; N = 19) and 22 months (t2; N = 16). Demographic and clinical features of the recruited patients are reported in Table 1 .Fig. 1 Flow chart.
Fig. 1
Table 1 Demographic and clinical variables of COVID-19 and OSAS patients.
Table 1 t0 t1 t2 OSAS
N 21 19 16 8
Sex 6 F, 15 M 5 F, 14 M 4 F, 12 M 2 F, 6 M
Age, years 57 ± 15 [19–82] 57 ± 15; [19–82] 59 ± 15; [19–82] 61.38 ± 6.78; [54–74]
Education, years 11 ± 3; [[5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18]] 12 ± 3; [[5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17]] 11 ± 3; [[5], [6], [7], [8], [9], [10], [11], [12], [13]] 9.5 ± 3; [[5], [6], [7], [8], [9], [10], [11], [12], [13]]
Oxygen therapy 11/21 10/19 8/16
Hospital Stay, daysa 23.19 ± 21; [0–86] 24.11 ± 21.41; [0–86] 21.5 ± 20.1; [0–86]
Hyposmia/Hypogeusia 6/21 5/19 4/16
Note: for age, education and hospital duration, mean, standard deviation, and range are reported. OSAS: obstructive sleep apnea syndrome.
Moreover, with the exploratory aim of comparing the cognitive-emotional profile of COVID-19 patients to that of control individuals suffering from respiratory problems, we recruited 8 additional individuals diagnosed with OSAS at the Respiratory Unit of the IRCCS INRCA (Italian National Research Centre On Aging), Mandich Hospital (Merate, Italy). None of them reported SARS-COV-2 infection. They were all tested before receiving night oxygen support by means continuous positive airway pressure (CPAP). They were all diagnosed with severe OSAS, i.e., Apnea Hypopnea Index (AHI) > 30 (range: 30–93). The average nocturnal saturation was 88% (range: 84%–94%); two patients did not present nocturnal desaturation. The average night time under 90% of saturation (i.e., CT90) was 31% (range: 5%–85%).
The study was approved by the local Ethics Committee (Protocol N°: 3477) and was conducted in accordance with the ethical standards of the Declaration of Helsinki. All participants provided their written informed consent.
2.2 Neuropsychological evaluation
The neuropsychological evaluation took place at around 6 months (t0; mean number of days: 185.29 ± 114.15 SD), 16 months (t1 mean number of days: 473.26 ± 97.67 SD) and 22 months (t2; mean number of days: 664 ± 100 SD) after resolution of respiratory disease.
2.2.1 Cognitive assessment
Cognitive assessment at t0 was carried out by two neuropsychologists and was tailored to the patients' age and education, at the discretion of the psychologist. It included several psychometric tests normed in the Italian population: the Mini Mental State Evaluation (MMSE) [27,28]; Attentional Matrices [29]; Trail Making Test (TMT) [30]; forward and backward digit span [31], Rey Verbal Learning Test (RVLT) [32]; Babcock Story Recall Test (BSRT) [29,33]; copy and recall of Rey-Osterrieth's complex fig. [34]; Frontal Assessment Battery (FAB) [[35], [36]]; verbal fluency by letter (i.e., phonemic fluency) and category (semantic fluency) [[37], [38]]; Weigl's Sorting Task [39]; Raven's Matrices [29,40].
At t1 and t2, in addition to the tests administered at t0, we sought to provide a more fine-grained characterization of the executive and attentional profile that may be part of long-COVID, by including: a computerized reaction time test (Open-source Open-access Reaction Time test, OORT) [41], both in its simple modality (i.e., reacting to a target with no distractors) and in a go-no-go modality; the oral version of the Symbol Digit Modalities Test (SDMT) [42]; the Clock Drawing Test (CDT) [43]; the Modified Five Point Test for non-verbal fluency [[44], [45]], and the alternate verbal fluency (Costa et al., 2013). In order to calculate the fluency-based Composite Shifting Index, Costa's version [38] of phonemic and semantic fluency tasks were administered. Moreover, at t2 the copy and recall of Taylor's fig. [46] replaced Rey-Osterrieth's to avoid learning effects. Likewise, parallel lists of words were used on the RVLT.
Raw scores were corrected for age and level of education, and whenever available, they were converted into equivalent scores (ES) [47,48], on an ordinal scale ranging from 0 to 4. ES from 2 to 4 are indicative of a non-pathological performance, whereas ES 1 and 0 can be regarded, respectively, as “borderline performance” (i.e., not pathological, but reflecting a significant decrease) and a pathological performance. For those tests which were not normed according to the ES method (i.e., the SDMT [42] and the CDT [43]), cut-off values were considered. For the statistical analyses (see below), corrected scores were used, whenever possible. ES were considered when different tests were used to test the same cognitive function in different patients. For example, logical reasoning was always tested with Raven's matrices, however, at the discretion of the neuropsychologist at t0, the more demanding Raven Standard Progressive Matrices [40] could be administered to some younger or more educated patients, whereas Raven Progressive Colored Matrices [29], relatively easier and shorter, to older participants. The use of standardized ES overcomes the problems of different score ranges.
2.2.2 Psycho-affective and quality of life questionnaires
Finally, at t1 and t2, we investigated participants' psychological state, as well as quality of life by means of self-administered questionnaires. The Beck Depression Inventory (BDI) [49] was administered to assess the presence and severity of depressive symptoms. BDI scores (range = 0–33) were classified as in: 0–9 = no depressive symptoms, 10–19 = mild symptoms, 20–29 = moderate symptoms; 30–33 = severe symptoms [50]. Moreover, participants completed the Impact of Event Scale-Revised (IES-R) [51] to assess the presence of post-traumatic stress disorders (PTSD) symptoms (range = 0–88); a score above 32 [52] indicated clinically relevant symptoms. We also administered the Psychological Well-Being Index (PGWBI) [53] considering both the total score (range = 0–110; higher scores, higher quality of life) and the six subscales Anxiety, Depression, Self-control, Health, Vitality, Positivity and Well-Being. Individual z-scores of the total score were calculated based on means and standard deviation of different age and sex groups [54]. Scores below −1.65 (i.e., below the theoretical 5th percentile) were regarded as a significant decrease of psychological well-being. Finally, all participants completed the short version of the World Health Organization Quality of Life Questionnaire (WHOQOL-BREF) [55]. We calculated both the total score (range = 26–130; higher score, higher quality of life) and the mean scores of the subscales (i.e., physical, psychological, social, and environmental) which were converted into a 0–100 scale.
OSAS patients were tested only once and were administered the same protocol of COVID-19 patients at t1, except for the IES-R.
The evaluation lasted about 105 min and was split into two daily sessions, whenever participant's fatigability made it necessary. Participants were allowed to take a short break among tests.
2.3 Analyses
All analyses were carried out with jamovi 1.6.23 [56]. Alpha was set 0.05.
For each timepoint, we calculated the percentage of patients showing pathological (i.e., ES score = 0 or scores below the cut-off) and borderline scores (i.e., ES = 1) at neuropsychological tests, as well as the psychological and quality-of-life-related profiles emerging from the questionnaires. Correlations between cognitive and psycho-affective measures at t1 and t2 were calculated by means of Spearman's rank correlation because several cognitive measures were not normally distributed.
Moreover, we tested longitudinal changes in cognitive and psycho-affective profiles, considering the need of oxygen support in the acute stage of the disease. In more detail, scores of neuropsychological tests (i.e., corrected scores or ES) and questionnaires were taken as dependent variables of Linear Mixed Models (LMMs), with timepoint (i.e., t0, t1, and t2) and oxygen therapy (i.e., O+, O-) as factors. The choice of models with no more than two predictors was made in light of the small size of our sample. Random intercepts were allowed for subjects, who were set as cluster variable. Degrees of freedom and p-values were calculated with Satterthwaite's method. Post-hoc comparisons were corrected with Bonferroni. In case the assumption of normality of residuals was violated (i.e., visual inspection of Q-Q plots and significant Kolmogorov-Smirnov test), longitudinal comparisons were carried out by means of simplified models, i.e., Friedman's Analysis of Variance (ANOVA) with Durbin-Conover post-hoc comparisons (for measures tested at three timepoints) or Wilcoxon's rank test (for measures tested at t1 and t2).
Finally, COVID-19 patients' cognitive and emotional/quality of life scores at each timepoint were compared to those of OSAS patients by means of non-parametric Mann-Whitney tests, as deemed more appropriate for small sample sizes.
3 Results
3.1 Cross-sectional analyses for each timepoint
At t0, three patients reported possible cognitive decline even before COVID infection and all patients reported subjective memory complaints and reduced attentional span/impaired concentration, as compared to pre-COVID. Twelve out of 21 patients (52%) showed deficits (ES = 0 or score below the cut-off) in at least one cognitive domain: the most impaired domain was verbal memory. Specifically, long-term memory (i.e., RAVLT delayed recall) was compromised in 5 patients (24%), 4 (19%) patients showed deficits in visuo-constructive abilities (i.e., copy of the Rey-Osterrieth complex figure), followed by immediate recall – the learning phase of the RAVLT (3 patients, 14%), verbal short-term memory (3 patients), verbal working memory (2 patients, 9%), and recall of a short story (2 patients). Five patients showed cognitive impairments in multiple domains. When borderline performances were also considered (i.e., ES = 1), the number of patients showing cognitive alterations raised to 14/21 patients (67%). Individual trends are reported in Table 2 .Table 2 Demographics and individual cognitive performance at different timepoints.
Table 2Note: Black: impairment in at least one cognitive function, i.e., equivalent score (ES) = 0 or score below the cut-off; dark grey: ES = 1 in at least one cognitive measure; light grey: ES > 1 or score above the cut-off in all cognitive measures. CPAP: continuous positive airway pressure; OTI: orotracheal intubation.
At t1, in terms of subjective memory impairment, 6/19 (32%) patients reported a subjective recovery, whereas 6/19 experienced an improvement although not comparable to pre-COVID levels, and 7/19 (37%) referred no improvements. In terms of subjective attentional problems and cognitive fatigability, 6/19 reported no more problems, 3/19 referred some improvements, and 10/19 still suffered from reduced attentional abilities. By repeating the tests administered at t0, 3/19 patients (16%) showed pathological performance in at least one domain; however, when considering borderline performances, the number raised to 10/19 (53%). Notably, the majority of tests with borderline performance were memory tests, i.e., short-term memory (measured with the digit span) and verbal learning (measured with the RAVLT). Whereas some patients improved from pathological performance at t0, others maintained a borderline performance. It is important to note that two patients with cognitive impairments at t0 dropped out, thus leading to a possible underestimation of the actual rate. Moreover, at t1, supplementary tests were administered for a deeper characterization of attentional and executive functions. The addition of these tests highlighted a pathological slowdown of reaction times and speed of processing in two additional patients (now 5/19;26% of patients with impaired performance), as well as borderline performances in other patients, mostly in simple and go-no-go reaction times (now 13/19; 68% of patients with cognitive alterations). See also Table S1, for individual data.
Depressive symptoms (i.e., BDI scores >9) were present in 6/19 (32%) of COVID-19 patients, namely, 3/19 (16%) with mild symptoms, 2/19 (11%) with moderate symptoms, and 1/19 (5%) with severe symptoms. Moreover, 7/19 (37%) showed clinically relevant PTSD-related symptoms (IES-R > 32). The analyses of the PGWBI, revealed that all participants' psychological well-being was below the average (range: from −2.9 to −0.35), with 9/19 (47%) participants significantly below the cut-off.
Correlational analyses (see Table S2 in the Supplementary Material) showed that higher scores on the IES-R (i.e., stronger PTSD-related symptoms) were associated with a worse performance on the attentional matrices (rs = −0.555; p = 0.014) and the Weigl's test (rs = −0.627; p = 0.005). Moreover, higher psychological well-being was positively correlated to verbal learning (RAVLT immediate recall; rs = 0.493; p = 0.034), and a perceived better quality of life (WHOQOL-BREF) was correlated with better executive functions, such as logical reasoning (Raven's matrices; rs = 0.515; p = 0.024) and abstraction on the Weigl's test (rs = 0.647; p = 0.004).
At t2, all patients reported a stable pattern of subjective cognitive functioning, compared to t1. Test scores indicated that only 2/16 (12,5%) showed pathological performance in at least one cognitive domain (patient 1: delayed recall of the RAVLT; patient 2: TMT-A and Weigl sorting task). Importantly, these patients reported subjective cognitive decline even before COVID-19. However, when considering borderline performances, the number of patients with a reduced cognitive performance was 5/16 (31%), when considering baseline tests, and 7/16 (43%) when considering additional executive and attentional measures. Overall, compared to t0 and t1, all percentages decreased. The pattern of tests with borderline performance varied among patients.
As for the psycho-affective variables, depressive symptoms were present in 4/16 (25%) of COVID-19 patients - all with mild symptoms – and 3/16 (18%) showed clinically relevant PTSD-related symptoms. In terms of psychological well-being, 10/16 (62%) patients scored below the cut-off.
Correlations with cognitive tests showed a persisting negative association, although weaker than at t1, of IES-R with the Weigl's test (rs = 0.521; p = 0.039) and a positive association with execution times of the TMT-A (rs = −0.654; p = 0.008), i.e., those with higher levels of traumatic stress symptoms, were slower in the attentional task.
3.2 Longitudinal analyses
Linear Mixed Models showed different effects of timepoint and oxygen therapy for different attentional measures (see Table 3 and Table S4 in the Supplementary Material).Table 3 Longitudinal change in cognitive and psycho-affective measures.
Table 3Measure Timepoints t0 t1 t2 pa Post-hoc
MMSE t0 - t1 - t2 29.1 ± 1.78 28.6 ± 1.68 28.8 ± 1.74 0.175
Att Matr t0 - t1 - t2 48.2 ± 7.84 51.5 ± 4.54 51.8 ± 5.81 0.120
TMT-A t0 - t1 - t2 38.3 ± 22.9 26.8 ± 13.4 22.8 ± 22.0 <0.001 t2 < t1 < t0
TMT-B t0 - t1 - t2 59.9 ± 45.9 45.1 ± 37.1 41.6 ± 34.7 0.012 t2 = t1 < t0
TMT-BA t0 - t1 - t2 28.4 ± 35.3 18.3 ± 29.4 23.9 ± 27.1 0.060
Span Forward t0 - t1 - t2 5.48 ± 1.08 5.82 ± 0.92 5.85 ± 0.95 0.341
Span Backward t0 - t1 - t2 4.13 ± 1.24 4.70 ± 1.02 4.40 ± 0.81 0.118
RAVLT Immediate t0 - t1 - t2 39.2 ± 8.62 42.3 ± 6.99 47.7 ± 7.26 0.002 t2 > t0
RAVLT Delayed t0 - t1 - t2 7.34 ± 3.51 8.56 ± 2.23 9.00 ± 2.92 0.144
BSRT (ES) t0 - t1 - t2 2.44 ± 1.38 2.79 ± 1.23 2.94 ± 0.99 0.196
Figure Recall t0 - t1 - t2 14.9 ± 6.55 20.2 ± 7.37 20.2 ± 5.11 0.005 t2 = t1 > t0
Figure Copy t0 - t1 - t2 32.3 ± 3.55 32.2 ± 2.09 33.2 ± 1.84 0.157
Flu Phon (ES) t0 - t1 - t2 3.14 ± 1.28 3.74 ± 0.45 3.56 ± 0.89 0.483b
Flu Phon Costa (CS) t1 - t2 – 41.9 ± 9.11 40.1 ± 10.1 0.429
Flu Sem (ES) t0 - t1 - t2 3.33 ± 1.28 4.00 ± 0 4.00 ± 0 0.050b
Flu Sem Costa (CS) t1 - t2 – 52.8 ± 8.90 52.5 ± 5.74 0.865
Flu Alternate t1 - t2 – 39.9 ± 11.2 43.6 ± 8.79 0.004 t2 > t1
Flu CSI t1 - t2 – 0.82 ± 0.17 0.91 ± 0.17 0.004 t2 > t1
FAB t0 - t1 - t2 16.1 ± 2.21 16.8 ± 1.78 16.6 ± 1.62 0.588
CDT t1 - t2 – 9.42 ± 1.44 9.34 ± 1.57 0.665c
Weigl Sorting Task t1 - t2 12.9 ± 2.19 13.5 ± 2.61 13.5 ± 2.73 0.405
OORT Simple t1 - t2 – 185 ± 78.3 168 ± 47.7 0.415
OORT Go-no-go t1 - t2 – 364 ± 98.7 325 ± 74.2 0.015 t2 < t1
SDMT t1 - t2 – 51.1 ± 11.8 51.9 ± 10.3 0.785
M5P - UDs t1 - t2 – 34.9 ± 10.1 38.8 ± 11.6 0.022 t2 > t1
Raven Matrices (ES) t0 - t1 - t2 3.24 ± 0.94 3.84 ± 0.38 3.75 ± 0.58 0.009 t1 > t2
BDI t1 - t2 – 10.1 ± 8.03 6.56 ± 5.42 0.089
IES-R t1 - t2 – 28.6 ± 25.9 15.9 ± 17.5 0.028c t2 < t1
PGWBI - tot - z t1 - t2 – −1.62 ± 0.59 −1.70 ± 0.55 0.636
PGWBI - Anxiety t1 - t2 – 10.0 ± 3.71 8.63 ± 2.55 0.193
PGWBI - Depression t1 - t2 – 5.32 ± 2.47 4.94 ± 0.93 0.751c
PGWBI - PW t1 - t2 – 9.53 ± 2.27 10.8 ± 3.00 0.186
PGWBI – Self-control t1 - t2 – 9.21 ± 1.18 9.69 ± 0.79 0.174
PGWBI - Health t1 - t2 – 7.63 ± 2.24 7.00 ± 1.93 0.282
PGWBI - Vitality t1 - t2 – 11.6 ± 1.95 12.4 ± 1.55 0.065
QoL - tot t1 - t2 – 92.5 ± 12.7 92.5 ± 13.8 0.939
QoL - Physical t1 - t2 – 60.3 ± 16.8 69.2 ± 13.0 0.051
QoL - Social t1 - t2 – 60.2 ± 16.3 63.5 ± 12.7 0.524
QoL - Psychological t1 - t2 – 74.1 ± 12.4 71.9 ± 14.9 0.025 t2 < t1
Note: athe p values refer to the main effect of “timepoint” in the Linear Mixed Models with timepoint and oxygenation as factors. For detailed results of the models, see Table S4 in the Supplementary Material. b Friedmann Analysis of Variance. c Wilcoxon Signed Rank Test.
MMSE: Mini-Mental State Examination; Att Mat: Attentional Matrices; TMT: Trail Making Test; RAVLT: Rey Auditory Verbal Learning Task; BSRT: Babcock Story Recall Test; Flu: fluency; Phon: phonological, by letter; Sem: semantic, by category; Costa: phonemic and semantic fluency according to Costa et al., 2013, administered at t1 and t2; ES: equivalent scores; CS: corrected stores; CSI: Composite Shifting Index; FAB: Frontal Assessment Battery; CDT: Clock Drawing Test; OORT: Open-access Open-source Reaction Times; SDMT: oral version of the Symbol Digit Modalities Test; M5P – UDs: unique designs of the Modified Five Point test; BDI: Beck Depression Inventory; IES-R: Impact of Event Scale Revised; PGWBI: Psychological General Well Being Index; PW: Positivity and Well-being; QoL: World Health Organization Quality of Life questionnaire.
A significant reduction of execution times was observed for TMT-A (F 2,27.4 = 18.15; p < 0.001) and TMT-B (F 2,28.4 = 5.17; p = 0.012; see Fig. 2 ). Moreover, an effect of oxygen therapy was also observed (F 1,17 = 7.54; p = 0.014), with O+ performing faster (33.1 ± 10.5 s) than O- (75.1 ± 11.1 s). Accordingly, O+ showed better shifting abilities than O-, as indicated by the TMT B-A index (F 1,17.2 = 7.57; p = 0.014). An improvement of reaction times in a go-no-go context (OORT; F 1,25.7 = 7.61; p = 0.015) was observed from t1 to t2.Fig. 2 Longitudinal improvements in attentional and memory tasks. Relevant post-hoc comparisons are depicted. Error bars represent the standard error. TMT = Trail Making Test; s: seconds; RAVLT: immediate recall of the Rey Auditory Verbal Learning Task. **p < .008.
Fig. 2
An improvement was also observed in verbal learning of the RAVLT (immediate recall; F 2,28.5 = 7.56) and non-verbal long-term memory, whereby the delayed recall of a complex figure improved over time (F 2,23.9 = 6.65; p = 0.005). See also Fig. 2.
As for the phonological fluency, the analyses highlighted no effect of timepoint, but an independent effect of oxygenation, namely, O+ scored better than O- (F 1,17.4 = 5.41; p = 0.032). The same pattern emerged for semantic fluency (F 1,15.8 = 8.08; p = 0.012). Interestingly, the analysis of alternate fluency, administered at t1 and t2 as supplementary test, showed an interaction of timepoint and oxygenation (F 1,12.7 = 4.96; p = 0.045), i.e., although O- patients performed overall worse (48.7 ± 1.86), than O+ (56.1 ± 1.80) they improved the most from t1 to t2 (p = 0.002).
Finally, longitudinal improvements were observed in the logical reasoning (Raven Matrices; F 2,25.7 = 5.63; p = 0.009), especially from t0 to t1(p = 0.009), and in graphic fluency of the M5P from t1 to t2 (F 1,13.1 = 12.31; p = 0.022).
Concerning, the psycho-affective scales, PTSD-related symptoms decreased from t1 to t2 (W = 77.5; p = 0.028). No changes in quality of life (WHOQOL-BREF total score) were found, with only a positive trend for the “Physical Health” (F 1;13.4 = 4.6; p = 0.051) and “Psychological” scales (F 1;14.3 = 6.31; p = 0.025), suggesting small improvements from t1 to t2.
In light of our findings showing that O+ patients performed better than O- in some cognitive tasks, we then looked for any difference in other demographic, clinical, or psychological variables. Of the 16 patients who completed all timepoints, 8 patients were O+ (6 CPAP +2 OTI) and 8 were O-. Mann-Whitney tests highlighted no significant differences with respect to age (p = 0.645), education (p = 1), days in hospital (p = 0.083), distance between hospital discharge and t0 (p = 0.636), and between t0 and t1 (p = 0.442), between t1 and t2 (p = 0.195). Furthermore, no differences were observed on the psychological and quality of life questionnaires (all ps > 0.431), besides a trend for PTSD-related symptoms (p = 0.081).
3.3 Exploratory analyses with OSAS patients
Only one OSAS patient reported attentional difficulties in everyday life, with sudden-onset sleep, and nobody showed pathological performance on the cognitive tests. Three patients obtained borderline scores on different cognitive domains (i.e., the copy of the Rey-Osterrieth's figure, non-verbal fluency, and go-no-go RTs). Two patients showed mild depressive symptoms, and three other patients showed a significant reduction of psychological well-being.
Mann-Whitney tests showed, at t0, that COVID-19 patients were significantly slower (38.3 ± 22.9 s) than OSAS (19.3 ± 6.82 s; U = 36; p = 0.035). At t1 and t2, no significant differences emerged in the cognitive tests (all ps > 0.106); the only difference was observed for the psychological domain of the WHOQOL-BREF, where, COVID-19 patients obtained higher scores at both t1 (74.1 ± 12.4; U = 22; p = 0.004) and t2 (71.9 ± 14.9; U = 28; p = 0.028). All results are reported in Table S5 of the Supplementary Material.
4 Discussion
In the present study, we provide longitudinal evidence of cognitive deficits in patients reporting subjective cognitive decline after COVID-19 infection. Overall, we observed a high prevalence of cognitive alterations around six months after the acute stage (i.e., t0) that progressively diminished more than 18 months after acute infection (i.e., t1 and t2). This pattern was also supported by longitudinal analyses showing improvements in attentional-executive and mnemonic functioning.
4.1 Cognitive functioning
Over 657 screened COVID-19 patients, around 5% of them reported subjective cognitive impairments after the infection. Another work (Hadad et al., 2022) reported slightly higher values (8%, i.e., 46 out of 523). Moreover, the prevalence of neuropsychological deficits at 6 months after acute recovery (52%) is slightly lower than previous works such as that by Ferrucci and colleagues [23] (63% at a 5-month follow-up), Miskowiak et al. [57] (59%–65% at a 4-month follow-up), and Poletti and colleagues (22) (75% at a 6-month follow-up). However, beyond methodological differences and different sample sizes, the actual estimates could be higher (see also below the reflections about borderline scores). The present study has focused the investigation on those patients reporting subjective cognitive impairments, thus possibly missing subtler cognitive alterations even in patients without subjective complaints. In this regard, it is worth mentioning that some patients may present anosognosia for their cognitive deficits [21] (however, see [24] for opposite findings). Moreover, only 21 out of 36 patients with subjective cognitive impairments decided to take part in the study.
Notably, the most impaired function resulted to be verbal learning, a finding in line with several neuropsychological works which found verbal memory as one of the most affected functions in the first six months following acute recovery [9,14,20,23].
Around 16 and 22 months after the acute disease (i.e., t1 and t2), by just repeating the tests administered at the first assessment, a much lower number of patients obtained pathological scores with an increased percentage of borderline ones, possibly reflecting a gradual improvement (see also the discussion of the longitudinal analyses). However, the addition of RTs test indicated a significant reduction in alertness in several patients. In our opinion, this highlights two crucial aspects: 1) the importance of computerized testing of attention as more sensitive measures than classic paper-and-pencil tests, and 2) the importance of considering both pathological and borderline scores (see also the work by Voruz and colleagues [21]). Although borderline scores could sometimes reflect situational variables (e.g., anxiety), we cannot fully exclude that they could represent red flags of persisting problems or even predictors of cognitive decline in the long run. Therefore, we believe they should not be underestimated.
Along the same line of reasoning, it is important to note that the only patients with persisting pathological scores more than 18 months after hospital discharge (i.e., at t2) were those who reported a subjective cognitive decline even before the pandemic. Although the lack of pre-COVID assessment does not allow us to appreciate the actual impact of the acute disease, we cannot exclude a link between SARS-COV-2 infection and cognitive decline [5]. Notably, studies on different viral infections, such as herpes viruses [58,59] and Spanish influenza [60], raised the possibility of increased risk of neurodegeneration following infections, likely mediated by hyperinflammation states. Along the same line, a systemic, pro-inflammatory cytokine storm has been hypothesized to underlie COVID-related neurological involvement [61]. It follows the need for future research to evaluate the trajectories of cognitive decline in COVID-19 survivors, in particular to longitudinally monitor individuals at risk of dementia. This line of research will benefit from the increasing availability of tele-neuropsychological tools [[62], [63], [64], [65]].
Moving beyond the discussion of pathological findings, longitudinal analyses demonstrated significant improvements from t0 in multiple cognitive domains. Ferrucci et al. [23], by testing their patients at 5 months and 1 year (roughly comparable to our t0 and t1), observed significant improvements in the verbal memory domain (i.e., in all measures of the Selective Reminding Test of the Brief Repeatable Battery of Neuropsychological tests). Interestingly, although we also found changes in the same cognitive domain, here, the greatest improvement emerged 20 months after acute recovery (i.e., at t2), especially in the immediate recall of the RAVLT - the encoding phase - with no effects on the delayed recall (i.e., the proper active retrieval from long-term storage). Furthermore, we observed an improvement in attentional-executive tests, as highlighted, for example, by the improvements on the TMT. Ferrucci et al. [23] came to similar conclusions by administering different tests, i.e., the SDMT and the complex Paced Auditory Serial Addition Test. Importantly, we showed that attentional-executive functions could keep improving after one year from the acute disease. Although methodological differences prevent direct comparison of the results, the observed trends point towards the same direction. Future works should ideally adopt a shared set of second-level cognitive tests to improve the comparability and favor aggregated meta-analysis, thus reducing the methodological gaps highlighted by recent reviews [14].
The present work also aimed at testing the effects of oxygen therapy. In light of recent works showing a negative impact of hypoxia in the acute stage on the following cognitive sequelae [20,23], we expected to find lower cognitive performances in patients requiring oxygen support. However, this was not the case for a number of attentional and executive tests. Other works have shown an unclear relationship between the incidence of cognitive deficits and the severity of acute respiratory deficits [[16], [66], [67]]. Manera and colleagues [68] found that patients admitted to intensive care because of respiratory distress symptoms performed better on the MMSE than those requiring non-invasive ventilation, suggesting that the former may have suffered less from cerebral hypoxia. However, the lack of objective oxygenation measure, relatively small sample sizes [21] and age effects [16] may have affected, at least in part, previous as well as the current findings.
Finally, to explore cognitive alterations due to low oxygenation levels, but in the absence of SARS-COV-2 infection, we compared COVID-19 and OSAS patients' cognitive-emotional profiles. Indeed, OSAS patients often report cognitive alterations in their everyday life (i.e., reduced attention, concentration and short-term memory) [69] that can be objectified through a neuropsychological assessment (see the meta-analysis by Stranks & Crowe [70]), often resulting in a reduced quality of life [71]. Here, we observed no cognitive impairments (besides few borderline scores) and a slower attentional performance (TMT-A execution times) of COVID-19 patients at t0 (i.e., the timepoint with the higher rate of cognitive alterations). We believe that longer-term effects of these pathologies (especially in older adults) are worth further investigation, considering the increased risk of cognitive decline related to impaired respiratory functions, common to both COVID-19 and OSAS.
4.2 Psycho-affective variables
Concerning the results on psycho-affective scales, we extend previous evidence on psychiatric sequelae six months after the acute stage [5]: more than 30% of our patients presented a certain degree of depressive symptoms and PTSD-related symptoms more than one year after acute recovery, with rates and severity decreasing over time. Notably, the correlations between psycho-affective and cognitive measures are in line with the recent findings of Voruz and colleagues [21], who recently confirmed the profound impact of psycho- affective variables on a wide spectrum of neuropsychological tests. Taken together, these results suggest that changes in cognitive functioning in COVID-19 patients cannot be fully understood without considering psychological aspects [22,57].
Finally, some limitations deserve consideration. The main limit of the present study is the small sample size that prevents from drawing definite conclusions about, e.g., the effects of COVID hypoxia on cognition, as well as differential cognitive profiles of COVID-19 and OSAS patients. Another limit regards the repeated administration of the same neuropsychological tests, potentially leading to practice effects and hampering the interpretation of repeated measurements. Here we used parallel versions whenever valid and normed forms were available (i.e., for the copy and recall of a complex figure and the RAVLT); however, we cannot fully exclude that test-retest effects may have partially influenced the previously discussed improvements.
In conclusion, our work confirms and extends previous literature, showing that cognitive alterations in COVID-19 patients may extend over one year from the resolution of acute respiratory problems but nonetheless face a significant recovery in several cases. Importantly: 1) such cognitive alterations are accompanied - and possibly affected - by significant and more prolonged psychological-emotional distress; 2) several patients may present with borderline scores in memory and executive tests that are worthy of attention; 3) neuropsychological monitoring is especially advised for patients with suspected or initial cognitive decline at higher risk of developing dementia.
Declaration of Competing Interest
The authors declare no conflict of interest.
Appendix A Supplementary data
Supplementary Tables
Image 1
Acknowledgement
This work was funded by an unconditional donation from Novatex Italia S.p.A and partially supported by the Italian Ministry of Health to Nadia Bolognini.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.jns.2022.120511.
==== Refs
References
1 Douaud G. Lee S. Alfaro-Almagro F. Arthofer C. Wang C. McCarthy P. …Smith S.M. SARS-CoV-2 is associated with changes in brain structure in UK Biobank Nature 604 7907 2022 697 707 35255491
2 Romero-Sánchez C.M. Díaz-Maroto I. Fernández-Díaz E. Sánchez-Larsen Á. Layos-Romero A. García-García J. …Segura T. Neurologic manifestations in hospitalized patients with COVID-19: the ALBACOVID registry Neurology 95 8 2020 e1060–e1070
3 Mao L. Wang M. Chen S. He Q. Chang J. Hong C. …Hu B. Neurological Manifestations of Hospitalized Patients with COVID-19 in Wuhan, China: A Retrospective Case Series Study. MedRxiv 2020
4 Moro E. Priori A. Beghi E. Helbok R. Campiglio L. Bassetti C.L. …EAN core COVID-19 Task Force The international European academy of neurology survey on neurological symptoms in patients with COVID-19 infection Eur. J. Neurol. 27 9 2020 1727 1737 32558002
5 Taquet M. Geddes J.R. Husain M. Luciano S. Harrison P.J. 6-month neurological and psychiatric outcomes in 236 379 survivors of COVID-19: a retrospective cohort study using electronic health records Lancet Psychiatry 8 5 2021 416 427 33836148
6 Dini M. Poletti B. Tagini S. Reitano M.R. Allocco E. Mazzocco K. …Ferrucci R. Resilience, psychological well-being and daily functioning following hospitalization for respiratory distress due to sars-cov-2 infection Healthcare 9 9 2021, September 1161 34574935
7 Ceban F. Ling S. Lui L.M. Lee Y. Gill H. Teopiz K.M. …McIntyre R.S. Fatigue and cognitive impairment in post-COVID-19 syndrome: a systematic review and meta-analysis Brain Behav. Immun. 101 2022 93 135 34973396
8 Crivelli L. Calandri I. Corvalán N. Carello M.A. Keller G. Martínez C. …Allegri R. Cognitive consequences of COVID-19: results of a cohort study from South America Arq. Neuropsiquiatr. 80 2021 240 247
9 Sambati L. Mattarozzi K. Mascia L. Tonetti T. Santoro R. Cretella L. …Guarino M. Cognitive and affective disorders in critical SARS-CoV-2 patients and caregivers J. Neurol. Sci. 429 2021
10 Heneka M.T. Golenbock D. Latz E. Morgan D. Brown R. Immediate and long-term consequences of COVID-19 infections for the development of neurological disease Alzheimers Res. Ther. 12 1 2020 1 3
11 Guedj E. Campion J.Y. Dudouet P. Kaphan E. Bregeon F. Tissot-Dupont H. …Eldin C. 18F-FDG brain PET hypometabolism in patients with long COVID Eur. J. Nucl. Med. Mol. Imaging 48 9 2021 2823 2833 33501506
12 Hellgren L. Thornberg U.B. Samuelsson K. Levi R. Divanoglou A. Blystad I. Brain MRI and neuropsychological findings at long-term follow-up after COVID-19 hospitalisation: an observational cohort study BMJ Open 11 10 2021 e055164
13 Han Y. Yuan K. Wang Z. Liu W.J. Lu Z.A. Liu L. …Lu L. Neuropsychiatric manifestations of COVID-19, potential neurotropic mechanisms, and therapeutic interventions Transl. Psychiatry 11 1 2021 1 20 33414379
14 Crivelli L. Palmer K. Calandri I. Guekht A. Beghi E. Carroll W. …Kivipelto M. Changes in Cognitive Functioning after COVID-19: A Systematic Review and Meta-Analysis 2022 Alzheimer’s & Dementia
15 Daroische R. Hemminghyth M.S. Eilertsen T.H. Breitve M.H. Chwiszczuk L.J. Cognitive impairment after COVID-19—a review on objective test data Front. Neurol. 12 2021 699582
16 Alemanno F. Houdayer E. Parma A. Spina A. Del Forno A. Scatolini A. …Iannaccone S. COVID-19 cognitive deficits after respiratory assistance in the subacute phase: a COVID-rehabilitation unit experience PLoS One 16 2 2021 e0246590
17 Almeria M. Cejudo J.C. Sotoca J. Deus J. Krupinski J. Cognitive profile following COVID-19 infection: clinical predictors leading to neuropsychological impairment Brain Behav. Immun. Health 9 2020 100163
18 Beaud V. Crottaz-Herbette S. Dunet V. Vaucher J. Bernard-Valnet R. Du Pasquier R. …Clarke S. Pattern of cognitive deficits in severe COVID-19 J. Neurol. Neurosurg. Psychiatry 92 5 2021 567 568 33219042
19 De Lorenzo R. Conte C. Lanzani C. Benedetti F. Roveri L. Mazza M.G. …Rovere-Querini P. Residual clinical damage after COVID-19: a retrospective and prospective observational cohort study PLoS One 15 10 2020 e0239570
20 Ferrucci R. Dini M. Groppo E. Rosci C. Reitano M.R. Bai F. …Priori A. Long-lasting cognitive abnormalities after COVID-19 Brain Sci. 11 2 2021 235 33668456
21 Voruz P. Allali G. Benzakour L. Nuber-Champier A. Thomasson M. Jacot de Alcântara I. …Péron J.A. Long COVID neuropsychological deficits after severe, moderate, or mild infection Clin. Transl. Neurosci. 6 2 2022 9
22 Poletti S. Palladini M. Mazza M.G. De Lorenzo R. Furlan R. Ciceri F. …Benedetti F. Long-term consequences of COVID-19 on cognitive functioning up to 6 months after discharge: role of depression and impact on quality of life Eur. Arch. Psychiatry Clin. Neurosci. 272 5 2022 773 782 34698871
23 Ferrucci R. Dini M. Rosci C. Capozza A. Groppo E. Reitano M.R. …Priori A. One-year cognitive follow-up of COVID-19 hospitalized patients Eur. J. Neurol. 29 7 2022 2006 2014 35285122
24 Miskowiak K.W. Fugledalen L. Jespersen A.E. Sattler S.M. Podlekareva D. Rungby J. …Johnsen S. Trajectory of cognitive impairments over 1 year after COVID-19 hospitalisation: pattern, severity, and functional implications Eur. Neuropsychopharmacol. 59 2022 82 92 35561540
25 Del Brutto O.H. Wu S. Mera R.M. Costa A.F. Recalde B.Y. Issa N.P. Cognitive decline among individuals with history of mild symptomatic SARS-CoV-2 infection: a longitudinal prospective study nested to a population cohort Eur. J. Neurol. 28 10 2021 3245 3253 33576150
26 Méndez R. Balanzá-Martínez V. Luperdi S.C. Estrada I. Latorre A. González-Jiménez P. …Menéndez R. Long-term neuropsychiatric outcomes in COVID-19 survivors: a 1-year longitudinal study J. Intern. Med. 291 2 2022 247 251 34569681
27 Magni E. Binetti G. Bianchetti A. Rozzini R. Trabucchi M. Mini-mental state examination: a normative study in Italian elderly population Eur. J. Neurol. 3 3 1996 198 202 21284770
28 Measso G. Cavarzeran F. Zappalà G. Lebowitz B.D. Crook T.H. Pirozzolo F.J. …Grigoletto F. The mini-mental state examination: normative study of an Italian random sample Dev. Neuropsychol. 9 2 1993 77 85
29 Spinnler H. Tognoni G. Standardizzazione e taratura italiana di test neuropsicologici Ital. J. Neurol. Sci. 8 Suppl 1987 1 120
30 Giovagnoli A.R. Del Pesce M. Mascheroni S. Simoncelli M. Laiacona M. Capitani E. Trail making test: normative values from 287 normal adult controls Ital. J. Neurol. Sci. 17 4 1996 305 309 8915764
31 Monaco M. Costa A. Caltagirone C. Carlesimo G.A. Forward and backward span for verbal and visuo-spatial data: standardization and normative data from an Italian adult population Neurol. Sci. 34 5 2013 749 754 22689311
32 Carlesimo G.A. Caltagirone C. Gainotti G.U.I.D. Fadda L. Gallassi R. Lorusso S. …Parnetti L. The mental deterioration battery: normative data, diagnostic reliability and qualitative analyses of cognitive impairment Eur. Neurol. 36 6 1996 378 384 8954307
33 Capitani E. Della Sala S. Laiacona M. Marchetti C. Standardizzazione ed uso clinico di un test di memoria di prosa 1994 Giunti Organizzazioni Speciali
34 Caffarra P. Vezzadini G. Dieci F. Zonato F. Venneri A. Rey-Osterrieth complex figure: normative values in an Italian population sample Neurol. Sci. 22 6 2002 443 447 11976975
35 Aiello E.N. Esposito A. Gramegna C. Gazzaniga V. Zago S. Difonzo T. …Bolognini N. The frontal assessment battery (FAB) and its sub-scales: validation and updated normative data in an Italian population sample Neurol. Sci. 43 2 2022 979 984 34184168
36 Appollonio I. Leone M. Isella V. Piamarta F. Consoli T. Villa M.L. …Nichelli P. The frontal assessment battery (FAB): normative values in an Italian population sample Neurol. Sci. 26 2 2005 108 116 15995827
37 Novelli G. Papagno C. Capitani E. Laiacona M. Tre test clinici di ricerca e produzione lessicale. Taratura su sogetti normali Archivio di Psicologia, Neurologia e Psichiatria 1986
38 Costa A. Bagoj E. Monaco M. Zabberoni S. De Rosa S. Papantonio A.M. …Carlesimo G.A. Standardization and normative data obtained in the Italian population for a new verbal fluency instrument, the phonemic/semantic alternate fluency test Neurol. Sci. 35 3 2014 365 372 23963806
39 Laiacona M. Inzaghi M.G. De Tanti A. Capitani E. Wisconsin card sorting test: a new global score, with Italian norms, and its relationship with the Weigl sorting test Neurol. Sci. 21 5 2000 279 291 11286040
40 Basso A. Capitani E. Laiacona M. Raven’s coloured progressive matrices: normative values on 305 adult normal controls Funct. Neurol. 2 2 1987 189 194 3666548
41 Rigoli M. Facchin A. Cardile D. Beschin N. Luzzatti C. Open-source open-access reaction time test (OORTT): an easy tool to assess reaction times Neurol. Sci. 42 6 2021 2461 2469 33095365
42 Nocentini U. Giordano A. Vincenzo S.D. Panella M. Pasqualetti P. The Symbol Digit Modalities Test--Oral version: Italian normative data Funct. Neurol. 21 2 2006 93 96 16796824
43 Mondini S. Mapelli D. Vestri A. Bisiacchi P. ENB esame neuropsicologico breve. Raffaello Cortina, MI 2003
44 Battista P. Griseta C. Tortelli R. Guida P. Castellana F. Rivolta D. Logroscino G. The modified five-point test (MFPT): normative data for a sample of Italian elderly Neurol. Sci. 42 6 2021 2431 2440 33074452
45 Cattelani R. Dal Sasso F. Corsini D. Posteraro L. The modified five-point test: normative data for a sample of Italian healthy adults aged 16–60 Neurol. Sci. 32 4 2011 595 601 21327400
46 Casarotti A. Papagno C. Zarino B. Modified Taylor complex figure: normative data from 290 adults J. Neuropsychol. 8 2 2014 186 198 23647550
47 Capitani E. Laiacona M. Composite neuropsychological batteries and demographic correction: standardization based on equivalent scores, with a review of published data J. Clin. Exp. Neuropsychol. 19 6 1997 795 809 9524875
48 Capitani E. Laiacona M. Outer and inner tolerance limits: their usefulness for the construction of norms and the standardization of neuropsychological tests Clin. Neuropsychol. 31 6–7 2017 1219 1230 28598726
49 Beck A.T. Ward C. Mendelson M. Mock J. Erbaugh J.J.A.G.P. Beck depression inventory (BDI) Arch. Gen. Psychiatry 4 6 1961 561 571 13688369
50 Beck A.T. Steer R.A. Carbin M.G. Psychometric properties of the Beck depression inventory: twenty-five years of evaluation Clin. Psychol. Rev. 8 1 1988 77 100
51 Weiss D. Marmar C. The impact of event scale—revised. W Wilson J. Keane T. Assessing Psychological Trauma and PTSD: A Handbook for Practitioners 1997
52 Creamer M. Bell R. Failla S. Psychometric properties of the impact of event scale—revised Behav. Res. Ther. 41 12 2003 1489 1496 14705607
53 Dupuy H. The Psychological General Well-Being (PGWB) Index 1984 (Assessment of quality of life in clinical trials of cardiovascular therapies)
54 Grossi E. Mosconi P. Groth N. Niero M. Apolone G. Questionario Psychological General Well-Being Index. Versione Italiana 2002
55 Whoqol Group Development of the World Health Organization WHOQOL-BREF quality of life assessment Psychol. Med. 28 3 1998 551 558 9626712
56 The jamovi project jamovi Retrieved from: https://www.jamovi.org 2021
57 Miskowiak K.W. Johnsen S. Sattler S.M. Nielsen S. Kunalan K. Rungby J. …Porsberg C.M. Cognitive impairments four months after COVID-19 hospital discharge: pattern, severity and association with illness variables Eur. Neuropsychopharmacol. 46 2021 39 48 33823427
58 Mancuso R. Sicurella M. Agostini S. Marconi P. Clerici M. Herpes simplex virus type 1 and Alzheimer’s disease: link and potential impact on treatment Expert Rev. Anti-Infect. Ther. 17 9 2019 715 731 31414935
59 Warren-Gash C. Forbes H.J. Williamson E. Breuer J. Hayward A.C. Mavrodaris A. …Smeeth L. Human herpesvirus infections and dementia or mild cognitive impairment: a systematic review and meta-analysis Sci. Rep. 9 1 2019 1 10 30626917
60 McCall S. Vilensky J.A. Gilman S. Taubenberger J.K. The relationship between encephalitis lethargica and influenza: a critical analysis J. Neuro-Oncol. 14 3 2008 177 185
61 De Virgiliis F. Di Giovanni S. Lung innervation in the eye of a cytokine storm: neuroimmune interactions and COVID-19 Nat. Rev. Neurol. 16 11 2020 645 652 32843733
62 Aiello E.N. Esposito A. Giannone I. Diana L. Appollonio I. Bolognini N. Telephone interview for cognitive status (TICS): Italian adaptation, psychometrics and diagnostics Neurol. Sci. 43 5 2022 3071 3077 34792669
63 Aiello E.N. Esposito A. Giannone I. Diana L. Woolley S. Murphy J. …Appollonio I. ALS cognitive behavioral screen-phone version (ALS-CBSTM-PhV): norms, psychometrics, and diagnostics in an Italian population sample Neurol. Sci. 43 4 2022 2571 2578 34601697
64 Aiello E.N. Preti A.N. Pucci V. Diana L. Corvaglia A. di San Pietro C.B. …Bolognini N. The Italian telephone-based verbal fluency battery (t-VFB): standardization and preliminary clinical usability evidence Front. Psychol. 13 2022
65 Aiello E.N. Pucci V. Diana L. Niang A. Preti A.N. Delli Ponti A. …Bolognini N. Telephone-based frontal assessment battery (t-FAB): standardization for the Italian population and clinical usability in neurological diseases Aging Clin. Exp. Res. 34 7 2022 1635 1644 35699839
66 Priftis K. Velardo V. Vascello M.G.F. Villella S. Galeri S. Spada M.S. Algeri L. Limited evidence for neuropsychological dysfunction in patients initially affected by severe COVID-19 Neurol. Sci. 2022 1 3
67 Hadad R. Khoury J. Stanger C. Fisher T. Schneer S. Ben-Hayun R. …Adir Y. Cognitive dysfunction following COVID-19 infection J. NeuroVirol. 2022 1 8
68 Manera M.R. Fiabane E. Pain D. Aiello E.N. Radici A. Ottonello M. …Pistarini C. Clinical features and cognitive sequelae in COVID-19: a retrospective study on N= 152 patients Neurol. Sci. 43 1 2022 45 50 34779965
69 Kapur V.K. Auckley D.H. Chowdhuri S. Kuhlmann D.C. Mehra R. Ramar K. Harrod C.G. Clinical practice guideline for diagnostic testing for adult obstructive sleep apnea: an American Academy of sleep medicine clinical practice guideline J. Clin. Sleep Med. 13 3 2017 479 504 28162150
70 Stranks E.K. Crowe S.F. The cognitive effects of obstructive sleep apnea: an updated meta-analysis Arch. Clin. Neuropsychol. 31 2 2016 186 193 26743325
71 Scarpina F. Bastoni I. Cappelli S. Priano L. Giacomotti E. Castelnuovo G. …Mauro A. Psychological well-being in obstructive sleep apnea syndrome associated with obesity: the relationship with personality, cognitive functioning, and subjective and objective sleep quality Front. Psychol. 12 2021 588767
| 36473347 | PMC9707027 | NO-CC CODE | 2022-12-06 23:15:07 | no | J Neurol Sci. 2023 Jan 15; 444:120511 | utf-8 | J Neurol Sci | 2,022 | 10.1016/j.jns.2022.120511 | oa_other |
==== Front
Int J Disaster Risk Reduct
Int J Disaster Risk Reduct
International Journal of Disaster Risk Reduction
2212-4209
The Authors. Published by Elsevier Ltd.
S2212-4209(22)00690-2
10.1016/j.ijdrr.2022.103471
103471
Article
Annual prevalence of non-communicable diseases and identification of vulnerable populations following the Fukushima disaster and COVID-19 pandemic
Murakami Michio a∗
Nomura Shuhei bcd
a Center for Infectious Disease Education and Research, Osaka University, Osaka, Japan
b Department of Health Policy and Management, School of Medicine, Keio University, Tokyo, Japan
c Department of Global Health Policy, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
d Tokyo Foundation for Policy Research, Tokyo, Japan
∗ Corresponding author.
29 11 2022
1 2023
29 11 2022
84 103471103471
21 5 2022
23 11 2022
27 11 2022
© 2022 The Authors
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Disasters, pandemics, and their response measures can have secondary effects on the physical and psychological health of affected populations. Using health insurance receipt data from 2009 to 2020, we assessed changes in the prevalence of major non-communicable diseases (NCDs), including hypertension, hyperlipidemia, diabetes, and mental disorders, among affected populations before and after the Fukushima disaster and coronavirus disease (COVID-19) outbreak in Japan. Furthermore, age and sex groups with the largest increases in prevalence after these events were identified. The participants of this study were employees of large companies and their dependent family members who were insured by health insurance societies (HIS). The dataset was provided by JMDC Inc. The annual age-adjusted prevalence of each disease belonging to the HIS scheme was used to calculate the ratio of disease prevalence before and after the events. After the Fukushima disaster, hypertension, hyperlipidemia, and diabetes generally increased over a 9-year period in Fukushima Prefecture. The increase in the prevalence rate of these three NCDs and mental disorders was the highest among females aged 40–74 years compared to males and the other age groups. The prevalence of all four diseases increased after the COVID-19 outbreak in Japan, with a marked increase in males aged 0–39 years. Changes in prevalence ratios of NCDs after the COVID-19 outbreak among the areas affected by the Fukushima disaster were lower than in the whole of Japan. It is important to provide tailor-made public health support among populations in accordance with the type of disasters and pandemic.
Keywords
COVID-19
Diabetes mellitus
Fukushima nuclear accident
Hyperlipidemias
Hypertension
Mental disorders
==== Body
pmc1 Introduction
Disasters, pandemics, and their countermeasures impact the lives of affected people with secondary effects on their physical and psychological health. Major non-communicable diseases (NCDs), such as hypertension, hyperlipidemia, diabetes, and mental disorders, have been reported to increase among people who have experienced natural or man-made disasters, such as the Chornobyl nuclear power plant accident in 1986 and Hurricane Katrina in 2005 [1,2]. Several studies have also revealed that the Great East Japan Earthquake and Tsunami (GJET) and the Fukushima Daiichi Nuclear Power Station accident in March 2011 (hereafter referred to as the “Fukushima disaster”) also caused an increase in NCDs among residents who evacuated for short to long periods [[3], [4], [5]]. In addition, long-term deterioration of mental health, including increases in psychological distress and suicide rates, has been observed, especially among residents affected by the Fukushima disaster, although there is a trend toward improvement over time [6,7]. Furthermore, in Minamisoma City, located within 10–40 km of the nuclear power station, a significant increase in the age-adjusted prevalence of diabetes was observed among both those who did not evacuate and those who did, during a 7-year period following the disaster [8]. High psychological distress has also been reported among non-evacuees in host communities where evacuees were accommodated [9]. Changes in living conditions due to evacuation after a disaster as well as various socioeconomic changes in the surrounding communities, may cause secondary physical and psychological health effects to evacuees and non-evacuees. However, the spatial and temporal extent of post-disaster secondary health effects have not been established, as there are only a limited number of long-term studies on both evacuees and non-evacuees over a wide area.
Following the emergence and worldwide spread of coronavirus disease (COVID-19), behavioral restrictions, such as lockdowns or stay-at-home measures, have been implemented in various countries, including Japan, to curb infection spread. During the pandemic, some studies have shown improvement in lifestyle, dietary, and drinking behaviors [[10], [11], [12]]; however, increases in body weight and worsening of metabolic biomarkers, such as glycemic values, among specific populations have also been reported [13,14]. Furthermore, there has been a significant increase in mental disorders among the global population, especially in females and younger age groups [15]. Similarities and differences in secondary health effects among multiple disasters and pandemics remain unclear due to limited analyses of data obtained before and after such events. In addition, identifying the age and sex groups in which NCDs occurred can provide beneficial information to promote effective support for vulnerable populations regarding the secondary health effects following disasters and pandemics. Assessment of the secondary impact of multiple disasters and pandemics on NCDs among the affected people and identification of their vulnerable populations is expected to lead to better post-disaster policy recommendations, including promoting effective health and prevention strategies.
We used health insurance receipt data to assess changes in the prevalence of hypertension, hyperlipidemia, diabetes, and mental disorders in the population affected by the Fukushima disaster and COVID-19 in Japan. First, we investigated how the prevalence of these diseases changed over 9 years, from January 2009 to December 2020, following the Fukushima disaster. The Fukushima, neighboring prefectures of Iwate and Miyagi prefectures, as well as the whole of Japan and three major regions within Fukushima Prefecture, were considered to assess the wide-area secondary health effects caused by the disaster. We then analyzed the prevalence of these diseases in Japan and Fukushima Prefecture by age and sex to identify the populations that were the most significantly affected. Next, we examined the annual changes in the prevalence of these diseases across Japan before and after the COVID-19 outbreak and identified the most significantly affected age and sex groups. We then analyzed the associations between population density or COVID-19 infection rates and changes in the prevalence of these diseases by the prefecture to determine prefectural characteristics that are susceptible to these secondary health effects. Finally, we compared the annual changes in the prevalence of these diseases before and after the COVID-19 outbreak among Fukushima, Iwate, and Miyagi prefectures and three major regions within Fukushima Prefecture after COVID-19 with the whole of Japan to assess the secondary health effects caused by the COVID-19 among the areas affected by the Fukushima disaster.
Our hypotheses are as follows.1. The prevalence of NCDs increased for a long period in wide areas affected by the Fukushima disaster.
2. The prevalence of NCDs increased in the whole of Japan after the COVID-19 outbreak.
3. Sex and age groups that increased the prevalence of NCDs differed between the Fukushima disaster and the COVID-19 outbreak.
4. The impact of the COVID-19 outbreak on the prevalence of NCDs was smaller among the areas that experienced the Fukushima disaster than the whole of Japan.
2 Methods
2.1 Ethics
This study was approved by the Osaka University Research Ethics Committee (approval number Ci21002). Since we received only anonymized aggregate data, we did not need to obtain informed consent.
2.2 Fukushima disaster
The reported number of people directly killed or missing due to GJET on March 11, 2011, was 18,423 (at the end of February 2022 [16]). The reported number of disaster-related deaths caused indirectly by the GJET (i.e., secondary victims) was 3784 (at the end of September 2021) [17]. Of the 18,423 people, 5785 people in Iwate Prefecture, 10,757 in Miyagi Prefecture, and 1810 in Fukushima Prefecture were killed or missing (Fig. 1 ). There were 470 disaster-related deaths in Iwate Prefecture, 929 in Miyagi Prefecture, and 2329 in Fukushima Prefecture. The number of people living in evacuation shelters immediately after the GJET was estimated to be more than 450,000 nationwide, including more than 400,000 in the Iwate, Miyagi, and Fukushima prefectures [18].Fig. 1 Locations of areas.
Fig. 1
Following the Fukushima disaster, mandatory evacuation orders were issued to residents within a 20-km radius around the Fukushima Daiichi Nuclear Power Station and a 10-km radius around the Daini Nuclear Power Station on March 12, 2011, followed by evacuation orders on April 22, 2011, for areas where an additional exposure level would reach 20 mSv/y. Evacuation orders were issued for Hirono Town, Naraha Town, Tomioka Town, Kawauchi Village, Okuma Town, Futaba Town, Namie Town, Katsurao Village, Iitate Village, Minamisoma City, Tamura City, Kawamata Town, and the hotspot areas in Date City. Of these, Tamura City, Kawamata City, and Date City are in Nakadori (central area), whereas the other municipalities are located in Hamadori (coastal area). It was estimated that more than 140,000 people were evacuated from these radii [19]. The affected people were evacuated to municipalities in Fukushima Prefecture outside the evacuation areas, such as Iwaki City (Hamadori), as well as to areas outside Fukushima Prefecture.
In 2012, the evacuation order was lifted in Kawauchi Village and Hirono Town, where the evacuation was decided by each municipality soon after the disaster. Furthermore, since 2014, the government has gradually lifted the evacuation order, and by the end of 2019, the evacuation order was lifted for the following municipalities: Tamura City, Kawauchi Village, Naraha Town, Katsurao Village, Minamisoma City, Iitate Village, Kawamata Town, Namie Town, Tomioka Town, and Okuma Town, excluding the difficult-to-return areas where radiation levels remain high located around the Fukushima Daiichi Nuclear Power Station.
2.3 COVID-19
The World Health Organization (WHO) declared COVID-19 a pandemic on March 11, 2020 [20]. In Japan, a declaration of a state of emergency was issued in seven prefectures on April 7, 2020 and extended to all prefectures on April 16, 2020. The state of emergency included requests to stay at home, instruction for schools to close temporarily, closure of businesses and stores that are not essential to daily life, and cancellation or postponement of events and exhibitions, but they were not mandatory. The state of emergency ended in 39 prefectures on May 14, 2020, in three more prefectures on May 21, 2020, and in all prefectures on May 25, 2020. This was the only emergency declaration for 2020, and the reported number of COVID-19-positive cases in Japan in 2020 was 234,109 [21].
2.4 Datasets and target population
In this study, we used health insurance receipt data provided by JMDC Inc. The target population covered in the JMDC data were employees of companies and their dependent family members who belonged to Employees’ Health Insurance (EHI) schemes. Please note that EHI can be divided into several types, depending on the size and type of organization to which the employee belongs. Of these, JMDC Inc. covers only employees working at large companies and their dependents who are insured by health insurance societies (HIS). Approximately 29 million people in the country are insured by HIS [22], and JMDC Inc. contracts a part of these insurers, covering 9 million people (as of 2022), and collect insurance claims containing completely anonymized receipt information. Other insurance schemes that are part of the EHI include those for employees of small and medium-sized companies and their families, and those for civil servants, private school teachers and seafarers, which JMDC Inc. does not cover.
In addition to the EHI schemes, Japan's insurance system has two major schemes: National Health Insurance (NHI) and Long-term Care Insurance (LTCI). All persons, regardless of citizenship, who have been recognized as having resided in Japan for more than three months are obliged to join one of three. NHI targets individuals who do not belong to the EHI scheme, including farmers, freelancers, non-regular employees, retired company employees, etc., and LTCI includes individuals aged 75 and above.
We used data on the annual number of patients for each targeted disease and other diseases (i.e., all patients visiting medical institutions regardless of their diagnosis) and the estimated prevalence belonging to the HIS scheme — hereinafter, it is simply described as “prevalence,” unless otherwise noted — by age and sex groups in each prefecture or municipality from January 2009 to December 2020. The prefectures and municipalities were based on the location of the medical institutions. The target diseases included hypertension, hyperlipidemia, diabetes, and mental disorders, based on the ICD 10 classification: I10–I15, E78, E10-E14, and F00–F99, respectively. The diseases were diagnosed by a physician, with no distinction made between primary and secondary diseases. The total number of patients in the JMDC database for hypertension, hyperlipidemia, diabetes, mental disorders, and other diseases in Japan from 2009 to 2020 is provided in Table S1.
Other studies have used the JMDC database to calculate the prevalence of specific diseases for the whole of Japan [23,24]. In this study, the prevalence (P d,t,r,a,s) of the target disease (d) in each prefecture or municipality (r) by age (a:0–19, 20–39, 40–64, and 65–74 years) and sex (s) groups for each year (t) were calculated according to eqs. (1), (2)).(1) Pd,t,r,a,s=Ad,t,r,a,sBt,r,a,s×Cd,t,a,s∑rDd,t,r,a,s
(2) Dd,t,r,a,s=Ad,t,r,a,sBt,r,a,s×Et,r,a,s
where A d,t,r,a,s is the actual number of patients with the target disease by each prefecture or municipality among the target age and sex population in the target year, B t,r,a,s is the actual number of patients who visited the medical institution regardless of their diagnosis in each prefecture or municipality among the target age and sex population in the target year, C d,t,a,s is the estimated number of patients nationwide among the target age and sex population in the target year and was estimated by multiplying the prevalence of each disease (the ratio of the number of patients to the number of people in the mother population covered by JMDC Inc.) by age (1-year interval) and sex groups and by the national population, D d,t,r,a,s is the estimated number of patients with the target disease in each prefecture and municipality in the target age and sex population in the target year, and E t,r,a,s is the number of people in each prefecture and municipality in the target age and sex population in the target year.
The Basic Resident Register data were used to determine the number of people in the prefectures for each year [25]. For the municipal population data, the national census performed in 2010, 2015, and 2020 was used for the years 2009–2010, 2011–2015, and 2016–2020, respectively [26]. Selection bias among the target population was corrected by C d,t,a,s and D d,t,r,a,s, as described in eq. (1).
Furthermore, using the commonly used 1985 model population as a reference population (i.e., distribution of the population by age and sex groups in 1985) [27], the age-adjusted prevalence was calculated for the 0–74, 0–39, and 40–74 years of age groups for males, females, and both males and females.
Diagnostic guidelines have been revised for various diseases, including the establishment of borderline hyperLDL-cholesterolemia in 2012 and the diagnosis of diabetes using HbA1c in 2010. Although these revisions may have impacted the prevalence, the bias was reduced by standardizing the assessment of the impact before and after the Fukushima disaster by normalizing the prevalence with that for the whole of Japan (see “2.5. Analysis” for details).
2.5 Analysis
Two analyses were performed in this study. First, to evaluate changes in prevalence after the Fukushima disaster, we considered the whole of Japan, Iwate, Miyagi, and Fukushima prefectures, as well as three regions within Fukushima Prefecture (i.e., Hamadori, Nakadori, and Aizu (mountainous area); Fig. 1, Table S2). These three regions, which are closer to the nuclear power station in the order Hamadori, Nakadori, and Aizu, divide Fukushima Prefecture into three parts from east to west. In other words, Hamadori was the region most affected by the disaster, and Aizu was the region least affected among regions in the prefecture. The prevalence of targeted diseases in the Hamadori, Nakadori, and Aizu regions was calculated by weighting data by the municipality based on B t,r,a,s. The arithmetic means and standard errors of the prevalence for each of the four time periods, such as 2009–2010, 2011–2013, 2014–2016, and 2017–2019, were calculated, and the prevalence for each time period was assumed to follow a normal distribution. To ensure that the prevalence was within the range of 0–1, two-tailed censoring was performed so that the arithmetic mean was at the center of the distribution. Monte Carlo simulations were performed with 10,000 iterations to determine the ratio or difference in prevalence (males and females, 0–74 years) between 2011 and 2013, 2014–2016, or 2017–2019 and 2009–2010 in the whole of Japan, Iwate, Miyagi, Fukushima prefectures, and Hamadori, Nakadori, and Aizu regions. These prevalence ratios and differences were further normalized to the values for the whole of Japan. Similarly, prevalence ratios and differences by age and sex groups were calculated for the whole of Japan and Fukushima Prefecture.
Second, to assess changes in the prevalence after the COVID-19 outbreak, data from 2015 to 2020 were used (Tables S2 and S3). For simplicity, we defined 2020 data as the period after the COVID-19 outbreak. We calculated the ratios and differences between the prevalence in 2020 (the period after the outbreak) and the prevalence in 2020 estimated from the 2015–2019 dataset. A linear regression equation was modeled using the dataset as a baseline to estimate the hypothetical 2020 prevalence in the absence of COVID-19. Based on the values and their 95% confidence intervals, calculations of the ratios and differences between the measured and estimated prevalence were performed with Monte Carlo simulations in 10,000 runs, assuming that the prevalence followed a normal distribution; two-tailed censoring was performed so that the prevalence fell within the range of 0–1, and the arithmetic mean was at the center of the distribution. The prevalence ratios and differences were estimated for age and sex groups in the whole of Japan: males and females (0–74 years), males and females (0–39 years), males and females (40–74 years), males (0–74 years), males (0–39 years), males (40–74 years), males (0–19 years), males (20–39 years), males (40–64 years), males (65–74 years), females (0–74 years), females (0–39 years), females (40–74 years), females (0–19 years), females (20–39 years), females (40–64 years), and females (65–74 years). Furthermore, the prevalence ratios and differences in 47 prefectures were calculated for males (0–39 years), the vulnerable population group identified in the above analysis, to assess the correlation with population density [25,28] or the reported number of COVID-19 infection cases per 100,000 people (COVID-19 infection rate) in 2020 [21] (Table S4). Spearman's rank correlation coefficient was estimated. The prevalence ratios (males and females, 0–74 years) in Iwate, Miyagi, Fukushima prefectures, and Hamadori, Nakadori, and Aizu regions and the values normalized to those for the whole of Japan were then calculated. The prevalence ratios and the values normalized to those for the whole of Japan by age and sex groups were calculated for Fukushima Prefecture.
Oracle Crystal Ball (Version 11.1.2.4.900) and IBM SPSS Statistics (Version 28) were used for the analysis.
3 Results
3.1 Age-adjusted prevalence over time
Fig. 2 shows the annual changes in age-adjusted prevalence (male and female, 0–74 years) of NCDs from 2009 to 2020 in the whole of Japan; Iwate, Miyagi, and Fukushima prefectures; and Hamadori, Nakadori, and Aizu regions. The age-adjusted prevalence of hypertension, hyperlipidemia, diabetes, and mental disorders in the whole of Japan was 0.102, 0.107, 0.061, and 0.075 in 2009 and 0.113, 0.121, 0.068, and 0.098 in 2020, respectively. The age-adjusted prevalence of hypertension, hyperlipidemia, diabetes, and mental disorders in Fukushima Prefecture was 0.106, 0.098, 0.044, and 0.072 in 2009 and 0.117, 0.119, 0.056, and 0.082 in 2020, respectively.Fig. 2 Annual age-adjusted prevalence of hypertension, hyperlipidemia, diabetes, and mental disorders by areas (males and females, 0–74 years). (a) Hypertension, (b) hyperlipidemia, (c) diabetes, (d) mental disorders. The prevalence represents those belonging to the health insurance societies scheme.
Fig. 2
Age-adjusted prevalence by age and sex groups for each disease in the whole of Japan and Fukushima Prefecture from 2009 to 2020 is shown in Fig. 3, Fig. 4 . Age-adjusted prevalence of hypertension, hyperlipidemia, and diabetes were generally higher in males than in females in the 40–74 years age group, and the differences tended to increase over time in the whole of Japan. In contrast, mental disorders were higher in females than in males in the 40–74 years age group, and the prevalence in the 0–39 years age group has been increasing in recent years. In Fukushima Prefecture, mental disorders showed an increasing trend in 2011–2012 for both age and sex groups, followed by a decreasing trend.Fig. 3 Annual age-adjusted prevalence of hypertension, hyperlipidemia, diabetes, and mental disorders by age group and sexes (the whole of Japan). (a) Hypertension, (b) hyperlipidemia, (c) diabetes, (d) mental disorders. The prevalence represents those belonging to the health insurance societies scheme.
Fig. 3
Fig. 4 Annual age-adjusted prevalence of hypertension, hyperlipidemia, diabetes, and mental disorders by age groups and sexes (Fukushima Prefecture). (a) Hypertension, (b) hyperlipidemia, (c) diabetes, (d) mental disorders. The prevalence represents those belonging to the health insurance societies scheme.
Fig. 4
3.2 Age-adjusted prevalence before and after the Fukushima disaster
Table 1 shows the ratios of the age-adjusted prevalence of NCDs in the 0–74 years group in the whole of Japan; Iwate, Miyagi, and Fukushima prefectures; and Hamadori, Nakadori, and Aizu regions before and after the Fukushima disaster. These differences are listed in Table S5. The table also shows the values normalized to those for the whole of Japan. After the Fukushima disaster, the increases in the prevalence of hypertension, hyperlipidemia, and diabetes were significantly higher among the Fukushima Prefecture than among the whole of Japan in all three time periods, including 2011–2013, 2014–2016, and 2017–2019 (excluding hyperlipidemia in 2011–2013). The prevalence ratios in 2011–2013 were 1.133 (95% uncertainty interval (UI): 1.081–1.188) for hypertension, 1.053 (0.980–1.132) for hyperlipidemia, 1.212 (1.082–1.344) for diabetes, and 1.084 (0.961–1.212) for mental disorders. Furthermore, in Hamadori, the prevalence of hypertension, diabetes, and mental disorders generally increased over the 9-year period, while in Nakadori, significant increases in the prevalence of hyperlipidemia and diabetes were observed in 2014–2016 and 2017–2019. In Aizu, there were also significant increases in the prevalence of hypertension in 2011–2013 and 2014–2016 and of diabetes in 2014–2016 and 2017–2019. In contrast, a slight downward trend was observed for hyperlipidemia in 2011–2013. No significant increases or decreases were observed for these four diseases in Iwate and Miyagi prefectures (excluding hypertension in 2014–2016 and diabetes in 2011–2013 in Iwate Prefecture).Table 1 Ratio of age-adjusted prevalence in three time periods to that in 2009–2010 (males and females, 0–74 years). Arithmetic mean (95% uncertainty interval). The prevalence represents those belonging to the health insurance societies scheme.
Table 1Area Hypertension Hyperlipidemia Diabetes Mental disorders
2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010 2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010 2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010 2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010
The whole of Japan 1.007 (0.971–1.045) 1.014 (0.983–1.047) 1.017 (0.986–1.049) 1.030 (0.994–1.067) 1.047 (1.024–1.071) 1.037 (1.004–1.070) 0.985 (0.957–1.013) 1.003 (0.970–1.038) 1.030 (0.999–1.062) 1.051 (0.979–1.129) 1.104 (1.026–1.189) 1.174 (1.092–1.263)
Iwate Prefecture 1.056 (0.988–1.127) 0.946 (0.899–0.998) 0.952 (0.897–1.011) 1.187 (0.971–1.480) 1.187 (0.973–1.473) 1.190 (0.975–1.476) 1.129 (1.052–1.214) 0.991 (0.917–1.071) 1.045 (0.976–1.121) 1.057 (0.712–1.705) 1.015 (0.688–1.628) 1.109 (0.751–1.781)
Miyagi Prefecture 0.991 (0.887–1.107) 0.965 (0.872–1.070) 0.925 (0.844–1.019) 1.126 (1.001–1.266) 1.090 (0.986–1.211) 1.093 (0.986–1.217) 1.071 (0.834–1.413) 1.047 (0.825–1.363) 1.019 (0.801–1.325) 0.997 (0.862–1.160) 1.010 (0.876–1.170) 1.037 (0.901–1.200)
Fukushima Prefecture 1.141 (1.107–1.175) 1.167 (1.141–1.194) 1.137 (1.118–1.155) 1.084 (1.021–1.155) 1.269 (1.194–1.350) 1.247 (1.163–1.340) 1.193 (1.067–1.319) 1.316 (1.289–1.343) 1.324 (1.291–1.357) 1.137 (1.034–1.244) 1.078 (1.031–1.128) 1.121 (1.073–1.171)
Hamadori 1.239 (1.113–1.365) 1.111 (1.092–1.131) 1.134 (1.101–1.168) 1.038 (0.855–1.275) 1.127 (0.942–1.364) 1.078 (0.907–1.302) 1.302 (0.963–1.782) 1.342 (1.032–1.806) 1.249 (0.963–1.679) 1.369 (1.031–1.721) 1.377 (1.236–1.532) 1.535 (1.371–1.714)
Nakadori 0.962 (0.876–1.062) 1.063 (0.968–1.175) 1.012 (0.920–1.119) 1.051 (0.887–1.249) 1.389 (1.199–1.632) 1.335 (1.147–1.576) 1.086 (0.859–1.333) 1.360 (1.221–1.521) 1.390 (1.254–1.551) 1.087 (1.041–1.135) 1.041 (0.963–1.121) 1.107 (1.079–1.136)
Aizu 1.134 (1.040–1.231) 1.121 (1.064–1.181) 1.084 (1.025–1.146) 0.948 (0.899–0.999) 0.984 (0.923–1.048) 1.042 (0.986–1.099) 1.034 (0.907–1.160) 1.210 (1.060–1.364) 1.280 (1.232–1.331) 1.092 (1.018–1.171) 1.110 (1.053–1.171) 1.136 (1.072–1.204)
Iwate Prefecture vs the whole of Japan 1.049 (0.972–1.130) 0.933 (0.878–0.992) 0.937 (0.877–1.002) 1.153 (0.940–1.442) 1.134 (0.929–1.410) 1.148 (0.938–1.431) 1.147 (1.063–1.239) 0.988 (0.907–1.076) 1.014 (0.940–1.094) 1.007 (0.672–1.622) 0.921 (0.617–1.482) 0.946 (0.634–1.518)
Miyagi Prefecture vs the whole of Japan 0.984 (0.876–1.104) 0.951 (0.856–1.060) 0.910 (0.827–1.007) 1.094 (0.966–1.237) 1.041 (0.939–1.159) 1.055 (0.946–1.178) 1.088 (0.845–1.434) 1.044 (0.820–1.364) 0.989 (0.778–1.293) 0.950 (0.806–1.122) 0.916 (0.777–1.079) 0.885 (0.753–1.040)
Fukushima Prefecture vs the whole of Japan 1.133 (1.081–1.188) 1.151 (1.107–1.197) 1.118 (1.080–1.158) 1.053 (0.980–1.132) 1.212 (1.136–1.295) 1.203 (1.113–1.298) 1.212 (1.082–1.344) 1.312 (1.260–1.364) 1.286 (1.236–1.338) 1.084 (0.961–1.212) 0.978 (0.897–1.064) 0.956 (0.876–1.040)
Hamadori vs the whole of Japan 1.230 (1.099–1.365) 1.096 (1.056–1.136) 1.116 (1.069–1.164) 1.008 (0.827–1.238) 1.076 (0.898–1.305) 1.040 (0.870–1.258) 1.322 (0.975–1.813) 1.338 (1.026–1.805) 1.212 (0.932–1.637) 1.304 (0.980–1.663) 1.249 (1.095–1.420) 1.309 (1.143–1.494)
Nakadori vs the whole of Japan 0.955 (0.863–1.062) 1.048 (0.948–1.163) 0.995 (0.900–1.105) 1.021 (0.856–1.217) 1.327 (1.142–1.561) 1.288 (1.100–1.522) 1.104 (0.870–1.355) 1.356 (1.211–1.523) 1.350 (1.213–1.511) 1.036 (0.951–1.123) 0.945 (0.847–1.048) 0.944 (0.872–1.017)
Aizu vs the whole of Japan 1.126 (1.025–1.231) 1.106 (1.041–1.175) 1.066 (0.999–1.136) 0.921 (0.863–0.981) 0.939 (0.877–1.004) 1.005 (0.944–1.069) 1.050 (0.919–1.183) 1.207 (1.053–1.364) 1.243 (1.181–1.306) 1.040 (0.939–1.147) 1.007 (0.919–1.100) 0.969 (0.881–1.063)
Regarding age and sex groups, there was a significant increase in the prevalence of these four diseases in Fukushima Prefecture during all three time periods in females aged 40–74 years: ratios in 2011–2013 were 1.280 (95%UI:1.214–1.350) for hypertension, 1.149 (1.055–1.257) for hyperlipidemia, 1.441 (1.139–1.806) for diabetes, and 1.283 (1.199–1.371) for mental disorders (Table 2 , ratios and Table S6, difference). An increase in the prevalence of hypertension and diabetes was observed in females aged 0–39 years. In contrast, among males aged 40–74 years, the prevalence of hyperlipidemia and diabetes increased significantly in 2014–2016 and 2017–2019, whereas the prevalence of mental disorders decreased in 2014–2016 and 2017–2019. Among males aged 0–39 years, there was an increase in the prevalence of hyperlipidemia in 2014–2016 and 2017–2019, whereas there was no significant change in the prevalence of other diseases (except hypertension in 2014–2016).Table 2 Ratio of age-adjusted prevalence in three time periods to that in 2009–2010 by age groups and sex (Fukushima Prefecture vs the whole of Japan). Arithmetic mean (95% uncertainty interval). The prevalence represents those belonging to the health insurance societies scheme.
Table 2Ages and sexes Hypertension Hyperlipidemia Diabetes Mental disorders
2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010 2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010 2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010 2011–2013 vs 2009–2010 2014–2016 vs 2009–2010 2017–2019 vs 2009–2010
Males and females (0–74 y) 1.133 (1.081–1.188) 1.151 (1.107–1.197) 1.118 (1.080–1.158) 1.053 (0.980–1.132) 1.212 (1.136–1.295) 1.203 (1.113–1.298) 1.212 (1.082–1.344) 1.312 (1.260–1.364) 1.286 (1.236–1.338) 1.084 (0.961–1.212) 0.978 (0.897–1.064) 0.956 (0.876–1.040)
Males (0–74 y) 1.037 (0.937–1.143) 1.013 (0.950–1.076) 0.997 (0.938–1.057) 1.067 (0.959–1.180) 1.380 (1.311–1.449) 1.363 (1.284–1.445) 1.035 (0.939–1.131) 1.099 (1.039–1.160) 1.088 (1.043–1.133) 1.015 (0.886–1.155) 0.937 (0.820–1.062) 0.948 (0.860–1.040)
Females (0–74 y) 1.272 (1.210–1.337) 1.324 (1.237–1.415) 1.259 (1.197–1.324) 1.151 (1.036–1.283) 1.219 (1.094–1.368) 1.221 (1.091–1.372) 1.453 (1.149–1.823) 1.577 (1.311–1.919) 1.514 (1.272–1.824) 1.165 (1.074–1.260) 1.089 (1.010–1.171) 1.046 (0.970–1.123)
Males and females (0–39 y) 1.023 (0.888–1.182) 0.990 (0.862–1.142) 1.137 (0.978–1.320) 1.037 (0.846–1.270) 1.334 (1.129–1.587) 1.439 (1.165–1.771) 1.285 (1.217–1.357) 1.209 (1.107–1.313) 1.416 (1.256–1.582) 1.019 (0.899–1.143) 1.000 (0.872–1.137) 1.010 (0.899–1.126)
Males and females (40–74 y) 1.139 (1.084–1.196) 1.160 (1.112–1.207) 1.117 (1.078–1.157) 1.053 (0.985–1.125) 1.201 (1.133–1.273) 1.182 (1.108–1.261) 1.206 (1.068–1.343) 1.320 (1.266–1.376) 1.280 (1.229–1.330) 1.137 (1.003–1.281) 0.972 (0.902–1.047) 0.928 (0.866–0.994)
Males (0–39 y) 0.945 (0.785–1.139) 0.799 (0.677–0.949) 0.947 (0.794–1.136) 0.924 (0.787–1.064) 1.227 (1.136–1.320) 1.232 (1.073–1.401) 1.044 (0.839–1.321) 1.074 (0.850–1.373) 1.214 (0.970–1.537) 1.050 (0.904–1.208) 1.021 (0.851–1.202) 1.064 (0.936–1.195)
Males (40–74 y) 1.042 (0.939–1.146) 1.026 (0.961–1.091) 1.000 (0.939–1.061) 1.092 (0.980–1.208) 1.404 (1.326–1.482) 1.385 (1.303–1.469) 1.033 (0.933–1.136) 1.099 (1.040–1.160) 1.080 (1.035–1.126) 0.998 (0.860–1.150) 0.890 (0.796–0.997) 0.878 (0.791–0.976)
Females (0–39 y) 1.092 (0.858–1.329) 1.195 (1.048–1.350) 1.216 (1.099–1.337) 1.222 (0.777–2.101) 1.291 (0.812–2.230) 1.505 (0.942–2.594) 1.609 (1.131–2.345) 1.327 (0.936–1.940) 1.605 (1.135–2.340) 1.009 (0.900–1.123) 1.011 (0.895–1.135) 0.994 (0.884–1.112)
Females (40–74 y) 1.280 (1.214–1.350) 1.330 (1.237–1.428) 1.262 (1.200–1.327) 1.149 (1.055–1.257) 1.218 (1.112–1.337) 1.209 (1.103–1.330) 1.441 (1.139–1.806) 1.605 (1.347–1.924) 1.515 (1.290–1.801) 1.283 (1.199–1.371) 1.150 (1.069–1.233) 1.091 (1.037–1.144)
3.3 Age-adjusted prevalence before and after the COVID-19 outbreak
Tables 3 and S7 show the ratios and differences in age-adjusted prevalence of each disease by sex and age group among the whole of Japan before and after the COVID-19 outbreak. The prevalence of all four diseases generally increased, with the ratios among males and females aged 0–74 years being 1.058 (95% UI: 1.032–1.087) for hypertension, 1.075 (1.013–1.143) for hyperlipidemia, 1.038 (1.012–1.066) for diabetes, and 1.033 (1.027–1.039) for mental disorders. Among the age and sex groups, the ratios of actual to estimated values for these four diseases were particularly high among males aged 20–39 years and 0–39 years (ratio among males aged 0–39 years: 1.098 (1.072–1.126) for hypertension, 1.143 (1.009–1.309) for hyperlipidemia, 1.060 (1.040–1.080) for diabetes, and 1.041 (1.032–1.050) for mental disorders). In contrast, the differences in prevalence were large for males and females aged 40–64, 64–74, and 40–74 years.Table 3 Ratio of observed to estimated prevalence in 2020 by age groups and sex (the whole of Japan). Arithmetic mean (95% uncertainty interval). The prevalence represents those belonging to the health insurance societies scheme.
Table 3Ages and sexes Hypertension Hyperlipidemia Diabetes Mental disorders
Males and females (0–74) 1.058 (1.032–1.087) 1.075 (1.013–1.143) 1.038 (1.012–1.066) 1.033 (1.027–1.039)
Males and females (0–39) 1.096 (1.077–1.116) 1.132 (0.997–1.294) 1.042 (1.029–1.055) 1.038 (1.028–1.048)
Males and females (40–74) 1.056 (1.027–1.084) 1.069 (1.017–1.123) 1.038 (1.009–1.069) 1.027 (1.018–1.037)
Males (0–74) 1.050 (1.023–1.077) 1.076 (1.022–1.135) 1.039 (1.015–1.065) 1.035 (1.022–1.048)
Males (0–39) 1.098 (1.072–1.126) 1.143 (1.009–1.309) 1.060 (1.040–1.080) 1.041 (1.032–1.050)
Males (40–74) 1.047 (1.017–1.079) 1.067 (1.024–1.112) 1.037 (1.009–1.066) 1.027 (1.003–1.051)
Males (0–19) 1.044 (1.018–1.070) 1.092 (0.943–1.274) 1.041 (1.011–1.071) 1.002 (0.989–1.016)
Males (20–39) 1.104 (1.074–1.134) 1.146 (1.011–1.310) 1.062 (1.039–1.084) 1.073 (1.054–1.094)
Males (40–64) 1.062 (1.023–1.104) 1.083 (1.029–1.143) 1.053 (1.017–1.090) 1.029 (1.000–1.060)
Males (65–74) 1.013 (1.002–1.025) 1.024 (1.006–1.043) 1.005 (0.979–1.031) 1.013 (0.994–1.034)
Females (0–74) 1.072 (1.047–1.097) 1.071 (1.011–1.139) 1.036 (1.004–1.069) 1.030 (1.022–1.039)
Females (0–39) 1.087 (1.016–1.164) 1.094 (0.984–1.224) 1.020 (0.975–1.069) 1.034 (1.021–1.046)
Females (40–74) 1.071 (1.048–1.094) 1.069 (1.013–1.129) 1.038 (1.008–1.070) 1.027 (1.016–1.038)
Females (0–19) 1.016 (0.935–1.108) 1.016 (0.952–1.086) 0.953 (0.862–1.061) 1.009 (0.997–1.022)
Females (20–39) 1.098 (1.024–1.180) 1.105 (0.984–1.249) 1.031 (0.980–1.084) 1.046 (1.029–1.062)
Females (40–64) 1.088 (1.043–1.135) 1.090 (1.013–1.175) 1.060 (1.007–1.117) 1.027 (1.018–1.036)
Females (65–74) 1.046 (1.013–1.079) 1.035 (1.003–1.067) 1.006 (0.981–1.034) 1.027 (1.004–1.052)
The ratios and differences in the prevalence of mental disorders before and after the COVID-19 outbreak by prefecture in males aged 0–39 years were significantly negatively correlated with the population density and COVID-19 infection rate in 2020 (Fig. 5 and Tables S8 and S9). In contrast, the population density and COVID-19 infection rates were not significantly associated with the prevalence ratios and differences in other diseases.Fig. 5 Correlations between population density or infection rate and the ratio of observed to the estimated age-adjusted prevalence of mental disorders in 2020 among 47 prefectures (males, 0–39 years). (a) Population density, (b) infection rate. ρ: Spearman's rank correlation coefficient. UI: uncertainty interval. The prevalence represents those belonging to the health insurance societies scheme.
Fig. 5
Table 4 shows the prevalence ratios in age-adjusted prevalence of each disease among Iwate, Miyagi, and Fukushima prefectures and three regions within Fukushima before and after the COVID-19 outbreak and standardized values for the whole of Japan. There were no significant changes for any of the diseases in males and females aged 0–74 years, except for a significant decrease in the prevalence of diabetes in Fukushima Prefecture: 0.969 (0.945–0.995). The prevalence of hypertension and diabetes by sex and age group decreased significantly in Fukushima Prefecture, especially among males (hypertension: males aged 0–74 years, males and females aged 0–39 years, males aged 0–39 years, males aged 40–74 years; diabetes: males aged 0–74 years, males and females aged 0–39 years, males and females aged 40–74 years, males aged 40–74 years).Table 4 Ratio of observed to estimated prevalence in 2020 (Iwate, Miyagi Fukushima prefectures, Hamadori, Nakadori, Aizu). Arithmetic mean (95% uncertainty interval). The prevalence represents those belonging to the health insurance societies scheme.
Table 4Area Ages and sexes Hypertension Hyperlipidemia Diabetes Mental disorders
Iwate Prefecture Males and females (0–74) 1.008 (0.956–1.066) 0.982 (0.924–1.044) 0.972 (0.924–1.022) 0.982 (0.938–1.031)
Miyagi Prefecture Males and females (0–74) 1.012 (0.921–1.118) 0.967 (0.900–1.042) 0.966 (0.890–1.051) 0.979 (0.945–1.014)
Fukushima Prefecture Males and females (0–74) 0.989 (0.953–1.026) 1.007 (0.926–1.098) 0.969 (0.945–0.995) 0.980 (0.939–1.022)
Hamadori Males and females (0–74) 0.943 (0.884–1.006) 1.011 (0.897–1.145) 1.010 (0.935–1.092) 0.901 (0.770–1.064)
Nakadori Males and females (0–74) 0.994 (0.929–1.067) 1.007 (0.900–1.136) 0.950 (0.894–1.013) 0.966 (0.846–1.112)
Aizu Males and females (0–74) 1.005 (0.974–1.038) 0.934 (0.826–1.064) 1.001 (0.806–1.277) 0.926 (0.854–1.006)
Fukushima Prefecture Males (0–74 y) 0.972 (0.968–0.976) 1.003 (0.934–1.079) 0.957 (0.941–0.974) 0.975 (0.915–1.041)
Fukushima Prefecture Females (0–74 y) 1.016 (0.927–1.114) 1.013 (0.918–1.122) 0.991 (0.899–1.096) 0.981 (0.938–1.027)
Fukushima Prefecture Males and females (0–39 y) 0.868 (0.822–0.917) 1.020 (0.872–1.213) 0.984 (0.969–0.999) 0.979 (0.902–1.065)
Fukushima Prefecture Males and females (40–74 y) 0.996 (0.958–1.035) 1.005 (0.932–1.089) 0.968 (0.941–0.997) 0.981 (0.946–1.017)
Fukushima Prefecture Males (0–39 y) 0.871 (0.810–0.939) 1.077 (0.861–1.384) 0.954 (0.886–1.032) 0.979 (0.905–1.065)
Fukushima Prefecture Males (40–74 y) 0.979 (0.972–0.986) 0.995 (0.944–1.048) 0.957 (0.942–0.972) 0.972 (0.923–1.023)
Fukushima Prefecture Females (0–39 y) 0.893 (0.706–1.178) 0.947 (0.804–1.132) 1.043 (0.907–1.213) 0.976 (0.895–1.068)
Fukushima Prefecture Females (40–74 y) 1.023 (0.935–1.126) 1.021 (0.926–1.132) 0.983 (0.891–1.092) 0.987 (0.910–1.073)
Iwate Prefecture vs the whole of Japan Males and females (0–74) 0.953 (0.898–1.013) 0.914 (0.840–0.992) 0.936 (0.883–0.991) 0.951 (0.907–0.998)
Miyagi Prefecture vs the whole of Japan Males and females (0–74) 0.956 (0.867–1.059) 0.900 (0.820–0.989) 0.930 (0.853–1.018) 0.948 (0.915–0.982)
Fukushima Prefecture vs the whole of Japan Males and females (0–74) 0.935 (0.893–0.977) 0.938 (0.844–1.037) 0.934 (0.899–0.969) 0.948 (0.908–0.990)
Hamadori vs the whole of Japan Males and females (0–74) 0.891 (0.832–0.957) 0.941 (0.822–1.078) 0.973 (0.897–1.057) 0.872 (0.745–1.031)
Nakadori vs the whole of Japan Males and females (0–74) 0.940 (0.872–1.013) 0.938 (0.825–1.067) 0.915 (0.856–0.982) 0.935 (0.819–1.077)
Aizu vs the whole of Japan Males and females (0–74) 0.950 (0.912–0.990) 0.869 (0.758–1.002) 0.964 (0.774–1.231) 0.896 (0.826–0.974)
Fukushima Prefecture vs the whole of Japan Males (0–74 y) 0.926 (0.902–0.951) 0.933 (0.854–1.019) 0.921 (0.894–0.948) 0.942 (0.882–1.007)
Fukushima Prefecture vs the whole of Japan Females (0–74 y) 0.948 (0.863–1.043) 0.947 (0.842–1.064) 0.957 (0.863–1.064) 0.952 (0.910–0.998)
Fukushima Prefecture vs the whole of Japan Males and females (0–39 y) 0.792 (0.749–0.839) 0.908 (0.735–1.124) 0.945 (0.927–0.963) 0.944 (0.868–1.025)
Fukushima Prefecture vs the whole of Japan Males and females (40–74 y) 0.944 (0.900–0.989) 0.941 (0.859–1.031) 0.933 (0.895–0.971) 0.954 (0.920–0.991)
Fukushima Prefecture vs the whole of Japan Males (0–39 y) 0.793 (0.735–0.858) 0.946 (0.724–1.241) 0.900 (0.833–0.976) 0.940 (0.868–1.023)
Fukushima Prefecture vs the whole of Japan Males (40–74 y) 0.935 (0.907–0.964) 0.933 (0.872–0.997) 0.923 (0.894–0.953) 0.946 (0.895–1.000)
Fukushima Prefecture vs the whole of Japan Females (0–39 y) 0.822 (0.640–1.088) 0.868 (0.710–1.068) 1.023 (0.884–1.196) 0.945 (0.865–1.035)
Fukushima Prefecture vs the whole of Japan Females (40–74 y) 0.956 (0.870–1.055) 0.956 (0.854–1.074) 0.948 (0.855–1.056) 0.961 (0.886–1.047)
In terms of standardized values for the whole of Japan, the change in the prevalence of hypertension was significantly lower among some sex and age groups in the areas affected by the Fukushima disaster than the whole of Japan: males and females aged 0–74 years in Fukushima Prefecture, Hamadori, and Aizu; males aged 0–74 years, males and females aged 0–39 years, males and females aged 40–74 years, males aged 0–39 years, males aged 40–74 years in Fukushima Prefecture. The changes in the prevalence of diabetes were also significantly lower: males and females aged 0–74 years in Iwate Prefecture, Fukushima Prefecture, and Nakadori; males aged 0–74 years, males and females aged 0–39 years, males and females aged 40–74 years, males aged 0–39 years, and males aged 40–74 years in Fukushima Prefecture. The change in prevalence of hyperlipidemia was significantly lower among males and females aged 0–74 years in Iwate and Miyagi prefectures, males aged 40–74 years in Fukushima Prefecture than the whole of Japan. The change in prevalence of mental disorders was significantly lower in males and females aged 0–74 years in Iwate, Miyagi, and Fukushima prefectures and Aizu 0–74, females aged 0–74 years, and males and females 40–74 years in Fukushima Prefecture than the whole of Japan.
4 Discussion
In this study, we evaluated the annual changes in the prevalence of hypertension, hyperlipidemia, diabetes, and mental disorders among employees of companies and their dependent family members belonging to the HIS scheme before and after the Fukushima disaster and the COVID-19 outbreak in Japan. Furthermore, we identified the vulnerable age and sex groups of people who experienced significant secondary health effects after the disaster and pandemic.
The analysis of the prevalence of these diseases after the Fukushima disaster by age and sex groups revealed that hypertension, hyperlipidemia, and diabetes generally increased in Fukushima Prefecture over the 9-year period following the disaster. Significant increases in hypertension, diabetes, and mental disorders were observed in Hamadori, hyperlipidemia and diabetes in Nakadori, and hypertension and diabetes in Aizu. Among the four diseases, the increased prevalence of diabetes was higher than that of the other diseases. Diabetes increased consistently in all regions.
After the Fukushima disaster, a significant increase in the prevalence of diabetes was consistently reported in multiple previous studies targeting various municipalities where evacuation orders were issued [[3], [4], [5],8]. The findings of our study are consistent with those of previous studies. For hyperlipidemia, previous studies reported an increase among evacuees within 4 years after the disaster but no significant increase in the 7-year data [3,8]. For hypertension, some studies that used a definition based on blood pressure alone have reported a risk decrease [3,8], whereas a study that used a definition based on blood pressure as well as the use of antihypertensive medication reported an increased risk [5]. Regarding psychological distress, a 5-fold risk increase was reported in the first year after the disaster, followed by an improvement trend [29]. Notably, this study highlighted that the worsening prevalence of these NCDs was not observed only in Hamadori, which was significantly affected by the Fukushima disaster, but also in Nakadori and Aizu. This is consistent with previous findings on non-evacuees [3,8]. The results suggest that the Fukushima disaster may have secondary physical health effects not only on evacuees but also on the wider population. The association between evacuation and increased prevalence of NCDs may be related to the exemption of medical expenses for the evacuees, in addition to behavioral changes related to lifestyle habits, such as exercise and diet, as well as to occupational and community changes [30]. However, the widespread increase in the prevalence of NCDs in Fukushima Prefecture cannot be explained by the exemption of medical expenses. The reasons for secondary physical health effects among non-evacuees are not clear; however, social support for non-evacuees is more likely to be overlooked in the aftermath of a disaster than that for evacuees. Considering that these diseases generally did not change in the Iwate and Miyagi prefectures that were affected by the GJET, but not largely by the Fukushima disaster, it is possible that the increase in the prevalence of these diseases observed over a wide area in Fukushima Prefecture was related to the long-term evacuation, hosting of evacuees, radiation anxiety, and socioeconomic changes in the residents of Fukushima Prefecture following the disaster, possibly resulting in lifestyle changes among residents. Hamadori, where a significant increase in the prevalence of mental disorders was observed, includes not only municipalities whose residents were ordered to evacuate, but also those that accepted the majority of evacuees. It has also been suggested that psychological distress among residents of host communities may be caused by the perception that evacuees are a threat to their social identity [9]. Strong radiation anxiety was observed not only in Hamadori but also Nakadori or Aizu, especially just after the Fukushima disaster [31], possibly resulting in behavioral changes (e.g., avoidance of radiation) and worsening mental health [32].
The increase in the prevalence of these four diseases in the post-disaster period was most pronounced among females and those aged 40–74 years. Psychological distress risk has been reported to be higher among elderly females [33]. Most of the previous studies after the Fukushima disaster have assessed the impact of evacuation on NCDs [[3], [4], [5],8], and there are limited studies that specifically discuss which age and sex groups experienced significant changes in prevalence after the disaster based on the same dataset. The strength of this study is that, based on the same dataset, changes in the prevalence after the disaster were assessed by age and sex groups to identify populations vulnerable to secondary health effects.
In contrast, after the COVID-19 outbreak, the prevalence of all four diseases increased nationwide, especially in males aged 20–39 and 0–39 years. Among the four diseases, the prevalence increase post-disaster in hyperlipidemia was particularly higher than that in the other diseases. To date, previous findings on NCDs risks after COVID-19 have not been coherent. While some population have improved their lifestyle habits [11,12], others have decreased their physical activity levels due to lockdown and other behavioral restrictions [34]. Molarius et al. [35] reported no statistically significant changes in the prevalence of high blood pressure before and after the COVID-19 outbreak in Sweden. A systematic review regarding the impact of the COVID-19-related lockdown on glycemic control among patients with type 1 and type 2 diabetes reported that glycemic values in patients with type 1 diabetes improved, but those in patients with type 2 diabetes worsened during the COVID-19 lockdown [14]. An increasing burden of mental disorders due to the COVID-19 pandemic has been reported among the public worldwide [15]. The increased prevalence of mental disorders among the young generation after the COVID-19 outbreak found in this study was consistent with a previous study [36] reporting that the increase in suicide after the COVID-19 outbreak was also risen especially among young generations in Japan possibly due to a complex reasons including unsecure working conditions and psychological depression associated with the reopening of school. In Japan, since the early days of the COVID-19 pandemic, active efforts have been made to raise awareness of exercise to promote physical and psychological health [37,38]. Despite these social awareness efforts, our study showed that the prevalence of NCDs were increasing in Japan, especially among young males.
Interestingly, the age and sex groups that showed large increases in the prevalence of these diseases differed between the Fukushima disaster and COVID-19 outbreak. After the Fukushima disaster, the affected people had often experienced family separation, fragmentation of communication, job loss, and behavioral changes due to radiation anxiety [32,39,40]. In contrast, after the COVID-19 outbreak, not only were there risk factors such as lack of exercise and worsening of dietary and drinking habits associated with measures such as stay-at-home requests, but also preventive factors such as reduced drinking opportunities and lifestyle modification. These differences between the Fukushima disaster and the COVID-19 outbreak may have made a difference in the high-risk group regarding NDCs. The pandemic may have resulted in polarization, with some individuals experiencing improved health and others experiencing worsened health. The young Japanese male population tends to have lower health practices than those of other age and sex groups [41], possibly resulting in significant secondary health effects due to the pandemic. In addition, the prevalence ratios of mental disorders before and after the outbreak were significantly negatively correlated with population density and COVID-19 infection rates. This suggests that the increased prevalence of mental disorders was not mainly attributed to the burden among patients with COVID-19 but also attributed to other social factors. Interestingly, the results of this study were in contrast to that of a previous study reporting that increases in major depressive and anxiety disorders in each country were significantly positively correlated with the rate of COVID-19 infection [15]. Although there were some differences in the timing of the declaration of the state of emergency, it was issued in all prefectures in 2020 in Japan. Therefore, the differences in the impact of social activity restrictions associated with these regulations were probably small among the prefectures. There are two possible reasons for the negative association between population density and the increased rate of mental disorders found in this study. First, the availability of mental health-promotion services and social infrastructure may differ among prefectures; urban areas may have had better access to mental health-promotion services and social infrastructure, even under restricted social activities during the COVID-19 pandemic. Second, there may be differences in the cultural worldviews of infection and related preventive behaviors among prefectures. It is possible that the social environment toward COVID-19 infection, shaped by this cultural worldview, had a negative impact on mental health among the population. Differences in strict cultural worldviews among prefectures in Japan and their relationship with mental disorders remain unclear and require further analysis.
The secondary effects of the COVID-19 outbreak in areas affected by the Fukushima disaster showed different trends from the whole of Japan. The prevalence of hypertension and diabetes in Fukushima Prefecture (especially among males) after the COVID-19 outbreak showed an improving trend compared to the pre-outbreak period. The changes in the prevalence of hypertension in Fukushima Prefecture (especially among men), diabetes in Fukushima Prefecture (especially among males) and Iwate Prefecture, hyperlipidemia in Iwate and Miyagi prefectures and Fukushima Prefecture (males aged 40–74 years old), and mental disorders in Iwate, Miyagi, and Fukushima prefectures were lower than those in the whole of Japan. This represents that the areas that experienced the Fukushima disaster were either less affected overall than the whole of Japan during the subsequent COVID-19 outbreak, or showed improvement in hypertension and diabetes. A previous study reported that after the COVID-19 outbreak, psychological distress was not worsened and problem drinking behavior was improved among residents who lived in municipalities spanning evacuation order areas during the Fukushima disaster [11]. There are two interpretations of the fact that areas that experienced the Fukushima disaster were less affected by the COVID-19 outbreak than other areas. First, NCDs had already been rising after the Fukushima disaster, so the impact of the COVID-19 outbreak was small. Disasters can cause fatalities among vulnerable populations and subsequently reduce survivors’ mortality rate (i.e., a harvesting effect) [42]. There may be similar effects on secondary health impacts. Second, the people affected by the Fukushima disaster may have unique and resilient characteristics and be more conscious about their health [43]. This study demonstrates the importance of providing support for secondary health effects on NCDs in the aftermath of disasters and pandemics. Since significantly affected age and sex groups are unique to each disaster or pandemic, tailor-made public health support is considered important. Specifically, it would help the central and local governments develop strategies such as creating a point of contact to increase support for vulnerable populations, publicizing information about that support to the populations, and establishing a support system of local health workers. Furthermore, while COVID-19 infections have a high fatality rate among the elderly [44], NCDs are more likely to affect young people. This means that infection control measures, including stay-at-home requests and other restrictions on socioeconomic activities, may be beneficial in reducing infection among the elderly under 75 years of age but may pose greater risks for the young regarding secondary health effects. Comprehensive public health measures in addition to infection control are required based on the characteristics of the population while considering the differences in the balance between risk-benefits and non-infectious secondary health effects among the populations.
This study has some limitations. First, the target population was employees of large companies and their dependent family members. The findings of this study cannot be generalized to the entire population and other countries. Furthermore, only the population aged 74 years and below were included in this study. Although COVID-19 isolation measures are known to worsen cognitive and mental health among people with dementia [45], this study was unable to assess the secondary health effects among the population aged 75 years and older. Second, we could not fully examine the possibility of changes in accessibility to healthcare facilities after the Fukushima disaster or the COVID-19 outbreak compared to accessibility to healthcare before the disaster and outbreak. However, it is unlikely that NCDs were overlooked before the disaster or outbreak since employees of companies, the target population of this study, generally receive regular health checkups. Third, we cannot rule out the possible discrepancies between the diagnosis obtained from the receipt data and the diagnosis in actual clinical settings. We reduced this bias by comparing the ratios of the prevalence of NCDs before and after the Fukushima disaster and COVID-19 outbreak, as well as the values in the whole of Japan. Fourth, this study was based on an ecological study design and did not identify the causes of changes in the prevalence of various diseases. Long-term cohort studies based on individuals exposed to disasters and pandemics are expected to assess the impact of disasters and pandemics on secondary health and identify the factors that influence this impact.
Despite these limitations, this study provided unique insights into the secondary health effects of the Fukushima disaster and COVID-19 outbreak and identified age and sex groups that were significantly affected by this disaster and pandemic.
5 Conclusions
This study used health insurance receipt data for employees of companies and their dependent family members from January 2009 to December 2020 to assess changes in the prevalence belonging to the HIS scheme of NCDs after the Fukushima disaster and COVID-19 outbreak in Japan compared to that before the disaster and outbreak and to identify age and sex groups particularly vulnerable to the resultant secondary health effects. Our study had strengths in that we used the same data to compare the secondary health effects of the Fukushima disaster and the COVID-19 outbreak on the affected people and to discuss the secondary health effects on people who experienced both the disaster and pandemic. The findings of this study are summarized as follows.● Age-adjusted prevalence of hypertension, hyperlipidemia, and diabetes increased over a 9-year period in the Fukushima Prefecture following the 2011 Fukushima disaster. An increase in the prevalence of mental disorders was also observed in the Hamadori region in the prefecture, which was most affected by the disaster.
● The increase in the prevalence of hypertension, hyperlipidemia, and diabetes was also observed in Nakadori and Aizu, which were less affected by the disaster, indicating that the Fukushima disaster had secondary health effects on wider populations.
● In contrast, no significant changes were observed for these four diseases in Iwate and Miyagi prefectures.
● After the Fukushima disaster, the prevalence ratios of these four diseases were particularly high among females aged 40–74 years.
● After the COVID-19 outbreak, the prevalence of these four diseases has increased significantly throughout Japan. In particular, the post-outbreak prevalence ratios of these four diseases were significantly higher among males aged 0–39 years than among females and males of other age groups, which differed from the age and sex groups vulnerable to secondary health effects after the Fukushima disaster.
● Post-outbreak prevalence ratios of mental disorders among males aged 0–39 years by prefecture were negatively and significantly associated with population density and COVID-19 infection rates.
● Changes in prevalence ratios of NCDs after the COVID-19 outbreak among the areas that experienced the Fukushima disaster were lower than in the whole of Japan.
● The population groups vulnerable to secondary health effects on NCDs after the Fukushima disaster or the COVID-19 outbreak are not common but are unique to the disaster or pandemic. Central or local governments must provide tailor-made public health support to the population in accordance with disasters and pandemics.
Notes
This article has already been registered for Preprints on medRxiv.
DOI is as follows: https://doi.org/10.1101/2022.05.18.22275293.
(https://www.medrxiv.org/content/10.1101/2022.05.18.22275293v1).
Author contribution
Michio Murakami: Conceptualization; Data curation; Methodology; Formal analysis; Visualization; Project administration; Writing –original draft, Shuhei Nomura: Conceptualization; Visualization; Writing –review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A Supplementary data
The following is the Supplementary data to this article:Multimedia component 1
Multimedia component 1
Data availability
The authors do not have permission to share data.
Acknowledgements
This work was supported by 10.13039/501100001691 JSPS KAKENHI grant number JP20H04354. We would like to thank Editage (www.editage.com) for English language editing and Dr. Kazuki Ide (Osaka University) for his advice.
Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.ijdrr.2022.103471.
==== Refs
References
1 Bromet E.J. Mental health consequences of the Chernobyl disaster J. Radiol. Prot. 32 2012 N71 N75 22394694
2 Moscona J.C. Peters M.N. Maini R. Katigbak P. Deere B. Gonzales H. Westley C. Baydoun H. Yadav K. Ters P. Jabbar A. Boulad A. Mahata I. Gadiraju T.V. Nelson R. Srivastav S. Irimpen A. The incidence, risk factors, and chronobiology of acute myocardial infarction ten years after Hurricane Katrina Disaster Med. Public Health Prep. 13 2018 217 222 29644946
3 Nomura S. Blangiardo M. Tsubokura M. Ozaki A. Morita T. Hodgson S. Postnuclear disaster evacuation and chronic health in adults in Fukushima, Japan: a long-term retrospective analysis BMJ Open 6 2016 e010080
4 Ebner D.K. Ohsawa M. Igari K. Harada K.H. Koizumi A. Lifestyle-related diseases following the evacuation after the Fukushima Daiichi nuclear power plant accident: a retrospective study of Kawauchi Village with long-term follow-up BMJ Open 6 2016 e011641
5 Ohira T. Nakano H. Nagai M. Yumiya Y. Zhang W. Uemura M. Sakai A. Hashimoto S. Changes in cardiovascular risk factors after the Great East Japan earthquake Asia Pac. J. Publ. Health 29 2017 47s 55s
6 Maeda M. Murakami M. Oe M. Fukushima nuclear disaster: multidimensional psychosocial issues and challenges to overcome them Nriagu J. Encyclopedia of Environmental Health second ed. 2019 Elsevier Oxford 121 131
7 Kuroda Y. Orui M. Hori A. Trends in suicide mortality in 10 years around the great east Japan earthquake: analysis of evacuation and non-evacuation areas in fukushima prefecture Int. J. Environ. Res. Publ. Health 18 2021
8 Nomura S. Murakami M. Ozaki A. Sawano T. Leppold C. Nishikawa Y. Saito H. Oikawa T. Tsubokura M. Comparative risk assessment of non-communicable diseases by evacuation scenario– a retrospective study in the 7 years following the Fukushima Daiichi nuclear power plant accident Glob. Health Action 14 2021 1918886
9 Kobayashi T. Yoshida K. Takebayashi Y. Goto A. Kumagai A. Murakami M. Social identity threats following the Fukushima nuclear accident and its influence on psychological distress Int. J. Disaster Risk Reduc. 37 2019 101171
10 Di Renzo L. Gualtieri P. Pivari F. Soldati L. Attinà A. Cinelli G. Cinelli G. Leggeri C. Caparello G. Barrea L. Scerbo F. Esposito E. De Lorenzo A. Eating habits and lifestyle changes during COVID-19 lockdown: an Italian survey J. Transl. Med. 18 2020 229 32513197
11 Murakami M. Kobayashi T. Oikawa Y. Goto S. Momoi M. Takebayashi Y. Ohira T. Yasumura S. Maeda M. Associations of the COVID-19 pandemic with the economic status and mental health of people affected by the Fukushima disaster using the difference-in-differences method: the Fukushima Health Management Survey SSM Popul Health 14 2021 100801
12 Samanta S. Banerjee J. Rahaman S.N. Ali K.M. Ahmed R. Giri B. Pal A. Dash S.K. Alteration of dietary habits and lifestyle pattern during COVID-19 pandemic associated lockdown: an online survey study Clin. Nutr. ESPEN 48 2022 234 246 35331497
13 Pellegrini M. Ponzo V. Rosato R. Scumaci E. Goitre I. Benso A. Belcastro S. Crespi C. De Michieli F. Ghigo E. Broglio F. Bo S. Changes in weight and nutritional habits in adults with obesity during the “lockdown” period caused by the COVID-19 virus emergency Nutrients 12 2020 1 11
14 Eberle C. Stichling S. Impact of COVID-19 lockdown on glycemic control in patients with type 1 and type 2 diabetes mellitus: a systematic review Diabetol. Metab. Syndrome 13 2021 95
15 Santomauro D.F. Mantilla Herrera A.M. Shadid J. Zheng P. Ashbaugh C. Pigott D.M. Abbafati C. Adolph C. Amlag J.O. Aravkin A.Y. Bang-Jensen B.L. Bertolacci G.J. Bloom S.S. Castellano R. Castro E. Chakrabarti S. Chattopadhyay J. Cogen R.M. Collins J.K. Dai X. Dangel W.J. Dapper C. Deen A. Erickson M. Ewald S.B. Flaxman A.D. Frostad J.J. Fullman N. Giles J.R. Giref A.Z. Guo G. He J. Helak M. Hulland E.N. Idrisov B. Lindstrom A. Linebarger E. Lotufo P.A. Lozano R. Magistro B. Malta D.C. Månsson J.C. Marinho F. Mokdad A.H. Monasta L. Naik P. Nomura S. O'Halloran J.K. Ostroff S.M. Pasovic M. Penberthy L. Reiner R.C. Jr. Reinke G. Ribeiro A.L.P. Sholokhov A. Sorensen R.J.D. Varavikova E. Vo A.T. Walcott R. Watson S. Wiysonge C.S. Zigler B. Hay S.I. Vos T. Murray C.J.L. Whiteford H.A. Ferrari A.J. Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic Lancet 398 2021 1700 1712 34634250
16 National Police Agency Police Measures and Damage from the Tohoku-Pacific Ocean Earthquake in 2011 (in Japanese) https://www.npa.go.jp/news/other/earthquake2011/pdf/higaijyoukyou.pdf
17 Reconstruction Agency Number of Disaster-Related Deaths 2021 (in Japanese) https://www.reconstruction.go.jp/topics/main-cat2/sub-cat2-6/20140526131634.html
18 Reconstruction Agency Changes in the Number of People Living in Evacuation Shelters 2022 (in Japanese) https://www.reconstruction.go.jp/topics/hikaku2.pdf
19 The National Diet of Japan The Official Report of the Fukushima Nuclear Accident Independent Investigation Commission 2012 https://www.nirs.org/wp-content/uploads/fukushima/naiic_report.pdf
20 Cucinotta D. Vanelli M. WHO declares COVID-19 a pandemic Acta Biomed. 91 2020 157 160 32191675
21 Ministry of Health Labour and Welfare COVID-19 Open Data 2022 (in Japanese) https://www.mhlw.go.jp/stf/covid-19/open-data.html
22 Ministry of Health Labour and Welfare Overview of the Health Care System in Japan Translated by the Authors (in Japanese) https://www.mhlw.go.jp/content/12400000/000377686.pdf
23 Ohisa M. Kimura Y. Matsuo J. Akita T. Sato T. Matsuoka T. Sakamune K. Katayama K. Huy Do S. Miyakawa Y. Tanaka J. Estimated numbers of patients with liver disease related to hepatitis B or C virus infection based on the database reconstructed from medical claims from 2008 to 2010 in Japan Hepatol. Res. 45 2015 1228 1240 25627814
24 Campos-Alberto E. Hirose T. Napatalung L. Ohyama M. Prevalence, comorbidities, and treatment patterns of Japanese patients with alopecia areata: a descriptive study using Japan medical data center claims database J. Dermatol. 2022 10.1111/1346-8138.16615
25 e-Stat, Population Demographic and Household Surveys Based on the Basic Resident Register 2021 (in Japanese) https://www.e-stat.go.jp/stat-search/files?page=1&toukei=00200241&tstat=000001039591
26 Statistics Bureau of Japan National Census 2021 (in Japanese) https://www.stat.go.jp/data/kokusei/2020/kekka.html
27 Ministry of Health Labour and Welfare Deaths by Prefecture - Age-Adjusted Mortality Rates by Prefecture in 2000 2002 (in Japanese) https://www.mhlw.go.jp/toukei/saikin/hw/jinkou/other/00sibou/1.html
28 Geospatial Information Authority of Japan National Area Survey by Prefecture and Municipality 2020 (in Japanese) https://www.gsi.go.jp/KOKUJYOHO/OLD-MENCHO-title.htm
29 Maeda M. Oe M. Mental health consequences and social issues after the Fukushima disaster, Asia Pac J Public Health 29 2017 36s 46s
30 Matsuyama Y. Tsuboya T. Bessho S.-i. Aida J. Osaka K. Copayment exemption policy and healthcare utilization after the Great East Japan Earthquake Tohoku J. Exp. Med. 244 2018 163 173 29503394
31 Suzuki S. Murakami M. Nishikiori T. Harada S. Annual changes in the Fukushima residents' views on the safety of water and air environments and their associations with the perception of radiation risks J. Radiat. Res. 59 2018 ii31 ii39 29528420
32 Maeda M. Oe M. Suzuki Y. Psychosocial effects of the Fukushima disaster and current tasks: differences between natural and nuclear disasters J. National Inst. Public He. 67 2018 50 58
33 Oe M. Fujii S. Maeda M. Nagai M. Harigane M. Miura I. Yabe H. Ohira T. Takahashi H. Suzuki Y. Yasumura S. Abe M. Three-year trend survey of psychological distress, posttraumatic stress, and problem drinking among residents in the evacuation zone after the Fukushima Daiichi Nuclear Power Plant accident [The Fukushima Health Management Survey], Psychiatry Clin. Neurosci. 70 2016 245 252
34 Tison G.H. Avram R. Kuhar P. Abreau S. Marcus G.M. Pletcher M.J. Olgin J.E. Worldwide effect of COVID-19 on physical activity: a descriptive study Ann. Intern. Med. 173 2020 767 770 32598162
35 Molarius A. Persson C. Living conditions, lifestyle habits and health among adults before and after the COVID-19 pandemic outbreak in Sweden - results from a cross-sectional population-based study BMC Publ. Health 22 2022 171
36 Tanaka T. Okamoto S. Increase in suicide following an initial decline during the COVID-19 pandemic in Japan Nat. Human Behav. 5 2021 229 238 33452498
37 Yamaguchi K. Takebayashi Y. Miyamae M. Komazawa A. Yokoyama C. Ito M. Role of focusing on the positive side during COVID-19 outbreak: mental health perspective from positive psychology Psychol. Trauma: Theory Res. Pract. Policy. 12 2020 S49 S50
38 Aung M.N. Yuasa M. Koyanagi Y. Aung T.N.N. Moolphate S. Matsumoto H. Yoshioka T. Sustainable health promotion for the seniors during COVID-19 outbreak: a lesson from Tokyo J. Infect. Dev. Ctries. 14 2020 328 331 32379708
39 Sone T. Nakaya N. Sugawara Y. Tomata Y. Watanabe T. Tsuji I. Longitudinal association between time-varying social isolation and psychological distress after the Great East Japan Earthquake Soc. Sci. Med. 152 2016 96 101 26851408
40 Suzuki Y. Yabe H. Yasumura S. Ohira T. Niwa S. Ohtsuru A. Mashiko H. Maeda M. Abe M. Psychological distress and the perception of radiation risks: the fukushima health management survey, bull World Health Organ 93 2015 598 605
41 Noguchi Y. Ueda K. Fukumoto K. Harada K. Ueda A. Wei C.-N. Relationship between perceived health status and health practices in the general adult population in Japan Open J. Prev. Med. 5 2015 11
42 Toulemon L. Barbieri M. The mortality impact of the August 2003 heat wave in France: investigating the ‘harvesting’ effect and other long-term consequences Popul. Stud. 62 2008 39 53
43 Takebayashi Y. Maeda M. Orui M. Nakajima S. Momoi M. Ito A. Sato H. Yasumura S. Ohto H. Resilience factors contributing to mental health among people affected by the Fukushima disaster: development of Fukushima Resilience Scale Front. Public Health 8 2020 159 32435630
44 COVID-19 Forecasting Team Variation in the COVID-19 infection-fatality ratio by age, time, and geography during the pre-vaccine era: a systematic analysis Lancet 399 2022 1469 1488 35219376
45 Suárez-González A. Rajagopalan J. Livingston G. Alladi S. The effect of COVID-19 isolation measures on the cognition and mental health of people living with dementia: a rapid systematic review of one year of quantitative evidence eClin Med. 39 2021
| 36465703 | PMC9707028 | NO-CC CODE | 2022-12-01 23:23:40 | no | Int J Disaster Risk Reduct. 2023 Jan 29; 84:103471 | utf-8 | Int J Disaster Risk Reduct | 2,022 | 10.1016/j.ijdrr.2022.103471 | oa_other |
==== Front
Respir Physiol Neurobiol
Respir Physiol Neurobiol
Respiratory Physiology & Neurobiology
1569-9048
1878-1519
Elsevier B.V.
S1569-9048(22)00159-8
10.1016/j.resp.2022.104000
104000
Article
Morphological and functional findings in COVID-19 lung disease as compared to Pneumonia, ARDS, and High-Altitude Pulmonary Edema
Zubieta-Calleja Gustavo R. a
Zubieta-DeUrioste Natalia a
de Jesús Montelongo de Jesus MontelongoFelipe b
Sanchez Manuel Gabriel Romo b
Campoverdi Aurio Fajardo c
Rocco Patricia Rieken Macedo de
Battaglini Denise f⁎
Ball Lorenzo fg
Pelosi Paolo fg
a High Altitude Pulmonary and Pathology Institute (HAPPI - IPPA), La Paz, Bolivia
b Hospital General de Ecatepec “Las Américas”, ISEM, Mexico
c Critical Care Unit, Viña del Mar, Chile
d Laboratory of Pulmonary Investigation, Carlos Chagas Filho Institute of Biophysics, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
e COVID-19 Virus Network, Ministry of Science, Technology, and Innovation, Brasilia, Brazil
f Anesthesia and Intensive Care, San Martino Policlinico Hospital, IRCCS for Oncology and Neurosciences, Genoa, Italy
g Department of Surgical Sciences and Integrated Diagnostics, University of Genoa, Genoa, Italy
⁎ Corresponding author.
29 11 2022
29 11 2022
1040002 10 2022
18 11 2022
27 11 2022
© 2022 Elsevier B.V. All rights reserved.
2022
Elsevier B.V.
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Coronavirus disease-2019 (COVID-19) may severely affect respiratory function and evolve to life-threatening hypoxia. The clinical experience led to the implementation of standardized protocols assuming similarity to severe acute respiratory syndrome (SARS-CoV-2). Understanding the histopathological and functional patterns is essential to better understand the pathophysiology of COVID-19 and then develop new therapeutic strategies. Epithelial and endothelial cell damage can result from the virus attack, thus leading to immune-mediated response. Pulmonary histopathological findings show the presence of Mallory bodies, alveolar coating cells with nuclear atypia, reactive pneumocytes, reparative fibrosis, intra-alveolar hemorrhage, moderate inflammatory infiltrates, micro-abscesses, microthrombus, hyaline membrane fragments, and emphysema-like lung areas. COVID-19 patients may present different respiratory stages from silent to critical hypoxemia, are associated with the degree of pulmonary parenchymal involvement, thus yielding alteration of ventilation and perfusion relationships. This review aims to: discuss the morphological (histopathological and radiological) and functional findings of COVID-19 compared to acute interstitial pneumonia, acute respiratory distress syndrome (ARDS), and high-altitude pulmonary edema (HAPE), four entities that share common clinical traits, but have peculiar pathophysiological features with potential implications to their clinical management.
Abbreviations
AIP, acute interstitial pneumonia
ARDS, acute respiratory distress syndrome
CO2, carbon dioxide
COVID-19, coronavirus disease 2019
CT, computed tomography
DAD, diffuse alveolar damage
DO2, oxygen delivery
ECCO2R, extracorporeal carbon dioxide removal
ECMO, extracorporeal membrane oxygenation
FiO2, fraction of inspired oxygen
HAPE, high altitude pulmonary edema
Hb, hemoglobin
HIF, hypoxia inducible factor
IIP, idiopathic interstitial pneumonia
LUS, lung ultrasound
MODS, multiple organ dysfunction
PaCO2, arterial partial pressure of carbon dioxide
PAO2, alveolar partial pressure of oxygen
PaO2, arterial partial pressure of oxygen
PEEP, positive end-expiratory pressure
SARS-CoV-2, severe acute respiratory syndrome coronavirus-2
SpO2, peripheral saturation of oxygen
V/Q, ventilation/ perfusion
VILI, ventilator-induced lung injury
Keywords
HAPE
SARS-CoV-2
COVID-19
lung sequelae
viral pneumonia
ARDS
==== Body
pmc1 Introduction
Coronavirus disease 2019 (COVID-19) is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus (Mitra et al., 2020) and is characterized by flue-type (variable) symptoms that could evolve to severe hypoxemia and subsequent death (Lin et al., 2020). Over 5% of infected patients develop pneumonia followed by acute respiratory distress syndrome (ARDS) thus requiring mechanical ventilation (Ferreira et al., 2021). Mortality has been reported as 0.3 deaths per 1,000 cases in patients aged between 5-17 years, 305 death per 1,000 cases in patients aged 85 or older (Wiersinga et al., 2020). A high incidence of different clinical presentations (Lavinio et al., 2021, Robba et al., 2021b), as well as cardiac (De Marzo et al., 2021), neurologic (Battaglini et al., 2020b), hepatic and renal injury (Lopes-Pacheco et al., 2021), have been reported yielding the definition of Multiple Organ Dysfunction Syndrome (MODS)-CoV-2 (Robba et al., 2020b). COVID-19 pneumonia histopathological findings have been reported different from (Batah and Fabro, 2021) or overlapped (Xu et al., 2020) with Acute Interstitial Pneumonia (AIP), ARDS, or High Altitude Pulmonary Edema (HAPE) (Pomara et al., 2020, Zubieta-Calleja and Zubieta-DeUrioste, 2021a, Zubieta-Calleja and Zubieta-DeUrioste, 2021b). These four diseases share similar clinical presentations, but also histopathological characteristics, and radiographic images. However, each of them has peculiar characteristics. Morphological and functional characteristics of these pathologies can vary depending on the specific environment (Zubieta-Calleja et al., 2021, Zubieta-Calleja and Zubieta-DeUrioste, 2022). This review aims to discuss the morphological (histopathological and radiological) and functional findings of COVID-19 compared with pneumonia, ARDS and HAPE as well as possible implications relevant to therapeutic strategies.
2 Histopathological findings
2.1 Acute Interstitial Pneumonia
Different microorganisms can cause pneumonia (or pneumonitis), classically defined as an acute disease marked by inflammation of lung tissue accompanied by infiltration of alveoli and often bronchioles with white blood cells (such as neutrophils) and fibrinous exudate (Travis et al., 2013) ( Table 1). The most recent anatomopathological classification dates back to 2013, including the broad definition of idiopathic interstitial pneumonia (IIP), incorporating more than 200 acute or chronic lung diseases with various degrees of inflammation and fibrosis (Travis et al., 2013). AIP is characterized by a severe idiopathic interstitial disease that includes the presence of clinical symptoms, bilateral lung infiltrates on radiographs, and the absence of identifiable etiology (Thille et al., 2013, Travis et al., 2013). In AIP, type I and I epithelial cells may present necrosis, erosion, fibrin exudation and ubiquitin-positive intracytoplasmic inclusion bodies (Mallory bodies) that are ubiquitin positive. AIP may evolve to ARDS and fibrosis (Thille et al., 2013).Table 1 Comparison of anatomopathological findings in AIP, ARDS, HAPE, and COVID-19.
Table 1Disease Anatomopathological findings
AIP Lung tissue derangement with alveolar collapse.
Hyaline membrane formation.
Polymorphonuclear and monocytes infiltration.
Pneumocytes characteristics in AIP include epithelial necrosis, erosion, fibrin exudation, Mallory bodies. Pneumocytes Ub+.
Early fibroblastic interstitial fibrosis, septal and para-septal reparative fibrosis.
ARDS Lung tissue derangement with alveolar collapse. Intra-alveolar and interstitial edema. Necrotic alveolar epithelial cells and exudates of serum proteins from the damaged and leaky capillaries.
Alveolar type I cell necrosis, alveolar type II cell proliferation, interstitial proliferation of fibroblast and myofibroblast or organizing interstitial fibrosis.
Hyaline tissue formation.
Reactive pneumocytes with nuclear atypia and hyperplasia. Mallory like inclusions in type 2 pneumocytes (Ub+).
Scattered neutrophilic infiltrates localized to terminal bronchioles and surrounding alveoli with evident confluence of infiltrates between adjacent lobules.
Thrombi are common in small/ medium pulmonary arterioles.
Abundant fibroblast in the interstitium, fibrosis and interstitial thickening by fibroblasts.
HAPE Diffuse alveolar edema including red blood cells, polymorphonuclear cells, and macrophages. Widened septa due to interstitial edema.
Occasional alveolar hyaline membranes.
Type II pneumocytes could be damaged (due to stress failure) and pulmonary surfactant impaired.
Congested pulmonary arterioles and capillaries. Intra-alveolar hemorrhage is present in scattered areas of lung parenchyma, particularly in the more severe cases.
Thrombi and fibrin clots plugging arterioles and capillaries.
COVID-19 Lung tissue derangement with alveolar collapse. Intra-alveolar hemorrhage. Alveolar rupture.
Hyaline tissue formation.
Polymorphonuclear and monocytes infiltration. Active viral replication within type 2 pneumocytes
Reactive pneumocytes with nuclear atypia and hyperplasia. Mallory like intracytoplasmic inclusions in type 2 pneumocytes. Masson’s bodies.
Microthrombi, vasculitis or vascular thrombosis.
Early fibroblastic interstitial fibrosis, septal and para-septal reparative fibrosis.
Table Legend: AIP, acute interstitial pneumonia; ARDS, Acute Respiratory Distress Syndrome; COVID-19, Coronavirus disease 2019; HAPE, high-altitude pulmonary edema.
2.2 ARDS
Diffuse alveolar damage (DAD) is considered the histological hallmark for the acute phase of ARDS (Table 1). DAD can present as typical histologic evolution of AIP, but the clinical evolution of AIP and DAD are different. Indeed, AIP and DAD constitute two independent entities that often coexist (Cardinal-Fernández et al., 2017). Three phases of traditional ARDS have been identified: 1) acute phase (<48 h) with interstitial and alveolar edema, 2) subacute phase (days 3-7) with the formation of hyaline membranes and fibrin deposits, and 3) proliferative phase (>7 days) with abundant fibroblast in the interstitium, fibrosis, and hyperplasia of type II pneumocytes (Mukhopadhyay and Parambil, 2012). Hyaline membrane formation and DAD with fibrin exudation and intraluminal granulation can be maintained due to epithelial cell damage (Guerin et al., 2015). Lung histopathology differs in “pulmonary and “extrapulmonary” ARDS (Pelosi et al., 2003), although both are characterized by increased lung weight compared to normal lungs (1.700-2.500 gr vs 800 gr, respectively) and presence of DAD, albeit with different distribution. Pulmonary ARDS primarily affects the alveolar epithelium, with damage occurring mainly in the intra-alveolar space, with alveolar flooding and areas of consolidation (Menezes et al., 2005, Rocco and Pelosi, 2008). In extrapulmonary ARDS, vascular endothelial cells are the first target of damage, with a subsequent increase in vascular permeability. The main pathologic alteration due to an indirect insult may be micro-vessel congestion and interstitial edema, with relative sparing of intra-alveolar spaces. The histopathological distinction between “pulmonary” and “extrapulmonary” ARDS has not been always feasible due to the frequent overlapping of these two clinical situations (Gattinoni et al., 2001, Goodman et al., 1999). DAD might evolve to fibrosis regardless of the underlying lung condition; however, its frequency, specific findings and pathogenesis might be disease-specific (Ackermann et al., 2020, Ball et al., 2021a, Kamp et al., 2022).
2.3 High Altitude Pulmonary Edema
HAPE is a non-cardiogenic pulmonary edema that develops after exposure to high altitudes (> 1,700 – 2,000 m above sea level) (Giesenhagen et al., 2019). It is assumed that there exists a genetic predisposition to HAPE and this might be associated with an exaggerated vascular response in patients susceptible to hypoxia and poor outcome in those with COVID-19 (Garrido et al., 2022). Histological findings of HAPE include marked diffused alveolar edema, congested pulmonary arterioles and capillaries, and occasional alveolar hyaline membranes (Droma et al., 2001) (Table 1). In most severe cases, histological findings may be compatible with DAD. In particular, the alveolar spaces are often distended with edema extended to alveoli, alveolar ducts, sacs, and respiratory bronchioles. The edema includes red blood cells, polymorphonuclear cells, and macrophages. Additionally, alveolar capillaries and pulmonary arterioles are congested with stasis. Intra-alveolar hemorrhage can be present in scattered areas of lung parenchyma in the more severe cases. Thrombi and fibrin clots plugging arterioles and capillaries can occasionally be found. The alveolar structures are usually maintained, but the septa are widened due to interstitial edema, while hyaline membranes can be eventually observed (Droma et al., 2001). HAPE is clearly associated with pulmonary artery hypertension, which manifests with changes in the pulmonary vascular endothelium characterized by 1) a transition form, from a quiescent state to an activated state with acquisition of adhesive capacity, 2) an aberrant proliferative and apoptosis-resistant phenotype, 3) a pro-inflammatory phenotype with the release of cytokines and chemokines, and 4) a fibrotic phenotype with the release of growth factors, angiotensin, and leptin (Huertas et al., 2018). Type II epithelial cells could be damaged and pulmonary surfactant impaired (Droma et al., 2001). Nevertheless, these lung histological changes reverse within 24 hours in younger patients and up to several days in older patients, never resulting in fibrosis.
2.4 COVID-19 pneumonia
Lung histopathology in COVID-19 pneumonia (Edler et al., 2020, Hanley et al., 2020, Pomara et al., 2020) is characterized by an inflammatory process (neutrophil and mononuclear cell infiltration), alveolar destruction and collapse, hyaline tissue formation, microhemorrhages, and micro-abscess possibly due to superinfection, which can result in extensive disruption of the lung parenchyma (Zubieta-Calleja and Zubieta-DeUrioste, 2021b). Pathological findings in the cells include intra-alveolar Masson Bodies, which are a typical hallmark of organizing pneumonia. Nuclear atypia of epithelial cells is also present (Zubieta-Calleja and Zubieta-DeUrioste, 2021b) (Table 1). Severe COVID-19 pneumonia shares some pathological similarities with “pulmonary” ARDS characterized by lung inflammation, presence of consolidation, some areas of atelectasis, changes in vascular permeability, and hypoxemia (Ackermann et al., 2020, Ranieri et al., 2012). Autopsy findings in severe COVID-19 pneumonia patients reveals the presence of DAD, fibrosis, bullae-filled necrotic lung tissue, chronic lung inflammation, edema in the bronchial mucosa, and thromboembolic events (Barisione et al., 2021, Barton et al., 2020, Edler et al., 2020, Wichmann et al., 2020). Three phases, similar to the traditional DAD, have been identified: 1) the acute/early exudative phase that is characterized by intra-alveolar edema and interstitial widening, with a peak of hyaline membrane (both diffuse and focal) formation 4-5 days after initial insult (Angeles Montero-Fernandez and Pardo-Garcia, 2021); 2) an organizing/mid proliferative phase, which is characterized by cellular fibroblastic proliferation, type II epithelial cell hyperplasia, and squamous metaplasia; and 3) late/fibrous phase DAD with honeycombing (Barisione et al., 2020). The hyaline membranes are in this stage fused into the alveolar septa and not easily recognizable (Angeles Montero-Fernandez and Pardo-Garcia, 2021). Post-mortem transbronchial lung cryobiopsies during mechanical ventilation confirmed the presence of "pneumocyte loss” (pneumolysis) with discontinuation of the alveolar epithelial lining, hyaline membranes, intra-alveolar fibrinous exudate, early interstitial fibrosis, obliteration of the alveolar structure by fibrosis, type 2 epithelial cell hyperplasia and atypia, Mallory-like intracytoplasmic inclusions in type 2 epithelial cells, micro-honeycombing, foci of bronchopneumonia, associated with vascularlysis, vasculitis or vascular thrombosis (Barisione et al., 2020). Vascular microthrombosis was described in conventional ARDS but occurs more frequently in COVID-19 (Ackermann et al., 2020).
2.4.1 Parenchymal destruction in severe COVID-19 pneumonia
Parenchymal destruction is the most severe evolution of COVID-19 pneumonia (Zubieta-Calleja and Zubieta-DeUrioste, 2021b). Irreversible remodeling of lung parenchyma, with fibrotic tissue adhering to the parietal pleura, was firstly described in pulmonary tuberculosis (Hall., 1938). It was therefore observed in COVID-19, in the context of a disease evolving to progressive hypoxemia, subsequent hypercapnia, and altered shunt (Zubieta-Calleja et al., 2021). A gel-like fluid has been found in autopsy lung samples in COVID-19, and the presence of hyaluronan has been reported (Hellman et al., 2020). COVID-19 patients present initially a predominance of advanced lesions in the lower lobes of both lungs. The probable explanation could be associated with aerodynamics and lung elasticity that initially respond with lower lobes distention on inhalation due to diaphragmatic contraction and direct main bronchi orientation, driving the SARS-CoV-2 directly to impact the lungs in that area (Yan et al., 2020). These findings in COVID-19 differentiates itself from HAPE, which occurs on a rapid ascent to high altitude in some individuals who can present an arterial partial pressure of oxygen (PaO2) as low as 30 mmHg but can fully recover within a few days. Also, in COVID-19, there is lung pneumocyte destruction followed by inflammation. HAPE-like edema, and some minor hemorrhage, giving rise to severe lung damage. However, HAPE-like edema in COVID-19 could be concomitantly present since the exclusion and destruction of respiratory units reduces the gas exchange surface significantly, simulating a high-altitude abrupt ascent. The normal remaining lung tissue can suffer edema as it results from extreme hypoxia, pulmonary hypertension, and stress failure of capillaries. The brownish color of the lungs probably results from the iron deposit due to hemolysis in the interstitium where fibrosis is forming.
3 Radiological findings
3.1 Acute Interstitial Pneumonia
Chest radiograph in AIP shows bilateral opacifications with air bronchograms and sparing of the costophrenic angles. Consolidation can appear during the organizing phase with a possible ground glass appearance and irregular lung opacities (Primack et al., 1993). Chest computed tomography (CT) findings of AIP patients reveal ground glass attenuation areas with a mosaic pattern and air space consolidation in dependent areas. This pattern results from alveolar septal edema, and hyaline membrane formation at histology, which are typical of the exudative phase (Palmucci et al., 2014). These signs are commonly bibasilar, sometimes diffuse or localized to the upper lobes. Intra-alveolar fibrosis with consolidation can be identified in the organizing phase with possible bronchiectasis formation and cyst. The latter are more common in the non-dependent lung areas (Palmucci et al., 2014) ( Table 2).Table 2 Comparison of radiological findings in AIP, ARDS, HAPE, and COVID-19.
Table 2Disease Radiological findings
AIP Ground grass attenuation areas with a mosaic pattern and air space consolidation in dependent areas, bibasilar, diffuse or localized in the upper lobes. Intra-alveolar fibrosis with consolidation in the organizing phase with possible bronchiectasis formation and cyst, especially in the non-dependent lung areas.
In “pulmonary” ARDS appearance is asymmetric, with a mix of consolidation and ground-glass opacification, possible pleural effusion, and emphysema. Predominance of air bronchograms and pneumomediastinum
In “extrapulmonary” ARDS predominantly symmetric ground-glass opacification are present with possible pleural effusion and emphysema. Ground-glass opalization and consolidation are greater in the central third of the lung than in the sternal or vertebral third without significant craniocaudal predominance. Consolidation is mainly distributed in the middle and basal levels as well as the vertebral position.
ARDS Heterogenous foci of consolidation, ground-glass opacities (with gravitational gradient), and crazy paving, with predominance in the posterior and basal areas. Possible air bronchograms and small pleural effusions.
In the organizing and fibrotic phase, traction bronchiectasis and reticulation may develop (anterior predominance).
HAPE Unilateral or Bilateral asymmetrical alveolo-interstitial opacities with relative sparing of the periphery and apices, usually marginated, with patchy distribution and initial confluence in the perihilar region, and prominent horizontal fissure, interlaced with normally aerated areas.
Possible lobar consolidation limited by horizontal fissure. Focal patchy air-space opacities/ consolidations. Pulmonary edema usually occurs in areas of high blood flow, with a patchy distribution.
COVID-19 Phenotype 1 (or L) usually presents as multiple focal over perfused ground glass opacities and normally aerated areas and is prevalent of mild to moderate diseases. Possible diversion of ventilation toward non-dependent aerated lung regions and reduction in pulmonary perfusion due to increased airway pressure, collapse of capillaries and/or microthrombosis and formation of no recruitable atelectasis.
Phenotype 2 (or H) shows a patchy ARDS-like pattern, with inhomogeneously distributed and hyper/hypo-perfused areas that is prevalent of severe disease. Increased lung weight and consolidated and non-aerated lung regions mainly distributed in the dependent lung regions. In these clinical conditions, areas with low V/Q persist but associated with areas of “true shunt”.
Phenotype (F) which represents a final evolution to fibrosis.
Table Legend: AIP, acute interstitial pneumonia; ARDS, Acute Respiratory Distress Syndrome; COVID-19, Coronavirus disease 2019. HAPE, high-altitude pulmonary edema.
3.2 ARDS
Studies have described morphological differences at chest CT scan in patients with pulmonary and extrapulmonary ARDS (Gattinoni et al., 2001, Goodman et al., 1999). “Pulmonary” ARDS appears to be asymmetric, with a mix of consolidation and ground-glass opacification, while “extrapulmonary” ARDS presents a symmetric ground-glass opacification. However, to date there has been no tomographic characteristic able to predict the etiology of ARDS (Desai et al., 1999), since direct and indirect injuries can coexist, making interpretation of the morphological pattern more difficult. In the organizing and fibrotic phase, traction bronchiectasis and reticulation may develop with an anterior predominance (Obadina et al., 2013) (Table 2). Typical patterns at CT scan in “pulmonary” and “secondary” ARDS are shown in Fig. 1 . Fig. 1 Typical pattern of “pulmonary” (left) and “secondary” (right) ARDS at CT scan.
Fig. 1
3.3 High Altitude Pulmonary Edema
Chest radiograph of patients with HAPE presents distinct patterns which can often be confused with cardiopulmonary diseases (Nair et al., 2021). Unilateral or bilateral asymmetrical alveolar-interstitial opacities with relative sparing of the periphery and apices have been identified in some patients (Zubieta-Calleja and Zubieta-DeUrioste, 2021a). The opacities could be marginated, and the distribution patchy, with initial confluence in the perihilar region, and prominent horizontal fissure. Other images showed lobar consolidation, limited by a horizontal fissure. Focal patchy air-space opacities/ consolidations, thus mimicking a tuberculosis-type pattern. Additionally, bilateral symmetrical perihilar or diffuse opacities were noted in some cases, although a monolateral diffuse opacity can be also identified (Nair et al., 2021). Perfusion scans and magnetic resonances of patients with HAPE show that pulmonary edema occurs in areas of high blood flow, with a patchy distribution (Dehnert et al., 2006, Hopkins et al., 2005). Typical patterns of HAPE identified at CT scan are shown in Fig. 2 and Table 2.Fig. 2 Typical patterns of HAPE identified at CT scan of a young woman after her arrival in La Paz, Bolivia at 3,500 m. Note the patchy and bilateral presentation of edema, areas where oxygen transport to pulmonary capillaries is severely compromised, generating low V’/Q’ areas responsible for the severe hypoxemia that can be encountered. The patient evolved favorably in 3 days and was able to fly back home on a scheduled flight, without sequalae.
Fig. 2
3.4 COVID-19 pneumonia
Distinct COVID-19 pneumonia phenotypes have been identified in the chest CT, including 1) a phenotype 1 (or L) (Gattinoni et al., 2020, Robba et al., 2020a) that presents as multiple focal over perfused ground glass opacities receiving more perfusion than ventilation and normally aerated areas, and is prevalent of mild to moderate diseases. In this case, possible diversion of ventilation toward non-dependent aerated lung regions and reduction in pulmonary perfusion due to increased airway pressure, the collapse of capillaries and/or micro-thrombosis and formation of non-recruitable consolidation can appear; 2) a phenotype 2 (or H) (Gattinoni et al., 2020, Robba et al., 2020a) that shows a patchy ARDS-like pattern, with inhomogeneously distributed and hyper/hypo-perfused areas that are associated with severe disease (Gattinoni et al., 2020, Orlandi et al., 2021). In this case, the lung weight can be increased with consolidated and non-aerated lung regions mainly distributed in the dependent lung regions. Areas with low ventilation/perfusion (V/Q) persist, but are associated with “true shunt” areas; 3) A phenotype (F) which represents a final evolution to fibrosis (Gattinoni et al., 2020, Robba et al., 2020a, Tonelli et al., 2021). The incidence rate of each COVID-19 phenotype was 62% and 10% for 1 and 2 phenotypes, respectively, in severe COVID-19 patients. Chest CT and lung ultrasound imaging (LUS) differed between survivors (mostly phenotype 1) and non-survivors (mostly phenotype 2) and were associated with ICU mortality (Orlandi et al., 2021). Moreover, chest CT positively correlates with lung ultrasound findings suggesting that it can be a valuable monitoring tool in critically ill settings for daily monitoring (Orlandi et al., 2021) (Table 2). Fig. 3 presents typical patterns of COVID-19 phenotype 1 and phenotype 2 at CT scan.Fig. 3 Typical pattern of COVID-19 phenotype 1 and 2. This figure depicts typical patterns of COVID-19 with preserved elastance (phenotype 1, left) and increased elastance (phenotype 2, right) at CT scan.
Fig. 3
4 Gas-exchange: general principles
AIP, ARDS, HAPE, and COVID-19 pneumonia present similar symptoms. However, a careful evaluation could identify some minimal or marked differences in the diagnostic criteria, definition, gas-exchange abnormalities, and clinical manifestations. In normal lung, gas-exchange is determined by the greater amount of normally aerated and perfused alveoli, with low venous admixture and wasted ventilation. Matching between ventilation (V’) and perfusion (Q’) of lung regions is the main determinant of gas-exchange (Petersson and Glenny, 2014). Venous admixture, mainly impairing oxygenation, is determined by lung areas with lower V’/Q’ ratios and/or true shunt (Petersson and Glenny, 2014). In normal lungs, a continuous distribution of alveoli from lower (0.001) to higher (10) V/Q areas are present following a gaussian shape, while it may present skewness in pathological conditions. In normal lungs, lung areas with extremely low V’/Q’are less frequent, while those with V’/Q’ between 1 to 0.1 are predominant. On the other hand, lung areas of “true shunt” are minimal and mostly due to anatomical shunt (3 to 5%). This means that in normal conditions, breathing in air, the oxygen content and arterial partial pressure of oxygen (PaO2) are only slightly lower than alveolar PO2 (PAO2) leading to low PAO2-PaO2 difference (Petersson and Glenny, 2014). In pathological conditions, when lung areas present extremely low V’/Q’ ratios, the oxygen content, as well as PaO2 decrease, while the PAO2-PaO2 difference increases (Fajardo et al., 2020). Thus, administering higher inspiratory oxygen fractions, in the presence of increased venous admixture mainly due to lower V’/Q’, the oxygen content and PaO2 increase while the PAO2-PaO2 difference decreases. This is explained by the fact that alveoli with low V’/Q’ areas are open to inspiratory gases and may be reached by increased FiO2 (Petersson and Glenny, 2014). On the other hand, in pathological conditions, when extremely high “true shunt” predominated, while breathing ambient air, the oxygen content as well as PaO2 decrease while the PAO2-PaO2 increases similarly or even greater compared to low V’/Q’. However, administering higher inspiratory oxygen fractions in the presence of increased venous admixture mainly due to higher “true shunt”, the oxygen content, as well as PaO2, minimally increase while the PAO2-PaO2 gradient minimally decreases (Karbing et al., 2020) ( Fig. 4). This is explained by the fact that in the presence of increased venous admixture mainly is due to increased “true” shunt, although some alveoli may contain a high fraction of inspired oxygen (FiO2), but the oxygen within them cannot diffuse into the capillaries that are closed (shunt) (Fig. 4). Wasted ventilation is due to high V’/Q’ areas as well as increased dead space (areas ventilated but not perfused), resulting predominantly in impairment of CO2 removal. In these lung regions, alveolar oxygenation depends on inspired oxygen fraction, but the effectiveness of carbon dioxide (CO2) washout is impaired by lower (high V’/Q’) or absent (dead space) perfusion. These areas represent part of the lung that do not contribute to increased oxygen content or PaO2 while impairing CO2 washout yield increased inspiratory wasted effort (Karbing et al., 2020). A high PaCO2 reduces the hypoxia tolerance, further aggravating the deleterious effects of hypoxia (Zubieta-Calleja et al., 2013). Fig. 5 depicts the response of V’/Q’ to increased FiO2, while Fig. 6 depicts possible V’/Q’ distribution in different pathological conditions.Fig. 4 Changes in gas exchange according to ventilation/ perfusion. This figure depicts changes in gas exchange according to ventilation (V’)/perfusion (Q’) alterations. On the left, normal gas exchange with normal V’/Q’. Above, venous admixture with three possible conditions: 1) low V’/Q’ with normal alveolar ventilation and increased capillary perfusion; 2) low V’/Q’ with decreased alveolar ventilation and normal capillary perfusion; and 3) true shunt with absent alveolar ventilation and variable capillary perfusion. Below, wasted ventilation with three possible conditions: 1) high V’/Q’ with normal alveolar ventilation and decreased capillary perfusion; 2) high V’/Q’ with normal capillary perfusion, and 3) dead space with normal capillary ventilation but absent perfusion.
Fig. 4
Fig. 5 Changes in gas exchange according to ventilation perfusion after increasing FiO2. This figure depicts changes in gas exchange according to ventilation (V’)/perfusion (Q’) alterations after increasing the fraction of inspired oxygen (FiO2). Above, venous admixture with three possible conditions: 1) low V’/Q’ with normal alveolar ventilation and increased capillary perfusion that rising FiO2 will determine marked increase of PaO2 and PAO2; 2) low V’/Q’ with decreased alveolar ventilation and normal capillary perfusion that rising FiO2 will determine increase of PaO2 and marked increase of PAO2; and 3) true shunt with absent alveolar ventilation and variable capillary perfusion not modified by FiO2. Below, wasted ventilation with three possible conditions: 1) high V’/Q’ with normal alveolar ventilation and decreased capillary perfusion that rising FiO2 will determine marked increase of PAO2; 2) high V’/Q’ with normal capillary perfusion that rising FiO2 will determine marked increase of PAO2, and 3) dead space with normal alveolar ventilation but absent capillary perfusion that with FiO2 will not change.
Fig. 5
Fig. 6 Ventilation/ perfusion alteration in AIP, ARDS, HAPE, and COVID-19. This figure depicts ventilation (V’)/ perfusion (Q’) changes in five possible conditions. 1) AIP characterized mainly by wasted ventilation and true shunt due to alveolar consolidation with less low V’/Q’; 2) ARDS characterized mainly by mainly wasted ventilation and less low V’/Q’, with high true shunt due to alveolar collapse, and the perfusion follow a gravitational gradient; 3) HEPA characterized by wasted ventilation and less low V’/Q’, with high true shunt due to alveolar collapse; 4) COVID-19 phenotype 1 characterized by wasted ventilation and low V’/Q’ and less true shunt due to alveolar consolidation but perfusion follow an antigravitational gradient; and 5) COVID-19 phenotype 2 characterized by wasted ventilation and low V’/Q’ and higher true shunt mainly due to consolidation, and the perfusion follow an antigravitational gradient.
Fig. 6
5 Acute Interstitial Pneumonia
In AIP, hypoxemia is mainly associated with increased lung weight, lung consolidation, and increased intrapulmonary shunt. Consolidation may be localized or widely distributed along the ventral-dorsal gradient or in the dependent lung regions in the supine position. On the other hand, the perfusion is mainly gravity-dependent (similar to ARDS), yielding increased venous admixture due to true shunt (Dakin et al., 2011). The lung areas of low V’/Q’ are present but not markedly affecting gas-exchange impairment. On the other hand, increased wasted ventilation due to the combination of high alveolar dead space and high V’/Q’ may be present mainly due to diversion of ventilation toward non-dependent aerated lung regions and reduction in pulmonary perfusion due to increased airway pressure or micro-thrombosis. V’/Q’ ratio in patients with interstitial pneumonia was firstly described many years ago. V’/Q’ profile in sub-pleural lung zones with high V’/Q’ ratios and dead space (wasted ventilation) showed flattened peaks with broadened V’/Q’ distribution (Suga et al., 2009) (Fig. 6).
5.1 ARDS
In ARDS, higher regional perfusion and true shunt are in the dependent lung regions [11] and edema follows a ventral-dorsal gradient [12] with prevalent low V’/Q’ and shunt (Fig. 6). The pressure applied on dependent lung regions favors progressive atelectasis formation (Pelosi et al., 1994), while positive end-expiratory pressure (PEEP) keeps the dependent alveoli open, improving gas-exchange and respiratory system compliance (Gattinoni et al., 1994). Hypoxemia is mainly associated with increased lung weight and atelectasis, yielding increased intrapulmonary shunt. Atelectasis is localized in the dependent lung regions in the supine position. In this clinical condition, the perfusion is mainly gravity-dependent, and the venous admixture is primarily due to true shunt (Dakin et al., 2011). In pulmonary ARDS, there are some lung regions with lower V’/Q’, which do not markedly affect gas-exchange. Moreover, there are areas of increased dead space.
5.2 High Altitude Pulmonary Edema
The pathophysiology of HAPE involves a reduced environmental partial pressure of inspired oxygen causing hypoxic pulmonary vasoconstriction that leads to an increased pulmonary arterial pressure with focalized higher pulmonary vascular resistances, higher heart rate and lower arterial oxygen saturation (Chanana et al., 2020). The etiopathology of HAPE is not clear and several hypothesis have been described: 1) genetic predisposition; 2) changes in renin-angiotensin-aldosterone system (RAS), since aldosterone increases the retention of extracellular fluid (Chanana et al., 2020); 3) modifications in nitric oxide (NO), which usually maintains the pulmonary vascular homeostasis; 4) different expression of hypoxia-inducible factor (HIF)-1 that exponentially increases when the cellular oxygen is low (Chanana et al., 2020); 5) viral infection which may explain the similar pattern observed in COVID-19 pneumonia (Zubieta-Calleja, 1989). The tongue in HAPE has very similar characteristics to the tongue in COVID-19. Clinical symptoms of HAPE include dyspnea, frothy pink sputum, tachypnea, persistent cough, cyanosis, and tachycardia (Chanana et al., 2020). The clinical diagnosis of HAPE is based on the presence of breathlessness, cough, chest discomfort and headache along with tachycardia, tachypnea, and a SpO2 < 86% for an altitude of 3500 m in a patient arriving to high altitude within up to 5 days. (Chawla and Tripathi, 2015). Regarding gas exchange and ventilation distribution, HAPE presents mainly wasted ventilation and less low V/Q and high true shunt due to alveolar collapse (Fig. 6).
5.3 COVID-19 pneumonia
Patients with COVID-19 pneumonia may present “silent hypoxemia” or “asymptomatic hypoxemia” with a peripheral saturation of oxygen (SpO2) <95% at sea level (SpO2 <85% at 3,600 m), accompanied by shortness of breath upon mild exercise, fever, tachycardia, tachypnea, hyperventilation, with intercostal and diaphragmatic fatigue, the sensation of drowning, gasping and subsequent death [16]. SARS-CoV-2 infection affects not only the lung (Zubieta-Calleja et al., 2021) but also other organs (Robba et al., 2020b). Clinical diagnosis is based on the presence of SpO2 <90% (PaO2 <60 mmHg) at sea level and lower as altitude increases. In COVID-19, gas exchange is determined by the evolutive stages of the disease. In phenotype 1, lung weight is minimally increased compared to normal lungs (1.100 gr vs 800 gr, respectively), and hypoxemia is mainly associated with increased ground-glass regions and low V/Q areas are inhomogeneously distributed within the lung parenchyma. Low V/Q areas might be associated with lower ventilation and possibly some degree of hypoxic vasoconstriction and/or greater perfusion in normally and/or poorly aerated areas. The areas of low V’/Q’ are markedly increased all over the lung from ventral to dorsal and from cranial to caudal lung regions (Pelosi et al., 1994) ( Fig. 6 ). A possible explanation could be a decreased ventilatory drive, that may contribute to the silent hypoxemia. On the other hand, increased wasted ventilation due to the combination of high alveolar dead space and low V’/Q’ is present in up to 40-50% of the total lung tissue mass, mainly due to diversion of ventilation toward non-dependent aerated lung regions and reduction in pulmonary perfusion due to increased airway pressure, collapse of capillaries and/or micro-thrombosis and formation of no recruitable atelectasis (Pelosi et al., 1994). This can also be justified by the increase in mucinous filling and pulmonary and vascular alterations (Ball et al., 2021c). In phenotype 2, lung weight is markedly increased (1,600 -1,700 gr), and hypoxemia is mainly associated with increased consolidated and non-aerated lung regions mainly distributed in the dependent lung regions. Under these clinical conditions, areas with low V’/Q’ persist but also combined with areas of “true shunt”. On the other hand, the amount of increased wasted ventilation is high and comparable with what is observed in phenotype 1 (L) ( Fig. 6 ).
6 Intensive care management
All these four pathophysiological entities present similar but also peculiar characteristics of treatment. There is not specific treatment for AIP and ARDS being the management of these two conditions is mainly supportive (Tasaka et al., 2022, Xaubet et al., 2013). In AIP and ARDS of infective etiology, broad-spectrum antibiotics are recommended initially until infectious etiology is confirmed, thereby selecting narrower spectrum antibiotics. No clear benefit of corticosteroids has been confirmed in both AIP and ARDS (Chang et al., 2022). However, recent data suggest potential benefits of dexamethasone (Villar et al., 2020) or methylprednisolone and hydrocortisone in patients with ARDS (Chang et al., 2022). Further trials are warranted to confirm their efficacy. Supportive therapies for AIP are like those applied for ARDS, using a stepwise approach to increase the complexity of treatment according to the increased severity of lung disease. In mechanically ventilated subjects, a lung protective ventilator strategy should be adopted (Tasaka et al., 2022, Xaubet et al., 2013). In HAPE, the first-line treatment is the administration of oxygen and the rapid descent to lower altitudes. Medical therapy includes nifedine, whereas a phosphodiesterase type 5 inhibitor may be helpful but is not formally included in guidelines. Diuretics are not recommended (Wilkins et al., 2015). Treatment possibilities of COVID-19 pneumonia should consider both anatomopathological and clinical findings. The separation in different phenotypes according to respiratory compliance has been recently challenged, concluding that no change in conventional lung-protective ventilation strategies is warranted (Reddy et al., 2022). We interpreted different phenotypes as an evolution and progression of the disease and lung injury and not as separate entities. In case of “silent hypoxemia”, in which the degree of hypoxemia may not correspond to the severity of lung images and histopathological findings, early therapy may include the use of high-flow nasal oxygen or continuous positive end-expiratory pressure (CPAP) (Robba et al., 2021a). Noninvasive respiratory support can be considered as the first step toward COVID-19 management in phenotype 1, while early intubation and invasive mechanical ventilation should be considered for phenotype 2 or worsening of phenotype 1. A recent meta-analysis concluded that the timing of intubation may have no effect on mortality and morbidity in critically ill COVID-19 (Papoutsi et al., 2021). During invasive mechanical ventilation, COVID-19 patients showed a decrease in pulmonary and gas volume and increased regions of poorly aerated and non-aerated lung tissue more than patients treated with non-invasive respiratory support (Ball et al., 2021b). In COVID-19 phenotype-1 (L), a relatively preserved lung compliance is present with few areas that can be recruited and low V’/Q’ areas. In this case, using a moderate PEEP level to redistribute pulmonary blood flow from non-ventilated to more ventilated areas could be suggested. Phenotype 2 (H) might benefit from invasive mechanical ventilation when non-invasive respiratory treatment is not effective, with progressive oxygen deterioration, prompt intubation is required. The respiratory treatment suggested is like that reported in “primary” ARDS with protective ventilation and moderate levels of PEEP individualized on patients’ need, with cautious use of recruitment maneuvers. Prone position may be effective to improve oxygenation, but improvement in oxygenation is often associated with redistribution of perfusion and not effective alveolar recruitment (Rossi et al., 2021). Indeed, when the ventilation of dependent areas is impaired in the supine position, prone rescue positioning can be applied. The gravitational effect makes the dependent lungs areas more perfused following V’/Q’ mismatch, thus resulting in hypoxemia, while prone position allows a more homogeneous distribution of ventilation and perfusion, reducing the risk of ventilator-induced lung injury (VILI), but improvement in oxygenation is often associated with redistribution of perfusion and not effective alveolar recruitment in COVID-19 (Battaglini et al., 2021, Rossi et al., 2021). The amount of consolidated tissue was higher in patients assessed during the third week and determined the oxygenation responses following pronation and recruitment maneuvers. In most refractory and severe cases, extracorporeal carbon dioxide removal (ECCO2R) and extracorporeal membrane oxygenation (ECMO) could be used in order to reduce Plateau pressures and tidal volumes and prevent VILI while adequately oxygenating the lungs and improving oxygen supply to all tissues (Battaglini et al., 2021). Other treatment options for improving gas exchange, include the use of erythropoietin in the treatment of severe COVID-19 pneumonia (Ehrenreich et al., 2020, Soliz et al., 2020) which can help improve the oxygen transport to tissues of the compromised oxygen transport triad (pneumo-dynamic pump, hemodynamic pump, and hemoglobin) (Zubieta-Calleja et al., 2021, Zubieta-Calleja and Zubieta-DeUrioste, 2021a). Erythropoietin could also exert indirect effects not limited to the increase of red blood cells, hematocrit and hemoglobin but also to the protective effects of the hormone in different tissues like the heart, the vessels, and the brain (Ehrenreich et al., 2020, Hadadi et al., 2020, Sahebnasagh et al., 2020). In fact, a correlation between acute lung injury and brain hypoxia has been described (Battaglini et al., 2020a). Hence, reduced systemic oxygenation can be associated with reduced brain tissue oxygenation and poor outcome (Battaglini et al., 2020a). Furthermore, the increase of hemoglobin becomes a fundamental factor of hypoxia tolerance, as previously shown by the Formula of Tolerance to hypoxia = Hb/PaCO2 * 3.01 (Hb, hemoglobin; PaCO2, arterial partial pressure of CO2). This formula states that higher hemoglobin and a lower PaCO2 give greater tolerance to hypoxia (Zubieta-Calleja et al., 2013). Indeed, PaO2 and oxygen delivery (DO2) can be optimized by modulating Hb concentration, pHa, PaCO2, cardiac output, and arterial content of oxygen. Thus, increasing hemoglobin and cardiac output could be considered in this specific clinical setting in order to improve oxygen delivery (Battaglini et al., 2020a). In addition to these parameters, others should be considered when facing hypoxia at high altitude. According to the alveolar gas equation, the alveolar partial pressure of oxygen is dramatically decreased at high altitude, but a higher diffusion capacity is presumed, with a supposed physiological adaptation to altitude (Zubieta-Calleja et al., 2013). Some literature suggests that physiological adaptation to altitude may counterbalance the hypoxic environment thus protecting from severe SARS-CoV-2 infection, probably mediated by reduced survival of the virus at high altitude, and down-regulation of ACE2 (Arias-Reyes et al., 2020). Modulation of the inflammatory response is a matter of debate in both ARDS and COVID-19 (Chaudhuri et al., 2021). COVID-19 can trigger a “cytokines storm” in lung tissues and systemic circulation with the hyperactivation of the inflammatory-immune system. The accumulation of immune cells at infected sites can result in ARDS. Accordingly, a “cytokines storm” has been previously described in several conditions, including AIP and ARDS (Sinha et al., 2020). In HAPE, hyperinflammatory status has been attributed to the formation of pulmonary edema. The possible implication for treatment could be the recognition of those patients at higher risk of hyperinflammatory response to recognize hyper/ hypo-inflammatory sub-phenotypes such as those currently under investigation in ARDS. This is matter of debate since HAPE is a poorly defined pathophysiological entity, hence not clearly supporting the use of this terminology and its implication for therapy to date (Sinha et al., 2020). A universal agreement on the use of corticosteroids in non-COVID ARDS is still lacking, but national guidelines started to consider the use of low-dose steroids in ARDS, while recommending against the use of high doses that may result in immune suppression and increased risk of opportunistic infections (Tasaka et al., 2022). COVID-19 has boosted the pharmacological research investigating the role of several immunomodulating agents in respiratory failure, including steroids, anti-interleukin monoclonal antibodies and drugs with indirect action on the immune response (Battaglini et al., 2022). While the use of steroids in hospitalized COVID-19 patients requiring oxygen supplementation is an established practice, these are typically contraindicated in initial, mild COVID-19. The use of other immunomodulating agents requires careful patient selection in COVID-19 and, more importantly, caution when translating to ARDS the findings of studies conducted in COVID-19 related respiratory failure (Battaglini et al., 2022).
7 Potential overlap of pathways involved
7.1 Renin-angiotensin system
The renin-angiotensin system (RAS) has been identified as potential effector in the pathophysiology of pulmonary diseases. Particularly, angiotensin II exerts multiple function for maintaining the homeostasis of blood pressure, vasoconstriction, and sodium balance. Angiotensin converting enzyme2 (ACE2) also plays an important role. In fact ACE deficient mice or the use of inhibitors led to remodeling of pulmonary arterioles (Kuba et al., 2006) and Deletion/Deletion (D/D) polymorphism of ACE gene was associated with more severity of pulmonary symptoms (Kuba et al., 2006). ACE inhibitors may attenuate endothelial dysfunction and fibrosis in interstitial pneumonia. Both clinical and experimental animal studies have implicated ACE in the pathogenesis of ARDS (Marshall et al., 2002, Raiden et al., 2002), suggesting its role on pulmonary vascular tone/permeability, epithelial cell survival, and fibroblast activation. D/D gene polymorphism again was more abundant in patients with ARDS than general ICU/ general population, and was associated with higher mortality (Marshall et al., 2002). Recombinant ACE2 protein improved the symptoms and fibrotic changes in ARDS, playing a protective role (Kuba et al., 2006). Similarly, HAPE recognizes the role of ACE2 and RAS in the pathogenesis of increased arterial pulmonary pressure, as occurs in response to hypoxia (Kleinsasser et al., 2012). The importance of the RAS in lung diseases has recently re-emerged with the identification of ACE2 in SARS-CoV-2. SARS-CoV-2 enters lung cells via the ACE2 receptor. ACE2 receptors are recognized by macrophages and other pro-inflammatory cells, spreading to other organs and infecting ACE2-expressing cells at local sites, potentially causing multi-organ disease (Al-Kuraishy et al., 2020).
7.2 Neuropilin-1
Neuropilin-1 (NPN-1) is a cell surface receptor that binds vascular endothelial growth factor (VEGF), potentially influencing endothelial permeability, chemotaxis, and proliferation. Inhibition of NPN-1 revealed attenuation of VEGF-mediated permeability of pulmonary artery endothelial cells, probably regulating endothelial barrier dysfunction in response to VEGF (Becker et al., 2005). In patients with ARDS, VEGF receptors 1 and 2 are up-regulated, whereas NPN-1 is down-regulated in comparison with patients without ARDS (Medford et al., 2009). Some studies demonstrated that changes in the plasma level of VEGF and soluble VEGF receptor-1 as well as NPN-1 are associated with HAPE (Zhang et al., 2018). In COVID-19, NPN-1 protein interacts with the host cell receptor is the Spike-1 (S1) subunit of the coronavirus, thus favoring SARS-CoV-2 entry into the cells, and in particular to the central nervous system (Al-kuraishy et al., 2021).
7.3 Hypoxia-inducible factor
Hypoxia-inducible factor (HIF)-1 is a transcriptional activator dependent on oxygen, which plays an important role in maintaining oxygen homeostasis by regulating the expression of genes associated with adaptation to hypoxia. HIF-1 induces the release of inflammatory factors, thus exacerbating lung injury (Liu et al., 2020). During hypoxic conditions, HIF-1 induces VEGF expression regulating pulmonary vascular permeability and promotes alveolar type II cells apoptosis. In the progression of ARDS, HIF-1 enhances glucose metabolism in epithelial cells, thus increasing lung ventilation and improving pulmonary edema and respiratory distress (Liu et al., 2020). During normoxia, HIF-1α levels were recognized as markers of HAPE susceptibility (Soree et al., 2016). In COVID-19, HIF-1α promoted SARS-CoV-2 infection and aggravated inflammatory responses (Tian et al., 2021).
8 Future directions
COVID-19 may be complicated by the so called post-COVID-19 syndrome which manifests with symptoms duration for more than 3 months and progressive pulmonary fibrosis (Mehta et al., 2022). The pathogenesis of post-COVID-19 syndrome include direct tissue injury associated with autoimmunity antibodies. This new entity probably results from the interplay between epithelial and endothelial injury, immune-genetic susceptibility, ARDS, and/or ventilator-induced lung injury, hyperinflammation, and hypercoagulability, that in turn activate pro-fibrotic processes and a fibro-proliferative cascade. The impact of post-COVID-19 syndrome is currently unknown, as well as the influence of new COVID-19 variants, vaccination, ventilatory setting, immunomodulation and inflammation on it. The identification of COVID-19 phenotypes, as well as the understanding of physiological, radiological, and anatomopathological implications, and their evolution are paramount for progressing therapeutic targets and personalized medicine (Mehta et al., 2022).
9 Conclusions
Remodeling and destruction of lung tissue may explain silent hypoxemia, intubation ineffectiveness, rapid progression to severe gasping, and finally, death in most COVID-19 patients. These processes are characterized by active viral replication of COVID-19 within type-2 pneumocytes associated with persistent infiltration of inflammatory cells, as well as a continuous repair process, but with an intense reactive fibrotic activity and superimposed opportunistic infections that finally lead to death. For this reason, extensive destruction of the lung parenchyma can be easily confounded with other diseases that may coexist in COVID-19 but can also help to explain the post-disease fibrosis, residual respiratory insufficiency, and exercise limitation.
Ethics approval and consent to participate
Not applicable
Consent for publication
Not applicable
Funding
None
Authors’ contribution
G.Z-C.: Writing original draft, conceptualization, methodology, visualization. N.Z-D.: Writing, review and editing, methodology. F.J.M.: Biopsies, pathological photomicrographs, and writing pathological findings. M.G.R-S: Pathology preparation, photomicrographs and writing pathological findings. M.E.: validation. A.F.C: Writing and editing. D.B., P.R.M. Writing, editing, validation. L.B.: Validation and editing. P.P.: Writing and editing.
Competing interests
None
Data availability
No data was used for the research described in the article.
Acknowledgments
We acknowledge the late Prof. Dr. Gustavo Zubieta-Castillo (Sr.), our mentor. We also thank Lucrecia DeUrioste and Rafaela Zubieta-DeUrioste for their collaboration and suggestions.
==== Refs
References
Ackermann M. Verleden S.E. Kuehnel M. Haverich A. Welte T. Laenger F. Vanstapel A. Werlein C. Stark H. Tzankov A. Li W.W. Li V.W. Mentzer S.J. Jonigk D. Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19 N. Engl. J. Med 383 2020 120 128 10.1056/NEJMoa2015432 32437596
Al-kuraishy H.M. Al-Gareeb A.I. Alzahrani K.J. Cruz-Martins N. Batiha G.E.-S. The potential role of neopterin in Covid-19: a new perspective Mol. Cell. Biochem. 476 2021 4161 4166 10.1007/s11010-021-04232-z 34319496
Al-Kuraishy H.M. Hussein N.R. Al-Naimi M.S. Al-Buhaidily A.K. Al-Gareeb A.I. Lungnier C. Renin–Angiotensin system and fibrinolytic pathway in COVID-19: One-way skepticism Biomed. Biotechnol. Res. J 4 2020 33 40 10.4103/bbrj.bbrj_105_20
Angeles Montero-Fernandez M. Pardo-Garcia R. Histopathology features of the lung in COVID-19 patients Diagnostic Histopathol 27 2021 123 127 10.1016/j.mpdhp.2020.11.009
Arias-Reyes C. Zubieta-DeUrioste N. Poma-Machicao L. Aliaga-Raduan F. Carvajal-Rodriguez F. Dutschmann M. Schneider-Gasser E.M. Zubieta-Calleja G. Soliz J. Does the pathogenesis of SARS-CoV-2 virus decrease at high-altitude Respir. Physiol. Neurobiol. 277 2020 103443 10.1016/j.resp.2020.103443
Ball L. Barisione E. Mastracci L. Campora M. Costa D. Robba C. Battaglini D. Micali M. Costantino F. Cittadini G. Patroniti N. Pelosi P. Fiocca R. Grillo F. Extension of Collagen Deposition in COVID-19 Post Mortem Lung Samples and Computed Tomography Analysis Findings Int. J. Mol. Sci 22 2021 7498 10.3390/ijms22147498 34299124
Ball L. Robba C. Herrmann J. Gerard S.E. Xin Y. Mandelli M. Battaglini D. Brunetti I. Minetti G. Seitun S. Bovio G. Vena A. Giacobbe D.R. Bassetti M. Rocco P.R.M. Cereda M. Rizi R.R. Castellan L. Patroniti N. Pelosi P. Bixio M. Gratarola A. Frisoni P. Loconte M. Molin A. Orefice G. Ciaravolo E. Costantino F. Battioni D. Buconte G. Casaleggio A. Cittadini G. Dogliotti L. Giasotto V. Perissi S. Pigati M. Santacroce E. Zaottini F. Dentone C. Taramasso L. Magnasco L. Bastianello M. Lung distribution of gas and blood volume in critically ill COVID-19 patients: a quantitative dual-energy computed tomography study Crit. Care 25 2021 214 10.1186/s13054-021-03610-9 34154635
Ball L. Robba C. Maiello L. Herrmann J. Gerard S.E. Xin Y. Battaglini D. Brunetti I. Minetti G. Seitun S. Vena A. Giacobbe D.R. Bassetti M. Rocco P.R.M. Cereda M. Castellan L. Patroniti N. Pelosi P. Ball L. Gratarola A. Loconte M. Molin A. Orefice G. Iannuzzi F. Costantino F. Battioni D. Bovio G. Buconte G. Casaleggio A. Cittadini G. Dogliotti L. Giasotto V. Pigati M. Santacroce E. Zaottini F. Dentone C. Taramasso L. Magnasco L. Valbusa A. Bastianello M. Computed tomography assessment of PEEP-induced alveolar recruitment in patients with severe COVID-19 pneumonia Crit. Care 25 2021 81 10.1186/s13054-021-03477-w 33627160
Barisione E. Grillo F. Ball L. Bianchi R. Grosso M. Morbini P. Pelosi P. Patroniti N.A. De Lucia A. Orengo G. Gratarola A. Verda M. Cittadini G. Mastracci L. Fiocca R. Fibrotic progression and radiologic correlation in matched lung samples from COVID-19 post-mortems Virchows Arch 478 2021 471 485 10.1007/s00428-020-02934-1 32989525
Barisione E. Grillo F. Ball L. Bianchi R. Grosso M. Morbini P. Pelosi P. Patroniti N.A. De Lucia A. Orengo G. Gratarola A. Verda M. Cittadini G. Mastracci L. Fiocca R. Fibrotic progression and radiologic correlation in matched lung samples from COVID-19 post-mortems Virchows Arch 2020 1 15 10.1007/s00428-020-02934-1
Barton L.M. Duval E.J. Stroberg E. Ghosh S. Mukhopadhyay S. COVID-19 Autopsies, Oklahoma, USA Am. J. Clin. Pathol 153 2020 725 733 10.1093/ajcp/aqaa062 32275742
Batah S.S. Fabro A.T. Pulmonary pathology of ARDS in COVID-19: A pathological review for clinicians Respir. Med. 176 2021 106239 10.1016/j.rmed.2020.106239
Battaglini D. Brunetti I. Anania P. Fiaschi P. Zona G. Ball L. Giacobbe D.R. Vena A. Bassetti M. Patroniti N. Schenone A. Pelosi P. Rocco P.R.M. Robba C. Neurological Manifestations of Severe SARS-CoV-2 Infection: Potential Mechanisms and Implications of Individualized Mechanical Ventilation Settings Front. Neurol. 11 2020 845 10.3389/fneur.2020.00845 32903391
Battaglini D. Cruz F. Robba C. Pelosi P. Rocco P.R.M. Failed clinical trials on COVID-19 acute respiratory distress syndrome in hospitalized patients: common oversights and streamlining the development of clinically effective therapeutics Expert Opin. Investig. Drugs. 2022 10.1080/13543784.2022.2120801
Battaglini D. Santori G. Chandraptham K. Iannuzzi F. Bastianello M. Tarantino F. Ball L. Giacobbe D.R. Vena A. Bassetti M. Inglese M. Uccelli A. Rocco P.R.M. Patroniti N. Brunetti I. Pelosi P. Robba C. Neurological Complications and Noninvasive Multimodal Neuromonitoring in Critically Ill Mechanically Ventilated COVID-19 Patients Front. Neurol. 11 2020 602114 10.3389/fneur.2020.602114
Battaglini D. Sottano M. Ball L. Robba C. Rocco P.R.M. Pelosi P. Ten golden rules for individualized mechanical ventilation in acute respiratory distress syndrome J. Intensive Med 1 2021 42 51 10.1016/j.jointm.2021.01.003
Becker P.M. Waltenberger J. Yachechko R. Mirzapoiazova T. Sham J.S.K. Lee C.G. Elias J.A. Verin A.D. Neuropilin-1 Regulates Vascular Endothelial Growth Factor–Mediated Endothelial Permeability Circ. Res. 96 2005 1257 1265 10.1161/01.RES.0000171756.13554.49 15920019
Cardinal-Fernández P. Lorente J.A. Ballén-Barragán A. Matute-Bello G. Acute Respiratory Distress Syndrome and Diffuse Alveolar Damage. New Insights on a Complex Relationship Ann. Am. Thorac. Soc 14 2017 844 850 10.1513/AnnalsATS.201609-728PS 28570160
Chanana N. Palmo T. Newman J.H. Pasha M.A.Q. Vascular homeostasis at high-altitude: role of genetic variants and transcription factors 204589402091347 Pulm. Circ. 10 2020 10.1177/2045894020913475
Chang X. Li S. Fu Y. Dang H. Liu C. Safety and efficacy of corticosteroids in ARDS patients: a systematic review and meta-analysis of RCT data Respir. Res. 23 2022 301 10.1186/s12931-022-02186-4 36333729
Chaudhuri D. Sasaki K. Karkar A. Sharif S. Lewis K. Mammen M.J. Alexander P. Ye Z. Lozano L.E.C. Munch M.W. Perner A. Du B. Mbuagbaw L. Alhazzani W. Pastores S.M. Marshall J. Lamontagne F. Annane D. Meduri G.U. Rochwerg B. Corticosteroids in COVID-19 and non-COVID-19 ARDS: a systematic review and meta-analysis Intensive Care Med 47 2021 521 537 10.1007/s00134-021-06394-2 33876268
Chawla A. Tripathi K.K. Objective criteria for diagnosing high altitude pulmonary edema in acclimatized patients at altitudes between 2700m and 3500m Med. J. Armed Forces India 71 2015 345 351 10.1016/j.mjafi.2015.09.002 26663962
Dakin J. Jones A.T. Hansell D.M. Hoffman E.A. Evans T.W. Changes in lung composition and regional perfusion and tissue distribution in patients with ARDS Respirology 16 2011 1265 1272 10.1111/j.1440-1843.2011.02048.x 21883676
De Marzo V. Di Biagio A. Della Bona R. Vena A. Arboscello E. Emirjona H. Mora S. Giacomini M. Da Rin G. Pelosi P. Bassetti M. Ameri P. Porto I. Prevalence and prognostic value of cardiac troponin in elderly patients hospitalized for COVID-19 J Geriatr Cardiol 18 2021 338 345 10.11909/j.issn.1671-5411.2021.05.004 34157060
Dehnert C. Risse F. Ley S. Kuder T.A. Buhmann R. Puderbach M. Menold E. Mereles D. Kauczor H.-U. Bärtsch P. Fink C. Magnetic Resonance Imaging of Uneven Pulmonary Perfusion in Hypoxia in Humans Am. J. Respir. Crit. Care Med 174 2006 1132 1138 10.1164/rccm.200606-780OC 16946125
Desai S.R. Wells A.U. Rubens M.B. Evans T.W. Hansell D.M. Acute Respiratory Distress Syndrome: CT Abnormalities at Long-term Follow-up Radiology 210 1999 29 35 10.1148/radiology.210.1.r99ja2629 9885583
Droma Y. Hanaoka M. Hotta J. Naramoto A. Koizumi T. Fujimoto K. Honda T. Kobayashi T. Kubo K. Pathological Features of the Lung in Fatal High Altitude Pulmonary Edema Occurring at Moderate Altitude in Japan High Alt. Med. Biol. 2 2001 515 523 10.1089/152702901753397081 11809092
Edler C. Schröder A.S. Aepfelbacher M. Fitzek A. Heinemann A. Heinrich F. Klein A. Langenwalder F. Lütgehetmann M. Meißner K. Püschel K. Schädler J. Steurer S. Mushumba H. Sperhake J.-P. Dying with SARS-CoV-2 infection—an autopsy study of the first consecutive 80 cases in Hamburg, Germany Int. J. Legal Med 134 2020 1275 1284 10.1007/s00414-020-02317-w 32500199
Ehrenreich H. Weissenborn K. Begemann M. Busch M. Vieta E. Miskowiak K.W. Erythropoietin as candidate for supportive treatment of severe COVID-19 Mol. Med. 26 2020 58 10.1186/s10020-020-00186-y 32546125
Fajardo A. Medina A. Modesto V. Gordo-Vidal F. El lado oculto de la hipoxemia. Rev. Chil. Anest 49 2020 784 794 10.25237/revchilanestv49n06-4
Ferreira J.C. Ho Y.L. Adler Maccagnan B. Besen P. Sa Malbouisson L.M. Utino Taniguchi L. Vitale Mendes P. Leite Vieira Costa E. Park M. Daltro-Oliveira R. Roepke M.L. R., M Silva-Jr, J. Carvalho Carmona M.J. Carvalho C.R.R. Protective ventilation and outcomes of critically ill patients with COVID-19: a cohort study on behalf of EPICCoV Study Group Ann. Intensive Care 11 2021 92 10.1186/s13613-021-00882-w 34097145
Garrido E. Maglia J.B. de, Sibila O. Zubieta-Calleja G. High-Altitude Pulmonary Edema in the Context of COVID-19 1–1 OBM Genet 6 2022 10.21926/obm.genet.2203163
Gattinoni L. Bombino M. Pelosi P. Lissoni A. Pesenti A. Fumagalli R. Tagliabue M. Lung structure and function in different stages of severe adult respiratory distress syndrome JAMA 271 1994 1772 1779 8196122
Gattinoni L. Caironi P. Pelosi P. Goodman L.R. What Has Computed Tomography Taught Us about the Acute Respiratory Distress Syndrome? Am. J. Respir. Crit. Care Med 164 2001 1701 1711 10.1164/ajrccm.164.9.2103121 11719313
Gattinoni L. Chiumello D. Caironi P. Busana M. Romitti F. Brazzi L. Camporota L. COVID-19 pneumonia: different respiratory treatments for different phenotypes Intensive Care Med 46 2020 1099 1102 10.1007/s00134-020-06033-2 32291463
Giesenhagen A.M. Ivy D.D. Brinton J.T. Meier M.R. Weinman J.P. Liptzin D.R. High Altitude Pulmonary Edema in Children: A Single Referral Center Evaluation J. Pediatr 210 2019 106 111 10.1016/j.jpeds.2019.02.028 31005280
Goodman L.R. Fumagalli R. Tagliabue P. Tagliabue M. Ferrario M. Gattinoni L. Pesenti A. Adult Respiratory Distress Syndrome Due to Pulmonary and Extrapulmonary Causes: CT, Clinical, and Functional Correlations Radiology 213 1999 545 552 10.1148/radiology.213.2.r99nv42545 10551239
Guerin C. Bayle F. Leray V. Debord S. Stoian A. Yonis H. Roudaut J.-B. Bourdin G. Devouassoux-Shisheboran M. Bucher E. Ayzac L. Lantuejoul S. Philipponnet C. Kemeny J.L. Souweine B. Richard J.-C. Open lung biopsy in nonresolving ARDS frequently identifies diffuse alveolar damage regardless of the severity stage and may have implications for patient management Intensive Care Med 41 2015 222 230 10.1007/s00134-014-3583-2 25476984
Hadadi A. Mortezazadeh M. Kolahdouzan K. Alavian G. Does recombinant human erythropoietin administration in critically ill COVID‐19 patients have miraculous therapeutic effects ? J. Med. Virol. 92 2020 915 918 10.1002/jmv.25839 32270515
Hall L.F. Pneumolysis in the Treatment of Pulmonary Tuberculosis Dis. Chest 4 1938 18 20 10.1378/chest.4.11.18
Hanley B. Lucas S.B. Youd E. Swift B. Osborn M. Autopsy in suspected COVID-19 cases J. Clin. Pathol 73 2020 239 242 10.1136/jclinpath-2020-206522 32198191
Hellman U. Karlsson M.G. Engström-Laurent A. Cajander S. Dorofte L. Ahlm C. Laurent C. Blomberg A. Presence of hyaluronan in lung alveoli in severe Covid-19: An opening for new treatment options ? J. Biol. Chem. 295 2020 15418 15422 10.1074/jbc.AC120.015967 32978255
Hopkins S.R. Garg J. Bolar D.S. Balouch J. Levin D.L. Pulmonary Blood Flow Heterogeneity during Hypoxia and High-Altitude Pulmonary Edema Am. J. Respir. Crit. Care Med 171 2005 83 87 10.1164/rccm.200406-707OC 15486339
Huertas A. Guignabert C. Barberà J.A. Bärtsch P. Bhattacharya J. Bhattacharya S. Bonsignore M.R. Dewachter L. Dinh-Xuan A.T. Dorfmüller P. Gladwin M.T. Humbert M. Kotsimbos T. Vassilakopoulos T. Sanchez O. Savale L. Testa U. Wilkins M.R. Pulmonary vascular endothelium: the orchestra conductor in respiratory diseases Eur. Respir. J 51 2018 1700745 10.1183/13993003.00745-2017 29545281
Kamp J.C. Neubert L. Ackermann M. Stark H. Werlein C. Fuge J. Haverich A. Tzankov A. Steinestel K. Friemann J. Boor P. Junker K. Hoeper M.M. Welte T. Laenger F. Kuehnel M.P. Jonigk D.D. Time-Dependent Molecular Motifs of Pulmonary Fibrogenesis in COVID-19 Int. J. Mol. Sci 23 2022 1583 10.3390/ijms23031583 35163504
Karbing D.S. Panigada M. Bottino N. Spinelli E. Protti A. Rees S.E. Gattinoni L. Changes in shunt, ventilation/perfusion mismatch, and lung aeration with PEEP in patients with ARDS: a prospective single-arm interventional study Crit. Care 24 2020 111 10.1186/s13054-020-2834-6 32293506
Kleinsasser A. Pircher I. Treml B. Schwienbacher M. Schuster M. Janzek E. Loibner H. Penninger J.M. Loeckinger A. Recombinant Angiotensin-Converting Enzyme 2 Suppresses Pulmonary Vasoconstriction in Acute Hypoxia Wilderness Environ. Med. 23 2012 24 30 10.1016/j.wem.2011.09.002 22441085
Kuba K. Imai Y. Penninger J. Angiotensin-converting enzyme 2 in lung diseases Curr. Opin. Pharmacol. 6 2006 271 276 10.1016/j.coph.2006.03.001 16581295
Lavinio A. Ercole A. Battaglini D. Magnoni S. Badenes R. Taccone F.S. Helbok R. Thomas W. Pelosi P. Robba C. Safety profile of enhanced thromboprophylaxis strategies for critically ill COVID-19 patients during the first wave of the pandemic: observational report from 28 European intensive care units Crit. Care 25 2021 155 10.1186/s13054-021-03543-3 33888132
Lin L. Lu L. Cao W. Li T. Hypothesis for potential pathogenesis of SARS-CoV-2 infection–a review of immune changes in patients with viral pneumonia Emerg. Microbes Infect. 9 2020 727 732 10.1080/22221751.2020.1746199 32196410
Liu Y. Xiang D. Zhang H. Yao H. Wang Y. Hypoxia-Inducible Factor-1: A Potential Target to Treat Acute Lung Injury Oxid. Med. Cell. Longev 2020 2020 1 13 10.1155/2020/8871476
Lopes-Pacheco M. Silva P.L. Cruz F.F. Battaglini D. Robba C. Pelosi P. Morales M.M. Caruso Neves C. Rocco P.R.M. Pathogenesis of Multiple Organ Injury in COVID-19 and Potential Therapeutic Strategies Front. Physiol. 12 2021 593223 10.3389/fphys.2021.593223
Marshall R.P. Webb S. Bellingan G.J. Montgomery H.E. Chaudhari B. McAnulty R.J. Humphries S.E. Hill M.R. Laurent G.J. Angiotensin Converting Enzyme Insertion/Deletion Polymorphism Is Associated with Susceptibility and Outcome in Acute Respiratory Distress Syndrome Am. J. Respir. Crit. Care Med 166 2002 646 650 10.1164/rccm.2108086 12204859
Medford A.R.L. Ibrahim N.B.N. Millar A.B. Vascular endothelial growth factor receptor and coreceptor expression in human acute respiratory distress syndrome J. Crit. Care 24 2009 236 242 10.1016/j.jcrc.2008.04.003 19327291
Mehta P. Rosas I.O. Singer M. Understanding post-COVID-19 interstitial lung disease (ILD): a new fibroinflammatory disease entity Intensive Care Med 2022 1 4 10.1007/s00134-022-06877-w
Menezes S.L.S. Bozza P.T. Faria Neto H.C.C. Laranjeira A.P. Negri E.M. Capelozzi V.L. Zin W.A. Rocco P.R.M. Pulmonary and extrapulmonary acute lung injury: inflammatory and ultrastructural analyses J. Appl. Physiol 98 2005 1777 1783 10.1152/japplphysiol.01182.2004 15649870
Mitra P. Misra S. Sharma P. COVID-19 Pandemic in India: What Lies Ahead Indian J. Clin. Biochem. 35 2020 257 259 10.1007/s12291-020-00886-6 32313407
Mukhopadhyay S. Parambil J. Acute Interstitial Pneumonia (AIP): Relationship to Hamman-Rich Syndrome, Diffuse Alveolar Damage (DAD), and Acute Respiratory Distress Syndrome (ARDS Semin. Respir. Crit. Care Med. 33 2012 476 485 10.1055/s-0032-1325158 23001802
Nair V. Singh P. Yanamandra U. Gupta A. Grewal R. Vardhan V. Saxena P. Mulajkar D. Radiographical Spectrum of High-altitude Pulmonary Edema: A Pictorial Essay Indian J. Crit. Care Med 25 2021 668 674 10.5005/jp-journals-10071-23827 34316147
Obadina E.T. Torrealba J.M. Kanne J.P. Acute pulmonary injury: high-resolution CT and histopathological spectrum Br. J. Radiol 86 2013 20120614 10.1259/bjr.20120614 23659926
Orlandi D. Battaglini D. Robba C. Viganò M. Bergamaschi G. Mignatti T. ML R. Lapolla A. Turtulici G. Pelosi P. COVID-19 phenotypes, lung ultrasound, chest computed tomography, and clinical features in critically ill mechanically ventilated patients Ultrasound Med. Biol. 47 2021 3323 3332 10.1016/j.ultrasmedbio.2021.07.014 34551862
Palmucci S. Roccasalva F. Puglisi S. Torrisi S.E. Vindigni V. Mauro L.A. Ettorre G.C. Piccoli M. Vancheri C. Clinical and radiological features of idiopathic interstitial pneumonias (IIPs): a pictorial review Insights Imaging 5 2014 347 364 10.1007/s13244-014-0335-3 24844883
Papoutsi E. Giannakoulis V.G. Xourgia E. Routsi C. Kotanidou A. Siempos I.I. Effect of timing of intubation on clinical outcomes of critically ill patients with COVID-19: a systematic review and meta-analysis of non-randomized cohort studies Crit. Care 25 2021 121 10.1186/s13054-021-03540-6 33766109
Pelosi P. D’Andrea L. Vitale G. Pesenti A. Gattinoni L. Vertical gradient of regional lung inflation in adult respiratory distress syndrome Am. J. Respir. Crit. Care Med 149 1994 8 13 10.1164/ajrccm.149.1.8111603 8111603
Pelosi P. D’Onofrio D. Chiumello D. Paolo S. Chiara G. Capelozzi V.L. Barbas C.S.V. Chiaranda M. Gattinoni L. Pulmonary and extrapulmonary acute respiratory distress syndrome are different Eur. Respir. J 22 2003 48s 56s 10.1183/09031936.03.00420803
Petersson J. Glenny R.W. Gas exchange and ventilation–perfusion relationships in the lung Eur. Respir. J 44 2014 1023 1041 10.1183/09031936.00037014 25063240
Pomara C. Li Volti G. Cappello F. COVID-19 Deaths: Are We Sure It Is Pneumonia? Please, Autopsy, Autopsy, Autopsy ! J. Clin. Med 9 2020 1259 10.3390/jcm9051259 32357503
Primack S.L. Hartman T.E. Ikezoe J. Akira M. Sakatani M. Müller N.L. Acute interstitial pneumonia: radiographic and CT findings in nine patients Radiology 188 1993 817 820 10.1148/radiology.188.3.8351354 8351354
Raiden S. Nahmod K. Nahmod V. Semeniuk G. Pereira Y. Alvarez C. Giordano M. Geffner J.R. Nonpeptide Antagonists of AT1 Receptor for Angiotensin II Delay the Onset of Acute Respiratory Distress Syndrome J. Pharmacol. Exp. Ther 303 2002 45 51 10.1124/jpet.102.037382 12235231
Ranieri V. Rubenfeld G. Thompson B. Ferguson N. Caldwell E. Fan E. Camporota L. Slutsky A. Acute Respiratory Distress Syndrome: the Berlin Definition JAMA 307 2012 2526 2533 10.1001/jama.2012.5669 22797452
Reddy M.P. Subramaniam A. Chua C. Ling R.R. Anstey C. Ramanathan K. Slutsky A.S. Shekar K. Respiratory system mechanics, gas exchange, and outcomes in mechanically ventilated patients with COVID-19-related acute respiratory distress syndrome: a systematic review and meta-analysis Lancet Respir. Med. 2022 10.1016/S2213-2600(22)00393-9
Robba C. Battaglini D. Ball L. Patroniti N. Loconte M. Brunetti I. Vena A. Giacobbe D.R. Bassetti M. Rocco P.R.M. Pelosi P. Distinct phenotypes require distinct respiratory management strategies in severe COVID-19 Respir. Physiol. Neurobiol. 279 2020 103455 10.1016/j.resp.2020.103455
Robba C. Battaglini D. Ball L. Pelosi P. Rocco P.R.M. Ten things you need to know about intensive care unit management of mechanically ventilated patients with COVID-19 Expert Rev. Respir. Med 17476348 2021 2021 1906226 10.1080/17476348.2021.1906226
Robba C. Battaglini D. Ball L. Valbusa A. Porto I. Della Bona R. La Malfa G. Patroniti N. Brunetti I. Loconte M. Bassetti M. Giacobbe D.R. Vena A. Silva C.L.M. Rocco P.R.M. Pelosi P. Coagulative Disorders in Critically Ill COVID-19 Patients with Acute Distress Respiratory Syndrome: A Critical Review J. Clin. Med 10 2021 140 10.3390/jcm10010140 33401632
Robba C. Battaglini D. Pelosi P. Rocco P.R.M. Multiple organ dysfunction in SARS-CoV-2: MODS-CoV-2 Expert Rev. Respir. Med. 14 2020 865 868 10.1080/17476348.2020.1778470 32567404
Rocco P.R. Pelosi P. Pulmonary and extrapulmonary acute respiratory distress syndrome: myth or reality? Curr. Opin. Crit. Care 14 2008 50 55 10.1097/MCC.0b013e3282f2405b 18195626
Rossi S. Palumbo M.M. Sverzellati N. Busana M. Malchiodi L. Bresciani P. Ceccarelli P. Sani E. Romitti F. Bonifazi M. Gattarello S. Steinberg I. Palermo P. Lazzari S. Collino F. Cressoni M. Herrmann P. Saager L. Meissner K. Quintel M. Camporota L. Marini J.J. Gattinoni L. Mechanisms of oxygenation responses to proning and recruitment in COVID-19 pneumonia Intensive Care Med 48 2021 1 11 10.1007/s00134-021-06562-4 34505911
Sahebnasagh A. Mojtahedzadeh M. Najmeddin F. Najafi A. Safdari M. Rezai Ghaleno H. Habtemariam S. Berindan-Neagoe I. Nabavi S.M. A Perspective on Erythropoietin as a Potential Adjuvant Therapy for Acute Lung Injury/Acute Respiratory Distress Syndrome in Patients with COVID-19 Arch. Med. Res. 51 2020 631 635 10.1016/j.arcmed.2020.08.002 32863034
Sinha P. Matthay M.A. Calfee C.S. Is a “Cytokine Storm” Relevant to COVID-19? JAMA Intern. Med 180 2020 1152 10.1001/jamainternmed.2020.3313
Soliz J. Schneider-Gasser E.M. Arias-Reyes C. Aliaga-Raduan F. Poma-Machicao L. Zubieta-Calleja G. Furuya W.I. Trevizan-Baú P. Dhingra R.R. Dutschmann M. Coping with hypoxemia: Could erythropoietin (EPO) be an adjuvant treatment of COVID-19 Respir. Physiol. Neurobiol. 279 2020 103476 10.1016/j.resp.2020.103476
Soree P. Gupta R.K. Singh K. Desiraju K. Agrawal A. Vats P. Bharadwaj A. Baburaj T.P. Chaudhary P. Singh V.K. Verma S. Bajaj A.C. Singh S.B. Raised HIF1α during normoxia in high altitude pulmonary edema susceptible non-mountaineers Sci. Rep. 6 2016 26468 10.1038/srep26468 27210110
Suga K. Kawakami Y. Koike M. Iwanaga H. Matsunaga N. Characteristic crescentic subpleural lung zones with high ventilation (V)/perfusion (Q) ratios in interstitial pneumonia on V/Q quotient SPECT Nucl. Med. Commun 30 2009 881 889 10.1097/MNM.0b013e328330571d 19668096
Tasaka S. Ohshimo S. Takeuchi M. Yasuda H. Ichikado K. Tsushima K. Egi M. ARDS Clinical Practice Guideline 2021 J. Intensive Care 10 2022 32 10.1186/s40560-022-00615-6 35799288
Thille A.W. Esteban A. Fernández-Segoviano P. Rodriguez J.-M. Aramburu J.-A. Peñuelas O. Cortés-Puch I. Cardinal-Fernández P. Lorente J.A. Frutos-Vivar F. Comparison of the Berlin Definition for Acute Respiratory Distress Syndrome with Autopsy Am. J. Respir. Crit. Care Med 187 2013 761 767 10.1164/rccm.201211-1981OC 23370917
Tian M. Liu W. Li X. Zhao P. Shereen M.A. Zhu C. Huang S. Liu S. Yu X. Yue M. Pan P. Wang W. Li Y. Chen X. Wu K. Luo Z. Zhang Q. Wu J. HIF-1α promotes SARS-CoV-2 infection and aggravates inflammatory responses to COVID-19 Signal Transduct. Target. Ther. 6 2021 308 10.1038/s41392-021-00726-w 34408131
Tonelli R. Marchioni A. Tabbì L. Fantini R. Busani S. Castaniere I. Andrisani D. Gozzi F. Bruzzi G. Manicardi L. Demurtas J. Andreani A. Cappiello G.F. Samarelli A.V. Clini E. Spontaneous Breathing and Evolving Phenotypes of Lung Damage in Patients with COVID-19: Review of Current Evidence and Forecast of a New Scenario J. Clin. Med 10 2021 975 10.3390/jcm10050975 33801368
Travis W.D. Costabel U. Hansell D.M. King T.E. Lynch D.A. Nicholson A.G. Ryerson C.J. Ryu J.H. Selman M. Wells A.U. Behr J. Bouros D. Brown K.K. Colby T.V. Collard H.R. Cordeiro C.R. Cottin V. Crestani B. Drent M. Dudden R.F. Egan J. Flaherty K. Hogaboam C. Inoue Y. Johkoh T. Kim D.S. Kitaichi M. Loyd J. Martinez F.J. Myers J. Protzko S. Raghu G. Richeldi L. Sverzellati N. Swigris J. Valeyre D. An Official American Thoracic Society/European Respiratory Society Statement: Update of the International Multidisciplinary Classification of the Idiopathic Interstitial Pneumonias Am. J. Respir. Crit. Care Med 188 2013 733 748 10.1164/rccm.201308-1483ST 24032382
Villar J. Ferrando C. Martínez D. Ambrós A. Muñoz T. Soler J.A. Aguilar G. Alba F. González-Higueras E. Conesa L.A. Martín-Rodríguez C. Díaz-Domínguez F.J. Serna-Grande P. Rivas R. Ferreres J. Belda J. Capilla L. Tallet A. Añón J.M. Fernández R.L. González-Martín J.M. Aguilar G. Alba F. Álvarez J. Ambrós A. Añón J.M. Asensio M.J. Belda J. Blanco J. Blasco M. Cachafeiro L. del Campo R. Capilla L. Carbonell J.A. Carbonell N. Cariñena A. Carriedo D. Chico M. Conesa L.A. Corpas R. Cuervo J. Díaz-Domínguez F.J. Domínguez-Antelo C. Fernández L. Fernández R.L. Ferrando C. Ferreres J. Gamboa E. González-Higueras E. González-Luengo R.I. González-Martín J.M. Martínez D. Martín-Rodríguez C. Muñoz T. Ortiz Díaz-Miguel R. Pérez-González R. Prieto A.M. Prieto I. Rivas R. Rojas-Viguera L. Romera M.A. Sánchez-Ballesteros J. Segura J.M. Serna-Grande P. Serrano A. Solano R. Soler J.A. Soro M. Tallet A. Villar J. Dexamethasone treatment for the acute respiratory distress syndrome: a multicentre, randomised controlled trial Lancet Respir. Med 8 2020 267 276 10.1016/S2213-2600(19)30417-5 32043986
Wichmann D. Sperhake J.-P. Lütgehetmann M. Steurer S. Edler C. Heinemann A. Heinrich F. Mushumba H. Kniep I. Schröder A.S. Burdelski C. de Heer G. Nierhaus A. Frings D. Pfefferle S. Becker H. Bredereke-Wiedling H. de Weerth A. Paschen H.-R. Sheikhzadeh-Eggers S. Stang A. Schmiedel S. Bokemeyer C. Addo M.M. Aepfelbacher M. Püschel K. Kluge S. Autopsy Findings and Venous Thromboembolism in Patients With COVID-19 Ann. Intern. Med. 173 2020 268 277 10.7326/M20-2003 32374815
Wiersinga W.J. Rhodes A. Cheng A.C. Peacock S.J. Prescott H.C. Pathophysiology, Transmission, Diagnosis, and Treatment of Coronavirus Disease 2019 (COVID-19) JAMA 324 2020 782 10.1001/jama.2020.12839 32648899
Wilkins M.R. Ghofrani H.-A. Weissmann N. Aldashev A. Zhao L. Pathophysiology and Treatment of High-Altitude Pulmonary Vascular Disease Circulation 131 2015 582 590 10.1161/CIRCULATIONAHA.114.006977 25666980
Xaubet A. Ancochea J. Bollo E. Fernández-Fabrellas E. Franquet T. Molina-Molina M. Montero M.A. Serrano-Mollar A. Guidelines for the Diagnosis and Treatment of Idiopathic Pulmonary Fibrosis Arch. Bronconeumol. (English Ed 49 2013 343 353 10.1016/j.arbr.2013.06.003
Xu Z. Shi L. Wang Y. Zhang J. Huang L. Zhang C. Liu S. Zhao P. Liu H. Zhu L. Tai Y. Bai C. Gao T. Song J. Xia P. Dong J. Zhao J. Wang F.-S. Pathological findings of COVID-19 associated with acute respiratory distress syndrome Lancet Respir. Med. 8 2020 420 422 10.1016/S2213-2600(20)30076-X 32085846
Yan R. Zhang Y. Li Y. Xia L. Guo Y. Zhou Q. Structural basis for the recognition of SARS-CoV-2 by full-length human ACE2 Science (80-.) 367 2020 1444 1448 10.1126/science.abb2762
Zhang S. Liu J. Jiang D. Wuren T. Ma S. Du Y. Yi X. Wu S. The plasma level changes of VEGF and soluble VEGF receptor-1 are associated with high-altitude pulmonary edema J. Med. Invest 65 2018 64 68 10.2152/jmi.65.64 29593196
Zubieta-Calleja, G., 1989. High Altitude Pathology at 12000 Ft.
Zubieta-Calleja, G., Ardaya, G., Zubieta DeUrioste, N., Paulev, P., Zubieta-Castillo, G., 2013. Tolerance to hypoxia [WWW Document]. zuniv.net. URL 〈https://zuniv.net/pub/TolerancetoHypoxiaFiziol.pdf〉.
Zubieta-Calleja G. Zubieta-DeUrioste N. The Oxygen Transport Triad in High-Altitude Pulmonary Edema: A Perspective from the High Andes Int. J. Environ. Res. Public Health 18 2021 7619 10.3390/ijerph18147619 34300070
Zubieta-Calleja G. Zubieta-DeUrioste N. Pneumolysis and “Silent Hypoxemia” in COVID-19 Indian J. Clin. Biochem. 36 2021 112 116 10.1007/s12291-020-00935-0 33191989
Zubieta-Calleja G. Zubieta-DeUrioste N. Venkatesh T. Das K.K. Soliz J. COVID-19 and Pneumolysis Simulating Extreme High-altitude Exposure with Altered Oxygen Transport Physiology; Multiple Diseases, and Scarce Need of Ventilators: Andean Condor’s-eye-view Rev. Recent Clin. Trials 15 2021 347 359 10.2174/1574887115666200925141108
Zubieta-Calleja G.R. Zubieta-DeUrioste N. High Altitude Pulmonary Edema, High Altitude Cerebral Edema, and Acute Mountain Sickness: an enhanced opinion from the High Andes – La Paz, Bolivia 3,500 m Rev. Environ. Health 2022 10.1515/reveh-2021-0172
| 36460252 | PMC9707029 | NO-CC CODE | 2022-12-08 23:16:23 | no | Respir Physiol Neurobiol. 2023 Mar 29; 309:104000 | utf-8 | Respir Physiol Neurobiol | 2,022 | 10.1016/j.resp.2022.104000 | oa_other |
==== Front
Infect Genet Evol
Infect Genet Evol
Infection, Genetics and Evolution
1567-1348
1567-7257
The Authors. Published by Elsevier B.V.
S1567-1348(22)00186-1
10.1016/j.meegid.2022.105389
105389
Article
The risk of COVID-19 can be predicted by a nomogram based on m6A-related genes
Lu Lingling ab1
Li Yijing c1
Ao Xiulan b1
Huang Jiaofeng a
Liu Bang ab
Wu Liqing b
Li Dongliang ab⁎
a Fuzong Clinical Medical College of Fujian Medical University, The 900th hospital. No.156 Xierhuan Road, Fuzhou, Fujian 350025, China
b Department of Hepatobiliary Disease, 900th Hospital of Joint Logistics Support Force, No.156 Xierhuan Road, Fuzhou, Fujian 350025, China
c Fuzhou First Hospital Affiliated to Fujian Medical University, Fuzhou, Fujian, China
⁎ Corresponding author at: Fuzong Clinical Medical College of Fujian Medical University, The 900th hospital of Joint Logistics Support Force. No.156 Xierhuan Road, Fuzhou, Fujian 350025, China.
1 Contributed equally to this work.
29 11 2022
12 2022
29 11 2022
106 105389105389
29 6 2022
24 10 2022
28 11 2022
© 2022 The Authors
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Background
The expression of m6A-related genes and their significance in COVID-19 patients are still unknown.
Methods
The GSE177477 and GSE157103 datasets of the Gene Expression Omnibus were used to extract RNA-seq data. The expression of 26 m6A-related genes and immune cell infiltration in COVID-19 patients were analyzed. Finally, we built and validated a nomogram model to predict the risk of COVID-19 infection.
Results
There were significant differences in 11 m6A regulatory factors between patients with COVID-19 and healthy individuals. The classification of disease subtypes based on m6A-related gene levels can be distinguished. COVID-19 patients in GSE177477 were classified into two categories based on m6A-related genes. The patients in cluster A were all symptomatic, while those in cluster B were asymptomatic. A significant correlation was also found between immune cells and m6A-related genes. Finally, seven m6A-related disease-characteristic genes, HNRNPA2B1, ELAVL1, RBM15, RBM15B, YTHDC1, HNRNPC, and WTAP, were screened to construct a nomogram model for predicting risk. The calibration curve, decision curve analysis, and clinical impact curve analysis were used to show that the nomogram model was effective and had a high net efficacy for risk prediction.
Conclusions
m6A-related genes were correlated with immune cells. The nomogram model effectively predicted COVID-19 risk. Moreover, m6A-related genes may be associated with the presence or absence of symptoms in COVID-19 patients.
Keywords
COVID-19
N6-methyladenosine
Risk
Immune
==== Body
pmc1 Introduction
A wide range of clinical symptoms are caused by the Coronavirus disease-2019 (COVID-19) and are exceedingly variable, including asymptomatic cases, pneumonia, and severe hypoxia (Fung et al., 2020). In individuals with COVID-19, the elderly, diabetes, hypertension, and obesity may significantly increase the risk of hospitalization and mortality (Muniyappa and Gubbi, 2020). The mechanisms behind the disparities in illness risk and outcomes across various groups remain obscure. Globally, elucidating the underlying pathophysiology and finding therapeutic approaches to COVID-19 has become imperative.
Several post-transcriptional RNA modifications have been identified in eukaryotic cells. N6-methyladenosine (m6A) is one of the most frequent modifications found across mammalian RNA transcripts and is common in messenger RNAs and some non-coding RNAs (Chen et al., 2020; Roundtree et al., 2017; Yang et al., 2018). m6A is incorporated into mRNA and is reversibly regulated by writers (methyltransferases) and erasers (demethylases) through cyclic enzymatic activity. m6A modification affects many events in mRNA metabolism, including maturation, folding, export, localization, translation, and degradation (Frye et al., 2016; Meyer and Jaffrey, 2017).
Many studies have confirmed the presence of m6A modifications in viral RNA transcripts and the importance of these modifications to the regulation of the viral life cycle (Dimock and Stoltzfus, 1977; Gokhale et al., 2016; Kane and Beemon, 1985; Lichinchi et al., 2016a). A relationship between m6A regulatory factors and COVID-19 has been found. Gokhale et al. (Gokhale et al., 2016) found that deletion of METTL3 and METTL4 increases the production of infectious viral particles of the hepatitis C virus (HCV) NS5A along with the HCV infection rate. In addition, silencing of METTL3 and METTL4 enhances the replication of the Zika virus (Lichinchi et al., 2016b). Recent studies have examined the dysregulation of m6A modification in host cells infected with SARS-CoV-2. Additionally, when SARS-CoV-2 infects host cells, m6A modification machinery is re-localized, and there is an increase in the cellular abundance of m6A (Liu et al., 2021). In addition, compared to healthy individuals, patients with severe COVID-19 exhibit a significant reduction in expression of the m6A writer gene METTL3 and inflammatory genes are upregulated in lung tissues (Li et al., 2021a). Moreover, downregulation of m6A-related genes has been observed in the blood leukocytes of COVID-19 patients (Qiu et al., 2021). In addition, RBM15 expression was upregulated and correlated with COVID-19 disease severity (Meng et al., 2021). Therefore, we speculated that m6A-related genes might play a key role in the infection process of SARS-CoV-2. Whether m6A-related genes are associated with disease risk requires further investigation.
In the present study, the expression levels of 26 m6A regulatory factors in COVID-19 patients and healthy individuals were analyzed. Differential gene expression levels in COVID-19 patients were used to classify patients. In addition, the differential expression of immune cells and the correlation between immune cells and m6A-related genes were analyzed. Finally, machine learning was used to screen disease-characteristic genes, which were then applied to construct a nomogram model to predict the risk of COVID-19.
2 Materials and methods
2.1 Sample sources
RNA-seq data was obtained from the GSE177477 (Masood et al., 2021) and GES157103 (Overmyer et al., 2021) datasets of the Gene Expression Omnibus (GEO) (Barrett et al., 2013), including data from 129 cases of patients with COVID-19 and 44 cases of healthy controls. There were 29 patients with COVID-19 in GSE177477. Eleven of the 29 patients were symptomatic while 18 were asymptomatic. The GSE157103 dataset included 100 COVID-19 patients and 26 healthy controls. As the data was from a publicly available GEO database, informed consent was guaranteed. R software (version 4.1.1) was used to analyze the expression of m6A-related genes.
2.2 Cluster analysis of patients with COVID-19
Sample clustering based to gene expression data is useful for the detection of disease subtypes. Partitioning Around Medoids (PAM) algorithm was used to construct a consensus matrix for patient grouping. Therefore, we conducted a cluster analysis of COVID-19-infected patients according to the levels of differentially expressed m6A-related genes, and then selected the optimal number of clusters for classification. Principal component analysis (PCA) was then performed to evaluate the effectiveness of the classification system. “ConsensusClusterPlus,” “limma,” “pheatmap,” “reshape2,” “ggpubr,” and “ggplot2” packages were used to perform classification and visualization.
Furthermore, m6A scores were evaluated for each patient. The m6A score is a score index based on the principal component analysis (PCA) algorithm, which combines the scores of patients' PC1 and PC2. According to the median, patients were divided into “m6A-score high” and “m6A-score low” groups.
2.3 The relationship between m6A-related genes and immune cells
Next, we analyzed the expression of 23 types of immune cells after clustering based on m6A-related gene expression. The correlation between m6A-related genes and immune cells has been previously studied. “GSEABase,” “limma,” “pheatmap,” “reshape2,” “ggpubr,” and “ggplot2” packages were applied to perform classification and visualization.
2.4 Construction and validation of the nomogram predictive model
First, machine learning, including random forest (RF) and support vector machine (SVM) classifiers, was applied to select disease-characteristic genes. Then, the accuracy of these two algorithms was evaluated according to the residual and reverse cumulative distributions of the residuals. An algorithm with higher accuracy was selected to further screen the characteristic genes. Additionally, the area under the ROC curve was used to measure the accuracy of the two algorithms.
Finally, the screened genes were used to establish a nomogram prediction model to evaluate disease risk. Validation analysis was performed using 1000 bootstrap resampling. A calibration curve was then drawn to measure the relationship between the nomogram model's prediction and actual risk. We assessed the net benefit using decision curve analysis (DCA) and measured the clinical usability using clinical impact curve analysis (CICA). The R software packages “rms” and “rmda” were applied to construct the model.
3 Results
3.1 Differential expression of m6A-related genes in COVID-19
RNA-seq data was obtained from the GSE177477 and GES157103 datasets of the GEO database. PCA results showed that the batch effect was eliminated (Fig. 1A). We used data from 173 patients, including 129 with COVID-19 and 44 healthy subjects. We analyzed the differential expression of 26 m6A regulators in different populations, including 9 writers, 15 readers, and 2 erasers. As shown in Fig. 1B, there were significant differences in 11 regulatory factors between the two groups: WTAP, RBM15, RBM15B, YTHDC1, YTHDF3, HNRNPC, FMR1, HNRNPA2B1, IGFBP2, ELAVL1, and IGF2BP1. Furthermore, WTAP, RBM15, HNRNPC, YTHDC1, FMR1, HNRNPA2B1, ELAVL1, and YTHDF3 were significantly upregulated in COVID-19 patients, while RBM15B, IGFBP2, and IGF2BP1 were significantly down-regulated. The expression of these 11 m6A-related genes in all subjects is shown in Fig. 1C.Fig. 1 m6A-related gene expression levels. (A) PCA showed elimination of batch effects. (B)Differential expression of m6A-related genes between patients with COVID-19 and the control group. (C) Heat map of differential gene expression levels in each sample.
Fig. 1
3.2 Cluster analysis
The best consensus matrix, k = 2, was obtained using the PAM algorithm (Fig. 2A). According to the expression of m6A-related genes, 129 patients with COVID-19 were divided into two categories: clusters A and B. The expression level of each m6A-related gene in each patient is shown in Fig. 2B. Significant differences was observed in the expression of m6A-related genes between clusters A and B (Fig. 2C). Finally, PCA results showed that clusters A and B could be distinguished according to m6A-related gene expression (Fig. 2D).Fig. 2 COVID-19 patients were grouped into clusters based on the expression levels of m6A-related genes. (A) The best consensus matrix (k = 2). (B) Heat map of m6A-related gene expression in COVID-19 patients. (C) Differential expression in clusters A and B. (D) Principal component analysis indicated that clustering was effective.
Fig. 2
Clustering of the 47 patients within the GSE177477 dataset alone revealed an interesting observation. The grouping obtained by clustering was consistent with that of the patients with or without concomitant symptoms. In other words, the patients in cluster A were all symptomatic, while those in cluster B were asymptomatic (Supplementary Fig. 1).
3.3 The relationship between m6A-related genes and immune cells
Next, we assessed the differential expression of 23 types of immune cells in clusters A and B. Fig. 3A illustrates the significant differences between the two groups with respect to 11 types of immune cells. Eosinophils (p < 0.01), natural killer cells (p < 0.05), neutrophils (p < 0.001), and type 2 T helper cells (p < 0.05) were significantly up-regulated in cluster B. However, activated CD8 T cells (p < 0.001), CD56bright natural killer cells (p < 0.01), CD56dim natural killer cells (p < 0.001), MDSCs (p < 0.001), monocytes (p < 0.01), regulatory T cells (p < 0.05), and type 1 T helper cells (p < 0.01) were significantly up-regulated in cluster A.Fig. 3 The relationship between m6A-related gene expression levels and immune cells. (A) The expression of immune cells in different clusters. (B) The correlation between m6A-related gene expression levels and immune cells. (C) The expression of immune cells in high and low RBM15B gene expression groups.
Fig. 3
The correlation between m6A and immune cells is shown in Fig. 3B. The gene RBM15B, which had a strong correlation, was selected for further analysis. Based on the RBM15B expression level, patients with COVID-19 were divided into two groups, high RBM15B expression and low RBM15B expression. Differences in immune cells were also compared between these two groups. Fourteen types of immune cells were significantly different between the RBM15B high expression and RBM15B low expression groups (Fig. 3C).
3.4 m6A score
We established a scoring system, the m6A-score, based on a composite score of m6A-related gene expression in COVID-19 patients. Each patient was assigned an m6A-score based on this analysis (Supplementary Table 1). The scores were compared between the different disease subtypes. The results indicated that the m6A score in m6A cluster B was higher than that in m6A cluster A (p < 0.001) (Fig. 4A).
3.5 RF and SVM algorithms for screening disease-feature genes
To examine the relationship between m6A-related genes and the disease risk of COVID-19 more accurately, we combined samples from two datasets, GSE177477 and GSE157103. The RF and SVM algorithms were used to screen the feature genes of the disease. We then compared the residual and reverse cumulative distributions of the residuals of these two algorithms. As shown in Fig. 4 B-C, regardless of the residual or reverse cumulative distribution of the residual, RF was better than SVM, which indicates that the RF algorithm is more accurate than the SVM algorithm. Furthermore, ROC curve analysis showed an AUC of 1 in RF and an AUC of 0.971 in SVM, which indicates that the screening of disease-characteristic genes was valid (Fig. 4D).Fig. 4 The m6A score and application of machine learning to screen disease signature genes. (A) m6A scores of clusters A and B. Residual (B) and reverse cumulative distribution of the residual (C) were used to compare the accuracy of RF and SVM algorithms. (D) RF and SVM algorithms were evaluated by ROC curves. (E) The optimal nTree was selected in the random forest analysis to screen disease-characteristic genes. (F) The importance score of the relevant gene was calculated.
Fig. 4
Then, a random forest algorithm with the nTree set to 500 was used to screen disease-characteristic genes, which revealed that the error was small and stable at an nTree of 188 (Fig. 4E). We then calculated the importance scores of 11 m6A-related genes. Finally, nine genes were screened based on an importance score of >5, including HNRNPC, WTAP, HNRNPA2B1, ELAVL1, RBM15, YTHDC1, YTHDF3, RBM15B, and FMR1 (Fig. 4F).
3.6 The nomogram prediction model
Based on the results of the random forest analysis and importance score, we selected seven genes to describe the impact of each gene on the risk of COVID-19 and set up a nomogram prediction model related to the disease (Fig. 5A). The seven m6A-related genes were HNRNPC, WTAP, HNRNPA2B2, ELAVL1, RBM15, YTHDC1, and RBM15B. The different expression levels of the seven genes were given corresponding scores. The total score could be used to predict the risk of COVID-19. In the risk prediction nomograms, YTHDC1 made the greatest contribution, suggesting that YTHDC1 was the most suitable for predicting the risk of disease infection. We then validated this model. The calibration curve showed good consistency between the predicted and observed results (Fig. 5B). DCA showed that the use of a nomogram model for predicting COVID-19 risk had a high net benefit (Fig. 5C). The CICA revealed that high-risk patients classified by the nomogram model had a high degree of coincidence with the actual positive patients (Fig. 5D).Fig. 5 Predictive models for the risk of COVID-19 patients. (A) The nomogram prediction model. (B) The calibration curve shows good consistency between the predicted results and the observed results. (C) Decision curve analysis shows a high net benefit. (D) Clinical impact curve analysis verifies the validity of the nomogram model.
Fig. 5
4 Discussion
Despite advancements in vaccination, SARS-CoV-2 continues to cause severe acute respiratory syndrome and lung damage in a considerable number of patients. It is crucial for the host antiviral defense that the immune system responds to viruses invading the host cells by generating interferon (IFN), inflammatory cytokines, and chemokines. Nevertheless, proinflammatory cytokines and chemokines can trigger cytokine storms in COVID-19 patients, leading to serious illnesses. SARS-CoV-2 is a member of the genus-Coronavirus, a positive-sense RNA virus (Lu et al., 2020). It has been revealed that m6A modifications are required to progress the RNA viral life cycle and the host cell's immune response (Dimock and Stoltzfus, 1977; Gokhale et al., 2016; Kane and Beemon, 1985; Lichinchi et al., 2016a). There is also evidence that SARS-CoV-2 genome translation to replication is regulated by the interaction between m6A-marked RNA and hnRNPA1 (Kumar et al., 2022).
Here, we analyzed m6A methylation modification in 129 COVID-19 patients and 44 healthy controls and found that eight genes, including WTAP, were significantly up-regulated in COVID-19, while three genes, including RBM15B, IGFBP2, and IGF2BP1, were significantly down-regulated. Changes in the expression of m6A-related genes indicate that the m6A methylome is aberrant in COVID-19 patients. A study on the Zika virus showed a decrease in the replication of the Zika virus when silencingm6A erasers, including FTO (Lichinchi et al., 2016b). WTAP participates in the m6A methyltransferase complex and m6A methylation modification by combining METTL3-METTL14 heterodimers (Fu et al., 2014). WTAP knockdown reduced methyl marks on m6A and decreased human Coronavirus OC43(HCoV-OC43) replication (Burgess et al., 2021). This evidence implies that a severe response may be due to uncontrollable virus replication. TGF-β controls the methylation of m6 A mRNA through SMAD2/3 interaction (Bertero et al., 2018). TGF-β was significantly different among groups with different disease states of COVID-19 (Ghazavi et al., 2021). Healthy controls do not exhibit any significant TGF-β1-expressing SARS-CoV-2-reactive Th cells (Ferreira-Gomes et al., 2021).
Next, according to the expression of m6A-related genes, COVID-19 patients were divided into two groups, cluster A and cluster B. Interestingly, the subgroup of GSE 177477 using the m6A-related gene cluster was the same as the group of patients with or without symptoms. This may be due to a deviation caused by an insufficient sample size, which is worthy of our attention. The transmission rate in asymptomatic patients is similar to that in symptomatic patients (Li et al., 2020). Our results provide an explanation for genetic effects on SARS-CoV-2 infection outcomes.
There was a significant difference in immune cells between clusters A and B in COVID-19 patients. Additionally, m6A is correlated with immune cells. A close association between m6A regulators and immune cell infiltration has been demonstrated in other studies (Cheng et al., 2022; Guo et al., 2021; Zhang et al., 2020). Although it is not clear by what mechanism m6A-related genes regulate the immune response, this has provided us with a new perspective. It was found that depletion of METTL3 in host cells decreased the m6A level of SARS-CoV-2, further increased RIG-I binding, and enhanced the expression of downstream innate immune signaling pathways (Li et al., 2021a). Dendritic cells (DCs) with METTL3 silencing display immature properties, and prolong the survival of allografts (Wu et al., 2020). Deletion of METTL3 in T cells affects T cell differentiation and homeostatic expansion (Li et al., 2017). Heterogeneous nuclear ribonucleoprotein C (hnRNPC) regulates neutrophil recruitment through its interaction with urokinase-type plasminogen activator receptors (uPARs) (Bhandary et al., 2009). Overexpression of HNRNPC in Parkinson's disease can inhibit the expression of inflammatory factors, such as IFN-β, IL-6, and TNF-α(Quan et al., 2021). Heterogeneous ribonucleoproteins (hnRNPs), such as hnRNP-K can regulate macrophage recruitment and activation (Ostareck and Ostareck-Lederer, 2019). With the continuous increase in the number of COVID-19 infections, the study of the mechanisms of the disease immune response have become particularly important. Response characteristics are helpful for understanding pathogenesis, controlling infections and for developing vaccines. Cellular immunity caused by Coronavirus infections mainly involves antigen-presenting cells (e.g., macrophages) and T cells (Tang et al., 2020). Studies have shown that antigen presentation is inhibited in symptomatic COVID-19 patients (Masood et al., 2021). T cell responses can be detected in most COVID-19 patients (Braun et al., 2020; Grifoni et al., 2020; Thevarajan et al., 2020). T cell activation characterized by CD38 expression, can be used as a marker of the acute phase of COVID-19 (Sekine et al., 2020). Braun et al. found that peak reactive CD4 T cells were detected in 83% of COVID-19 patients and 35% of healthy controls (Braun et al., 2020). It is reasonable to speculate that there may be T cell cross-reactions which provide some protection against infection by SARS-CoV-2. This also helps to explain the different clinical outcomes of people infected with the virus. In severe COVID-19, sustained cytokine production and hyper-inflammation can be observed (Giamarellos-Bourboulis et al., 2020). Some studies have also found that plasma cell-like dendritic cells (pDCs) are rapidly consumed in severe COVID-19 patients (Laing et al., 2020).
Finally, seven disease-characteristic genes were screened using the random forest algorithm, including HNRNPA2B1, ELAVL1, RBM15, RBM15B, YTHDC1, HNRNPC, and WTAP. A nomogram prediction model based on these seven m6A-related genes was established to assess the risk of COVID-19. DCA and CICA further confirmed that the accuracy of the risk-scoring model. Some reports have established risk-scoring models based on m6A regulatory factors, but mainly in oncology, including ovarian cancer, clear cell renal cell carcinoma, and neuroblastoma (Li et al., 2021b; Wang et al., 2020; Zhao et al., 2020). Nomogram prediction models for COVID-19 are based on baseline clinical characteristics. These nomogram models can predict the survival, or overall mortality, of patients (Dong et al., 2021; Moon et al., 2021). However, there are few prediction models based on m6A-related genes. Our study is the first to establish a nomogram model based on m6A regulatory factors to effectively predict the risk of COVID-19. This provides a new approach to clinical practice. This nomogram provides an easy-to-use individualized assessment tool. Using a nomogram, clinicians can calculate an overall score based on the expression of these seven m6A-related genes. The higher the total nomogram score, the greater the COVID-19 risk. The total score corresponds to different levels of risk in the nomogram, which provides clinical value for individualized decision-making.
There are still some limitations in the present study. Firstly, our sample came from two datasets, and so the number of patients was limited. To reduce deviation, multi-center samples and an expansion of the sample size will be required. Secondly, there was a lack of basic clinical characteristics of the patients, which may have affected the accuracy of the nomogram model. In addition, there may also be important variables among these clinical factors. Finally, the specific mechanisms of m6A-related genes in COVID-19 must be verified through further experiments.
5 Conclusions
In conclusion, the nomogram model based on seven m6A-related genes, including HNRNPA2B1, YTHDC1, ELAVL1, RBM15, RBM15B, WTAP, and HNRNPC, effectively predicted the risk of COVID-19. Among these, YTHDC1 was the most suitable for predicting the risk of infection. The m6A-related genes were significantly related to immune cells. Moreover, m6A-related genes may be associated with the presence or absence of symptoms in COVID-19 patients.
Ethics approval and consent to participate
Since all the research data are from open online databases, all informed consent can be guaranteed.
Consent for publication
Not applicable.
Availability of data and materials
The data that support the findings of this study is available in the GSE177477 and GSE157103 dataset of Gene Expression Omnibus (GEO) at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE177477 and at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE157103, respectively.
Funding
This study was supported by grants from the 900th Hospital of the Joint Logistics Support Force (Grant Number:2020Q02), the 900th Hospital of the Joint Logistics Support Force Fund (Grant Number: 2020Z12), the 900th Hospital of the Joint Logistics Support Forceaa (Grant Number:2021JQ04), the 900th Hospital of the Joint Logistics Support Forcea (Grant Number: 2021MS12), and Guiding Project of Social Development of Fujian Province (Grant Number: 2021Y0062).
Authors' contributions
Study concept and design (ALL), acquisition of data (LL, YL, and XA), analysis and interpretation of data (LL, YL, XA, BL, and JH), and drafting of the manuscript (LL, JH, XA, and BL). All authors contributed to study supervision and critical revision of the manuscript. All authors have read and approved the final version of the manuscript.
The following are the supplementary data related to this article.Supplementary Fig. 1 COVID-19 patients in GSE 177477 were grouped into clusters based on the level of m6A-related genes. (A) the best consensus matrix k = 2. (B) Heat map of m6A-related gene expression in COVID-19 patients. (C) Differential expression in cluster A and B. (D) Principal component analysis indicated that clustering was effective.
Supplementary Fig. 1
Supplementary Table 1
m6A score in patients with COVID-19.
Supplementary Table 1
Declaration of Competing Interest
The authors declare that there is no conflict of interests.
Data availability
Data will be made available on request.
Acknowledgements
The authors thank the contributors to GEO for sharing open access to the expression profile data sets and gene-related information.
==== Refs
References
Barrett T. Wilhite S.E. Ledoux P. Evangelista C. Kim I.F. Tomashevsky M. Marshall K.A. Phillippy K.H. Sherman P.M. Holko M. Yefanov A. Lee H. Zhang N. Robertson C.L. Serova N. Davis S. Soboleva A. NCBI GEO: archive for functional genomics data sets--update Nucleic Acids Res. 41 2013 D991 D995 23193258
Bertero A. Brown S. Madrigal P. Osnato A. Ortmann D. Yiangou L. Kadiwala J. Hubner N.C. de Los Mozos I.R. Sadée C. Lenaerts A.S. Nakanoh S. Grandy R. Farnell E. Ule J. Stunnenberg H.G. Mendjan S. Vallier L. The SMAD2/3 interactome reveals that TGFβ controls m(6)a mRNA methylation in pluripotency Nature 555 2018 256 259 29489750
Bhandary Y.P. Velusamy T. Shetty P. Shetty R.S. Idell S. Cines D.B. Jain D. Bdeir K. Abraham E. Tsuruta Y. Shetty S. Post-transcriptional regulation of urokinase-type plasminogen activator receptor expression in lipopolysaccharide-induced acute lung injury Am. J. Respir. Crit. Care Med. 179 2009 288 298 19029002
Braun J. Loyal L. Frentsch M. Wendisch D. Georg P. Kurth F. Hippenstiel S. Dingeldey M. Kruse B. Fauchere F. Baysal E. Mangold M. Henze L. Lauster R. Mall M.A. Beyer K. Röhmel J. Voigt S. Schmitz J. Miltenyi S. Demuth I. Müller M.A. Hocke A. Witzenrath M. Suttorp N. Kern F. Reimer U. Wenschuh H. Drosten C. Corman V.M. Giesecke-Thiel C. Sander L.E. Thiel A. SARS-CoV-2-reactive T cells in healthy donors and patients with COVID-19 Nature 587 2020 270 274 32726801
Burgess H.M. Depledge D.P. Thompson L. Srinivas K.P. Grande R.C. Vink E.I. Abebe J.S. Blackaby W.P. Hendrick A. Albertella M.R. Kouzarides T. Stapleford K.A. Wilson A.C. Mohr I. Targeting the m(6)a RNA modification pathway blocks SARS-CoV-2 and HCoV-OC43 replication Genes Dev. 35 2021 1005 1019 34168039
Chen J. Jin L. Wang Z. Wang L. Chen Q. Cui Y. Liu G. N6-methyladenosine regulates PEDV replication and host gene expression Virology 548 2020 59 72 32838947
Cheng L. Li H. Zhan H. Liu Y. Li X. Huang Y. Wang L. Zhang F. Li Y. Alterations of m6A RNA methylation regulators contribute to autophagy and immune infiltration in primary Sjögren’s syndrome Front. Immunol. 13 2022 949206
Dimock K. Stoltzfus C.M. Sequence specificity of internal methylation in B77 avian sarcoma virus RNA subunits Biochemistry 16 1977 471 478 189800
Dong Y.M. Sun J. Li Y.X. Chen Q. Liu Q.Q. Sun Z. Pang R. Chen F. Xu B.Y. Manyande A. Clark T.G. Li J.P. Orhan I.E. Tian Y.K. Wang T. Wu W. Ye D.W. Development and validation of a nomogram for assessing survival in patients with COVID-19 pneumonia Clin. Infect. Dis.: Off. Publ. Infect. Dis. Soc. Am. 72 2021 652 660
Ferreira-Gomes M. Kruglov A. Durek P. Heinrich F. Tizian C. Heinz G.A. Pascual-Reguant A. Du W. Mothes R. Fan C. Frischbutter S. Habenicht K. Budzinski L. Ninnemann J. Jani P.K. Guerra G.M. Lehmann K. Matz M. Ostendorf L. Heiberger L. Chang H.D. Bauherr S. Maurer M. Schönrich G. Raftery M. Kallinich T. Mall M.A. Angermair S. Treskatsch S. Dörner T. Corman V.M. Diefenbach A. Volk H.D. Elezkurtaj S. Winkler T.H. Dong J. Hauser A.E. Radbruch H. Witkowski M. Melchers F. Radbruch A. Mashreghi M.F. SARS-CoV-2 in severe COVID-19 induces a TGF-β-dominated chronic immune response that does not target itself Nat. Commun. 12 2021 1961 33785765
Frye M. Jaffrey S.R. Pan T. Rechavi G. Suzuki T. RNA modifications: what have we learned and where are we headed? Nat. Rev. Genet. 17 2016 365 372 27140282
Fu Y. Dominissini D. Rechavi G. He C. Gene expression regulation mediated through reversible m6A RNA methylation Nat. Rev. Genet. 15 2014 293 306 24662220
Fung S.Y. Yuen K.S. Ye Z.W. Chan C.P. Jin D.Y. A tug-of-war between severe acute respiratory syndrome coronavirus 2 and host antiviral defence: lessons from other pathogenic viruses Emerg. Microbes Infect. 9 2020 558 570 32172672
Ghazavi A. Ganji A. Keshavarzian N. Rabiemajd S. Mosayebi G. Cytokine profile and disease severity in patients with COVID-19 Cytokine 137 2021 155323
Giamarellos-Bourboulis E.J. Netea M.G. Rovina N. Akinosoglou K. Antoniadou A. Antonakos N. Damoraki G. Gkavogianni T. Adami M.E. Katsaounou P. Ntaganou M. Kyriakopoulou M. Dimopoulos G. Koutsodimitropoulos I. Velissaris D. Koufargyris P. Karageorgos A. Katrini K. Lekakis V. Lupse M. Kotsaki A. Renieris G. Theodoulou D. Panou V. Koukaki E. Koulouris N. Gogos C. Koutsoukou A. Complex immune dysregulation in COVID-19 patients with severe respiratory failure Cell Host Microbe 27 2020 992–1000.e1003
Gokhale N.S. McIntyre A.B.R. McFadden M.J. Roder A.E. Kennedy E.M. Gandara J.A. Hopcraft S.E. Quicke K.M. Vazquez C. Willer J. Ilkayeva O.R. Law B.A. Holley C.L. Garcia-Blanco M.A. Evans M.J. Suthar M.S. Bradrick S.S. Mason C.E. Horner S.M. N6-Methyladenosine in Flaviviridae viral RNA genomes regulates infection Cell Host Microbe 20 2016 654 665 27773535
Grifoni A. Weiskopf D. Ramirez S.I. Mateus J. Dan J.M. Moderbacher C.R. Rawlings S.A. Sutherland A. Premkumar L. Jadi R.S. Marrama D. de Silva A.M. Frazier A. Carlin A.F. Greenbaum J.A. Peters B. Krammer F. Smith D.M. Crotty S. Sette A. Targets of T cell responses to SARS-CoV-2 coronavirus in humans with COVID-19 disease and unexposed individuals Cell 181 2020 1489–1501.e1415
Guo W. Tan F. Huai Q. Wang Z. Shao F. Zhang G. Yang Z. Li R. Xue Q. Gao S. He J. Comprehensive Analysis of PD-L1 Expression, Immune Infiltrates, and m6A RNA Methylation Regulators in Esophageal Squamous Cell Carcinoma. 12 2021
Kane S.E. Beemon K. Precise localization of m6A in Rous sarcoma virus RNA reveals clustering of methylation sites: implications for RNA processing Mol. Cell. Biol. 5 1985 2298 2306 3016525
Kumar R. Khandelwal N. Chander Y. Nagori H. Verma A. Barua A. Godara B. Pal Y. Gulati B.R. Tripathi B.N. Barua S. Kumar N. S-adenosylmethionine-dependent methyltransferase inhibitor DZNep blocks transcription and translation of SARS-CoV-2 genome with a low tendency to select for drug-resistant viral variants Antivir. Res. 197 2022 105232
Laing A.G. Lorenc A. Del Barrio I. Del Molino Das A. Fish M. Monin L. Muñoz-Ruiz M. McKenzie D.R. Hayday T.S. Francos-Quijorna I. Kamdar S. Joseph M. Davies D. Davis R. Jennings A. Zlatareva I. Vantourout P. Wu Y. Sofra V. Cano F. Greco M. Theodoridis E. Freedman J.D. Gee S. Chan J.N.E. Ryan S. Bugallo-Blanco E. Peterson P. Kisand K. Haljasmägi L. Chadli L. Moingeon P. Martinez L. Merrick B. Bisnauthsing K. Brooks K. Ibrahim M.A.A. Mason J. Lopez Gomez F. Babalola K. Abdul-Jawad S. Cason J. Mant C. Seow J. Graham C. Doores K.J. Di Rosa F. Edgeworth J. Shankar-Hari M. Hayday A.C. A dynamic COVID-19 immune signature includes associations with poor prognosis Nat. Med. 26 2020 1623 1635 32807934
Li H.B. Tong J. Zhu S. Batista P.J. Duffy E.E. Zhao J. Bailis W. Cao G. Kroehling L. Chen Y. Wang G. Broughton J.P. Chen Y.G. Kluger Y. Simon M.D. Chang H.Y. Yin Z. Flavell R.A. M(6)a mRNA methylation controls T cell homeostasis by targeting the IL-7/STAT5/SOCS pathways Nature 548 2017 338 342 28792938
Li R. Pei S. Chen B. Song Y. Zhang T. Yang W. Shaman J. Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV-2) Science (New York, N.Y.) 368 2020 489 493 32179701
Li N. Hui H. Bray B. Gonzalez G.M. Zeller M. Anderson K.G. Knight R. Smith D. Wang Y. Carlin A.F. Rana T.M. METTL3 regulates viral m6A RNA modification and host cell innate immune responses during SARS-CoV-2 infection Cell Rep. 35 2021 109091
Li Q. Ren C.C. Chen Y.N. Yang L. Zhang F. Wang B.J. Zhu Y.H. Li F.Y. Yang J. Zhang Z.A. A risk score model incorporating three m6A RNA methylation regulators and a related network of miRNAs-m6A regulators-m6A target genes to predict the prognosis of patients with ovarian cancer Front. Cell Dev. Biol. 9 2021 703969
Lichinchi G. Gao S. Saletore Y. Gonzalez G.M. Bansal V. Wang Y. Mason C.E. Rana T.M. Dynamics of the human and viral m(6)A RNA methylomes during HIV-1 infection of T cells Nat. Microbiol. 1 2016 16011 27572442
Lichinchi G. Zhao B.S. Wu Y. Lu Z. Qin Y. He C. Rana T.M. Dynamics of human and viral RNA methylation during Zika virus infection Cell Host Microbe 20 2016 666 673 27773536
Liu J. Xu Y.P. Li K. Ye Q. Zhou H.Y. Sun H. Li X. Yu L. Deng Y.Q. Li R.T. Cheng M.L. He B. Zhou J. Li X.F. Wu A. Yi C. Qin C.F. The m(6)A methylome of SARS-CoV-2 in host cells Cell Res. 31 2021 404 414 33510385
Lu R. Zhao X. Li J. Niu P. Yang B. Wu H. Wang W. Song H. Huang B. Zhu N. Bi Y. Ma X. Zhan F. Wang L. Hu T. Zhou H. Hu Z. Zhou W. Zhao L. Chen J. Meng Y. Wang J. Lin Y. Yuan J. Xie Z. Ma J. Liu W.J. Wang D. Xu W. Holmes E.C. Gao G.F. Wu G. Chen W. Shi W. Tan W. Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding Lancet (Lond. Engl.) 395 2020 565 574
Masood K.I. Yameen M. Ashraf J. Shahid S. Mahmood S.F. Nasir A. Nasir N. Jamil B. Ghanchi N.K. Khanum I. Razzak S.A. Kanji A. Hussain R. M E.R Hasan Z. Upregulated type I interferon responses in asymptomatic COVID-19 infection are associated with improved clinical outcome Sci. Rep. 11 2021 22958 34824360
Meng Y. Zhang Q. Wang K. Zhang X. Yang R. Bi K. Chen W. Diao H. RBM15-mediated N6-methyladenosine modification affects COVID-19 severity by regulating the expression of multitarget genes Cell Death Dis. 12 2021 732 34301919
Meyer K.D. Jaffrey S.R. Rethinking m(6)A readers, writers, and erasers Annu. Rev. Cell Dev. Biol. 33 2017 319 342 28759256
Moon H.J. Kim K. Kang E.K. Yang H.J. Lee E. Prediction of COVID-19-related mortality and 30-day and 60-day survival probabilities using a nomogram J. Korean Med. Sci. 36 2021 e248
Muniyappa R. Gubbi S. COVID-19 pandemic, coronaviruses, and diabetes mellitus Am. J. Physiol. Endocrinol. Metab. 318 2020 E736–e741
Ostareck D.H. Ostareck-Lederer A. RNA-binding proteins in the control of LPS-induced macrophage response Front. Genet. 10 2019 31 30778370
Overmyer K.A. Shishkova E. Miller I.J. Balnis J. Bernstein M.N. Peters-Clarke T.M. Meyer J.G. Quan Q. Muehlbauer L.K. Trujillo E.A. He Y. Chopra A. Chieng H.C. Tiwari A. Judson M.A. Paulson B. Brademan D.R. Zhu Y. Serrano L.R. Linke V. Drake L.A. Adam A.P. Schwartz B.S. Singer H.A. Swanson S. Mosher D.F. Stewart R. Coon J.J. Jaitovich A. Large-scale multi-omic analysis of COVID-19 severity Cell Syst. 12 2021 23 40.e27 33096026
Qiu X. Hua X. Li Q. Zhou Q. Chen J. M(6)a regulator-mediated methylation modification patterns and characteristics of immunity in blood leukocytes of COVID-19 patients Front. Immunol. 12 2021 774776
Quan W. Li J. Liu L. Zhang Q. Qin Y. Pei X. Chen J. Influence of N6-Methyladenosine modification gene HNRNPC on cell phenotype in Parkinson’s disease Park. Dis. 2021 2021 9919129
Roundtree I.A. Evans M.E. Pan T. He C. Dynamic RNA modifications in gene expression regulation Cell 169 2017 1187 1200 28622506
Sekine T. Perez-Potti A. Rivera-Ballesteros O. Strålin K. Gorin J.B. Olsson A. Llewellyn-Lacey S. Kamal H. Bogdanovic G. Muschiol S. Wullimann D.J. Kammann T. Emgård J. Parrot T. Folkesson E. Rooyackers O. Eriksson L.I. Henter J.I. Sönnerborg A. Allander T. Albert J. Nielsen M. Klingström J. Gredmark-Russ S. Björkström N.K. Sandberg J.K. Price D.A. Ljunggren H.G. Aleman S. Buggert M. Robust T cell immunity in convalescent individuals with asymptomatic or mild COVID-19 Cell 183 2020 158–168.e114
Tang Y. Liu J. Zhang D. Xu Z. Ji J. Wen C. Cytokine storm in COVID-19: the current evidence and treatment strategies Front. Immunol. 11 2020 1708 32754163
Thevarajan I. Nguyen T.H.O. Koutsakos M. Druce J. Caly L. van de Sandt C.E. Jia X. Nicholson S. Catton M. Cowie B. Tong S.Y.C. Lewin S.R. Kedzierska K. Breadth of concomitant immune responses prior to patient recovery: a case report of non-severe COVID-19 Nat. Med. 26 2020 453 455 32284614
Wang Z. Cheng H. Xu H. Yu X. Sui D. A five-gene signature derived from m6A regulators to improve prognosis prediction of neuroblastoma Cancer Biomark.: Sect. A Dis. Markers 28 2020 275 284
Wu H. Xu Z. Wang Z. Ren Z. Li L. Ruan Y. Dendritic cells with METTL3 gene knockdown exhibit immature properties and prolong allograft survival Genes Immun. 21 2020 193 202 32457372
Yang Y. Hsu P.J. Chen Y.S. Yang Y.G. Dynamic transcriptomic m(6)A decoration: writers, erasers, readers and functions in RNA metabolism Cell Res. 28 2018 616 624 29789545
Zhang B. Wu Q. Li B. Wang D. Wang L. Zhou Y.L. M(6)A regulator-mediated methylation modification patterns and tumor microenvironment infiltration characterization in gastric cancer Mol. Cancer 19 2020 53 32164750
Zhao Y. Tao Z. Chen X. Identification of a three-m6A related gene risk score model as a potential prognostic biomarker in clear cell renal cell carcinoma PeerJ 8 2020 e8827
| 36460278 | PMC9707050 | NO-CC CODE | 2022-12-01 23:20:29 | no | Infect Genet Evol. 2022 Dec 29; 106:105389 | utf-8 | Infect Genet Evol | 2,022 | 10.1016/j.meegid.2022.105389 | oa_other |
==== Front
Int J Lat Am Relig
International Journal of Latin American Religions
2509-9957
2509-9965
Springer International Publishing Cham
184
10.1007/s41603-022-00184-4
Religion, Spirituality, and Health in Latin America
Religious Coping, Experiential Avoidance, Self-Compassion, and Post-Traumatic Stress by COVID-19: a Serial Mediation Study
http://orcid.org/0000-0001-5966-3907
Fuentes-Ferrada Reiner 12
http://orcid.org/0000-0001-7611-802X
Cerda-Planas Catalina [email protected]
34
http://orcid.org/0000-0001-5400-3168
Fernández María Beatriz 567
1 grid.7119.e 0000 0004 0487 459X Mind Body Lab, Facultad de Medicina, Instituto de Estudios Psicológicos, Universidad Austral de Chile, Valdivia, Chile
2 Millennium Nucleus to Improve, the Mental Health of Adolescents and Youths, Imhay, Santiago, Chile
3 grid.441800.9 0000 0001 2227 4350 Instituto Teológico Egidio Viganó, Universidad Católica Silva Henríquez, Santiago, Chile
4 grid.7870.8 0000 0001 2157 0406 Centro de Estudios de la Religión, Pontificia Universidad Católica de Chile, Santiago, Chile
5 grid.7870.8 0000 0001 2157 0406 Institute of Sociology, Pontificia Universidad Católica de Chile, Santiago, Chile
6 Millenium Institute for Care Research (MICARE), Santiago, Chile
7 grid.7870.8 0000 0001 2157 0406 Center UC for Studies in Age and Aging (CEVE-UC), Santiago, Chile
28 11 2022
112
22 8 2022
15 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The COVID-19 pandemic has brought consequences to mental health, with religiosity being a relevant coping factor in reducing the negative impact of the health crisis. Based on a convenience sample of Chilean adults, this study sought to explore the relationship between religious coping and post-traumatic stress due to COVID-19, hypothesizing that this relationship would be mediated by experiential avoidance and self-compassion. A non-experimental cross-sectional design was used, applying an online survey to 300 adults who lived in Chile. The results show that all variables are related and that experiential avoidance (EA) and self-compassion play a serial mediating role in the relationship between negative religious coping (NRC) and post-traumatic stress. Furthermore, the results showed that religious coping, experiential avoidance, self-compassion, and COVID-19 post-traumatic stress are significantly interrelated. Moreover, a serial mediation was found among the variables: higher experiential avoidance and lower level of self-compassion mediate the impact of COVID-19 post-traumatic stress in people with negative religious coping. At the end of the article, the implications of the results and how these variables interact in a serial mediation mechanism that sheds further light on the relationships between negative religious coping, mental health, and adverse situations such as COVID-19 are discussed.
Keywords
Religious coping
Experiential avoidance
Self-compassion
Post-traumatic stress
COVID-19
==== Body
pmcIntroduction
In the context of the COVID-19 pandemic, mental health has been strongly altered by changes in the routine and lockdown contexts, which has significantly increased the prevalence of adverse outcomes in mental health, such as bad mood, worry, depression, anxiety, and suicidal thoughts and behaviors (Hit and Struggling 2020). In turn, in the first seven months of the pandemic, a higher prevalence of depression, anxiety, and stress was observed than in the months before the pandemic (Lakhan et al. 2020). It has also been indicated that the impact on health is disproportionate among the population and is mediated by social and cultural determinants (Baqui et al. 2020; Laurencin and McClinton 2020; Zhang and Schwartz 2020). Among them, religiosity has been postulated as a protective factor for mental health in symptomatic aspects such as suicide, anxiety, depression (Bekelman et al. 2007; Mofidi et al. 2007), and bipolar disorder (Koenig 2007; Moritz et al. 2006). However, some authors suggest the opposite: religiosity may be a risk factor associated with anxious and depressive symptoms (Simkin and Etchevers 2014) and negatively related to mental health and quality of life (Newberg 2010).
Although there is an agreement in the literature on the negative effects of COVID-19 on mental health, the basic processes of how people cope with stressful situations, considering psychological and religious mechanisms, are not entirely clear. One area of interest in looking at this dimension involves focusing on variables of experience processing or coping strategies, mainly cognitive, in which the person evaluates resources and threats to answer to or face stressful situations (Lazarus and Folkman 1984).
The following are some psychological and religious coping variables that have reduced the impact of COVID-19 on mental health. A relevant cognitive coping variable that is predictive of mental health is Experiential Avoidance (EA). EA refers to all deliberate efforts to avoid or escape from the content of a particular experience, such as thoughts, emotions, and physical sensations that are experienced as aversive, even when this leads to actions that are inconsistent with one’s values and goals (Hayes et al. 1996). EA is a basal clinical variable and a transdiagnostic category of different physical and mental illnesses (Gloster et al. 2017). In the early quarantine period, EA has been identified as one of the main predictors of mental health distress in adults, such as depression, anxiety, stress, loneliness, and negative emotions (Ferreira et al. 2021). In turn, AE is significantly related to fatigue severity, pain intensity, and pain disability during COVID-19 in Latino adults (Mayorga et al. 2022).
Another important variable in mental health as a psychological coping mechanism is self-compassion, understood as the internal attitude of being warm and kind to oneself instead of self-criticizing when failing or suffering (Neff 2016). This variable predicted healthcare professionals’ quality of life and well-being during the pandemic (Lluch-Zans et al. 2022). It has also been shown that compassion-based interventions can help alleviate stress in the pandemic context of adult women (Wu 2021) and contribute positively to the life satisfaction of men in COVID-19 quarantine (Li et al. 2021).
Finally, a coping strategy associated with how people use their sacred-related experiences to respond to crises or suffering events is religious coping (RC) (Pargament and Raiya 2007). During the pandemic, it was observed that people with more depressive and anxious symptoms were associated with higher use of negative RC (Zarrouq et al. 2021). Similarly, a negative relationship between positive religious coping and perceived stress was observed in a sample of quarantined adults (Mahamid and Bdier 2021). Furthermore, RC has also been shown to moderate the relationship between hope and well-being in the general population exposed to strict lockdowns during the COVID-19 pandemic (Counted et al. 2022). On the other hand, in a sample of older adult women, it was observed that high levels of religious coping were associated with lower anxiety about death in the pandemic context (Rababa et al. 2021).
RC, EA, and self-compassion appear to be potentially related variables. First, all three variables behave as a process of managing and coping with stressful events. Second, all three variables are closely related to well-being and mental health (Akbari et al. 2022; Ferrari et al. 2019; Fabricatore et al. 2004). Third, these three variables tend to be specifically associated with stress (Bardeen et al. 2013; Hormansyah and Hidayah 2019; Ano and Vasconcelles 2005). Fourth, while both EA and self-compassion are also related to post-traumatic stress, only EA has been identified as a predictor of risk for post-traumatic stress symptoms following a traumatic event (Kumpula et al. 2011; Henschel et al. 2021). Therefore, we hypothesize that it would have a more preponderant role as a mediating variable than self-compassion since a recent systematic review of self-compassion and post-traumatic stress disorders only shows evidence that supports the association between increased self-compassion and decreased post-traumatic stress symptoms (Winders et al. 2020), but not its predictability.
Similarly, positive RC appeared to be a protective variable that favors growth and positive changes after a traumatic event (Gómez and García 2021). Conversely, the negative RC subscale appeared to be the most consistently associated with indicators of poor psychological functioning, such as anxiety and depression (Pargament et al. 2011). Finally, all these variables appear to be associated with the pandemic’s effects on mental health during the COVID-19 pandemic. Nevertheless, despite this affinity between functioning and mental health effects, there is no work in the literature devoted to exploring the mechanisms underlying these variables.
The present study aims to explore the relationships between negative RC, EA, self-compassion, and post-traumatic stress caused by COVID-19 and to observe the multi-serial mediation of EA and self-compassion in the relationship between NRC and COVID post-traumatic stress in a sample of Chilean adults.
Method
Data Collection Process
The sample was a nonrepresentative and convenience sampling that included persons over 18 years old who lived in Chile. They were contacted and invited to participate using academic contact networks of the researchers and partner institutions (Centro UC Estudios de la Religión, Centro de Salud Universitario Universidad Austral de Chile, and Centro UC Estudios de la Vejez y Envejecimiento), and also through personal networks, mainly due to practical criteria of application and feasibility of the data collection process. In addition, given the novelty of the research field in the country, there was not enough prior knowledge that could theoretically or empirically support the criteria for the sample’s definition.
The invitation was distributed through social media (Facebook, Whatsapp, and Instagram) and email, which included a brief presentation of the study and the link to the informed consent (Google Form completed online), where the potential participants found all the information referred to the project and the link to participate in the survey, in case they freely decided to do so. By clicking on the link, all participants expressed their consent to participate. The procedure was evaluated and approved by the Ethical Committee for Social Sciences, Arts, and Humanities of the Pontificia Universidad Católica de Chile in August 2021. Because of the pandemic, the whole data collection process was done through a self-applied online survey (encuestaonline.com), which consisted of 21 questions and 140 items applied between September and December 2021.
Instruments
The Spanish version of the Coronavirus Psychological Impact Questionnaire (Cuestionario de Impacto Psicológico del Coronavirus) was used to assess COVID post-traumatic stress (Sandin et al. 2020) using the Post-Traumatic Stress Symptoms Scale. This instrument had eight items evaluated on a 5-point Likert scale ranging from 1 (never or almost never) to 5 (always or almost always), obtaining the total score with the sum of its items. In the present sample, the reliability of the total scores on this scale was satisfactory (Cronbach’s α = 0.92).
The Spanish validation of the Action and Acceptance Scale (AAQ-II, Bond 2011) was used to measure experiential avoidance (Ruiz 2013), which in its original version had a reliability coefficient of 0.82. The scale had seven items evaluated on a 7-point Likert scale ranging from 1 = “never” to 7 = “always,” and the total score was obtained by summing its items. In the present sample, the reliability of this scale achieved a very good level (Cronbach’s α = 0.95).
The Chilean version of the Self-Compassion Scale was used to evaluate Self-Compassion (Araya et al. 2017), which has 26 items evaluated on a 5-point Likert scale ranging from 1 = “almost never” to 5 = “almost always.” In this study, the positive items of self-pity were inverted: the higher the score, the lower the level of self-pity. The total score was obtained by summing all items. The scale’s reliability was satisfactory (Cronbach’s α = 0.82).
Finally, religious coping was measured using the brief-RCOPE scale, according to Pargament (1998). The validated version for the Chilean context (García et al. 2020) was used. The original version of the instrument has two subscales: positive religious coping (PRC) and NRC, each with seven items, using a Likert scale from 1 (not at all) to 5 (very much) and obtaining the total score with the sum of its items. Only the negative religious coping subscale was used in the sample analyzed here, which also obtained a satisfactory reliability level (α = 0.87).
Statistical Analysis
First, descriptive statistics and correlations were conducted. Second, a serial mediation model was used to examine the direct, indirect, and total effects among the following variables: NRC (independent), COVID post-traumatic stress (dependent), EA (mediator 1), and self-compassion (mediator 2). All estimates were performed using the statistical software STATA (version 17).
Sample
The final sample consisted of 300 participants who met the selection criteria (adults who lived in Chile), excluding those who participated in the survey but did not answer it completely (N = 70). The respondents were aged between 18 and 87 years old with a mean value of 49.65 years (SD = 16.50) and were distributed into four age groups as follows: 16% were youngsters (18–29 years old), 25% were young adults (30–45 years), 26.3% of them were adults (46–59 years), and 32.7% older adults/seniors (more than 60 years old). Females represented 65.7% of the sample. Those who lived in Santiago represented 60% of the sample, while those in Valdivia were 16.7% and 23.3% in other cities. Slightly over 10% had only some level of school education (most of them complete, 7.3%), while 7.3% completed technical superior education. In total, 15.3% had incomplete superior education, and almost 30% completed it. Over a third of the sample (35.7%) had some postgraduate studies. Furthermore, 60.3% of participants identified as Catholics, 14% as Evangelicals, and 21% had no religious affiliation. Other religions represented 4.6% of the sample. For further details about these sociodemographic background variables, see Table 1.Table 1 Sample characterization (N = 300)
Variable N % Valid %
Age Youngsters (18–29 years) 48 16.0 16.0
Young adults (30–45 years) 75 25.0 25.0
Adults (46–59 years) 79 26.3 26.3
Older adults—seniors (more than 60) 98 32.7 32.7
Sex Female 197 65.7 65.7
Male 103 34.3 34.3
City Santiago 180 60.0 60.0
Valdivia 50 16.7 16.7
Other 70 23.3 23.3
Educational level Secondary incomplete 9 3.0 3.0
Secondary complete 22 7.3 7.3
Higher-technical incomplete 5 1.7 1.7
Higher-technical complete 22 7.3 7.3
Higher-university incomplete 46 15.3 15.3
Higher-university complete 89 29.7 29.7
Postgraduate 107 35.7 35.7
Religious affiliation Catholic 181 60.3 60.3
Evangelical 42 14.0 14.0
Protestant 10 3.3 3.3
Jewish 1 0.3 0.3
Without religion 63 21.0 21.0
Other 3 1.0 1.0
Age: M = 49.65; SD = 16.50
Results
Table 2 shows the descriptive statistics of all variables and their correlations. As the table displays, high scores on negative religious coping were positively correlated with higher levels of EA (r = 0.632, p < 0.001) and COVID post-traumatic stress (r = 0.415, p < 0.001) and with lower self-compassion (r = 0.545, p < 0.001). At the same time, EA was correlated with COVID post-traumatic stress (r = 0.439, p < 0.001) and self-compassion (r = 0.722, p < 0.001). Finally, self-compassion was correlated with COVID post-traumatic stress (r = 0.454, p < 0.001).Table 2 Descriptive statistics and Pearson correlation coefficient values regarding study variables
Variables Mean SD 1 2 3 4
1. Negative religious coping 11.345 5.623 1.000
2. Experiential avoidance 19.14 11.598 0.632*** 1.000
3. Self-compassion (more to less) 72.128 20.078 0.544*** 0.722*** 1.000
4. COVID post-traumatic stress 17.916 8.679 0.415*** 0.439*** 0.454*** 1.000
*p < 0.05, **p < 0.01, ***p < 0.001
Figure 1 shows the results of the estimated serial mediation model. The total effect of NRC on COVID post-traumatic stress was statistically significant (c = 0.646, SE = 0.101, t = 6.37, and p < 0.001). Also, the direct effects of NRC on EA and self-compassion were significant (B = 1.240, SE = 0.111, t = 11.1, p < 0.001, and B = 0.441; SE = 0.220, t = 2.01, and p < 0.05, respectively). The direct effect of EA as the first mediating variable on the second mediating variable of Self-compassion is also significant (B = 1.113, SE = 0.113, t = 9.97, and p < 0.001). A review of the direct effects of mediating variables on COVID post-traumatic stress showed that the effect of self-compassion was significant (B = 0.129, SE = 0.041, t = 3.12, and p < 0.01), while the mediating variable EA was not significant. When NRC and all other mediating variables were simultaneously entered into the model, the direct effect of this first variable on COVID post-traumatic stress also was statistically significant (c’ = 0.252, SE = 0.124, t = 2.03, and p < 0.05).Fig. 1 Serial-multiple mediation of experiential avoidance and self-compassion in the relationship between negative religious coping and COVID post-traumatic stress with B non-standardized. *p < 0.05, **p < 0.01, ***p < 0.001
Finally, the total indirect effect of NRC through EA and self-compassion on COVID post-traumatic stress was statistically significant (c − c′ = 0.394, SE = 0.088, t = 4.43, and p < 0.001).
Discussion
Currently, the COVID-19 health crisis and its effects on the population continue to be of interest and relevance in wellness and mental health issues. In this context, this study explored the relationships between negative RC, EA, self-compassion, and post-traumatic stress due to COVID-19. Our results also reveal the mechanisms underlying the relationship between negative RC and post-traumatic stress by COVID-19, highlighting the role of two mediating variables, EA and self-compassion.
First, the results indicate that all variables are correlated. According to the literature, it can be stated that both EA and self-compassion play a role in the management of post-traumatic stress (Henschel et al. 2021; Winders et al. 2020). For this reason, we explored the serial mediating role of EA and self-compassion in the relationship between negative RC and post-traumatic stress by COVID-19. Second, the results show that EA and self-compassion have a sequential mediating role in the relationship between negative RC and post-traumatic stress by COVID-19. In turn, self-compassion is also shown to be a mediating variable between RC and stress, suggesting that in people with negative coping styles, more self-compassionate coping may mediate the impact of RC on post-traumatic stress by COVID-19. However, EA does not appear to mediate the relationship between RC and COVID-19 post-traumatic stress. These findings are consistent with works that consider experiential avoidance and self-compassion as variables associated with different mental health indicators, such as anxiety and depression (Yela et al. 2020), and physical health, such as chronic pain (Costa and Pinto-Gouveia 2013).
Regarding negative religious coping, we found that it was positively correlated with post-traumatic stress caused by COVID-19. In the present study, people who used negative religious coping and were less compassionate toward themselves had greater symptomatology of post-traumatic stress caused by COVID-19. Therefore, these results support previous work indicating that self-compassion plays an important role in coping with adverse situations (Allen and Leary 2010), in contexts of stress (Stutts et al. 2018), and in post-traumatic stress (Winders et al. 2020).
As for EA, we found significant and positive relationships with post-traumatic stress caused by COVID-19, consistent with the theoretical approach that the strategy of controlling or avoiding unwanted psychological experiences is an unhelpful psychological response to improve emotional functioning (Tull and Gratz 2008). However, EA alone did not mediate the relationship between negative religious coping and post-traumatic stress caused by COVID-19, but together with serial self-compassion for three main reasons. First, although research supports a strong negative relationship between self-compassion and avoidance, the self-compassion scale has been related to two of the three dimensions of psychological coping, denial, and emotional disconnection, which are oriented to avoidance (Neff et al. 2005). However, negative religious coping assumes that its items assess not only a psychological coping style but also incorporate maladaptive religious strategies that reflect tension, conflict, or struggle with the sacred, which do not always correspond to an avoidance style. Second, negative religious coping or spiritual struggle, although in the immediate, may correlate with psychological distress; in the long term, it could also result in a possibility for learning, growth, and transformation (Pargament et al. 2011).
On the contrary, experiential avoidance functions in both the short and long term as a maladaptive psychological strategy (Kelly et al. 2019; Moroz and Dunkley 2019). Thus, we believe this difference in functioning over time could affect the mediating role of experiential avoidance. Third, we consider that serial mediation is possible since self-compassion brings an affective quality to the relationship with the self (critical judgments and kindness toward the self). This complements harmoniously with EA, which points to how we relate to the contents of consciousness (thoughts, emotions, or sensations). Thus, the relationship between the self and the contents of the consciousness could be analogous to the relationship we can establish with the sacred. In other words, serial mediation could account for how we relate to our contents of consciousness and how we relate to ourselves and mediates how we cope with adverse situations such as COVID-19 when we have a maladaptive relationship of conflict or struggle with the sacred.
Thus, it is suggested to promote interventions that work on decreasing experiential avoidance and strengthening self-compassion, such as mindfulness-based interventions (Golden et al. 2021) or acceptance and commitment therapy (Hayes 2019), which may collaborate in symptomatic mental health relief by COVID-19 in people who use maladaptive religious coping strategies.
It is important to point out that our study has some important limitations to consider. First, we cannot establish causal relationships between variables due to our cross-sectional design. Second, the self-report modality of the questionnaires could result in some of the responses being biased. For future studies, it would be important to explore the relationship between religiosity and mental health by considering an instrument of emotional awareness or internal affective states, such as the Five Facets of Mindfulness Questionnaire (FFMQ) (Baer et al. 2006), to clarify whether the results are biased by a report of one’s own emotional experience that is difficult to access in a context of religious tension or conflict. Third, the convenience sampling used to select cases can be considered another limitation that prevents us from generalizing the results to the Chilean population.
In conclusion, the serial mediation model shows that greater use of experiential avoidance in sequence with lower levels of self-compassion predicts the likelihood that people with negative religious coping will suffer from COVID-19 post-traumatic stress. The contribution of these results adds a psychological processing background that may shape how people organize their religious beliefs and experiences and how these contents are used to cope with post-traumatic stress.
Author Contribution
All authors contributed to the study’s conception and design. Material preparation, data collection, and analysis were performed by Catalina Cerda-Planas and Reiner Fuentes. The first draft of the manuscript was written by Reiner Fuentes, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Data Availability
The raw data supporting the conclusions of this article will be made available by the authors without undue reservation (available on request).
Declarations
Ethics Approval
All procedures carried out in studies with human participants were in accordance with the ethical standards of the Ethics Committee of the Universidad Católica Chile and with the Declaration of Helsinki of 1964 and its subsequent amendments or comparable ethical standards.
Consent to Participate
Informed consent was obtained from all individual adult participants included in the study.
Consent for Publication
The authors declare that they consent to the publication of this manuscript.
Competing Interests
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Akbari M Seydavi M Hosseini ZS Krafft J Levin ME Experiential avoidance in depression, anxiety, obsessive-compulsive related, and post-traumatic stress disorders: a comprehensive systematic review and meta-analysis J Contextual Behav Sci 2022 24 1 53 10.1016/j.jcbs.2022.03.007
Allen AB Leary MR Self-compassion, stress, and coping Soc Pers Psychol Compass 2010 4 2 107 118 10.1111/j.1751-9004.2009.00246.x
Ano GG Vasconcelles EB Religious coping and psychological adjustment to stress: a meta-analysis J Clin Psychol 2005 61 4 461 480 10.1002/jclp.20049 15503316
Araya C Moncada L Fauré J Mera L Musa G Cerda J Brito G Adaptation and preliminary validation of the self-compassion scale in a Chilean context Latinoamerican J Posit Psychol 2017 3 47 58
Baer R, Smith G, Hopkins J, Krietemeyer J, Toney L (2006) Using self-report assessment methods to explore facets of mindfulness. Assessment 13:27–45
Bardeen JR Fergus TA Orcutt HK Experiential avoidance as a moderator of the relationship between anxiety sensitivity and perceived stress Behav Ther 2013 44 3 459 469 10.1016/j.beth.2013.04.001 23768672
Baqui P Bica I Marra V Ercole A van Der Schaar M Ethnic and regional variations in hospital mortality from COVID-19 in Brazil: a cross-sectional observational study Lancet Glob Health 2020 8 8 e1018 e1026 10.1016/S2214-109X(20)30285-0 32622400
Bekelman DB Dy SM Becker DM Wittstein IS Hendricks DE Yamashita TE Gottlieb SH Spiritual well-being and depression in patients with heart failure J Gen Intern Med 2007 22 470 477 10.1007/s11606-006-0044-9 17372795
Bond FW Hayes SC Baer RA Carpenter KM Guenole N Orcutt HK Zettle RD Preliminary psychometric properties of the acceptance and action questionnaire–II: a revised measure of psychological inflexibility and experiential avoidance Behav Ther 2011 42 4 676 688 10.1016/j.beth.2011.03.007 22035996
Costa J Pinto-Gouveia J Experiential avoidance and self-compassion in chronic pain J Appl Soc Psychol 2013 43 8 1578 1591 10.1111/jasp.12107
Counted V Pargament KI Bechara AO Joynt S Cowden RG Hope and well-being in vulnerable contexts during the COVID-19 pandemic: does religious coping matter? J Posit Psychol 2022 17 1 70 81 10.1080/17439760.2020.1832247
Fabricatore AN Handal PJ Rubio DM Gilner FH Stress, religion, and mental health: Religious coping in mediating and moderating roles Int J Psychol Relig 2004 14 2 91 108 10.1207/s15327582ijpr1402_2
Ferrari M Hunt C Harrysunker A Abbott MJ Beath AP Einstein DA Self-compassion interventions and psychosocial outcomes: a meta-analysis of RCTs Mindfulness 2019 10 8 1455 1473 10.1007/s12671-019-01134-6
Ferreira MJ Sofia R Carreno DF Eisenbeck N Jongenelen I Cruz JFA Dealing with the pandemic of COVID-19 in Portugal: on the important role of positivity, experiential avoidance, and coping strategies Front Psychol 2021 12 647984 10.3389/fpsyg.2021.647984 34248745
García FE Oyanedel JC Páez D Arias PR Psychometric properties of the brief religious coping scale (Brief-RCOPE) in Chilean adults exposed to stressful events J Relig Health 2020 60 475 487 10.1007/s10943-019-00976-7
Gloster AT, Meyer AH, Lieb R (2017) Psychological flexibility as a malleable public health target: Evidence from a representative sample. J Contextual Behav Sci 6(2):166–171
Golden HL Vosper J Kingston J Ellett L The impact of mindfulness-based programmes on self-compassion in nonclinical populations: a systematic review and meta-analysis Mindfulness 2021 12 1 29 52 10.1007/s12671-020-01501-8
Gómez M, García F (2021) Posttraumatic growth and psychosocial variables among victims of political violence. LIBERABIT. Revista Peruana De Psicología 27(1):e447. 10.24265/liberabit.2021.v27n1.05
Hayes SC Acceptance and commitment therapy: towards a unified model of behavior change World Psychiatry 2019 18 2 226 10.1002/wps.20626 31059616
Hayes SC, Wilson KG, Gifford EV, Follette VM, Strosahl K (1996) Experiential avoidance and behavioral disorders: A functional dimensional approach to diagnosis and treatment. J Consult Clin Psychol 64(6):1152
Henschel AV Williams JL Hardt MM The role of experiential avoidance and emotion regulation in DSM-5 post-traumatic stress symptomatology J Loss Trauma 2021 26 6 527 539 10.1080/15325024.2020.1841506
Hit H, Struggling F (2020) Worry, gratitude & boredom: as COVID-19 affects mental, financial health, who fares better; who is worse? In: Data Digest. Angus Reid Institute. https://angusreid.org/covid19-mental-health/. Accessed 1 Aug 2022
Hormansyah RD, Hidayah N (2019) Does self-compassion affect the stress on married women workers? In 4th ASEAN Conference on Psychology, Counselling, and Humanities (ACPCH 2018). Atlantis Press 106–108. 10.2991/acpch-18.2019.27
Kelly MM DeBeer BB Meyer EC Kimbrel NA Gulliver SB Morissette SB Experiential avoidance as a mediator of the association between post-traumatic stress disorder symptoms and social support: a longitudinal analysis Psychol Trauma 2019 11 3 353 10.1037/tra0000375 30525774
Koenig HG Spirituality in patient care 2007 Philadelphia Templeton Foundation Press
Kumpula MJ Orcutt HK Bardeen JR Varkovitzky RL Peritraumatic dissociation and experiential avoidance as prospective predictors of post-traumatic stress symptoms J Abnorm Psychol 2011 120 3 617 627 10.1037/a0023927 21604826
Lakhan R Agrawal A Sharma M Prevalence of depression, anxiety, and stress during COVID-19 pandemic Journal of Neurosciences in Rural Practice 2020 11 4 519 525 10.1055/s-0040-1716442 33144785
Laurencin CT McClinton A The COVID-19 pandemic: a call to action to identify and address racial and ethnic disparities J Racial Ethn Health Disparities 2020 7 3 398 402 10.1007/s40615-020-00756-0 32306369
Lazarus RS, Folkman S (1984) Stress, appraisal, and coping. Springer, New York
Li A, Wang S, Cai M, Sun R, Liu X (2021) Self-compassion and life-satisfaction among Chinese self-quarantined residents during COVID-19 pandemic: a moderated mediation model of positive coping and gender. Personality and individual differences 170:110457. https://doi.org/10.1016%2Fj.paid.2020.110457
Lluch-Sanz C, Galiana L, Vidal-Blanco G, Sansó N (2022) Psychometric properties of the self-compassion scale-short form: study of its role as a protector of Spanish nurses professional quality of life and well-being during the COVID-19 pandemic. Nurs Rep 12(1):65–76. 10.3390/nursrep12010008
Mahamid FA Bdier D The association between positive religious coping, perceived stress, and depressive symptoms during the spread of coronavirus (COVID-19) among a sample of adults in Palestine: across sectional study J Relig Health 2021 60 1 34 49 10.1007/s10943-020-01121-5 33389439
Mayorga NA Manning KF Garey L Viana AG Ditre JW Zvolensky MJ The role of experiential avoidance in terms of fatigue and pain during COVID-19 among Latinx adults Cogn Ther Res 2022 46 470 479 10.1007/s10608-022-10292-2
Mofidi M DeVellis RF DeVellis BM Blazer DG Panter AT Jordan JM The relationship between spirituality and depressive symptoms: testing psychosocial mechanisms J Nerv Ment Dis 2007 195 8 681 688 10.1097/NMD.0b013e31811f4038 17700301
Moritz S Quan H Rickhi B Liu M Angel M Vintila R Sawa R Sorianao J Toews J A home study-based spirituality education program decreases emotional distress and increases quality of life-a randomized, controlled trial Altern Ther Health Med 2006 12 6 26 35
Moroz M Dunkley DM Self-critical perfectionism, experiential avoidance, and depressive and anxious symptoms over two years: a three-wave longitudinal study Behav Res Ther 2019 112 18 27 10.1016/j.brat.2018.11.006 30466031
Neff KD Hsieh YP Dejitterat K Self-compassion, achievement goals, and coping with academic failure Self and Identity 2005 4 3 263 287 10.1080/13576500444000317
Neff KD The self-compassion scale is a valid and theoretically coherent measure of self-compassion Mindfulness 2016 7 264 274 10.1007/s12671-015-0479-3
Newberg A Goleman D La espiritualidad, el cerebro y la salud La espiritualidad a debate 2010 Barcelona Kairos 275 297
Pargament K Feuille M Burdzy D The brief RCOPE: current psychometric status of a short measure of religious coping Religions 2011 2 1 51 76 10.3390/rel2010051
Pargament KI Raiya HA A decade of research on the psychology of religion and coping: things we assumed and lessons we learned Psyche and Logos 2007 28 2 742 766
Pargament KI Smith BW Koenig HG Perez L Patterns of positive and negative religious coping with major life stressors J Sci Study Relig 1998 37 4 710 724 10.2307/1388152
Rababa M Hayajneh AA Bani-Iss W Association of death anxiety with spiritual well-being and religious coping in older adults during the COVID-19 pandemic J Relig Health 2021 60 1 50 63 10.1007/s10943-020-01129-x 33284402
Ruiz FJ, Luciano C, Cangas AJ, Beltrán I (2013) Measuring experiential avoidance and psychological inflexibility: the Spanish version of the acceptance and action questionnaire-II. Psicothema 25(1):123–129. 10.7334/psicothema2011.239
Sandín B Valiente RM García-Escalera J Chorot P Impacto psicológico de la pandemia de COVID-19: Efectos negativos y positivos en población española asociados al periodo de confinamiento nacional Rev Psicopatol Psicol Clin 2020 25 1 1 22 10.5944/rppc.27569
Simkin H Etchevers M Religiosidad, espiritualidad y salud mental en el marco del Modelo de los Cinco Factores de la Personalidad Acta Psiquiatr Psicol Am Lat 2014 60 4 265 275
Stutts LA Leary MR Zeveney AS Hufnagle AS A longitudinal analysis of the relationship between self-compassion and the psychological effects of perceived stress Self and Identity 2018 17 6 609 626 10.1080/15298868.2017.1422537
Tull M Gratz K Further examination of the relationship between anxiety sensitivity and depression: the mediating role of experiential avoidance and difficulties engaging in goal-directed behaviour when distressed Anxiety Disorders 2008 22 199 210 10.1016/j.janxdis.2007.03.005
Winders SJ Murphy O Looney K O’Reilly G Self-compassion, trauma, and post-traumatic stress disorder: a systematic review Clin Psychol Psychother 2020 27 3 300 329 10.1002/cpp.2429 31986553
Wu X The effectiveness of compassion-focused group therapy on perceived stress in the COVID-19 epidemic among the academic women Int J Early Child Spec Educ 2021 30 2 633 639
Yela JR Crego A Gómez-Martínez MÁ Jiménez L Self-compassion, meaning in life, and experiential avoidance explain the relationship between meditation and positive mental health outcomes J Clin Psychol 2020 76 9 1631 1652 10.1002/jclp.22932 31981226
Zarrouq B Abbas N Hilaly JE An investigation of the association between religious coping, fatigue, anxiety and depressive symptoms during the COVID-19 pandemic in Morocco: a web-based cross-sectional survey BMC Psychiatry 2021 21 264 10.1186/s12888-021-03271-6 34022849
Zhang CH, Schwartz GG (2020) Spatial disparities in coronavirus incidence and mortality in the United States: an ecological analysis as of May 2020. J Rural Health 36(3):433–445
| 0 | PMC9707074 | NO-CC CODE | 2022-12-01 23:20:20 | no | Int J Lat Am Relig. 2022 Nov 28;:1-12 | utf-8 | null | null | null | oa_other |
==== Front
JAYS
Journal of Applied Youth Studies
2204-9193
2204-9207
Springer Nature Singapore Singapore
87
10.1007/s43151-022-00087-3
Editorial
Fresh Understandings and Challenges for Youth Studies Research
Cuervo Hernan [email protected]
1
Wood Bronwyn E. 2
Black Rosalyn 3
1 grid.1008.9 0000 0001 2179 088X University of Melbourne, Melbourne, Australia
2 grid.267827.e 0000 0001 2292 3111 Victoria University of Wellington, Wellington, New Zealand
3 grid.1021.2 0000 0001 0526 7079 Deakin University, Burwood, Australia
28 11 2022
15
16 11 2022
© The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
==== Body
pmcAfter 3 years, this is our last issue as Editors-in-Chief of the Journal of Applied Youth Studies (JAYS). We want to take this opportunity to make some reflections of the journey that JAYS has undergone in this time.
Since 2020, we have published 14 issues, 74 research articles, and 14 book review articles. We have received submissions from every corner of the world, and we have published the work of household names in youth studies scholarship, such as Carles Feixa, Ken Roberts, Judith Bessant, Johanna Wyn, Sharlene Swartz, Lucas Walsh, Ana Miranda, and Airi-Alina Allaste, to name a few. But we are as proud to have been a publishing outlet for several early career and postgraduate researchers and scholars from the Global South and non-English speaking countries. Such scholars sometimes can face additional challenges to place their fine research work in journals that are dominated by giant publishing companies in the Global North and by the English language as the medium and gatekeeper of academic communication. In that vein, 43% of our published research articles were authored by early career and postgraduate researchers. And 29% of research articles came from researchers working in higher education institutions located in the Global South (e.g. Nepal, Mexico, South Africa, India, Argentina, Zimbabwe, Peru, China, and Nigeria). This broad-reaching production of high standard scholarship has provided us with the unique possibility to learn more about the state of youth studies research and the lives of young people around the world, rather than a youth studies derived almost entirely from the Global North.
Contributors to JAYS have enabled us to gain a vantage point from which to view global youth research. In the journal’s pages, authors reaffirmed the relevance of some of the major theoretical and empirical themes in the field. For example, traditional debates around transitions (Andres & Wyn 2010; Furlong and Cartmel 2007; Roberts 2003), including the problematic of NEET and the relationship between youth and the state (Furlong 2006; Mizen 2004; Roberts 2011); ideas and practices of belonging (Habib and Ward 2019; Harris et al. 2021; Wood and Black 2018); youth citizenship and participation (Black et al. 2011; Chevalier 2016; Wood 2014); issues of wellbeing (Hanckel 2016; Hendry 2020; Wright and McLeod 2014); young people’s representations of the future (Cook 2016; Honwana 2012; Leccardi 2005); and the relevance of decentring the metrocentric perspectives (Cuervo 2014; Farrugia 2016); and Eurocentric approaches (Cuervo and Miranda 2019; Swartz 2022) entrenched in youth sociology were discussed in the journal.
Many of these debates have been enriched and expanded in JAYS through, for example, the treatment of the challenges faced by youth in situation of NEET, including their invisibility to public policies (Assmann and Broschinski 2021; Vieira et al. 2021). In this case, researchers pointed out at the homogenisation that the category of NEET produces on youth in this context, where heterogenous transitions and situations are represented as a common experience. Further, colleagues have extended discussions around youth citizenship and participation by placing the emphasis on the role of new media on youth engagement (Chen and Stilinovic 2020), by looking at everyday practices that shape youth wellbeing (Fu and Li 2022), and by generating new debates around which should be the age limit for young people to have a say in electoral matters that affect their present and future (Bessant 2020). Scholars also took to the task of opening dialogues around youth work and youth workers practices, as a way of engaging with those at the forefront of supporting those most vulnerable in society. We were fortunate to receive articles conceptualising the intersection of Derrida’s theorisation of the gift as a deconstruction of youth work practices (Gee 2020), as well as an analysis the experiences of those undertaking a degree on youth work at higher education institution and how these experiences might impact on those intended to receive their expert support (Tallon et al. 2022).
Scholars contributing to JAYS also reaffirm the unstoppable trend to render visible the lives of others. In this case, youth lives constructed in rural spaces and in the Global South received significant attention in the journal. For example, place was intersected with gender to highlight the challenges and influences imposed by parents on young women in rural Nepal in their transitions to metropolitan areas in the search of constructing futures (Basnet 2022). Similarly, other colleagues concentrated on the intersection of regional places and gender to discuss how vulnerability can be a precondition for resistance and agency in young women’s lives (te Riele and Shelley 2021). Two important contributions sought to provide theoretical and policy guidelines to map the construction of research and debate agendas from the Global South that could help to redress the Eurocentric approach to youth sociology (see Miranda and Alfredo 2022; Swartz 2022).
While it is impossible to encapsulate the variety of debates and agendas that circulate in the 74 research articles published under our tenure in JAYS, it is obviously important to bring to the fore how COVID-19 shape the scholarly work produced in the field and represented in JAYS. Our work in the journal started just a few months before the pandemic came to alter global lives. There has been a variety of analyses on the impact of COVID-19 on the lives of young people—some of these have been played out in JAYS. For example, using intersectionality as a method and an analytical tool, Couch et al. (2021) explored how young people of refugee background in Australia support their families and communities through public health promotion, volunteering, and innovation. Woodrow and Moore (2021) showed how the close of the “night-time economies” during the pandemic in Great Britain, impacted on the leisure of intersectionally disadvantaged young people who were subject to criminalisation, exclusion, and stigmatisation. Finally, in a powerful analysis, O’Keeffe (2022) revealed how young people in Melbourne, Australia, who experienced 262 days in lockdown, challenged adult and official (e.g. council) interpretations of the public space by building DIY dirt jumps across the city. This unstructured construction of public space brought to sharp relief youth agency and participation on issues that matter to their lives.
And there is more to come. The aspects of youth requiring further research are only growing as the world emerges from the pandemic. From health to education, social media to the economy, the world young people are growing up in has accelerated in social change in many different directions. A concerted body of research is needed to explore, analyse, and explain what it means to be young in this world of change.
Some new areas requiring extensive research include:The acceleration of time—Harmut Rosa (2014) suggests that changes to the temporal structure of society profoundly influencing modern life. The acceleration of technology, social change, and the pace of life have significant impacts on relationships and how we experience the world—how is this shaping young people’s lives and futures?
Growing levels of material poverty and struggle for basic goods (fuel, food, and power) are a new phenomenon if the Global North—what can be learnt from youth in the Global South (Swartz 2022) and how will this shape the long-term futures of all youth?
The pandemic has accelerated educational change and greatly exacerbated inequalities between children of low- and high-income families. Some reports estimate that on average, students globally are 8 months behind where they would have been absent the pandemic, but the impact varies widely, with countries some countries as little as 4 months behind, while others, are 15 months behind (McKinsey Education Practice 2022). Inequalities have also grown in all school systems—for examples Black children in the USA have fallen much further behind in reading and writing than White children. Previously about 50% of the world’s children “learning poor” were unlikely to meet basic literacy and numeracy levels—now a conservative estimate is that 10% more will join that group.
Young people’s own feelings about the future in times when uncertainty has taken on global proportions. Some researchers note that young people tend to be more optimistic about the future than older people, even in the face of crises such as a global pandemic, but they also highlight the need for a deeper analysis of the role of hope in young people’s lives (see Keating and Melis 2022).
These aspects of social change have very significant implications for young people and many new areas of research and innovation will emerge in response which we hope to see in future issues of JAYS.
To conclude we want to thank the Springer team that has been supporting us in this new endeavour of providing a new outlet to the burgeoning youth research field. In particular we would like to thank Carola Vermeeren, who with dexterity steered the journal to a brilliant launch in the first year, and to Katerina Kalamari, who currently efficiently help us in the production and assemblage of articles and issues. A big thank you to Earl Ritfeld and Catherine Murphy who have so brilliantly supported us and authors in the whole submission and publication process in the past 2 years. And a special thank you to Lawrence Liu who has all along encouraged and supported our ideas in this journey. Thank you to our editorial board and a special thank you to the many reviewers that gifted their time, creativity, and labour to this journal and the youth studies field. We are sure the new editors will continue, and exceed, our project of carving a new space of research, intellectual exchange, and fruitful debate in this rich space that is youth studies research.
==== Refs
References
Andres L Wyn J The making of a generation: the children of the 1970s in adulthood 2010 University of Toronto Press
Assmann ML Broschinski S Mapping young NEETs across Europe: exploring the institutional configurations promoting youth disengagement from education and employment JAYS 2021 4 95 117 10.1007/s43151-021-00040-w
Basnet N Parental decisions and influence on young women’s education to work transitions and possible selves futures in Nepal JAYS 2022 5 135 150 10.1007/s43151-022-00074-8
Bessant J From denizen to citizen: contesting representations of young people and the voting age JAYS 2020 3 223 240 10.1007/s43151-020-00014-4
Black R Walsh L Taylor F Young people on the margins: what works in youth participation Youth Stud Aust 2011 30 1 42 48
Chen PJ Stilinovic M New media and youth political engagement JAYS 2020 3 241 254 10.1007/s43151-020-00003-7
Chevalier T Varieties of youth welfare citizenship: towards a two-dimension typology J Eur Soc Policy 2016 26 1 3 19 10.1177/0958928715621710
Cook J Young people’s strategies for coping with parallel imaginings of the future Time Soc 2016 25 3 700 717 10.1177/0961463X15609829
Couch J Liddy N McDougall J ‘Our voices aren’t in lockdown’—refugee young people, challenges, and innovation during COVID-19 JAYS 2021 4 239 259 10.1007/s43151-021-00043-7
Cuervo H Critical reflections on youth and equality in the rural context J Youth Stud 2014 17 4 544 557 10.1080/13676261.2013.844781
Cuervo H, Miranda A (eds) (2019) Youth, inequality and social change in the Global South. Springer, Singapore
Farrugia D The mobility imperative for rural youth: the structural, symbolic and non-representational dimensions rural youth mobilities J Youth Stud 2016 19 6 836 851 10.1080/13676261.2015.1112886
Fu J Li M Citizenship practices of chinese students on an Australian campus: a case study JAYS 2022 5 197 215 10.1007/s43151-022-00078-4
Furlong A Not a very NEET solution: representing problematic labour market transitions among early school-leavers Work Employ Soc 2006 20 3 553 569 10.1177/0950017006067001
Furlong A, Cartmel F (2007) Young people and social change. McGraw-Hill Education (UK), London
Gee R Informal education as a Derridean gift: a deconstructive reading of the principles guiding youth work practice within neoliberal policy regimes JAYS 2020 3 103 113 10.1007/s43151-020-00021-5
Habib S Ward M Youth, place and theories of belonging 2019 Routledge
Hanckel B Mitigating risk and facilitating access to capabilities: the role of affect in the design of an ICT-tool for queer youth in Asia Emot Space Soc 2016 18 35 43 10.1016/j.emospa.2016.01.007
Harris A, Cuervo H, Wyn J (2021) Thinking about belonging in youth studies. Palgrave, London
Hendry NA Young women’s mental illness and (in-) visible social media practices of control and emotional recognition Soc Media Soc 2020 6 4 1 10 10.1177/2056305120963832
Honwana AM (2012) The time of youth: work, social change, and politics in Africa. Kumarian Press, Sterling
Keating A, Melis G (2022) Youth attitudes towards their future: the role of resources, agency and individualism in the UK. Journal of Applied Youth Studies 5(1):1–18
Leccardi C Facing uncertainty: temporality and biographies in the new century Young 2005 13 2 123 146 10.1177/1103308805051317
McKinsey Education Practice (2022). How COVID-19 caused a global learning crisis. Retrieved from https://www.mckinsey.com/industries/education/our-insights/how-covid-19-caused-a-global-learning-crisis. Accessed 10-11-2022
Miranda A Alfredo MA Transitions in the post-pandemic COVID-19 context: building youth policies in the Global South JAYS 2022 5 233 247 10.1007/s43151-022-00082-8
Mizen P The changing state of youth 2004 Palgrave Macmillan
O’Keeffe P Young peoples’ construction of DIY dirt jumps in Melbourne, Australia, throughout the COVID-19 lockdowns JAYS 2022 5 117 134 10.1007/s43151-022-00075-7
Roberts K Change and continuity in youth transitions in Eastern Europe: lessons for Western sociology Sociol Rev 2003 51 4 484 505 10.1111/j.1467-954X.2003.00432.x
Roberts S Beyond ‘NEET’ and ‘tidy’ pathways: considering the ‘missing middle’ of youth transition studies J Youth Stud 2011 14 1 21 39 10.1080/13676261.2010.489604
Rosa H (2014) From work-life to work-age balance? Acceleration, alienation, and appropriation at the workplace. In: Korunka C, Hoonakker P (eds) The impact of ICT on quality of working life. Springer, Dordrecht, pp 43–61
Swartz S (2022) A charter for Global South youth studies scholars. JAYS 1-8.10.1007/s43151-022-00084-6
Tallon R Hay A Smith L Evaluating a degree qualification in youth work: a qualitative Aotearoa New Zealand study JAYS 2022 5 151 165 10.1007/s43151-022-00077-5
te Riele K Shelley B Vulnerability as a precondition for resistance and agency: young women in a regional community JAYS 2021 4 381 399 10.1007/s43151-021-00052-6
Vieira MM Pappámikail L Ferreira T NEETs in Europe: from plural (in)visibilities to public policies JAYS 2021 4 89 94 10.1007/s43151-021-00050-8
Wood BE Researching the everyday: young people’s experiences and expressions of citizenship Int J Qual Stud Educ 2014 27 2 214 232 10.1080/09518398.2012.737047
Wood BE Black R Halse C Spatial, relational and affective understandings of citizenship and belonging for young people today: towards a new conceptual framework Interrogating belonging for young people in schools 2018 Palgrave 165 185
Woodrow N Moore K The liminal leisure of disadvantaged young people in the UK before and during the COVID-19 pandemic JAYS 2021 4 475 491 10.1007/s43151-021-00064-2
Wright K, McLeod J (eds) (2014) Rethinking youth wellbeing: critical perspectives. Springer, Singapore
| 0 | PMC9707075 | NO-CC CODE | 2022-12-01 23:20:20 | no | JAYS. 2022 Nov 28;:1-5 | utf-8 | null | null | null | oa_other |
==== Front
Sci Educ (Dordr)
Sci Educ (Dordr)
Science & Education
0926-7220
1573-1901
Springer Netherlands Dordrecht
36466315
407
10.1007/s11191-022-00407-2
Editorial
Trust in Science and Science Education—Part 2
Erduran Sibel [email protected]
grid.4991.5 0000 0004 1936 8948 Department of Education, University of Oxford, 15 Norham Gardens, Oxford, OX2 6PY UK
29 11 2022
2022
31 6 14251427
7 11 2022
© The Author(s), under exclusive licence to Springer Nature B.V. 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
issue-copyright-statement© Springer Nature B.V. 2022
==== Body
pmcThis issue of Science & Education is the second part of a thematic set of papers on “Trust in Science and Science Education.” October 2022 issue included 13 papers and an editorial, covering some epistemic and societal perspectives about science, and what they imply for trust in science. In this issue, 12 papers question a range of topics including the actors, voices, and contexts of science, drawing attention to issues such as social justice, ethnicity, and gender. Educational examples are discussed including in the context of informal learning environments, and trust in science is related to critical rationalism and realism. A theme across some papers is the importance of scientific knowledge in ethical and political education. In the following paragraphs, a brief review of the content of the papers is presented with implications for future research.
In considering the role of the social context of science, Fernandez, Benitez, and Romero-Maltrana argue that even though the social character of science is evident, it does not entail epistemic relativism with respect to scientific knowledge. Rather, the social context of science itself increases the reliability of scientific outputs. The authors believe that science education needs to pay closer attention to the scientific community as a social group. Smith draws on the media accounts of the COVID-19 vaccination debate in Australia, arguing that science education should make explicit the ways in which the community of science interacts to produce and verify knowledge, and equip students to recognize uncertainty and dissent as being central to science. Allchin on the other hand posits that the core question for the citizen-consumer is not philosophically “why trust science?” but sociologically “who speaks for science?” The next two papers focus more closely on issues related to social justice. Waight and colleagues contrast narratives of successes of science and technology with narratives of vaccine hesitancy and mistrust in science among particularly minoritized groups. They argue for reframing the scientific, technological, and science education communities to address racism and injustice. Focusing on learning in museum contexts, Piequeras and colleagues present a study based on an out-of-school educational activity, where teenagers explore and discuss hominid reconstructions. Their results show that the exhibition displays human evolution in the form of a linear sequence from a primitive African prehistory to a more advanced European present. Behind this depiction of human evolution lies stereotypic notions of ethnicity and gender.
Educational interventions that integrated trust-related themes in science teaching and learning are discussed by the following three papers. Bateman presents a study where students explored a complex, real-world geological problem about which expert scientists had previously made competing claims through geologic maps. Provided with data, expert uncertainty ratings, and the previous claims, students made new geologic maps of their own and presented arguments about their claims in written form. She argues for explicit instruction around uncertainty can be introduced during undergraduate courses. Nation and Feldman examine the complex nature between science teacher beliefs and the impact on their instructional practices of climate change–centered curriculum. The teachers involved in their study did not espouse these beliefs within their instruction of the curriculum and instead remained “neutral” when teaching about climate change. Estigarribia and colleagues discuss an educational innovation through design-based research, which was oriented to the treatment of information disseminated in the media and social networks related to COVID-19. They identified criteria that students applied to trust or not to trust the circulating information and a series of strategies to corroborate the information.
The final set of papers raise questions about rationality and critical realism in relation to trust in science. Duarte and colleagues note that despite advances in discussions about the nature of science, there is still a paucity of discussion on the ontological dimension of science in science education research. The authors argue that the reestablishment of trust in science and its teaching requires an effort to understand the contemporary social contradictions, problems, and challenges. Ferguson highlights the dichotomy between trust in scientific evidence and scientists as experts, versus critiques of the limitations of scientific knowledge and a mistrust in scientists as biased professionals and political agents. The author presents the argument that society can both trust in scientific evidence and question scientific bias in the same space, holding these two seemingly opposite positions in productive tension. The author further advocates that we should teach students to do the same. Critical realism is presented as an ontology and epistemology to frame science education with a focus on the development of critical scientific literacy by teaching students what is real and what is arbitrary about science. Chmielewski too considers whether the critical rationalist philosophy of science may provide a rationale for trusting scientific knowledge. The author considers insights of Karl Popper’s social and political philosophy in order to see whether they may be of help in offsetting the distrust of science and visits how the theoretical principles of the critical rationalist philosophy of science may afford a foundation for building trust in science. Toscano and Millar investigate rationality and its relationship to trust in science in the context of three proposed spaces of science education: the formal, informal, and casual. They discuss how each space privileges particular forms of rationality as a means for understanding trust in science.
Overall, the second and final set of papers provides a range of perspectives for the history, philosophy, and sociology of science as well as science education communities to draw on about trust in science and science education. In our age of misinformation and fake news, the ability to engage in scientific reasoning and particularly evidence-based reasoning has become ever more important. Yet, as several articles in the two thematic issues illustrate, fostering scientific habits of mind may be necessary but not sufficient in navigating the complex set of science-related problems that the planet faces including the climate emergency. The continuing legacy of racism, sexism, and colonialism around the world marginalize some individuals and communities from active and equitable engagement with and in science, disadvantaging not only them but also the very institution of science itself given the set of voices contributing to scientific debate becomes restricted as a consequence. A constructive way forward will be through nuanced considerations about the opportunities as well as the tensions surrounding the epistemic, ontological, and social dimensions of science in the context of science education where polarized dichotomies and simplifications are avoided.
Data Availability
There is no data to refer to in relation to this editorial.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
| 36466315 | PMC9707077 | NO-CC CODE | 2022-12-16 23:17:46 | no | Sci Educ (Dordr). 2022 Nov 29; 31(6):1425-1427 | utf-8 | Sci Educ (Dordr) | 2,022 | 10.1007/s11191-022-00407-2 | oa_other |
==== Front
Sleep Vigil
Sleep Vigil
Sleep and Vigilance
2510-2265
Springer Nature Singapore Singapore
36466045
220
10.1007/s41782-022-00220-5
Editorial
Reforming Sleep Health: A Need to Focus on Sleep Health Policy to Reduce Disease Burden and Promote Health Equity and Equality
https://orcid.org/0000-0003-2583-563X
Gupta Ravi [email protected]
1
https://orcid.org/0000-0002-8686-7259
Pandi-Perumal Seithikurippu R. [email protected]
2
Mallick Hruda Nanda [email protected]
3
1 grid.413618.9 0000 0004 1767 6103 Department of Psychiatry and Division of Sleep Medicine, All India Institute of Medical Sciences, Veerbhadra Marg, Rishikesh, 249203 India
2 grid.412431.1 0000 0004 0444 045X Saveetha Medical College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, 600077 India
3 Faculty of Medicine and Health Sciences, SGT University, Gurugram, India
29 11 2022
2022
6 2 255257
© The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
issue-copyright-statement© Springer Nature Singapore Pte Ltd. 2022
==== Body
pmc“Reform is not a onetime action; it is a permanent attitude” - António Manuel de Oliveira Guterres, UN Secretary-General, 2017
Across the human lifespan, sleep duration varies between individuals and populations. As we grow and mature, our sleep patterns, composition, and architecture drastically change. An average adult habitually spends around one-third of their life in sleep, while newborns spend ~ 16 h of sleep in 24 h. Time spent in sleep reduces as we age, but the need for sleep remains constant. The sleep–wake cycle is influenced by hosts of other factors viz., social, economic, cultural, ecological, psychological, and environmental, besides psychiatric and other medical disorders [1].
Sleep is an important physiological process in itself for the body and partial and total sleep deprivation is associated with morbidity as well as mortality [1]. Not only short sleep but also prolonged sleep has been found to have similar results emphasizing the fact that like other physiological processes e.g. blood pressure, heart rate, and blood sugar to name a few, sleep duration has to fall a narrow range to maintain good health [2].
Subjective and objective sleep studies in small-scale subsistence societies have expanded our knowledge of the variation among cross-cultural sleep patterns, revealing the adaptability and the natural sleep–wake pattern of human sleep. As sleep duration is traded in exchange for both leisure or success, shortened sleep duration and delayed sleep–wake phase have almost become a norm in the modern world, which are known to be associated with an increase in blood sugar, circulating insulin levels, and resulting in insulin resistance and weight gain [1, 3, 4]. Sleep deprivation also increases the risk for psychiatric disorders viz., depression, anxiety, substance use disorder, dementia, and other medical disorders e.g., hypertension, coronary artery disease, atherosclerosis, and stroke to name a few through various intertwining pathways [5]. As a result, various sleep disorders such as sleep apnea, insomnia, circadian rhythm sleep disorders, sleep-related movement disorders, parasomnias, and hypersomnia can occur. These sleep disorders increase the risk of the psychiatric and other medical disorders mentioned above, resulting in a vicious cycle that leads to poor health [6–9].
In modern 24-h society, when a sizeable portion of the population across all age groups is chronically sleep deprived, sleep disturbances may be one of the factors contributing to the growth in the prevalence of type 2 diabetes mellitus (T2DM), coronary artery disease (CAD), cancer, and depression [1, 10–12]. Considering the bidirectional relationship between medical disorders and sleep disorders, it is likely that the incidence of sleep disorders could also be showing an increasing trend. However, this has never been examined systematically as sleep disorders remain ignored category for the global as well as national regulatory agencies.
Data from cross-sectional studies suggested that nearly 425 million people across the globe are suffering from obstructive sleep apnea (OSA) with the greatest number in China followed by the United States of America, Brazil, and India [13]. Similarly, nearly half of the university students were found to suffer from insomnia in the South-East Asia (ASEAN) region [14]. World has experienced a major pandemic since 2020 which brought the world to a standstill for a reasonable period and resulted in a loss of productivity, and economy in addition to that adverse consequences on human health and untimely deaths. Even though the pandemic was started because of the viral agent, sleep disorders showed a close association with it. OSA was found to increase the risk for SARS-CoV-2 infection whereas insomnia appeared to result from it [15].
Undiagnosed, underdiagnosed, and untreated sleep disorders have personal, societal, and public-health consequences.Individual effects include poor performance (e.g. academic, sports), poor quality of life, increased utilization of healthcare resources for other medical disorders as discussed above, suicide, loss of wages, reduced productivity, work-absenteeism, and chances of injury, disability, or loss of life due to industrial as well as motor vehicle accidents [16, 17]. From the public health point of view, reduced individual productivity culminates in overall hampered economic growth, greater expenditure towards healthcare services, and loss of innocent lives because of vehicular accidents. The list of accidents that are attributed to poor sleep health is long starting from three-mile island, challenger explosion, Exxon oil spill, the American Airline Flight 1420 crash, Hoboken train crash, and Air India Flight 812 crash, to name a few [18, 19]. Reports suggest that nearly 40% of road traffic accidents occur due to poor sleep health and two third pilots in India report falling asleep in the cockpit [19, 20]. The estimated costs of untreated insomnia and sleep apnea fall in the range of billions of US dollars per annum [16, 17]. Fortunately, the picture is not all grim and a silver lining is apparent backed by the cost–benefit analysis which has shown that optimal treatment of insomnia and OSA reduce healthcare, economic, social, and global disease burden to a great extent [16, 17].
As already discussed, sleep duration and timing are influenced by social, economic, cultural, ecological, psychological, and environmental, besides psychiatric and other medical disorders [1]. Fortunately, both these factors are modifiable and thus can help in reducing economic and public-health burdens arising not only out of sleep disorders but also associated with other medical disorders. But to attain that, sleep disorders should be given priority in global as well as national healthcare programs and policies.
Considering the grave consequences of sleep disorders, sleep health must be addressed to bring about health equity and equality among the population divided by race, socioeconomic status, and other determinants [21]. Sleep health aims to improve regularity, timing, and duration of sleep ensures good efficiency and satisfaction from sleep, and ultimately aims to promote alertness during the waking period [1]. Concepts of primary and secondary preventive measures also apply to sleep disorders. This may be addressed through awareness campaigns, and environmental changes e.g., addressing noise, light, air, and light pollutions, improving living conditions; improving opportunities for optimal treatment of sleep disorders through human resource development in Sleep Medicine, infrastructure development in terms of sleep laboratories and accessibility to treatment for OSA; and through regulatory measures for promoting sleep viz., addressing to balance sleep-work time, incentivizing sleep-health promotion programs and avoiding commercial driving and pilot licenses to persons suffering from untreated sleep disorders [21].
South-East Asia (including India and China) is home to nearly half of the world's population and is still considered a developing region. Environmental conditions in these countries are concerning, and socioeconomic disparities are severe, paving the way for poor sleep health and its medical, societal, and economic consequences [21, 22]. If we want to improve our citizens' overall health, we must prioritize sleep health through a variety of measures [23]. Therefore, a multi-pronged approach is required, with coordination among various stakeholders (e.g. inclusion of governmental and non-governmental organizations (NGOs), citizenship charters, or community engagement groups, as well as patient groups). This is best accomplished by creating a framework for a sleep-health policy that addresses various aspects of sleep health across age and sex [1, 22]. This must include, but are not limited to school start time, working hours (duration, and timing), shiftwork (e.g. night shift—reverse rotating shifts, motor vehicle accidents (MVAs) and traffic safety.
What we need is solid sleep-health reform. This cannot be accomplished solely through the efforts of scientists who can generate data on the negative effects of poor sleep health or data showing how improving sleep health improves overall health or through the efforts of existing sleep physicians who provide care to individual patients. It takes the political will for policymakers (bureaucrats and politicians) to initiate and sustain dialogues with professional scientific bodies and sleep-health experts.
Data availability
Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Grandner MA Fernandez F-X The translational neuroscience of sleep: a contextual framework Science 2021 374 568 573 10.1126/science.abj8188 34709899
2. Chaput J-P Dutil C Featherstone R Ross R Giangregorio L Saunders TJ Janssen I Poitras VJ Kho ME Ross-White A Carrier J Sleep duration and health in adults: an overview of systematic reviews Appl Physiol Nutr Metab 2020 45 10 (Suppl. 2) S218 S231 10.1139/apnm-2020-0034 33054337
3. De La Iglesia HO Fernández-Duque E Golombek DA Lanza N Duffy JF Czeisler CA Valeggia CR Access to electric light is associated with shorter sleep duration in a traditionally hunter-gatherer community J Biol Rhythms 2015 30 4 342 350 10.1177/0748730415590702 26092820
4. Martins AJ Isherwood CM Vasconcelos SP Lowden A Skene DJ Moreno CR The effect of urbanization on sleep, sleep/wake routine, and metabolic health of residents in the Amazon region of Brazil Chronobiol Int 2020 37 9–10 1335 1343 10.1080/07420528.2020.1802287 32777972
5. Besedovsky L Lange T Haack M The sleep-immune crosstalk in health and disease Physiol Rev 2019 99 3 1325 1380 10.1152/physrev.00010.2018 30920354
6. Fang H Tu S Sheng J Shao A Depression in sleep disturbance: a review on a bidirectional relationship, mechanisms and treatment J Cell Mol Med 2019 23 4 2324 2332 10.1111/jcmm.14170 30734486
7. Tietjens JR Claman D Kezirian EJ De Marco T Mirzayan A Sadroonri B Goldberg AN Long C Gerstenfeld EP Yeghiazarians Y Obstructive sleep apnea in cardiovascular disease: a review of the literature and proposed multidisciplinary clinical management strategy J Am Heart Assoc 2019 8 1 e010440 10.1161/JAHA.118.010440 30590966
8. Uddin MS Tewari D Al Mamun A Kabir MT Niaz K Wahed MI Barreto GE Ashraf GM Circadian and sleep dysfunction in Alzheimer’s disease Ageing Res Rev 2020 1 60 101046 10.1016/j.arr.2020.101046
9. Simon EB Vallat R Barnes CM Walker MP Sleep loss and the socio-emotional brain Trends Cogn Sci 2020 24 6 435 450 10.1016/j.tics.2020.02.003 32299657
10. Liu J Ren ZH Qiang H Wu J Shen M Zhang L Lyu J Trends in the incidence of diabetes mellitus: results from the Global Burden of Disease Study 2017 and implications for diabetes mellitus prevention BMC Public Health 2020 20 1 1 2 10.1186/s12889-020-09502-x 31898494
11. Ralapanawa U Sivakanesan R Epidemiology and the magnitude of coronary artery disease and acute coronary syndrome: a narrative review J Epidemiol Global Health 2021 11 2 169 10.2991/jegh.k.201217.001
12. Liu Q He H Yang J Feng X Zhao F Lyu J Changes in the global burden of depression from 1990 to 2017: findings from the Global Burden of Disease study J Psychiatr Res 2020 1 126 134 140 10.1016/j.jpsychires.2019.08.002
13. Benjafield AV Ayas NT Eastwood PR Heinzer R Ip MS Morrell MJ Nunez CM Patel SR Penzel T Pépin JL Peppard PE Estimation of the global prevalence and burden of obstructive sleep apnoea: a literature-based analysis Lancet Respir Med 2019 7 8 687 698 10.1016/S2213-2600(19)30198-5 31300334
14. Chowdhury AI Ghosh S Hasan MF Khandakar KA Azad F Prevalence of insomnia among university students in South Asian Region: a systematic review of studies J Prev Med Hyg 2020 61 4 E525 33628956
15. Pataka A Kotoulas S Sakka E Katsaounou P Pappa S Sleep dysfunction in COVID-19 patients: prevalence, risk factors, mechanisms, and management J Personaliz Med 2021 11 11 1203 10.3390/jpm11111203
16. Taddei-Allen P Economic burden and managed care considerations for the treatment of insomnia Am J Manag Care 2020 26 4 Suppl S91 S96 32282179
17. Morsy NE Farrag NS Zaki NF Badawy AY Abdelhafez SA El-Gilany AH El Shafey MM Pandi-Perumal SR Spence DW BaHammam AS Obstructive sleep apnea: personal, societal, public health, and legal implications Rev Environ Health 2019 34 2 153 169 10.1515/reveh-2018-0068 31085749
18. Avaialble at https://www.somnologymd.com/2019/11/disasters-caused-by-lack-of-sleep/. Last accessed on 15 Nov 2022
19. Available at https://www.telegraph.co.uk/world-news/2022/09/23/two-thirds-indian-pilots-admit-falling-asleep-cockpit/#:~:text=In%202010%2C%20an%20Air%20India,of%20the%20two%20hour%20flight. Last accessed on 15 Nov 2022
20. Available at https://www.thehindu.com/news/national/kerala/sleep-deprived-drivers-responsible-for-40-of-road-accidents-say-transport-officials/article61629032.ece Last accessed 15 Nov 2022
21. Hale L Troxel W Buysse DJ Sleep health: an opportunity for public health to address health equity Annu Rev Public Health 2020 2 41 81 10.1146/annurev-publhealth-040119-094412
22. Accessed at https://frontline.thehindu.com/dispatches/india-ranks-at-the-bottom-in-a-list-180-countries-in-the-2022-environmental-performance-index/article65497256.ece on 13 Nov 2022
23. Akhtar N Mallick H Recommendations for a National Sleep Policy in India Natl Med J India 2019 32 59 60 10.4103/0970-258X.272131 31823948
| 36466045 | PMC9707081 | NO-CC CODE | 2022-12-06 23:15:21 | no | Sleep Vigil. 2022 Nov 29; 6(2):255-257 | utf-8 | Sleep Vigil | 2,022 | 10.1007/s41782-022-00220-5 | oa_other |
==== Front
J Gen Intern Med
J Gen Intern Med
Journal of General Internal Medicine
0884-8734
1525-1497
Springer International Publishing Cham
36443625
7973
10.1007/s11606-022-07973-9
Original Research
Evaluating Diversity and Inclusion Content on Graduate Medical Education Websites
Wei Chapman MD 1
http://orcid.org/0000-0002-6928-4182
Bernstein Simone A. MD [email protected]
2
Gu Alex MD 3
Mehta Aadit BS 4
Sharma Dhruv BA 4
Mortman Ryan BS 5
Verduzco-Gutierrez Monica MD 6
Chretien Katherine C. MD 7
1 grid.412833.f 0000 0004 0467 6462 Department of Medicine, Staten Island University Hospital and Northwell Health, 475 Seaview Ave, Staten Island, NY 10305 USA
2 grid.4367.6 0000 0001 2355 7002 Department of Psychiatry, Washington University School of Medicine and Barnes-Jewish Hospital, 660 S. Euclid Ave, Campus Box 8509, St. Louis, MO 63110 USA
3 grid.253615.6 0000 0004 1936 9510 Department of Orthopaedic Surgery, George Washington University School of Medicine and Health Sciences, 2300 M St, Washington, DC 20037 USA
4 grid.253615.6 0000 0004 1936 9510 George Washington University School of Medicine and Health Sciences, 2300 M St, Washington, DC 20037 USA
5 grid.253615.6 0000 0004 1936 9510 George Washington University School of Medicine and Health Sciences, 2300 M St, Washington, DC 20037 USA
6 grid.267309.9 0000 0001 0629 5880 Department of Rehabilitation Medicine, Long School of Medicine, University of Texas Health Science Center San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229 USA
7 grid.21107.35 0000 0001 2171 9311 Department of Medicine, Johns Hopkins University School of Medicine, 1600 McElderry Street, Suite 230, Baltimore, MD 21205 USA
28 11 2022
14
14 11 2021
17 11 2022
© The Author(s), under exclusive licence to Society of General Internal Medicine 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Abstract
Websites are important tools for programs to provide future residency applicants with freely accessible information regarding their program, including diversity, equity, and inclusion (DEI) initiatives.
Objective
To describe the variability of DEI content in residency programs and compare DEI website content by specialty.
Methods
Using the 2021 Accreditation Council for Graduate Medical Education (ACGME) list of residency programs, residency training websites were identified and evaluated. Information was extracted from program websites as indicators of DEI content. Chi-square analysis and one-way ANOVA were chosen to assess for statistical differences.
Results
In total, 4644 program websites representing 26 specialties were assessed. Among all the programs, the average DEI completeness of a program website was 6.1±14.6% (range 0–100%). While 6.2% of all programs had a diversity webpage, only 13.3% included a commitment to DEI, and few programs (2.7%) provided information about underrepresented in medicine (URiM) faculty.
Conclusions
Graduate medical education programs can enhance information for current and prospective applicants about DEI initiatives on their websites. Including DEI initiatives on residency websites may improve diversity recruitment efforts.
KEY WORDS
Diversity
Inclusion
Residency
Medical education
Websites
==== Body
pmcINTRODUCTION
Diversity, equity, and inclusion (DEI) in the physician workforce, including in graduate medical education, remains a critical issue, with widespread efforts by multiple medical associations to increase the representation of those underrepresented in medicine (URiM) at all levels.1–4 Improving DEI in this sector is vital in promoting higher quality and culturally responsive care.1 The Accreditation Council for Graduate Medical Education (ACGME) has proposed new standards to systematically and intentionally increase a diverse workforce.1 Previous survey studies show that URiM applicants rank a program more favorably if the residency program has URiM initiatives, faculty, leadership, and residents.5,6 This information was most frequently obtained from in-person experiences, such as during interviews.7
Because of the COVID-19 pandemic, many traditional recruitment methods changed, leaving applicants to seek residency program information through the internet as their primary resource.8 While many programs have implemented social media recruitment strategies, including virtual fairs and social media accounts, program websites have continued to be essential for applicant recruitment and program information.8,9 Gaeta et al. found that 78% of applicants reported that program websites influenced their decision to apply to a particular program during the application cycle.10 Additionally, Steele et al. conducted a survey that showed residency applicants still predominantly use program websites to access information relative to other digital platforms.9 Thus, websites can be essential for programs to provide future residency applicants with freely accessible information regarding their program, including DEI initiatives. DEI information within residency program websites may reflect a program’s overall commitment to DEI, as well as a commitment to share information transparently.
Our study aimed to determine and assess the variability of DEI content among residency program websites and compare them by specialty.
METHODS
A complete list of residency programs in all specialties was gathered from the ACGME Directory in March 2021.11 Residency program demographics included location. Three authors (AM, DS, RM) independently collected variables pairwise from program websites and discussed any discrepancies. Discrepancies were resolved when a consensus was reached between the three authors collecting the data.
Based on previous research and conference consensus statements, the following seven variables were extracted from program websites as indicators of DEI content: webpage(s) dedicated to diversity, a DEI commitment statement, diversity initiatives (e.g., summer research for medical students, lectures, mentorship), a link to the institutional DEI page, information about bias training, URiM faculty, and fourth-year rotations for URiM applicants.1,2,5,6,12–15 While there is no gold standard for evaluating DEI content on websites, two variables were chosen (webpage dedicated to diversity and DEI commitment statement) because they were analyzed in previous studies.12–14 Based on the framework provided by Gonzaga et al., DEI initiative information at a program, department, and institution level was important in demonstrating sincere DEI commitment. Thus, we included five departmental-level variables (diversity initiatives, link to institutional DEI page, bias training, URiM faculty, and fourth-year rotations). Biographical information, such as photos, was assessed in previous literature but was not included in our current study since only some websites provide explicit information that residents or faculty members are URiM.12,14 Completeness, which combined all variables to determine the percentage present for each program website, was analyzed to determine the differences in DEI content among specialties and reported as an average percentage with a standard deviation. Completeness was calculated as follows: (Sum of the presence of variables / 7) × 100%.
To compare diversity completeness by specialty, we used chi-square analyses, a non-parametric analysis tool, to compare the presence of each website variable among all specialties. One-way ANOVA tests were used to assess completeness (Statistical Package for the Social Sciences 25; Armonk, NY). A p < 0.05 was considered statistically significant. A one-way ANOVA test was chosen to compare completeness between specialties because this data had a normal distribution (skewness statistic: 2.8 ± 0.04; kurtosis statistic: 8.4 ± 0.08).
The study did not meet criteria for human subject research as defined by the Department of Health and Human Services and thus did not require Institutional Review Board review.
RESULTS
In total, 4644 programs were assessed representing 26 specialties. 6.2% of all programs had a dedicated diversity webpage, with Child Neurology websites having the most DEI webpages (18.7%). Although 13.3% of programs included a commitment to DEI, Thoracic Surgery program websites contained the most explicit commitments to DEI (41.4%).
Few programs (2.7%) provided information about URiM faculty. A notable exception was Neurosurgery, where 58.1% of websites contained this information. Few program websites (0.8%) provided information regarding URiM-tailored visiting student rotations.
Regarding links to the program institution’s office of diversity, 14.1% were found on program webpages. Thoracic Surgery program websites most frequently included a link to the institutional office of diversity. Few programs (1.2%) discussed bias training. Among all programs, the average DEI completeness of a program website was 6.1±14.6% (range 0–100%).
Table 1 outlines website completeness by specialty. For the two specialties with the largest number of programs, family medicine (679 programs) and internal medicine (575 programs), there was lack of completeness in the diversity information on the websites (2.3 ± 10.2% and 2.8 ± 10.9%, respectively). Of the 4644 programs, only seven were assessed to have 100% completeness: two programs in dermatology, one program in neurosurgery, two programs in otolaryngology, and two programs in urology. Overall, 80.2% of total programs had a completeness of 0%, with the lowest percentages in Neurosurgery (33.3), Thoracic Surgery (58.6), and Child Neurology (58.7). Table 1 Completeness of Diversity, Equity, and Inclusion Website Variables by Specialty
Specialty Total (N) Most complete program website, % (N) Least complete program website, % (N) Percent of programs with completeness of 0 (%) Average completeness of websites (mean % ± SD)
Overall 4644 100 (7) 0 (3726) 80.2 6.1±14.6
Anesthesiology 150 71.4 (1) 0 (109) 72.7 8.2±16.0
Child neurology 75 57.1 (1) 0 (44) 58.7 11.6±16.7
Dermatology 140 100.0 (2) 0 (102) 72.9 9.3±19.5
Diagnostic radiology 183 57.1 (3) 0 (142) 77.6 5.3±12.0
Emergency medicine 263 71.4 (1) 0 (224) 85.2 4.8±12.9
Family medicine 679 85.7 (2) 0 (633) 93.2 2.3±10.2
General surgery 330 71.4 (5) 0 (280) 84.8 4.6±13.2
Internal medicine 565 85.7 (1) 0 (519) 91.9 2.8±10.9
Neurology 157 57.1 (2) 0 (114) 72.6 8.9±15.3
Neurosurgery 117 100.0 (1) 0 (39) 33.3 16.7±19.0
Nuclear medicine 30 57.1 (2) 0 (18) 60.0 13.8±18.6
Obstetrics and gynecology 279 71.4 (1) 0 (225) 80.6 5.8±13.5
Ophthalmology 123 85.7 (1) 0 (88) 71.5 8.9±17.4
Orthopaedics 192 71.4 (1) 0 (162) 84.4 3.9±10.9
Otolaryngology 118 100 (2) 0 (81) 68.6 11.3±22.1
Pathology 138 57.1 (2) 0 (112) 81.2 4.6±11.2
Pediatrics 201 71.4 (2) 0 (158) 78.6 7.5±16.1
Physical medicine and rehabilitation 97 71.4 (2) 0 (69) 71.1 9.3±18.6
Plastic surgery (integrated) 82 71.4 (2) 0 (55) 67.1 11.3±19.7
Preventative medicine 61 57.1 (2) 0 (41) 67.2 10.1±15.5
Psychiatry 257 71.4 (1) 0 (217) 84.4 4.6±12.0
Radiation oncology 89 71.4 (2) 0 (68) 76.4 7.5±16.3
Radiology (interventional) 89 71.4 (1) 0 (63) 70.8 8.8±17.2
Thoracic surgery (integrated) 29 42.9 (2) 0 (17) 58.6 12.8±15.9
Urology 137 100.0 (2) 0 (101) 73.7 8.9±19.1
Vascular surgery 63 57.1 (3) 0 (45) 71.4 8.4±15.7
DISCUSSION
This study assessed the DEI-related content of residency program websites and found many areas where DEI content could be enhanced.
It is well documented that many racial and ethnic disparities exist within the US healthcare system.14 One of the mechanisms to alleviate such disparities is having a more diverse workforce. Diversity in graduate medical education is important as physicians from URiM groups are more likely to care for patients of the same background.16 Furthermore, diversity in training helps provide different perspectives and increased awareness of social determinants of populations, thus lending itself to optimal outcomes17. Outlining diversity initiatives and describing a commitment to DEI help showcase a welcoming residency space.12
When deciding where to apply, the 2017 National Resident Matching Program Applicant Survey revealed that approximately 37% of applicants viewed an institution’s cultural, racial-ethnic, and gender diversity as a factor when selecting programs.18 Therefore, a website dedicated to DEI, where programs can describe their initiatives in one central location, can be an asset for recruitment of a diverse residency class. Another way to enhance program websites is to include information about URiM faculty, something known to be appreciated by applicants but uncommonly found.19 Similarly, having support from an established office of DEI is important for recruiting URiM applicants since these offices can help coordinate recruitment efforts and events for the residency program.4 A framework to address the ACGME standard on creating a diverse and inclusive workforce outlines five key steps: setting diversity as a priority; seeking out candidates; implementing inclusive recruitment practices; investing in trainee success; and building the pipeline. Having diversity listed as a priority on a program website is a simple way to fulfill the first step of this framework and communicate this priority.20 Other institutions have enacted similar strategies for improving diversity and inclusion.21,22 During the COVID-19 pandemic, Hoff et al. implemented virtual recruitment initiatives, including having a dedicated DEI website outlining their goals and achievements, and increased URiM resident recruitment by 21%.21
Two critical factors in retaining and recruiting URiM physicians in the workforce are information related to URiM faculty and DEI-related opportunities.5,7 Our findings also showcase multiple ways for residency programs to enhance diversity initiatives based on information located on websites. These include research grants for URiM medical students, URiM mentorship, and scholarly lectures. Scholarships that support visiting student rotations or research can attract candidates from all socioeconomic backgrounds, as the cost can be the main deterrent for students in performing an away rotation.23Despite the importance of website information and the relatively low barriers to adding content to program websites, our study found that most program websites did not contain any of the assessed variables.
This study has limitations. Program websites may not reflect actual DEI initiatives. This paper reviews the outward and public display of DEI but recognizes there may be programming and priorities that are not mentioned on program websites. Also, DEI website content may have been missed and not located or assessed, although there were multiple independent website reviewers. It is not known whether there was an increase in the number of URiMs being interviewed or matching in specialties at programs with a greater number of DEI initiatives shared on program websites. This study is also cross-sectional, and findings may change over time. Furthermore, while a public commitment to diversity is important, measuring the impact of initiatives will be essential as next research steps.
CONCLUSION
Graduate medical education programs can enhance the DEI content provided to current and prospective applicants on their websites to reflect their commitment to DEI. Internal Medicine programs’ websites, in particular, scored lower in DEI content than other specialties, suggesting opportunities for improvement. Ensuring DEI website content reflects actual initiatives and commitment may help programs diversify their workforce.
Declarations
Conflict of interest
The authors declare that they do not have a conflict of interest.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Gonzaga AMR, Appiah-Pippim J, Onumah CM, et al. A Framework for Inclusive Graduate Medical Education Recruitment Strategies: Meeting the ACGME Standard for a Diverse and Inclusive Workforce. Acad Med. 2020;95(5): 10.1097/ACM.0000000000003073.
2. Clayborne EP Martin DR Goett RR Diversity pipelines: The rationale to recruit and support minority physicians J Am Coll Emerg Physicians Open 2021 2 1 e12343 33532751
3. Wesson DE King TE Jr Todd RF Achieving diversity in academic internal medicine: recommendations for leaders Am J Med. 2006 119 1 76 81 10.1016/j.amjmed.2005.10.032 16431199
4. West MA Hwang S Maier RV Ensuring Equity, Diversity, and Inclusion in Academic Surgery: An American Surgical Association White Paper Ann Surg. 2018 268 3 403 407 10.1097/SLA.0000000000002937 30004923
5. Peek ME Kim KE Johnson JK “URM candidates are encouraged to apply”: A national study to identify effective strategies to enhance racial and ethnic faculty diversity in academic departments of medicine Acad Med. 2013 88 3 405 412 10.1097/ACM.0b013e318280d9f9 23348090
6. Agawu A Fahl C Alexis D The Influence of Gender and Underrepresented Minority Status on Medical Student Ranking of Residency Programs J Natl Med Assoc. 2019 111 6 665 673 31668360
7. Wong D Kuprasertkul A Khouri RK Jr Assessing the Female and Underrepresented Minority Medical Student Experience in the Urology Match: Where Do We Fall Short? Urology. 2021 147 57 63 10.1016/j.urology.2020.08.076 33065172
8. Tang O Ruddell JH Hilliard RW Improving the online presence of residency programs to ameliorate COVID-19’s impact on residency applications Postgrad Med. 2021 133 4 404 408 10.1080/00325481.2021.1874195 33412975
9. Steele TN Galarza-Paez L Aguilo-Seara G Social media impact in the Match: A survey of current trends in the United States Arch Plast Surg. 2021 48 1 107 113 10.5999/aps.2020.00836 33503753
10. Gaeta TJ Birkhahn RH Lamont Aspects of residency programs’ web sites important to student applicants Acad Emerg Med. 2005 12 1 89 92 10.1197/j.aem.2004.08.047 15635145
11. Accreditation Data System (ADS). ACGME. https://apps.acgme.org/ads/Public/Programs/Search. Accessed March 8, 2021.
12. Sanchez AN, Martinez CI, Lara AM, et al. Evaluation of Diversity and Inclusion Presence Amongst United States Physical Medicine and Rehabilitation Residency Program Websites. Am J Phys Med Rehabil. 2021; 10.1097/PHM.0000000000001693.
13. Smith J, Chiu AG, Sykes KJ, et al. Diversity in Academic Otolaryngology: An Update and Recommendations for Moving From Words to Action. Ear Nose Throat J. 2020; 145561320922633. 10.1177/0145561320922633
14. Driesen AMDS Romero Arenas MA Arora TK Tang A Nfonsam VN O’Grady CL Riall TS Morris-Wiseman LF Do General Surgery Residency Program Websites Feature Diversity? J Surg Educ. 2020 77 6 e110 e115 10.1016/j.jsurg.2020.06.014 32600889
15. Institute of Medicine, Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. 2003.Washington, DC: National Academic Press.
16. Association of American Medical Colleges. Addressing racial disparities in health care: A targeted action plan for academic medical centers. https://members.aamc.org/eweb/upload/addressing%20racial%20disparaties.pdf. Published 2009.
17. Khan NR Taylor CM Rialon KL Resident Perspectives on the Current State of Diversity in Graduate Medical Education J Grad Med Educ. 2019 11 2 241 243 10.4300/JGME-D-19-00062.1 31024664
18. National Resident Matching Program. Results of the 2017 NRMP Applicant Survey by Preferred Specialty and Applicant Type. National Resident Matching Program, Washington, DC. September 2017.
19. Aagaard EM Julian K Dedier J Soloman I Tillisch J Pérez-Stable EJ Factors affecting medical students’ selection of an internal medicine residency program J Natl Med Assoc 2005 97 1264 1270 16296217
20. Gonzaga AM Appiah-Pippim J Onumah CM Yialamas MA A Framework for Inclusive. Graduate Medical Education Recruitment Strategies Academic Medicine. 2020 95 5 710 716 10.1097/acm.0000000000003073 31702694
21. Hoff ML Liao NN Mosquera CA An Initiative to Increase Residency Program Diversity Pediatrics. 2022 149 1 e2021050964 10.1542/peds.2021-050964 34972222
22. Nationwide Children’s Hospital. Diversity and Inclusion. https://www.nationwidechildrens.org/for-medical-professionals/education-and-training/residency-programs/general-pediatrics/resident-diversity-inclusion-coalition. Accessed August 11, 2022
23. Higgens E Newman L Halligan K Miller M Schwab S Kosowicz L Do audition electives impact Match success? Med Educ Online. 2016 13 21
| 36443625 | PMC9707083 | NO-CC CODE | 2022-12-01 23:20:20 | no | J Gen Intern Med. 2022 Nov 28;:1-4 | utf-8 | J Gen Intern Med | 2,022 | 10.1007/s11606-022-07973-9 | oa_other |
==== Front
Acta Neurol Belg
Acta Neurol Belg
Acta Neurologica Belgica
0300-9009
2240-2993
Springer International Publishing Cham
36443623
2147
10.1007/s13760-022-02147-0
Original Article
The effects of augmented and virtual reality gait training on balance and gait in patients with Parkinson’s disease
http://orcid.org/0000-0001-9626-4462
Gulcan Kubilay [email protected]
1
http://orcid.org/0000-0001-8464-1929
Guclu-Gunduz Arzu [email protected]
2
http://orcid.org/0000-0002-6134-4865
Yasar Evren [email protected]
1
http://orcid.org/0000-0001-6598-4184
Ar Ulas [email protected]
1
http://orcid.org/0000-0003-0188-5245
Sucullu Karadag Yesim [email protected]
3
http://orcid.org/0000-0001-6450-6666
Saygili Fettah [email protected]
4
1 Health Ministry of Türkiye Republic, Ankara Bilkent City Hospital, Physical Medicine and Rehabilitation Hospital, Ankara, Türkiye
2 grid.25769.3f 0000 0001 2169 7132 Faculty of Health Sciences, Department of Physiotherapy and Rehabilitation, Gazi University, Emek Mah, Biskek Cad. No: 2 Cankaya, 06490 Ankara, Türkiye
3 grid.488643.5 0000 0004 5894 3909 Ankara Bilkent City Hospital Neurology Clinic, University of Health Sciences, Ankara, Türkiye
4 grid.34517.34 0000 0004 0595 4313 Faculty of Health Sciences, Department of Physiotherapy and Rehabilitation, Aydin Adnan Menderes University, Aydın, Türkiye
28 11 2022
19
6 9 2022
14 11 2022
© The Author(s) under exclusive licence to Belgian Neurological Society 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Objective
Augmented reality (AR) and virtual reality (VR) facilitate motor learning by enabling the practice of task-specific activities in a rich environment. Therefore, AR and VR gait training may improve balance and gait in Parkinson’s Disease (PD).
Methods
Thirty patients with PD were randomly divided into study (n = 15) and control (n = 15) groups. The study group was given AR and VR gait training combined with conventional training. The control group was given conventional training only. The training was applied to both groups 3 days a week for 6 weeks. Motor symptoms with the Unified Parkinson Disease Rating Scale-Motor Examination (UPDRS-III), balance with posturography and Berg Balance Scale (BBS), perceived balance confidence with Activity-Specific Balance Confidence Scale (ABC), gait with spatio-temporal gait analysis, and functional mobility with Timed Up and Go Test (TUG) were assessed.
Results
At the end of the study; UPDRS-III, posturography measurements, BBS, ABC, spatio-temporal gait parameters, and TUG improved in the study group (p < 0.05), while BBS, ABC, and only spatial gait parameters (except for step width) improved in the control group (p < 0.05). There was no change in posturography measurement, temporal gait parameters, and TUG in control group (p > 0.05). When the developed parameters in both groups were compared, the amount of improvement in BBS and ABC was found similar (p > 0.05), while the improvement in the other parameters was found higher in the study group (p < 0.05).
Conclusion
It was concluded that AR and VR gait training provides the opportunity to practice walking with different tasks in increasingly difficult environments, thus improving balance and walking by facilitating motor learning.
Keywords
Augmented reality
Virtual reality
Parkinson’s disease
Balance
Gait
==== Body
pmcIntroductıon
Balance and gait disorders in PD show a disease-specific pattern. In patients with Parkinson’s Disease (PD), it is seen that the limits of stability decreased, automatic postural strategies are delayed, walking slows down, the step length decreases, the duration of the stance phase and the number of steps increases, and the step length decreases when approaching obstacles, and freezing episodes occurred [1, 2].
Previous studies indicate that physiotherapy approaches improve balance and gait in PD [3, 4]. In these systematic reviews, it is emphasized that approaches aiming to improve balance and walking in PD should be based on motor learning and neural plasticity. This is possible with training in which task-specific activities are studied intensively in a diversified environment, and feedback is given on task performance to achieve effective functional movement. On the other hand, it is not possible to create a diversified environment, increase the difficulty of the task, give sufficient feedback, and practice intensively in the clinical environment. Augmented reality (AR) and Virtual reality (VR) can provide the needs necessary to provide these motor learning principles, which can be difficult to meet with traditional therapy [5]. VR applications, which enable patients to interact with an artificial virtual environment, also allow the physiotherapist to evaluate patients with objective methods and observe developments. Developing technology aims to increase the reality and immersion by adding sensory stimuli to the virtual environment in VR applications, which brings out AR systems. In AR, virtual objects placed in the virtual environment or added tactile, visual, and auditory stimuli increase the interaction of individuals with the virtual environment.
When the literature is examined, it is seen that VR practices generally focus on standing balance training, while the use of VR during gait training is a newer technology, and studies are limited in PD [6–9]. In these studies, it was shown that the spatio-temporal features of walking [6, 8] and functional mobility [7, 9] improved with VR training. In only one of these studies, in which different VR walking platforms were used, it was seen that VR and AR were used together in gait training [9].
As a result, it is seen that there is a need for studies examining the effects of augmented and virtual reality training in PD with objective measurement methods. Therefore, our study aimed to examine the effects of AR and VR gait training on balance and gait in early to mid-stage PD.
Methods
Participants
Between July and December of 2021, 30 patients with PD who applied to the Department of Physical Medicine and Rehabilitation, Republic of Türkiye Ministry of Health, Ankara Bilkent City Hospital and who were diagnosed by a neurologist participated in the study. The inclusion criteria were as follows: (1) Hoehn and Yahr (H&Y) ≤ 3; (2) being 40 years or older; and (3) that the individuals agreed to be included in the study after adequate information was given about the study. The exclusion criteria were as respectively: (1) standardized Mini-Mental State Examination score < 24; (2) the presence of cardiovascular, vestibular, musculoskeletal, or additional neurological disease; (3) having a sensorial impairment (i.e., auditory and visual loss). In addition, the study was approved by the Republic of Türkiye Ministry of Health, Ankara Bilkent City Hospital Clinical Research Center Ethics Committee (E2-21–632). The participants were randomized to the study (n = 15) and control (n = 15) groups (Fig. 1). The control group received conventional physiotherapy, and the study group received AR and VR training. Conventional training (except standing balance and walking exercises) was also applied to the study group. Both groups received routine medication for PD.Fig. 1 Flow chart of the study
Interventions
Conventional training
Conventional training was applied to the patients in the control group. It was planned with the following objectives in mind: increasing mobility by improving the impaired kinesthetic sensation; improving flexibility of soft tissues, body alignment, mobility in and out of bed; and improving balance and gait [10]. The training was applied 3 days a week for 6 weeks, with each exercise session lasting approximately one hour, accompanied by a physiotherapist. Training started with exercises in the supine position and continued with exercises in sitting and standing. After the walking exercises, the training was terminated with stretching and relaxation exercises. Each of the exercises was done in 2 sets of 10 repetitions. In gait training, each exercise was performed for 4 min.
Augmented and virtual reality gait training
Gait and balance training was given with AR and VR gait training using the C-Mill VR + (Motek Medical, Amsterdam, The Netherlands) device [11]. The training was carried out 3 days a week for 6 weeks with each exercise session lasting approximately 1.5 h accompanied by a physiotherapist. Training started with conventional exercises (except for balance and walking exercises), continued with C-Mill VR + training, and ended with stretching and relaxation exercises. In all of the gait training, it was basically aimed to increase the step length and improve the swing phase and the walking adaptability to the changes in the environment (surface, speed, auditory stimulus). Training procedure was planned as follows: informing individuals about the device and exercises, using safety vests to minimize the risk of falling during training, providing visual and auditory feedback to individuals by the device and physiotherapist during training, increasing the individual specific difficulty level of each exercise in the training content through different variables, and adjusting the initial difficulty level of the training between 12 and 15 points which indicates medium difficulty according to the Borg scale [12]. The 10 exercises (Fig. 2.) in AR and VR gait training were as, respectively: 1. Stepping stones, 2. Random stepping stones, 3. Obstacle avoidance, 4. Speed adaptation, 5. Slalom, 6. Monster game, 7. Balls track, 8. Auditory cueing, 9. Nature island, 10. The Italian Alps. Exercises were performed for 4 min resting the individuals for about 1 min between exercises.Fig. 2 AR and VR gait training
Data acquisition and assessment
Patients were evaluated twice, before and after 6 weeks. The evaluations started 30–45 min after the patients took their dopaminergic drugs. All assessments were made by a physiotherapist blinded to the study groups. The demographics and disease characteristics were recorded.
The clinical motor symptoms of the patients were evaluated with the motor assessment subscale of the Unified Parkinson’s Disease Rating Scale (UPDRS-III). The severity of motor symptoms was evaluated with 13 items in UPDRS-III. Items are scored between 0 and 4; 0 = no impairment, 4 = severe impairment. Higher scores indicate more serious impairment [13].
Static and dynamic standing balance was evaluated with the Huber 360° Evolution System (LPG Systems, Valence, France), which objectively evaluates postural sway and limits of stability [14]. The stability test was performed by recording the amount of sway of the Center of Pressure (CoP) under standing on double-leg and single-leg conditions. The test was performed under eyes open and eyes-closed conditions. The lower values mean that the amount of sway is low and the postural stability is better. In the limits of stability test, while on standing and with fixed feet position, patients were asked to shift their weight in a total of eight directions according to the trigonometric coordinate system. The high values mean that the amount of CoP sway is high, and the stability limits of the patient are good.
Functional balance was evaluated with Berg Balance Scale (BBS) [15].
The level of confidence that an individual feels during the activities of daily living related to balance was evaluated with Activities-Specific Balance Confidence Scale (ABC) [16].
Spatio-temporal gait analysis was performed with the C-Mill VR + (Motek Medical, Amsterdam, The Netherlands) device. In the test, patients were asked to walk on the platform of the C-Mill VR + device for 3 min at the highest confident speed they felt safe [11]. Right-left step and stride length, step width, stance-swing phases, and total double support phase duration were recorded.
Functional mobility was assessed with the Timed Up and Go test (TUG) [17].
Statistical analysis
Statistical analyzes were performed using SPSS v.23.0 (SPSS Inc, Chicago, IL, USA). Non-parametric tests were used because the number of cases was not in accordance with the normal distribution. Descriptive statistics were made by giving the median and interquartile range (25–75 interquartile range: IQR). Descriptive statistics of categorical variables were expressed as frequency and percentage (%). A comparison of data within the group was made using the Wilcoxon test. The amount of improvement after the training in the research and control groups was compared with the Mann–Whitney U Test. The level of statistical significance was set at p < 0.05. The sample size calculation was based on the significant improvement of the BBS after similar intervention study in PD [9]. To achieve %80 power with a two-sided level of 5%, the sample size required per group was 14 using G*Power software (version 3.1.9.7; Heinrich Heine Universitat, Dusseldorf, Germany).
Results
Table 1 shows the demographic and disease characteristics of the groups. According to the results of the study, UPDRS-III total score decreased in both groups after the training (p < 0.05, Table 2). There was no difference between the amount of decrease (p > 0.05, Table 3).Table 1 Demographic and clinical characteristics of the disease (Could you write the median and iqr in all tables on a bottom row?)
Study Group n = 15 median (IQR) Control Group n = 15 median (IQR) p
Gender
Female 2 (13.3) 2 (13.3) 1.000
Male 13 (86.7) 13 (86.7)
Age, (year) 61 (57–66) 60 (51–67) 0.884
BMI, (kg/m2) 27.70 (26.80–29.10) 27.50 (25.70–30.40) 0.836
Disease duration, (year) 6 (1–10) 6 (1–9) 0.771
H&Y stage, (0–5) 3 (2–3) 3 (2–3) 0.870
H&Y n (%) n (%)
1 1 (6.7) 2 (13.3) 0.763
2 6 (40) 4 (26.7)
3 8 (53.3) 9 (60)
BMI Body Mass Index, H&Y Hoehn and Yahr, *p < 0.05
Table 2 A comparison of before and after training
Study Group Control Group
Before trainig median (IQR) After trainig median (IQR) p Before trainig median (IQR) After trainig median (IQR) p
UPDRS-III, (0–108) 15.0 (10.0–23.0) 11.0 (9.0–17.0) 0.002* 20.0 (15.0–33.0) 17.0 (8.0–24.0) 0.001*
Double-leg Stability Test
Eyes Open Area (mm2) 159.45 (93.80–186.65) 112.08 (76.96–145.91) 0.005* 139.86 (99.56–287.69) 143.52 (88.17–524.77) 0.268
Length (mm) 471.24 (365.74–550.06) 285.03 (199.00–344.23) 0.001* 354.69 (243.25–548.86) 372.98 (253.84–564.87) 0.650
Speed (mm/s) 9.42 (7.31–11.00) 6.02 (4.02–6.72) 0.001* 7.09 (4.87–10.98) 6.99 (5.03–11.30) 0.776
Eyes Closed Area (mm2) 316.21 (165.50–410.47) 204.14 (92.19–271.00) 0.036* 253.01 (119.15–2481.66) 265.11 (156.13–899.23) 0.140
Length (mm) 580.92 (431.21–777.12) 377.63 (226.43–606.42) 0.003* 552.01 (413.16–806.96) 431.21 (323.94–809.13) 0.281
Speed (mm/s) 11.62 (8.62–15.54) 7.69 (5.44–12.13) 0.004* 11.04 (8.26–16.14) 7.69 (6.70–10.67) 0.125
Single-Leg Stability Test
Area, (mm2) Right 889.95 (590.65–1964.63) 684.43 (277.00–882.59) 0.031* 1681.59 (668.11–2561.94) 1313.02 (601.77–1964.63) 0.233
Left 720.75 (563.23–1065.66) 566.12 (373.70–713.49) 0.004* 1030.59 (656.76–2159.53) 1233.24 (611.81–3245.69) 0.307
Length, (mm) Right 1072.69 (709.46–1321.95) 554.05 (376.79–862.39) 0.008* 868.86 (759.40–1231.15) 855.53 (622.04 –1227.41) 0.570
Left 807.27 (543.46–1341.02) 518.53 (389.17–982.92) 0.009* 892.43 (646.21–1129.77) 672.13 (543.46–904.77) 0.078
Limits of Stability Test, (mm2) 21,663.63 (14,418.97–40,447.57) 36,678.34 (22,455.59–50,963.31) 0.001* 18,351.90 (10,650.80–34,461.20) 25,455.13 (13,023.49–38,498.43) 0.233
Berg Balance Scale, (0–56) 53.0 (48.0–55.0) 54.0 (50.0–56.0) 0.009* 50.0 (39.0–54.0) 52.0 (44.0–55.0) 0.002*
ABC, (0–100) 85.30 (67.50–97.30) 94.37 (82.50–98.00) 0.003* 58.0 (44.60–94.37) 62.0 (52.0–96.25) 0.001*
Step Length, (m) Right 0.33 (0.28–0.44) 0.50 (0.41–0.69) 0.001* 0.27 (0.24–0.41) 0.36 (0.31–0.48) 0.001*
Left 0.33 (0.27–0.44) 0.53 (0.38–0.70) 0.001* 0.28 (0.14–0.42) 0.38 (0.23–0.49) 0.001*
Stride Length (m) 0.66 (0.61–0.87) 1.02 (0.80–1.37) 0.001* 0.56 (0.34–0.84) 0.79 (0.52–0.97) 0.001*
Step Width (m) 0.15 (0.14–0.19) 0.13 (0.11–0.15) 0.010* 0.15 (0.13–0.18) 0.15 (0.12–0.17) 0.155
Stance Duration, (s) Right 0.33 (0.28–0.44) 0.50 (0.41–0.69) 0.001* 0.78 (0.69–0.90) 0.80 (0.65–0.91) 0.460
Left 0.33 (0.27–0.44) 0.53 (0.38–0.70) 0.001* 0.83 (0.70–0.93) 0.79 (0.71–0.91) 0.281
Swing Duration, (s) Right 0.38 (0.32–0.42) 0.41 (0.36–0.47) 0.001* 0.37 (0.28–0.43) 0.35 (0.27–0.41) 0.069
Left 0.37 (0.35–0.43) 0.42 (0.37–0.48) 0.003* 0.33 (0.23–0.41) 0.37 (0.23–0.43) 0.078
Total Double Support Duration, (s) 0.46 (0.37–0.51) 0.33 (0.26–0.39) 0.011* 0.47 (0.38–0.59) 0.44 (0.34–0.52) 0.057
TUG (s) 10.0 (8.0–11.0) 8.0 (5.0–10.0) 0.002* 11.0 (10.0–14.0) 11.0 (9.0–14.0) 0.118
UPDRS-III Unified Parkinson’s Disease Rating Scale-Motor assessment subscale, ABC Activities-Specific Balance Confidence Scale, TUG Timed Up and Go Test, *p < 0.05
Table 3 Comparison of the improvement amounts of parameters which developed in both groups
Study Group Δ median (IQR) Control Group Δ median (IQR) p
UPDRS-III, (0–108) – 4.00 [( – 5.00) to ( – 1.00)] – 5.00 [( – 7.00) to ( – 3.00)] 0.489
Berg Balance Scale (0–56) 2.0 [(1.0) to (3.0)] 2.0 [(1.0) to (3.0)] 0.433
ABC (0–100) 4.0 [(0.70) to (11.30)] 4.0 [(3.12) to (6.87)] 0.967
Step Length (m)
Right 0.16 [(0.12) to (0.24)] 0.06 [(0.03) to (0.13)] 0.001*
Left 0.17 [(0.11) to (0.24)] 0.06 [(0.03) to (0.13)] 0.005*
Stride Length (m) 0.35 [(0.27) to (0.42)] 0.17 [(0.07) to (0.24)] 0.001*
UPDRS-III Unified Parkinson’s Disease Rating Scale-Motor assessment subscale, ABC Activities-Specific Balance Confidence Scale, *p < 0.05
In the study group, there was an improvement in all parameters of open-closed eyes double-leg stability tests, single-leg stability test, limits of stability test, BBS, and ABC with AR and VR training (p < 0.05, Table 2). In the control group, there was only an improvement in BBS and ABC (p < 0.05, Table 2). There was no difference between the groups in terms of the number of developments in BBS and ABC (p > 0.05, Table 3).
Our findings showed that in the study group, after the training right-left step length, stride length, and right-left swing phase duration increased, while step width, right-left stance phase duration, and total double support phase duration decreased (p < 0.05, Table 2). However, in the control group, it was observed that the right-left step length and stride length increased (p < 0.05, Table 2), and no difference was found in other gait parameters (p > 0.05, Table 2). The right-left step length and stride length which are the parameters that developed in both groups, improved more in the study group than in the control group (p < 0.05, Table 3).
Discussion
In this study, which we aimed to analyze the effects of AR and VR gait training combined with conventional training on balance and gait in PD, it was observed that the severity of motor symptoms decreased, and balance, balance confidence perception and gait improved after 6 weeks of training.
As stated in the results, a similar amount of decrease in the severity of motor symptoms was observed in both groups. In our study, conventional training consisting of exercises aimed at improving bradykinesia, hypokinesia, rigidity, posture, postural control, and walking was applied to both groups. Therefore, the improvement in these symptoms is considered as a natural result of training. Considering that AR and VR are not accessible practices in every clinic, it is a positive result to achieve improvement with conventional training. Previous studies have also shown that both VR gait and conventional gait training are effective in reducing the severity of motor symptoms in patients with early to mid-stage PD [8, 18].
Although balance disorder becomes evident with disease progression, posturographic evaluations in individuals with early-stage PD show abnormalities in postural sway [19]. In our study, postural sway decreased, and the stability limits increased under standing on both single-double leg and eyes-closed conditions in the study group after the AR and VR gait training. There has also been improvement in clinical testing functional balance and the perception of confidence in balance-related activities in the study group. This similar improvement, which was obtained in different dimensions of balance, was only achieved in BBS and ABC in the control group.
Although the AR and VR gait training we have implemented is specific training for walking, the training has created a richer and increasingly difficult environment for patients to transfer weight in different directions with a dual-task and practice staying on a single leg than conventional training, so that static and functional balance has also improved. On the other hand, the improvement in functional tests only in the control group is an important indicator of how the content of the training affects the outcome measures.—Another reason for the improvement in all parameters of balance in the study group is that AR and VR training was given on a moving surface, with intense visual and auditory stimuli that both corrected the movement and motivated the patient.
When the literature is reviewed, it is seen that the number of studies evaluating the effects of AR and/or VR gait training on balance in PD using posturography is insufficient [9]. For this reason, our study stands out as a study that evaluates the effect of AR and VR gait training on postural stability in Parkinson's patients using posturography. Wang et al. used the C-Mill VR + system that we used in our study, and they only examined the developments in the postural stability tests but did not evaluate the stability limit. According to the results of their study, they showed that the postural sway rate did not change in PD without postural instability and gait disturbance, and the rate of postural sway increased in those with postural instability and gait disturbance with training. This result brings us to the opposite conclusion of our study because the AR and VR walking training applied by Wang et al. consisted of 5 exercises together with a warm-up and the training lasted only 7 sessions, and it was not combined with conventional exercises. Despite this different result, the functional balance evaluated by BBS in both groups is similar to the results of our study.
Gait disturbances are common in the mid to late stages of PD and lead to falls and losses of independence. In PD, cadence and double support phase duration increase, while step length, swing phase duration, and walking speed decrease [20]. One of the important disorders in walking in PD is the decrease in walking speed and deterioration of walking performance with the motor or cognitive dual-task added to walking. This situation is associated with impaired executive function [21].
In our study, when examined the gait after the training, it was seen that the patients in both groups walked with longer strides, but this improvement was higher in AR and VR gait training applied group. In addition, with AR and VR gait training, it was observed that the patients were able to walk with a narrower support surface, and the temporal parameters of gait improved, while it was determined that there was no improvement in these parameters of gait in the control group. Additionally, while functional mobility improved in the study group, it did not improve in the control group.
In our study, we had the chance to use more visual and auditory stimuli with the AR and VR training compared to the control group. In previous studies, it has been reported that the intense audio-visual cues and feedback mechanism of VR systems can facilitate motor learning when combined with exercise training [8, 22]. We can say that this situation provided an advantage in our study group compared to the control group, especially in improving the temporal characteristics of gait. Likewise, using the cues resulted in greater improvement in step and stride length in gait than conventional training.
The number of studies that analyze the effects of VR training on the spatio-temporal characteristics of walking with similar gait analysis methods to our study is very few [6, 8]. In other studies, it is seen that only its effects on functional mobility were examined [7, 9]. Mirelman et al. showed an increase in walking speed and stride length in PD patients who were given VR gait training on the treadmill. In another study, Calabro et al. showed that the spatio-temporal properties of walking with VR gait training applied on a treadmill provided more improvement than conventional training. In the studies of Feng et al. [7] and Wang et al. [9], who used the same AR and VR application as our study, study spatio-temporal gait analysis was not performed, and it was shown that functional mobility as assessed by TUG improved. As a result, in these studies, the researchers stated that the spatio-temporal features of walking and functional mobility are due to the VR environment’s ability to provide rich audio-visual stimuli, practice tasks that require cognitive demand, provide training at the level of difficulty for the patient, and motivate the patient. It is thought that the improvement in walking is related to these advantages of AR and VR gait training.
We think that the improvement we achieved in walking and functional mobility with AR and VR gait training is also related to the decrease in the severity of parkinsonian motor symptoms and the improvement in static and functional balance.
The most important limitation of our study is that although we provided dual-task walking training, we did not evaluate the improvement in dual-task performance, and the other limitation is that we did not examine the long-term effects of the training due to the COVID-19 pandemic.
Another limitation of our study is the low number of cases. We recommend that future studies be conducted by including more cases.
If we want to report our experiences and observations about AR and VR training, compared to conventional training, exercise sessions have become more enjoyable for patients thanks to features such as intense visual and auditory cues, goal-oriented games, and scoring systems included in the content of AR and VR training.
All of the patients participated in the training without interruption and stated that they wanted to continue the sessions afterward. We can say that this training has also changed the perspective of our patients who define exercise as boring. This seems to be an important advantage of AR and VR technologies to ensure that patients with chronic diseases such as PD continue to exercise. In conclusion, our findings demonstrate that AR and VR gait training combined with conventional training is an effective tool in the rehabilitation of balance and gait in patients with PD.
Funding
None.
Declarations
Conflict of interest
None.
Ethical approval
Republic of Türkiye Ministry of Health, Ankara Bilkent City Hospital Clinical Research Center Ethics Committee No. 2, with the decision number E2-21–632. The study was registered in the ClinicalTrial.gov PRC system (ClinicalTrial.gov ID: NCT05439967, Date of registration: 27/06/2022, “retrospectively registered”.)
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Stegemöller EL Buckley TA Pitsikoulis C Barthelemy E Roemmich R Hass CJ Postural instability and gait impairment during obstacle crossing in Parkinson’s disease Arch Phys Med Rehabil 2012 93 4 703 709 10.1016/j.apmr.2011.11.004 22318131
2. Caetano MJD Lord SR Allen NE Brodie MA Song J Paul SS Stepping reaction time and gait adaptability are significantly impaired in people with Parkinson's disease: implications for fall risk Parkinsonism Relat Disord 2018 47 32 38 10.1016/j.parkreldis.2017.11.340 29239746
3. Tan SB Williams AF Kelly D Effectiveness of multidisciplinary interventions to improve the quality of life for people with Parkinson's disease: a systematic review Int J Nurs Stud 2014 51 1 166 174 10.1016/j.ijnurstu.2013.03.009 23611510
4. van der Kolk NM King LA Effects of exercise on mobility in people with Parkinson’s disease Mov Disord 2013 28 11 1587 1596 10.1002/mds.25658 24132847
5. Yanovich E Ronen O The use of virtual reality in motor learning: a multiple pilot study review Advn Phys Educ 2015 5 03 188 10.4236/ape.2015.53023
6. Mirelman A Maidan I Herman T Deutsch JE Giladi N Hausdorff JM Virtual reality for gait training: can it induce motor learning to enhance complex walking and reduce fall risk in patients with Parkinson's disease? J Gerontol A Biol Sci Med Sci 2011 66 2 234 240 10.1093/gerona/glq201 21106702
7. Feng H Li C Liu J Wang L Ma J Li G Virtual reality rehabilitation versus conventional physical therapy for improving balance and gait in Parkinson’s disease patients: a randomized controlled trial Med Sci Monit 2019 25 4186 10.12659/MSM.916455 31165721
8. Calabrò RS Naro A Cimino V Buda A Paladina G Di Lorenzo G Improving motor performance in Parkinson’s disease: a preliminary study on the promising use of the computer assisted virtual reality environment (CAREN) Neurol Sci 2020 41 4 933 941 10.1007/s10072-019-04194-7 31858331
9. Wang Y Gao L Yan H Jin Z Fang J Qi L Efficacy of C-Mill gait training for improving walking adaptability in early and middle stages of Parkinson’s disease Gait Posture 2022 91 79 85 10.1016/j.gaitpost.2021.10.010 34656008
10. Morris ME Movement disorders in people with Parkinson disease: a model for physical therapy Phys Ther 2000 80 6 578 597 10.1093/ptj/80.6.578 10842411
11. Motek Medical BV. (2022). C-Mill, The Cutting-Edge Balance and Gait Treadmill, /, 2022 (Accesed :01.02.2022).
12. Borg GA Psychophysical bases of perceived exertion Med Sci Sports Exerc 1982 14 5 377 381 10.1249/00005768-198205000-00012 7154893
13. Opara J Małecki A Małecka E Socha T Motor assessment in Parkinsons disease Ann Agric Environ Med 2017 24 3 411 415 10.5604/12321966.1232774 28954481
14. LPG Systems, Huber 360® Evolution push the boundaries with the new Huber 360® evolution, www.lpgmedical.com/en/professional-area/huber/ , 2022 (Accesed : 06.11.2022).
15. Berg K Wood-Dauphine S Williams J Gayton D Measuring balance in the elderly: preliminary development of an instrument Physiother Can 1989 41 6 304 311 10.3138/ptc.41.6.304
16. Powell LE Myers AM The activities-specific balance confidence (ABC) scale J Gerontol A Biol Sci Med Sci 1995 50 1 28 34 10.1093/gerona/50A.1.M28
17. Morris S Morris ME Iansek R Reliability of measurements obtained with the timed “up & go” test in people with Parkinson disease Phys Ther 2001 81 2 810 818 10.1093/ptj/81.2.810 11175678
18. Maestri C Fundarò R Ferriero G Chimento P Taveggia G Casale R Self-selected speed gait training in Parkinson’s disease: robot-assisted gait training with virtual reality versus gait training on the ground Eur J Phys Rehabil Med 2019 55 4 456 462 30370751
19. Chastan N Debono B Maltête D Weber J Discordance between measured postural instability and absence of clinical symptoms in Parkinson’s disease patients in the early stages of the disease Mov Disord 2008 23 3 366 372 10.1002/mds.21840 18044726
20. Zanardi APJ da Silva ES Costa RR Passos-Monteiro E Dos Santos IO Kruel LFM Gait parameters of Parkinson’s disease compared with healthy controls: a systematic review and meta-analysis Sci Rep 2021 11 1 1 13 10.1038/s41598-020-80768-2 33414495
21. Larsen DSN Kegelmeyer DK Buford JA Kloos AD Heathcock JC Basso DM Neurologic rehabilitation. Neuroscience and neuroplasticity in physical therapy practice 2015 New York McGraw Hill Professional 283 323
22. Gómez-Jordana LI Stafford J Peper CE Craig CM Virtual footprints can improve walking performance in people with Parkinson’s disease Front Neurol 2018 9 681 10.3389/fneur.2018.00681 30174648
| 36443623 | PMC9707084 | NO-CC CODE | 2022-12-01 23:20:20 | no | Acta Neurol Belg. 2022 Nov 28;:1-9 | utf-8 | Acta Neurol Belg | 2,022 | 10.1007/s13760-022-02147-0 | oa_other |
==== Front
Place Brand Public Dipl
Place Branding and Public Diplomacy
1751-8040
1751-8059
Palgrave Macmillan UK London
284
10.1057/s41254-022-00284-0
Original Article
A battle for foreign perceptions: Ukraine’s country image in the 2022 war with Russia
Lee Seow Ting [email protected]
Seow Ting Lee
is Professor of Strategic and Health Communication at the University of Colorado Boulder. Her research examines organization-stakeholder relationships and communication strategies, with emphases on new media, ethics, and civic engagement. She holds a Ph.D. and an M.A. from the University of Missouri School of Journalism.
grid.266190.a 0000000096214564 Department of Advertising, Public Relations and Media Design, University of Colorado Boulder, 1511 University Avenue, UCB 478, Boulder, CO 80309 USA
28 11 2022
114
10 11 2022
14 11 2022
© The Author(s), under exclusive licence to Springer Nature Limited 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
This paper examines the 2022 War in Ukraine through the conceptual lens of country image within a broader framework of public diplomacy, soft power, and nation branding. It applies integrative and longitudinal perspectives to examine Ukraine’s country image and country image cultivation strategies over a period of three decades from its independence in 1991 to the 2022 Russo-Ukraine War. To understand Ukraine’s country image, or perceptions of Ukraine held by international stakeholders, a case study examined a convergence of multiple sources of evidence and methods. One method focuses on sentiment analysis through Leximancer, a computer-assisted qualitative data analysis software, to evaluate and compare Ukraine’s country image in international news media discourse across three periods of analysis separated by three major events: Ukraine’s independence in 1991, the 2014 Revolution of Dignity, and the 2022 Russo-Ukraine War. The findings suggest that Ukraine’s country image has strengthened over time, and has been effective in delegitimizing the 2022 Russian invasion and securing strong support from the international community.
Keywords
Country image
Public diplomacy
Soft power
Nation branding
Ukraine
Russia
War
Conflict
Leximancer
Case study
==== Body
pmcA favorable image of a country in a foreign public’s mind can benefit the country in many significant ways (Dubinsky 2019; Ingenhoff and Klein 2018; Kotler and Gertner 2002; Martin and Eroglu 1993; Wu and Wang 2018). It is imperative for a country to have a positive image that attracts, for example, tourism and investment—and in the case of Ukraine more recently, support and aid from the international community as the country battles a full-scale invasion by Russia in 2022.
As a young nation that gained independence in August 1991, Ukraine has struggled with its country image (Bureiko 2021; Markessinis 2010; Zarembo 2011). After independence, the government tried to shed its image of “a grey, cold, poverty-stricken, backward, alcohol-ridden, desolate former Soviet republic” (Markessinis 2010). Markessinis attributed Ukraine’s weak country image to its failure in creating a separate, unique international reputation by itself that would set it apart. As a result, “the country’s image is plagued with the regional stereotypes which pour into the country’s image void” (ibid). Ukraine, like other former Soviet and Yugoslav republics, experienced an identity crisis (Tereshchuk 2016). Foreign publics identified Ukraine closely with the former USSR. The two countries have been in conflict since 2014, but Russia’s full-scale invasion of Ukraine on February 24, 2022 drew significant international attention as Moscow’s attack on its Eastern European neighbor set off the largest armed conflict in Europe since the end of the Cold War.
Russia’s hopes for a quick victory in Ukraine have yet to materialize. As of November 2022, Ukraine has refused to capitulate despite heavy losses. The Ukrainians have been fighting this war on two fronts: hard power (military engagement with the Russians) and soft power (power of attraction and persuasion to secure support from the international community). Unlike hard power or military might, soft power is the power to attract co-opt rather than to coerce, thus influencing the preferences of others through appeal and attraction (Nye 2008). Ukrainian President Volodymyr Zelenskyy, who has roots in show business, delivered a speech at the 2022 Grammy Awards in which he rallied for global support for Ukraine (Kramer 2022). Although the actor-turned-wartime leader of Ukraine failed to secure an invitation to speak at the 2022 Academy Awards, his efforts to court the global community reflect a soft power approach to rally support for country’s defense in a war situation (Nye 2022a).
Relating country image to public diplomacy, nation branding, and soft power
This paper applies the conceptual lens of country image within a broader framework of public diplomacy, soft power, and nation branding. Country image can be understood through “the sum of beliefs and impressions people hold about places” (Kotler et al. 1993, p. 141). Image is a complex construct and includes both cognitive and affective evaluations (Alvarez and Campo 2014; Beerlí and Martín 2004; Maher and Carter 2011). Images represent “a simplification of large number of associations and pieces of information connected with a place” because “they are a product of the mind trying to process and pick out essential information from huge amounts of data about a place” (Kotler et al. 1993, p. 141). Country image has been defined as “the total of all descriptive, inferential and informational beliefs one has about a particular country” (Martin and Eroglu 1993), or “a simplification of large number of associations and pieces of information connected” with a country (Kotler and Gertner 2002).
Buhmann et al. (2018) noted that country image is an interdisciplinary construct that draws attention from a range of disciplines including business studies (e.g., international marketing and destination branding), social psychology (e.g., intergroup relations, collective and national identity research), sociology and political science (e.g., international relations, political anthropology, sociology of competition, public diplomacy), and communication science (e.g., mass media, social networks and media effects research, international communication, public relations, and communication management). However, country image studies within the context of war and conflict have focused on tourism (e.g., Alvarez and Campo 2014; Tabak and Avraham 2018) or country-of-origin associations related to consumers’ purchase intentions and decisions (e.g., Kang et al. 2018; Maher and Carter 2011). So far, no study has examined Ukraine’s country image within a broader framework of public diplomacy, soft power, and nation branding although much has been discussed in the press and trade journals about the perceived gains made by Ukraine in what were labeled conveniently as “soft power” advances (A new nation brand 2022; Adler 2022; Blake 2022; Clarke 2022; Ellwood 2022; Galeeva 2022; Kavanagh 2022; Knickmeyer and Arhirova 2022; Kramer 2022; McClory 2022; Nye 2022a, 2022b; Sarat and Aftergut 2022; Watson 2022).
The constructs of country image, public diplomacy, soft power, and nation branding continue to be debated by scholars, and remain largely segregated by disciplinary biases and conceptual disarray (Buhmann et al. 2018). Broadly defined, country image is the cognitive representation that individuals hold about a given country (Dubinsky 2019; Ingenhoff and Klein 2018; Wu and Wang 2018). As a perceptual construct, country image is an expression of personalized feelings of what individuals know and think about a country that are developed by representative products, national characteristics, economic and political background, and history of a country. A discussion of country image as a construct is incomplete without the related constructs of public diplomacy, nation branding, and soft power.
Public diplomacy broadly refers to strategic efforts to promote a country’s national interest, policies, and culture to strategically influence citizens of other nations by means of values, culture, art, food, music, media, language, and economic aid (Hachten and Scotton 2015). From a strategic perspective, the mandate of public diplomacy is to enhance a country’s image and reputation because country image is a core element of a state’s strategic equity (Dinnie 2008). As such, an important goal of public diplomacy is to enhance country image by “cultivating and managing a favorable international/world opinion toward a nation-state” (Wang 2006, p. 92).
When countries engage in public diplomacy, they are seeking to gain power, more specifically soft power—what Nye (2008) described as “the ability to affect others to obtain the outcomes one wants through attraction rather than coercion or payment” (p. 94) via intangible or indirect influences such as culture, values, and ideology. Soft power is a political theory of attraction (Nye 2004). Nye (2008) observed that a country’s soft power arises primarily from three sources: its culture (when it is attractive to others); its political values such as democracy and human rights (when it lives up to them); and its policies (when they are viewed as legitimate). To some scholars, country image—how country is perceived by international publics—is a form of soft power (e.g., Aronczyk 2013).
Also related to country image is nation branding, the purposeful cultivation of a country’s image on the international stage (Anholt 2007; Dinnie 2008; Fan 2006, 2010; Kahraman 2017; Lee 2021; Quelch and Jocz 2005). Father of nation branding Simon Anholt describes the process as “a strategic, policy-making approach, designed to help places build on the strengths that will earn them a better reputation” through “the process of designing, planning and communicating the name and the identity, in order to build and manage the reputation” (Anholt 2007, p. 4). It is a process through which a nation’s image “can be created or altered, monitored, evaluated and proactively managed in order to enhance the country’s reputation among a target international audience” (Fan 2010, p. 101). The scope of nation branding is broad, encompassing political, economic, and cultural dimensions with goals to enhance, change, or repair an existing image to make it more attractive and competitive (Anholt 2007; Fan 2006; Quelch and Jocz 2005).
Like public diplomacy, nation branding is a way to strengthen country image (Wu and Wang 2018). Nation branding is the purposeful cultivation of a country’s image on the international stage (Anholt 2007; Dinnie 2008; Fan 2006, 2010; Kahraman 2017; Lee 2021; Quelch and Jocz 2005). As a proactive action to strengthen and promote country image, nation branding is mainly driven by the will of the state actor or organizations on behalf of the state actor and thus “it is necessary to consider nation branding studies when examining country image in public diplomacy and political communication” (Wu and Wang 2018, p. 216). Although both public diplomacy and nation branding are concerned with the management of country image (Ingenhoff and Klein (2018), they have different disciplinary antecedents. Public diplomacy emerged from international relations and communications (i.e., international PR), whereas nation branding originated from marketing, and can be understood as the promotion and development of a country image through marketing activities (Anholt 2010; Fan 2006; Hynes et al. 2014). According to Fan (2006), the ultimate goal of nation branding is country image promotion.
To some scholars, nation branding and soft power are positively correlated (Fan 2006, 2010; Kahraman 2017; Lee 2021). According to Kahraman (2017), “[i]f nation branding as a technique for soft power is used strategically, it projects a positive image of the country and consequently the country will gain soft power” (p. 94). Fan (2006) defined public diplomacy as “a subset of nation branding that focuses on the political brand of a nation, whereas nation branding is about how a nation as whole reshapes international opinions” (p. 147). In explicating the relationships between and among the constructs of country image, nation branding, and soft power, Fan (2006) conjectured that a successful nation branding campaign “will help create a more favorable and lasting image among the international audience, thus further enhancing a country’s soft power” (p. 147).
This study adopts integrative and longitudinal perspectives to evaluate Ukraine’s country image and its efforts to cultivate its image. The findings can help clarify the conceptual relationships between and among country image, public diplomacy, soft power, and nation branding in a contemporary war context. Focusing on the 2022 war in Ukraine, this paper is driven by two research questions. The first research question asks: How has Ukraine’s country image changed? (RQ1). To evaluate Ukraine’s country image outcomes in the context of the 2022 war in Ukraine, it is helpful to take a longitudinal approach and begin from a baseline of its country image over a period of three decades from its independence in 1991 leading up to the 2022 Russo-Ukraine War.
The second research question, how has Ukraine sought to improve its country image? (RQ2), helps to contextualize Ukraine’s country image cultivation efforts, to better understand the outcomes in relation to the context in which the change effort occurs.
The two research questions are connected within a theory of change (Anderson 2005; Chen 1990; Weiss 1995) evaluation framework that helps us to understand the relationship between a given intervention and its outcomes. Within this analytical framework, RQ1 focuses on changes in Ukraine’s country image as outcomes that are related to the interventions that RQ2 addresses. More specifically, a logic model explicates the resources expended, the activities undertaken to produce the outcomes, and the tangible results of the activities (W. K. Kellogg Foundation 2004). Within the field, a theory of change is implicit and deserves more scholarly attention. Banks (2020), in discussing the challenges of evaluation in public diplomacy, observed that a theory of change can help identify changes (outcomes) we can reasonably expect public diplomacy to achieve.
Method
A case study approach was used. As a research strategy, case studies contribute to our knowledge of individual, group, organizational, social, political, and related phenomena (Yin 1984, 2003, 2014). Yin (2003) defined a case study as “an empirical inquiry that investigates a contemporary phenomenon within its real-life context especially when the boundaries between phenomenon and context are not clearly defined” (p. 13).
Case studies are “the preferred strategy when ‘how’ or ‘why’ questions are being posed, when the investigator has little control over events, and when the focus is on a contemporary phenomenon with some real-life context” (Yin 2003, p. 1). The essence of a case study, as explained by Schramm (1971), is to “illuminate a decision or a set of decisions: why they were taken, how they were implemented, and with what result” (p. 17).
One defining trait or strength of case studies is their reliance on “multiple sources of evidence, with data needing to converge in a triangulating fashion” (p. 14) that can be based on any mix of quantitative and qualitative evidence. Therefore, this case study focused on a convergence of multiple sources of evidence and methods: (1) Secondary data from soft power and nation branding indices, media coverage, and government documents; and (2) Primary data based on sentiment analyses of international news. In war, news media discourse is a core channel for soft power propagation to legitimize or delegitimize a war (Chouliaraki 2007). Given its perceptual nature, country image is typically associated with sentiments—individual emotions, subjective perceptions, evaluative opinions, and behaviors. Sentiment analyses help us understand people’s sentiments, what Mullen (2017) defined as subjective feelings, rather than facts, that include people’s attitude, emotions, and opinions.
RQ1 outcomes: sentiment analyses and Ukraine’s country image
Country image is affected by many factors, including personal experience and media portrayals (Wu and Wang 2018, p. 216). For many individuals, media portrayals of a country are more accessible compared to visits to personally experience a foreign country. A textual analysis of English-language texts published outside Ukraine was conducted using Leximancer, a data mining software and machine learning tool. The texts reflected a wide range of news publications of varying size and influence, and from a diverse population of countries and news media across the globe compiled by LexisNexis. They include BBC News, Egyptian Gazette, CNN, the Malay Mail, Sydney Morning Herald, Asahi Shimbun, LA Times, Straits Times, China Daily, New York Times, Fox News, Nepal Times, Palestine Chronicle, South China Morning Post, Rio Times, and Times of India.
Leximancer is particularly suited for analyses requiring the visualization and interpretation of large, complex corpuses of natural language text data. The software applies an iterative process of seeding word definitions from frequencies and co-occurrences of words within blocks of text. It then applies algorithms to analyze meanings within passages of text by extracting concepts, themes and how they relate to one another (Cretchley et al. 2010; Lemon and Jameson 2020; Tseng et al. 2015; Ward et al. 2014). In this study, Leximancer’s ability to scientometrically describe and analyze content of collections of textual documents was harnessed for sentiment analyses by extracting subjectivity to measure “the attitude of a speaker or writer toward a concept, whether they express something positively or negatively” (Ward et al. 2014).
Computer-assisted qualitative data analysis software (CAQDAS) such as Leximancer allows for an objective and efficient examination of the data, driven by the program (rather than the researcher) and with minimal researcher’s influence which might cloud the analysis (Wilk et al. 2021). The use of Leximancer is growing across a variety of disciplines. More specifically, Leximancer sentiment analyses have been used in many studies including international perception of China’s vaccine diplomacy (Lee 2021); Singapore’s hosting of the Trump-Kim Summit (Kim and Lee 2020), hospitality studies (Ma and Hsiao 2018), and social media sentiment about digital entrepreneurship (Wilk et al. 2021).
In a comparative, longitudinal approach, three periods of analysis were examined in this study, based on more than 2.7 million international news media texts over a period of three decades that were generated using the keyword “Ukraine.” The subsequent section presents the results for each period of analysis after the texts were seeded for positive, neutral, and negative sentiments.
Period 1: Post-independence (August 1991–February 2014)
The first period of analysis begins with Ukraine’s independence leading up to the Revolution of Dignity in February 2014 that overthrew the Ukrainian government that was led by the country’s pro-Moscow president Victor Yanukovych. Upon Yanukovych’s request, Russia intervened and the conflict between the two countries quickly escalated into the annexation of Crimea in late February 2014 by Russia and the Donbas War, in what many consider to be precursors to Russia’s full-scale invasion of Ukraine on February 24, 2022.
This period of analysis precedes the 2015 institutionalization of Ukrainian public diplomacy. Bureiko (2021) observed that after independence in August 1991, “public diplomacy in the official Ukrainian public and academic discourse was practically absent.” The Leximancer results support Bureiko’s observation that Ukrainian public diplomacy initiatives were isolated and lacked complexity, strategic vision, and financial support during this period. A sentiment analysis of 591,397 texts showed mostly negative sentiments about Ukraine with 58.7% of texts showing negative mentions, 20.4% positive mentions, and 20.9% neutral mentions.
A concept refers to a group of related words that travel together in the text. Evidence words include synonyms and adjectives. Leximancer concepts begin with seed words. The learning process then evolves the seed-word definitions into a full thesaurus that generate concepts that are word-like (e.g., create/creating/creation) or name-like (proper names such as Valerie or Singapore).
The overall likelihood scores in Leximancer indicated mostly negative sentiments (63.3%). “Backward” was the negative concept with the highest likelihood score at 73%. Likelihood scores describe the conditional probability of text segments within a concept also containing another concept (Leximancer 2011), meaning that 73% of text segments with “backward” contained negative sentiments. Other prominent negative concepts included “unsafe” (67%), “Russia” (63%), “poverty” (61%), “Communist” (59%), “conflict” (56%), and “history” (43%).
Four elements were reported positively. “Freedom” was the positive concept in Leximancer with the highest likelihood score at 45%, meaning that 45% of text segments with the term “freedom” contained positive sentiments. Other less prominent positive concepts included “agriculture” (41%), “affordable” (36%), and “culture” (31%) (see Table 1).Table 1 Sentiment analysis and likelihood scores of each Leximancer concept
Period 1
Post-independence Period 2
Pre-war Period 3
War
Positive concepts Positive concepts Positive concepts
Concept Likelihood (%) Concept Likelihood (%) Concept Likelihood (%)
Freedom
Agriculture
Affordable
Culture
45
41
36
31
Culture
History
Tourism
Modern
Grain exports
Independence
Affordable
Friendly
European
82
75
72
70
69
66
63
58
51
Zelenskyy
Heroic
Zelenska
Help
Military
Resistance
Resilience
Brave
Grammys
Recapture
Soft power
Social media
Humanitarian aid
94
91
90
89
88
86
85
85
78
78
66
64
56
Negative concepts Negative concepts Negative concepts
Concept Likelihood (%) Concept Likelihood (%) Concept Likelihood (%)
Backward
Unsafe
Russia
Poverty
73
67
63
61
Russia
Military
Safety
Communist
69
62
58
50
Russia
War
Losses
Casualties
95
85
81
79
Communist
Conflict
History
59
56
43
Conflict
Regional differences
49
46
Refugees
Casualties
United States
Inflation
Grain exports
77
72
62
60
59
Period 2: Early phase of Russo-Ukraine conflict (March 2014–February 2022)
The second period of analysis, focused on the aftermath of the February 2014 Revolution of Dignity in Ukraine, was marked by significant events that were part of the wider Russo-Ukraine conflict: continued armed engagement in various regions including Russia’s move to annex the Crimean Peninsula from Ukraine, the Crimea status referendum followed by Crimea’s independence in March 2014, and the creation of self-proclaimed breakaway states in Donetsk and Luhansk. The institutionalization of Ukrainian public diplomacy began in earnest in 2015, with an important milestone: a proposal by the foreign ministry to establish a specialized state institution called the Ukrainian Institute with the goal of improving perceptions of Ukraine around the world. The institute opened in June 2017 and became fully operational in 2019.
A sentiment analysis of the 1,514,157 texts between March 1, 2014 and February 23, 2022 (Russian launched a full-scale invasion of Ukraine on February 24, 2022) showed mostly positive sentiments about Ukraine with 54.3% of texts showing positive mentions, 34.1% negative mentions, and 11.6% neutral mentions. The overall likelihood scores in Leximancer indicated mostly positive sentiments (57.1%), a marked change from Period 1’s results. “Culture” was the positive concept with the highest likelihood score at 82%, meaning that 82% of text segments with “culture” contained positive sentiments. During this period, cultural diplomacy was a key priority for Ukraine despite persistent conflicts with Russia. Other prominent positive concepts included “history” (75%), “tourism” (72%), “modern” (70%), “grain exports” (69%), “independence” (65%), “affordable” (63%),” “friendly” (58%), and “European” (51%) (see Table 1). These concepts suggest a changing country image in the international news media discourse. It is worth noting that “history” was a negative concept (likelihood score of 43%) in Period 1 but in Period 2, it transformed into a positive concept (likelihood score of 75%). The positive concepts appear to resonate with Ukraine’s efforts, post-independence, to focus on cultural diplomacy, promotion of Ukrainian exports, and its tourism potential.
Six elements were reported negatively. “Russia” was the negative concept with the highest likelihood score at 69%, meaning that 69% of text segments with the term “Russia” contained negative sentiments. Other less prominent negative concepts included “military” (62%), “safety” (58%), “Communist” (50%), “conflict” (49%), and “regional differences” (46%) that speak to Ukraine’s old Soviet ties and prevailing tensions in the region.
Period 3: Russo-Ukraine war of 2022 (beginning February 2022)
The final period of analysis begins on February 24, 2022 with the full-scale Russian invasion of Ukraine and ends on September 30, 2022, when this paper was completed. A sentiment analysis of the 621,168 texts over 7 months revealed a majority of positive sentiments about Ukraine with 79.9% of texts showing positive mentions, 15.2% negative mentions and 4.9% neutral mentions. The data also revealed significant shifts in sentiments about Ukraine, when compared against the results from the two preceding periods of analysis.
The overall likelihood scores indicated a clear trend of positive sentiments at 80.3%. “Zelensky” was the positive concept with the highest likelihood score at 94%, meaning that 94% of text segments with “Zelensky” contained positive sentiments. Other prominent positive concepts are as follows: “heroic” (91%), “Zelenska” (Zelensky’s wife) (90%), “help” (89%), “military” (88%), “resistance” (86%), “resilience” (85%), “brave” (85%), “Grammy Awards” (78%), “recapture” (78%), “soft power” (66%), “social media” (64%), and “humanitarian aid” (56%) (see Table 1). These concepts speak to the leadership role played by Zelensky, the values that are being used by Ukraine to brand its beleaguered nation, and the significance of Ukraine’s social media communication about the war.
Nine elements were reported negatively. “Russia” was the negative concept with the highest likelihood score at 95%, meaning that 95% of text segments with the term “Russia” contained negative sentiments. Other prominent negative concepts included “war” (85%) “losses” (81%), “casualties” (79%), “refugees” (77%), “children” (72%), “United States” (62%), “inflation (60%), and “grain exports” (59%) (See Table 1). Russia emerged as the biggest loser in Period 3, suggesting a support for the Ukrainians’ efforts to delegitimize Russia’s attack. Moscow did not find favor in the preceding periods of analysis, but in Period 3, its reputation suffered significantly, consistent with the findings of soft power and nation branding indices such as the Anholt-Ipsos Nation Brand Index (NBI) 2022 and the Global Soft Power Index 2022 that shows an ascent in Ukraine’s country image during this period.
Reputations are difficult to measure but this has not stopped global indices such as the Anholt-Ipsos Nation Brand Index (NBI) and the Global Soft Power Index from continuing to rank nations annually using surveys and public opinion polls. Ukraine’s rankings in these nation brand and soft power indices support assertions (Bureiko 2021; Markessinis 2010; Zarembo 2011) about the country’s weak reputation. The NBI collects over 60,000 interviews online in 20 panel countries with adults aged 18 or over each year. In the 2021 Anholt-Ipsos NBI that ranked 60 countries, Ukraine was at the 48th place, having dropped five places from its 2020’s 43rd spot (Germany maintains top ‘nation brand’ 2021). Russia placed 27th in 2021 and 26th in 2020. In both years, Germany was ranked as the world’s strongest nation brand, for the seventh time overall and fifth year in a row.
In Brand Finance’s Global Soft Power Index 2021, Ukraine fell 15 places to 61st place, and lost its place in the top 50 (Ukraine drops out 2021). In the latest 2022 index, released in March 2022—less than a month after the Russian invasion, Russia’s soft power collapsed whereas Ukraine’s surged forward. Worldwide perceptions of Ukraine improved. The majority of survey respondents rated Ukraine’s conduct and response during the conflict positively. Net positive approval ratings ranged from + 23% in Brazil to + 60% in the U.K.—the highest in the study (Russia’s soft power collapses globally 2022). Ukraine’s Familiarity score increasing by 44%, Influence by 24%, and Reputation by 12%. The Index utilizes annual surveys to study global perceptions of nation brands by measuring the opinions of survey respondents worldwide on 120 nation brands. The global media spotlight on the war has also had a positive indirect effect on foreign perceptions of Ukraine across most other Global Soft Power Index pillars, even those unrelated to the war effort (USA bounces back 2022).
In the Leximancer analysis, the other negative concepts emphasized the consequences of war not only as domestic war losses and casualties but also impact on the global economy. The war has exacerbated supply shocks caused by COVID-19 on global commodity markets, trade, and financial linkages. The two countries are major suppliers of global commodities. Russia is a supplier of oil, gas, and metals, and, together with Ukraine, of wheat and corn. The OECD projected that the world economy will pay a “hefty price” for the war in Ukraine encompassing weaker growth, higher inflation, and potentially long-lasting damage to global supply chains (Horobin 2022).
In sum, the findings reveal an enhanced positive country image for Ukraine in Period 3. The results support the assertions made in the press and trade journals (A new nation brand 2022; Adler 2022; Blake 2022; Clarke 2022; Ellwood 2022; Galeeva 2022; Kavanagh 2022; Knickmeyer and Arhirova 2022; Kramer 2022; McClory 2022; Nye 2022a, 2022b; Sarat and Aftergut 2022; Watson 2022) that Ukraine has effectively advanced its interests in the war in the court of public opinion. The country image that the Ukrainian government has sought to associate with the symbolic representation of Ukraine at war and to delegitimize Russia’s military actions—its plight as an invaded nation, its heroic leaders, its resilience, its brave defense, its appeal to universal values of freedom and national sovereignty, and its need of international support, and the urgent need to prevent further fallout in the global economy—is vividly in place in the international news media discourse.
The U.S. is a negative concept (likelihood score of 62%). This result supports findings of a Global Soft Power Index survey that was conducted in March 2022 based on a representative sample of over 5000 respondents in 10 countries. In every surveyed nation, only a small number of respondents, between 1 and 10%, blame Ukraine for the invasion. The only Western nation to feature a significant proportion of people who blame the U.S. for the conflict is the U.S. itself—at 22% (Russia’s soft power collapses 2022).
This study’s Leximancer sentiment analyses have significant empirical value, especially in a field where evaluation is fraught with challenges. Outcomes often manifest themselves over the long term. Nye (2022a, b) suggested that soft power, for instance, tends to “operate subtly and over a longer time horizon.” Wagner (2014) argued that generating hard power “requires much less time as its resources are tangible” but soft power “takes relatively long to build as it is intangible.” However, as seen in the substantive country image gains generated by Ukraine within a short span of months, the power of values, persuasion and attraction could have the potential to be more powerful—less subtle and more immediate—in shaping foreign perceptions than previously envisioned. Instantaneous communication or the ability of social and digital media users—governments, corporations, organizations, crowd-funders, communities, individuals—to easily share information rapidly—augmented by visuals that help convey believability—may have very well amplified Ukraine’s country image in this case.
That said, it is difficult to conclusively attribute an outcome to a specific intervention. Outside of laboratory settings, causal relationships are difficult to prove. However, within a theory of change analytical framework, the country image outcomes of Ukraine can be better understood in relation to its country image interventions or the context (RQ2) in which the change effort occurs.
RQ2 Interventions: Ukraine’s country image cultivation
Ukraine has engaged in significant efforts to enhance its country image since its independence in 1991, consistent with its desire to change how the country was perceived by foreign publics. Through a logic model approach (W. K. Kellogg Foundation 2004) within a framework of theory of change (Anderson 2005; Chen 1990; Weiss 1995), this study investigates the country image resources expended by Ukraine, the activities undertaken to target the outcomes discussed earlier under RQ1, and the tangible results of the activities.
Institutionalization of public diplomacy
A key resource for Ukraine’s country image cultivation was cultural diplomacy. To differentiate itself from Russia after independence in 1991, Ukraine embarked on cultural diplomacy as a core strategy to strengthen its country image. However, it was a difficult start; “public diplomacy in the official Ukrainian public and academic discourse was practically absent; soft power initiatives were isolated and lacked complexity, strategic vision and financial support” (Bureiko 2021). Ukrainian public diplomacy began in earnest as late as 2015 with the institutionalization of public diplomacy as “a set of planned measures to convey specific messages about Ukraine to influence the formation of public opinion” in other countries (Bureiko 2021). The institutionalization is grounded in five core activities:The creation of a department of public diplomacy (later renamed the department of communications and public diplomacy) in 2015. It established a digital portal to communicate information about modern Ukraine to foreign publics at www.ukraine.ua. Since 2015, Ukraine has actively used Twitter to promote a number of information campaigns including #CrimeisUkraine, #JusticeforMH17, #LetMyPeopleGo, #StopNordStream2, and #KyivnotKiev (Bureiko 2021).
The establishment of the Ukrainian Institute (https://ui.org.ua/en/) in 2017 to improve understanding and perception of Ukraine around the world through cultural diplomacy and the country’s commitment to development to the values of freedom and democracy, and of international exchanges and dialogue. Then foreign minister Pavlo Klimkin described the institute as “the identification of Ukraine—what Ukraine is for the average Dutchman, Greek or Brazilian, what it is associated with, and of course, its attractiveness” (Ukrainian Institute: Soft power for promoting modern Ukraine 2017).
The formation of the Ukrainian Cultural Foundation (https://ucf.in.ua/) in 2017 to internationalize Ukrainian culture with state support for grant funding for culture and creative industries development and preservation of Ukrainian cultural heritage.
The launch of the Ukrainian Book Institute (https://book-institute.org.ua/uk) in 2016 to promote Ukrainian literature abroad through residency programs for authors and translation of Ukrainian literature into foreign languages.
The formal adoption of a national brand called “UkraineNow” in May 2018 and attendant narratives to promote Ukraine internationally as a “creative, dynamic and innovative” nation and “a country of freedom and dignity” where “millennia of history and culture are complemented by an ambitious vision of the future”—in the words of the Ukrainian Foreign Minister Dmytro Kuleba (The Ministry of Foreign Affairs of Ukraine launched the website 2021).
These five pre-war initiatives not only focused on activities of harnessing inter-institutional synergy and applied research and analytics but also, beyond image building, helped to consolidate Ukraine’s foreign policy priorities in cultural diplomacy, attraction of investment, promotion of Ukrainian exports, educational opportunities, tourism potential, and scientific-expert and digital diplomacies (Bureiko 2021).
The 2015 institutionalization of public diplomacy in Ukraine is a direct response to the forays of aggression against Ukraine by Russia, including the annexation of Crimea in 2014, the Donbas conflict, and what Ukrainians viewed as widespread anti-Ukraine misinformation initiated by Moscow. March 2021 marked a new milestone when the Ukrainian Foreign Ministry, for the first time in its history, formalized a five-year public diplomacy strategy (Ukrainian Public Policy Strategy 2021). The strategy document outlined seven action areas or activities: cultural, expert, culinary, scientific, educational, sport, and digital diplomacy. Prior to this, public diplomacy activities were assigned to cultural and information centers, which operated at foreign diplomatic missions. These centers were tasked disparately with disseminating information about Ukraine abroad, acquainting foreign audiences with Ukrainian history and culture, and informing them about Ukraine’s tourism opportunities and attractions, albeit with scant coordination (Bureiko 2021). The tangible results of the March 2021 strategy document included a consolidated framework of strategy, recalibration of activities, and greater engagement with foreign audiences.
The Zelensky factor
Ukrainian President Volodymyr Zelensky has emerged as Ukraine’s most significant country image cultivation resource. In tangible results, international news media reports have called him a hero and the voice and face of Ukrainian resilience leading a beleaguered nation in standing up against a powerful aggressor (Adler 2022; Blake 2022; Editorial: Why Volodymyr Oleksandrovych Zelenskyy 2022; Sarat and Aftergut 2022; How Volodymyr Zelensky found his roar 2022; Watson 2022). Blake (2022) argued that Zelensky is successful because he has “demonstrated, during the invasion, a style of political presentation that is the antithesis of that shown by Russian President Vladimir Putin.”
Ingenhoff and Klein (2018) found that political leaders can have a significant nation branding influence in shaping their country’s country image; the leader’s integrity played the strongest role followed by competence and charisma. Fan (2010) noted that “famous faces (celebrities)” is one important element that affects a country’s image. In the case of the Ukraine war of 2022, Zelensky “has proved especially adept at wielding soft power” (Nye 2022a). Zelensky’s country image cultivation activities are well-documented. The former comedian and television actor, with his informal attire, ubiquitous social media presence, and constant communication with Western media and parliaments (prompting standing ovations) and friendships with Western celebrities, has proved successful “in representing Ukraine as an attractive and heroic country” (Nye 2022a). The tangible result “was not just Western sympathy but a substantial increase in deliveries of the military equipment that Ukraine needed for the hard-power task at hand” (ibid).
In response to the U.S. government’s offer to evacuate him and his family—prime targets of Russian assassination attempts—out of the country, Zelensky told Washington: “The fight is here. I need ammunition, not a ride” (Braithwaite 2022). Zelensky has been highly adept at leveraging social media networks to humanize Ukraine’s plight and build support for the country’s defense (Khan 2022). He operated his social networks himself, appeared before the public directly, conveyed the messages personally in his own voice, and managed to repel Russian misinformation attempts about his whereabouts and his control of the country (Einhorn 2022). After rejecting Washington’s offer to spirit him out of Ukraine, he promptly tweeted that “Ukrainians are proud of their president” and posted a video titled, “Do not believe the fakes,” revealing that he was still in Kyiv, as depicted in the video. He said:“I am here. We are not putting down arms. We will be defending our country, because our weapon is truth, and our truth is that this is our land, our country, our children, and we will defend all of this. That is it. That's all I wanted to tell you. Glory to Ukraine.”
Zelensky may be the voice and face of Ukraine on news media and social media, but the Ukrainian government’s expertise in shaping its message through social media—in addition to video, tweets and memes—has helped it marshal the support of many countries and their citizens. In tangible results for the Ukrainians, governments have punished Russia with more sanctions than any other country in history (Shapero 2022). Ukraine’s social media campaigns have inspired foreign publics to express support for McDonalds, Coca-Cola, Starbucks, Dentsu, P&G, and other companies that suspended their businesses in Russia (Russ and Kumar 2022; Ukraine: Latest industry news 2022). Ukraine’s official social media channels have experienced a surge in followers since the start of the war. In just one month after the start of the war, followers of Ukraine’s Twitter account increased to 1.9 million followers, nearly six times as many as before the war (Social Blade 2022). Zelensky’s Instagram account added more than six million followers in just one month. Zelensky has been so popular on social media that hundreds of fake Zelenskys have emerged, using his image, name and “brand” to scam individuals into sending donations to purportedly aid Ukraine’s military and humanitarian efforts (Suciu 2022).
In terms of activities, Zelensky’s wife, Ukraine’s camera-shy first lady Olena Zelenska also stepped into the global stage. Zelenska made an unofficial trip to Washington, D.C. in July 2022 to become the rare first lady to address U.S. Congress. In her speech, she said that she was there to ask for more weapons for her country: “Weapons that will not be used to wage a war on somebody else’s land but to protect one’s home and the right to make up a life in that home” (Friedman 2022). Thanking the U.S. for its support, Zelenska applied soft power to ask for hard power; “I’m asking for weapons,” she said. “I’m asking for air defense systems in order for rockets not to kill children in their strollers” (Ukraine’s first lady appeals 2022). During her trip to Washington, D.C., Zelenska reaped tangible results including receiving a $30 million donation of computers and learning materials from HP Inc.’s Digital Equity for Ukraine initiative (First lady of Ukraine accepts 2022). Like her husband, Zelenska has given numerous media interviews to international media, including one to Time magazine where she was featured on the cover and spoke at length about the trauma of war (Shuster 2022). In July 2022, she hosted a high-profile summit moderated by British journalist Piers Morgan that featured prominent political figures and Hollywood celebrities including U.S. first lady Jill Biden, actors Richard Gere, Mila Kunis, Ashton Kutcher, and European Commission President Charles Michel (Nerozzi 2022).
Leveraging on existing public diplomacy machinery
When Russia invaded Ukraine, the latter was able to leverage on existing resources in the form of extant public diplomacy machinery created by the 2015 institutionalism of Ukrainian public diplomacy strategy. The department of public diplomacy (later renamed the department of communications and public diplomacy) quickly built on the Ukrainian digital portal, expanding into activities to communicate information about the war to foreign audiences at www.ukraine.ua. On July 27, the website featured this message:“On February 24, 2022, the ordinary life in Ukraine stopped, as did this website. Russia launched a full-scale attack on Ukraine at 5 am that day with an attempt to destroy the freedom and independence of our country. From that moment, Ukrainians ‘froze’ at the end of winter on that terrifying February morning. Spring has not come to Ukraine yet. You can find out how Ukraine is going through the war right now or discover what a dynamic and ambitious country we were before Russia launched a full-scale invasion on us.”
Visitors could click on a new link “Do not look away from the war” or proceed to “Explore peaceful Ukraine” for the original, pre-war content. Pop-up content redirected the visitor to a situation page featuring updates on the war and photographs of war-ravaged cities and communities. The same page also featured three links for visitor to “Support Ukraine,” “Donate to Ukraine,” and “Arm Ukraine.”
A similar messaging strategy was found on the Ukrainian Institute website (https://ui.org.ua/en/). It featured two photos of the Palace of Youth—the first taken before the war and the second featuring the bombed-out building spliced together, with the message: “What held a memory may now become a memory.” In April 2022, Ukraine launched a campaign, spearheaded by the Office of the President, to highlight the nation’s bravery. The series of campaign, created by Kyiv-based advertising agency Banda (a Cannes Lion winner), would appear in 15 countries worldwide. The campaign actively recruited global media owners to donate their media surfaces to highlight Ukraine’s bravery.
Other activities include the role played by the Ukraine House, a state-funded venue for Ukrainian cultural events, in expanding its mission to support Ukraine in the 2022 war against Russia. In September 2021, Zelensky officially opened the Ukraine House in Washington, D. C. It was designed to serve as a soft power hub in the U.S., to “provide the opportunity to significantly expand meetings, receptions, cultural and historic exhibitions, and other related activities in Washington” (Ukraine House opens 2022). When the war broke out, the Ukraine House proved to be a valuable asset, as local events to raise funds and awareness were organized at various Ukraine House locations across the globe including Washington, D.C. and Davos, Switzerland.
The role of G2P and P2P exchanges
While the role of governments is considered important in a traditional government-to-government (G2G) framework and in the government-to-public (G2P) structure of public diplomacy, newer public diplomacy relies more on people-to-people (P2P) exchanges in which voluntary participation of private citizens is saliently important (McPhail 2014; Rockower 2014). Ukraine’s soft power resources in the 2022 war have expanded into G2P and P2P exchanges. As argued by Rockower (2014), unlike traditional state diplomacy that relies on government-to-government (G2G) exchange, contemporary public diplomacy focuses on channels of government-to-public (G2P) exchange, and increasingly people-to-people (P2P) exchange. As the number of “help Ukraine” initiatives proliferated across the globe—and billions were raised through celebrity-fronted concerts, corporate initiatives, and private donations, the lines between G2P diplomacy and P2P diplomacy have also blurred. Ukraine has heavily depended on financial and military assistance from other nations, but the tangible results of crowdfunding efforts have also revealed the activities enacted by a large network of private individual donors and volunteers worldwide in augmenting Ukraine’s war chest (Private donations help 2022; Taylor 2022).
One example of P2P exchange is the indirect soft power role of foreign volunteer fighters in Ukraine and their activities on the battlefield and beyond it. In tangible results, the 2022 war in Ukraine has drawn a legion of volunteers to help Ukrainians repel the Russian invasion. Military veterans across the globe have been drawn to the war, “emboldened by their combat training and an eagerness to apply their skills in a conflict that, for many, feels like a struggle of good versus evil” (Horton et al. 2022). Many are self-appointed freedom fighters who responded to Zelensky’s plea through the International Legion of Defence of Ukraine (https://fightforua.org) that he created to recruit “foreign citizens wishing to join the resistance against the Russian occupants and fight for global security.” The website featured a photo of Zelensky accompanied by a personal plea: “Anyone who wants to join the defence of security in Europe and the world can come and stand side by side with the Ukrainians against the invaders of the twenty-first century” (Russia invaded Ukraine: Enlist to the international legion of defence of Ukraine n.d.). In March of 2022, Zelensky claimed that 16,000 foreigners had volunteered to fight in Ukraine (Bella and Timsit 2022).
Military volunteers are a form of hard-power resource, but they amplify Ukraine’s soft power resources and country image indirectly. Many of these volunteers use social media to livestream their day-to-day lives fighting on Ukraine’s front line. Their powerful messages and visuals appeal to their home audiences and beyond, creating a secondary and indirect source of soft power—and tangible results—to communicate Ukraine’s plight to the world and amplify its messages. For example, South Korean citizen Rhee Keun, also known on social media as Ken Rhee, was one of the many prominent volunteer fighters in Ukraine. He signed up at the Ukrainian embassy in Seoul when Zelensky asked for global volunteers. Rhee, a former member of South Korean naval special forces, posted regularly on social media photographs and videos of his experience in Ukraine (General 2022). The YouTuber, who has 700,000 followers and documented much of his Ukraine experience on his popular Instagram account, said he has no regrets when he was arrested upon his return to Seoul after being injured on the front lines in Kyiv. Seoul banned its citizens from traveling to Ukraine, and Rhee, who was hurt while leading a special operations patrol there, was met at Incheon Airport by 15 South Korean police officers waiting to arrest him on his return (S. Korean does not regret 2022). His dramatic return generated other tangible results of more international media coverage and viral interviews that helped to highlight the 2022 war in Ukraine.
Another example of P2P exchange reflects the indirect soft power activities played by communication professionals, with tangible results for Ukraine’s country image. It is significant that the public relations and advertising industries across the globe have also rallied to Ukraine’s aid, creating—in many cases, pro bono—impactful messaging and visuals that serve the Ukrainian government’s soft power and nation branding goals. In early May, a pro bono industry team effort led by TANK Worldwide and Grey Canada drew inspiration from President Zelensky’s speech to the Canadian parliament. The advertising campaign, targeted at Canadians, imagined Toronto city under attack like what happened in Ukraine, to send a powerful message on behalf of the Ukrainian government (Ukraine: Latest industry news 2022). On May 19, 2022, as part of its Global Awards, PR Week, an influential a trade magazine for the public relations industry, named Zelensky PR Week Global Communicator of the Year in recognition of his communication skills (Ukraine: Latest industry news 2022). In early March, Cannes Lion, the world’s largest gathering of the advertising and creative communications industry, banned Russian delegates and entries from its 2022 Awards (Ukraine: Latest industry news 2022).
Conclusion
This study is significant for addressing an understudied topic. It is the first study to apply the conceptual lens of country image within a broader framework of related constructs of public diplomacy, soft power, and nation branding to a twenty-first century, contemporary war context. Research on military conflicts typically emphasize hard power, consistent with Nye’s conceptual dichotomy of soft versus hard power, or attraction versus coercion. The findings of the primary data based on Leximancer sentiment analyses corroborated secondary data from existing soft power and nation brand indices that measure reputation, suggesting that Ukraine has strengthened its country image since its independence in 1991, and most recently in in the context of the 2022 Russo-Ukraine war to effectively deny the legitimacy of the Russian invasion and secure strong support from the international community. Rhetoric alone cannot change a country image. In a sense, “the ability to share information—and to be believed—becomes an important source of attraction and power” (Nye 2004, p. 31). Country image may be symbolic, but the power of argument and belief must be substantiated by substance—as seen in the resources expended by Ukraine, the activities undertaken to produce the outcomes, and the tangible results of the activities. The question is whether Ukraine’s enhanced country image would have a lasting effect beyond the 2022 war.
As a case study based on a comparative and longitudinal approach, this research tells a compelling story about the state of Ukraine’s country image and its country image cultivation strategies through a logic model within a theory of change framework, but it does not explicitly pinpoint causality beyond identifying the chronology of events and relating them to the outcomes from a baseline. Another limitation is the analysis of English-language news articles. Future research should address the direct causal relationships between and among country image, soft power, and nation branding, as well as address more diverse samples beyond English-language international media discourse.
Declarations
Conflict of interest
The author states that there is no conflict of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
A new nation brand was born. 2022. Bloom Consulting Journal. https://www.bloomconsulting.com/journal/a-new-nation-brand-was-born/.
Adler, N. 2022. Ukraine’s Zelenskyy: A wartime statesman for the social media age. Al Jazeera. https://www.aljazeera.com/features/2022/3/3/ukraines-zelenskyy-a-wartime-statesman-for-the-social-media-age.
Alvarez, M. D., and S. Campo. 2014. The influence of political conflicts on country image and intention to visit: A study of Israel's image. Tourism management 40: 70–78.
Anderson, A. 2005. The community builder’s approach to theory of change: A practical guide to theory development. Aspen Institute Roundtable on Community Change website http://www.theoryofchange.org/pdf/TOC_fac_guide.pdf.
Anholt S Competitive identity: The new brand management for nations, cities, regions 2007 Basingstoke, UK Palgrave Macmillan
Anholt S Places: Identity, image, and reputation 2010 New York Palgrave Macmillan
Aronczyk, M. 2013. Branding the nation: The global business of national identity. Oxford University Press.
Banks, R. (2020). Public diplomacy evaluation. In: N. Snow & Cull, N. J. (eds) The Routledge handbook of public diplomacy. Routledge, New York, pp. 64–76
Beerli, A., and J.D. Martin. 2004. Factors influencing destination image. Annals of Tourism Research 31: 657–681.
Bella, T. & Timsit, A. (2022, March 3). Zelensky says 16,000 foreigners have volunteered to fight for Ukraine against Russian invasion. Washington Post. Retrieved from https://www.washingtonpost.com/world/2022/03/03/zelensky-ukraine-16000-foreign-volunteers-russia/
Blake, M. (2022, March 4). How Zelenskyy emerged as the antithesis of Putin and proved you don’t need to be a strongman to be a great leader. The Conversation. Retrieved from https://theconversation.com/how-zelenskyy-emerged-as-the-antithesis-of-putin-and-proved-you-dont-need-to-be-a-strongman-to-be-a-great-leader-178485
Braithwaite, S. (2022, February 26). Zelensky refuses U.S. offer to evacuate, saying ‘I need ammunition, not a ride.’ CNN. Retrieved from https://www.cnn.com/2022/02/26/europe/ukraine-zelensky-evacuation-intl/index.html
Buhmann, A., Ingenhoff, D., White, C., & Kiousis, S. (2018). Charting the landscape in research on country image, reputation, brand, and identity: a transdisciplinary overview. In D. Ingenhoff, White, C., Buhmann, A., & Kiousis, S. (Eds.), Bridging Disciplinary Perspectives of Country Image, Reputation, Brand, and Identity (pp. 1–10). Routledge.
Bureiko, N. (2021). Ukraine’s public policy enters a new phase. Retrieved from https://neweasterneurope.eu/2021/06/23/ukraines-public-diplomacy-enters-a-new-phase/
Chen HT Theory-driven evaluations 1990 Thousand Oaks, CA Sage
Chouliaraki, L. (Ed.). (2007). The soft power of war. John Benjamins Publishing.
Clarke, M. & Ramscar, H. (2022, June 15). Ukraine: The fight for soft power values. Briefings for Britain. Retrieved from https://www.briefingsforbritain.co.uk/ukraine-the-fight-for-soft-power-values/
Cretchley J Galois C Chenery H Smith A Conversations between carers and people with schizophrenia: A qualitative analysis using Leximancer Qualitative Health Research 2010 20 12 1611 1628 10.1177/1049732310378297 20675536
Dinnie K Nation branding: Concepts, issues, practice 2008 Oxford, Burlington, MA Elsevier Butterworth-Heinemann
Dubinsky Y Analyzing the roles of country image, nation branding, and public diplomacy through the evolution of the modern Olympic movement Physical Culture and Sport 2019 84 1 27 40 10.2478/pcssr-2019-0024
Editorial: Why Volodymyr Oleksandrovych Zelenskyy is Chicago’s kind of leader. Chicago Tribune. 2022. https://www.chicagotribune.com/opinion/editorials/ct-edit-volodymyr-zelenskyy-ukraine-invasion-20220228-smdukknwwnhidp2a3vvx3mtzmu-story.html.
Einhorn, S. 2022. Ukraine-Russia war: On social media, Zelensky already won. Jerusalem Times. https://www.jpost.com/opinion/article-702750.
Ellwood, D. 2022. PD in wartime: Narratives, propaganda and smart power in the Ukraine conflict. CPD Blog. https://uscpublicdiplomacy.org/blog/pd-wartime-narratives-propaganda-smart-power-ukraine-conflict-part-2-inventing-global.
Fan Y Banding the nation: What is being branded? Journal of Vacation Marketing 2006 12 1 5 14 10.1177/1356766706056633
Fan Y Branding the nation: Towards a better understanding Place Branding and Public Diplomacy 2010 6 2 97 103 10.1057/pb.2010.16
First lady of Ukraine accepts computer donation from HP Inc. for Ukrainian children and healthcare workers. 2022. https://press.hp.com/us/en/press-releases/2022/ukraine-first-lady-accepts-hp-computer-donation.html.
Friedman, V. 2022. In Washington, Olena Zelenska dressed for Ukraine. New York Times. https://www.nytimes.com/2022/07/21/style/olena-zelenska-washington-style.html.
Galeeva, D. 2022. The role of soft power in the Ukraine conflict. Arab News. https://www.arabnews.com/node/2074626.
General, R. 2022. Korean former Navy SEAL Ken Rhee shares video of tense Ukraine mission evacuating wounded soldier. Yahoo News. https://www.yahoo.com/video/korean-former-navy-seal-ken-180256207.html.
Germany retains top ‘nation brand’ ranking, US and China experience significant decline. 2021. IPSOS. https://www.ipsos.com/en-us/news-polls/Germany-Retains-Top-Nation-Brand-Ranking-the-United-Kingdom-emerges-ahead-of-Canada-to-Round-Out-the-Top-Three-US-and-China-Experience-Significant-Decline.
Hachten WA Scotton JF The world news prism: Digital, social and interactive 2015 Hoboken Wiley
Horobin, W. 2022. Russia-Ukraine war threatens prolonged effect on global economy. Al Jazeera. https://www.aljazeera.com/economy/2022/6/8/russia-ukraine-war-threatens-long-lasting-impact-on-global-eco.
Horton, A., R. Nakhlawi, and S. Mekhennet. 2022. Ukraine war volunteers are coming home, reckoning with difficult fight. https://www.washingtonpost.com/national-security/2022/05/28/americans-fighting-in-ukraine/.
How Volodymyr Zelensky found his roar. 2022. The Economist. https://www.economist.com/europe/2022/02/26/how-volodymyr-zelensky-found-his-roar.
Hynes, N., B. Caemmerer, E. Martin, and E. Masters. 2014. Use, abuse or contribute!: A framework for classifying how companies engage with country image. International Marketing Review 31 (1): 79–97.
Ingenhoff D Klein S A political leader’s image in public diplomacy and nation branding International Journal of Communication 2018 12 4507 4532
Kahraman A Chitty N Ji L Rawnsley GD Hayden C A methodological approach to nation branding The Routledge handbook of soft power 2017 New York Routledge, Taylor and Francis 93 103
Kang, M., S. Kim., and G. Lee. 2018. How international political conflict hurts country image and further influences consumers’ purchase intention. International Business Research 11 (2): 95–102.
Kavanagh, J. 2022. The Ukraine War shows how the nature of war is changing. Carnegie endowment for international peace. https://carnegieendowment.org/2022/06/16/ukraine-war-shows-how-nature-of-power-is-changing-pub-87339.
Khan, I. 2022. Zelensky humanizes Ukraine’s plight in his social media messaging (2022, March 17). CNET. https://www.cnet.com/news/politics/zelenskyy-humanizes-ukraines-plight-in-his-social-media-messaging/.
Kim, H. S., & S.T. Lee. 2020. Peace talks: Public diplomacy and place branding in the 2018 Trump-Kim summit in Singapore. Place Branding and Public Diplomacy 17: 155–167. 10.1057/s41254-020-00163-6
Knickmeyer, E., and H. Arhirova. 2022. Taking spotlight, Ukraine first lady pleads for more US arms. AP News. https://apnews.com/article/russia-ukraine-biden-jill-congress-d8f03b9bf1262ed3853829631cfca371.
Kotler P Gertner D Country as brand, product, and beyond: A place marketing and brand management perspective Journal of Brand Management 2002 9 4 249 261 10.1057/palgrave.bm.2540076
Kotler P Haider D Reiner I Marketing places: Attracting investment, industry and tourism to cities, states and nations 1993 New York, NY The Free Press
Kramer, A.E. 2022. Zelensky, whose career started in show business, pressed Hollywood for an appearance at the Oscars. New York Times. https://www.nytimes.com/2022/03/27/world/europe/zelensky-oscars.html.
Lee ST Vaccine diplomacy: Nation branding and China’s COVID-19 soft power play Place Branding and Public Diplomacy 2021 10.1057/s41254-021-00224-4
Lemon LL Hayes J Enhancing trustworthiness of qualitative findings: Using Leximancer for qualitative data analysis triangulation The Qualitative Report 2020 25 3 604 614
Leximancer. 2011. Leximancer manual version 4. www.leximancer.com/site-media/lm/science/Leximancer_Manual_Version_4_0.pdf
Ma E Cheng M Hsiao A Sentiment analysis: A review and agenda for future research in hospitality contexts International Journal of Contemporary Hospitality Management 2018 30 11 3287 3308 10.1108/IJCHM-10-2017
Maher, A.A., and L.L. Carter. 2011. The affective and cognitive components of country image: Perceptions of American products in Kuwait. International Marketing Review 28 (6): 559–580. 10.1108/02651331111181411
Markessinis, A. 2010. Branding Ukraine. Nation Branding. https://nation-branding.info/2010/03/10/branding-ukraine/.
Martin IM Eroglu S Measuring a multi-dimensional construct: Country image Journal of Business Research 1993 28 3 191 210 10.1016/0148-2963(93)90047-S
McClory, J. 2022. The soft power of Ukraine’s blue and gold offers hope for victory. Nikkei Asia. https://asia.nikkei.com/Opinion/The-soft-power-of-Ukraine-s-blue-and-gold-offers-hope-for-victory.
McPhail TL Global communication: Theories, stakeholders, and trends 2014 4 Malden, MA Wiley Blackwell
Mullen, T. 2017. Introduction to sentiment analysis. https://lctmaster.org/files/MullenSentimentCourseSlides.pdf
Nerozzi, T.H.J. 2022. Ukrainian first lady Olena Zelenska hosts summit moderated by Piers Morgan. Fox News. https://www.foxnews.com/world/ukrainian-first-lady-olena-zelenska-hosts-summit-moderated-piers-morgan.
Nye, J. 2022a. Eight lessons from the Ukraine War. https://www.chinausfocus.com/peace-security/eight-lessons-from-the-ukraine-war.
Nye, J. 2022b. Soft power after Ukraine. Retrieved from https://www.project-syndicate.org/commentary/soft-power-after-russia-war-in-ukraine-by-joseph-s-nye-2022b-05
Nye J Soft power: The means to success in world politics 2004 New York Public Affairs
Nye J Public diplomacy and soft power Annals of the American Academy of Political and Social Science 2008 616 1 94 109 10.1177/0002716207311699
Private donations help support Ukraine’s fight against Russia. 2022. Voice of America. https://learningenglish.voanews.com/a/private-donations-help-support-ukraine-s-fight-against-russia/6635352.html.
Quelch J Jocz K Positioning the nation-state Place Brand & Public Diplomacy 2005 1 229 237 10.1057/palgrave.pb.5990024
Rockower, P. 2014. The state of gastrodiplomacy. Public Diplomacy Magazine,11. Winter. https://www.bidd.org.rs/the-state-of-gastrodiplomacy/.
Russ, H., and U.S. Kumar. 2022. McDonald's, Starbucks, Coke and Pepsi all halt sales in Russia. Reuters. https://www.reuters.com/markets/europe/morgan-stanley-backs-cautious-fed-rate-hike-ukraine-crisis-fuels-inflation-2022-03-08/.
Russia invaded Ukraine: Enlist to the international legion of defense of Ukraine. n.d. International Legion of Defense of Ukraine. https://fightforua.org.
Russia’s soft power collapses globally following invasion, attitudes towards Ukraine soar. 2022. https://brandfinance.com/press-releases/russias-soft-power-collapses-globally-following-invasion-attitudes-towards-ukraine-soar.
S. Korean does not regret breaking the law for Ukraine. 2022. Taipei Times. https://www.taipeitimes.com/News/world/archives/2022/06/28/2003780724.
Sarat, A., and D. Aftergut. 2022. Not all heroes wear capes. Some just stand up to dictators like Zelenskyy is with Putin. USA Today. https://www.usatoday.com/story/opinion/columnists/2022/03/02/zelenskyy-proving-heroes-exist-russia-invades-ukraine/9330217002/.
Schramm, W. 1971. Notes on case studies of instructional media projects. Working paper for Academy of Educational Development, Washington DC.
Shapero, J. 2022. Russia surpasses Iran to become world's most sanctioned country. Axios. https://www.axios.com/2022/03/08/russia-most-sanctioned-country.
Shuster, S. 2022. Ukraine's First Lady Olena Zelenska takes on the trauma of war. Time. https://time.com/6194177/olena-zelenska-ukraine-first-lady-interview/.
Social Blade. 2022. https://socialblade.com/twitter/user/ukraine.
Suciu, P. 2022. There are now hundreds of Volodymyr Zelensky impersonators on social media. Forbes. https://www.forbes.com/sites/petersuciu/2022/03/16/there-are-now-hundreds-of-volodymyr-zelenskyy-impersonators-on-social-media/?sh=369f0ee54a6e.
Tabak, L., and E. Avraham. 2018. Country image repair strategies during an asymmetrical conflict: an analysis of the Gaza conflict in 2014. International Journal of Strategic Communication 12 (3): 237–251.
Taylor, A. 2022. Ukraine has received nearly $900 million in donations as the Ukraine-Russia war rages on. Fortune. https://fortune.com/2022/04/15/how-much-donated-ukraine-war/.
Tereshchuk V Cultural diplomacy as a tool of Ukraine’s foreign policy: Achievements and challenges The Copernicus Journal of Political Studies 2016 2 33 46
The Ministry of Foreign Affairs of Ukraine launched the website UKRAINE.UA about modern Ukraine for foreign audiences. 2021. Ukraine Ministry of Foreign Affairs. https://mfa.gov.ua/news/dmitro-kuleba-zapustiv-sajt-ukraineua-pro-suchasnu-ukrayinu-dlya-inozemnih-auditorij.
Tseng C Wu B Morrison AM Zhang J Chen YC Travel blogs on China as a destination image formation agent: A qualitative analysis using Leximancer Tourism Management 2015 46 347 358 10.1016/j.tourman.2014.07.012
Ukraine drops out of Top 50. 2021. Global Soft Power Index 2021. Retrieved from https://brandfinance.com/press-releases/global-soft-power-index-ukraine-drops-out-of-top-50.
Ukraine House opens in Washington. 2022. U.S.-Ukraine Business Council. https://www.usubc.org/site/recent-news/ukraine-house-opens-in-washington.
Ukraine: Latest industry news. 2022. Advertising Association. https://adassoc.org.uk/our-work/ukraine-latest-industry-news/.
Ukraine’s first lady appeals to members of Congress for continued support in fight against Russian invasion. 2022. Yahoo News. https://news.yahoo.com/ukraine-first-lady-appeals-members-161126737.html.
Ukrainian Institute: Soft power for promoting modern Ukraine in the world. 2017. Ukrainian Foreign Ministry. https://www.kmu.gov.ua/news/249730313.
Ukrainian Public Diplomacy Strategy 2021–2025. 2021. Ukraine Foreign Ministry. https://mfa.gov.ua/storage/app/sites/1/%D0%A1%D1%82%D1%80%D0%B0%D1%82%D0%B5%D0%B3%D1%96%D1%97/public-diplomacy-strategy.pdf. Accessed 20 July 2022.
USA bounces back better to top of nation brand ranking. 2022. Global Soft Power Index 2022. Retrieved from https://brandfinance.com/press-releases/global-soft-power-index-2022-usa-bounces-back-better-to-top-of-nation-brand-ranking.
W.K. Kellogg-Foundation Logic model development guide: Using logic models to bring together planning, evaluation, & action 2004 Battle Creek, MI W. K Kellogg Foundation
Wagner, J. N. E. 2014. The effectiveness of soft and hard power in contemporary international relations. E-International Relations. https://www.e-ir.info/2014/05/14/the-effectiveness-of-soft-hard-power-in-contemporary-international-relations/.
Wang J Managing national reputation and international relations in the global era: Public diplomacy revisited Public Relations Review 2006 32 2 91 96 10.1016/j.pubrev.2005.12.001
Ward V West R Smith S McDermott S Keen J Pawson R House A The role of informal networks in creating knowledge among health-care managers: A prospective case study Health Services and Delivery Research 2014 10.3310/hsdr02120
Watson, I. 2022. It’s our brand: Ukraine’s government turns nation’s bravery into biggest cultural export. PR Week. https://www.prweek.com/article/1752815/its-brand-ukraines-government-turns-nations-bravery-biggest-cultural-export.
Weiss CH Nothing as practical as good theory: Exploring theory-based evaluation for comprehensive community initiatives for children and families New Approaches to Evaluating Community Initiatives: Concepts, Methods, and Contexts 1995 1 65 92
Wilk V Cripps H Capatina A Micu A Micu AE The state of #digitalentrepreneurship: A big data Leximancer analysis of social media activity International Entrepreneurship and Management Journal 2021 17 4 1899 1916 10.1007/s11365-020-00729-z
Wu D Wang J Ingenhoff D White C Buhmann A Kiousis S Country image in public diplomacy: From messages to relationships Bridging disciplinary perspectives of country image, reputation, brand, and identity 2018 London Routledge 212 229
Yin R Case study research: Design and methods 1984 1 Newbury Park Sage
Yin RK Designing case studies Qualitative Research Methods 2003 5 14 359 386
Yin R Case study research: Design and methods 2014 5 Thousand Oaks Sage
Zarembo, K. 2011. Soft power and Ukraine. Kyiv Post. https://www.kyivpost.com/article/opinion/op-ed/soft-power-and-ukraine-100406.html?flavour=mobile.
| 0 | PMC9707085 | NO-CC CODE | 2022-12-01 23:20:20 | no | Place Brand Public Dipl. 2022 Nov 28;:1-14 | utf-8 | null | null | null | oa_other |
==== Front
High Educ (Dordr)
High Educ (Dordr)
Higher Education
0018-1560
1573-174X
Springer Netherlands Dordrecht
974
10.1007/s10734-022-00974-x
Article
Student Ethnic Diversity Management in Ethiopian Universities: Practices, Challenges, and Way Forward
Adamu Abebaw Yirga [email protected]
grid.7123.7 0000 0001 1250 5688 Department of Educational Planning and Management, Addis Ababa University, Addis Ababa, Ethiopia
29 11 2022
116
21 11 2022
© The Author(s), under exclusive licence to Springer Nature B.V. 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Ethnic-related issues have been the historic and prevalent questions of Ethiopian society. Student ethnic diversity management is one of the major challenges facing public universities in Ethiopia. I interviewed 48 deans to examine the practice, challenges, and way forward of student ethnic diversity management. The study revealed that the practice of student ethnic diversity management is low and there are different internal and external factors that affect diversity management in higher education. Above all, the politicization of ethnicity is found to be the mother of all challenges associated with student ethnic diversity management in Ethiopia. This implies that addressing issues related to the ethnic political and administrative system at a national level is the highest priority, but is a long-term plan, and universities need to take different measures that help them to at least avoid or minimize the negative impacts of diversity.
Keywords
Diversity
Diversity management
Ethiopia
Higher education, University
http://dx.doi.org/10.13039/501100007941 Addis Ababa University
==== Body
pmcIntroduction
Diversity can be broadly conceived of as all the ways that people are different (Pant and Vijava, 2015; Adamu, 2014). The difference could be based on various attributes including, among other things, ethnicity, religion, race, language, culture, ability, social status, and sexual orientation. In the higher education context, diversity may be mostly considered as all those differences among students, staff, and faculty members. This study focused on student ethnic diversity in Ethiopian public universities.
Diversity is important at different levels of personal development, but it is thought to be significantly important during the university years because many students come to university in late adolescence and early adulthood, which is a critical stage of development in which individuals define themselves in relation to others and exercise various social roles before making permanent commitments to different issues including professions, close relationships, and social and political groups (Gurin et al., 2002). Diversity considerably enhances students’ development in the cognitive, affective, and interpersonal domains (Milem, 2003). Some of the individual benefits of diversity include greater openness to diversity (Pascarella et al., 1996), increased cultural knowledge and understanding of diversity (Antonio, 2001), increased intellectual engagement and personal development (Umbach & Kuh, 2006), developed complex and critical thinking (Gurin et al., 2002), developed perspective-taking skills (Gottfredson et al, 2008), reduced levels of stereotyping and ethnocentrism (Milem, 2003), understanding self and other (Pascarella et al., 1996), growth in intellectual self-concept (Gurin et al., 2002) and academic skills (Milem, 2003), and better prepared for living and working in a diverse society (Gurin et al., 2002).
A diverse student body can be valuable because it potentially leads to a wider range of thoughts, ideas, and opinions (Adamu, 2014), increases the likelihood of socializing with diverse groups and discussing various diversity-related issues (Gurin et al., 2002), and creates an enriched academic environment (Shaw, 2005). This, in turn, contributes to developing mutual understanding and positive intergroup relations by challenging students to refine their way of thinking and reducing prejudices toward outgroup members (Pettigrew, 1998).
It is necessary to note that although the mere presence of students from diverse backgrounds is an opportunity for meaningful interactions among diverse peers (Roksa et al., 2017), it does not guarantee to benefit from diversity as the outcome of an increased diverse student body is not necessarily positive. Study indicates that diversity-related challenges lead to less cohesiveness, less effective communication, increased anxiety, and greater discomfort, tension, and conflict among diverse groups (Cox, 2001). When diversity is not properly addressed or is totally ignored, it can have negative effects, such as “increased egocentrism, and negative relationships characterized by hostility, rejection, divisiveness, scapegoating, bullying, stereotyping, prejudice, and racism” (Johnson & Johnson, 2000, p. 15). This shows that a diverse student body is a necessary but not sufficient condition to benefit from diversity (Hurtado, 2007; Milem, 2003; Umbach & Kuh, 2006). This implies that the way diversity is addressed in different contexts determines its impact (Milem et al., 2005). Therefore, managing diversity becomes one of the crucial tasks of universities.
There is no universally agreed-upon definition of diversity management (Foster & Harris, 2005), and yet generally managing diversity can be defined as “the process of creating and maintaining a positive environment where the differences of all personnel are recognized, understood and valued, so that all can achieve their full potential” (Lumadi, 2008, p.8). In this study, managing diversity is likewise conceived of as a continuing process that intends to create and foster a positive learning and living environment on campus that benefits all individuals and groups regardless of their ethnic differences.
Ethnic, religious, and linguistic diversities are considered to be significant distinguishing features of Ethiopian society (Adamu, 2014), and diversity-related issues are rooted in the social and political history of the country. In Ethiopia, there are more than 80 languages and Amharic is the working language of the federal government. There are also different religions including Christianity (Catholic, Orthodox, and Protestant), Islam, Judaism, and traditional religions.
In Ethiopia, the first higher education was established in 1950 during the Haileselassie regime. As any government does, the Haileselassie regime promised to bring development to the country. However, the regime’s failure to address several social and political problems and the denial of equality created widespread dissatisfaction. And yet, it was difficult to challenge the political and administrative system as there was a lack of political consciousness among Ethiopians, most of whom were illiterate and working for landlords under a feudal system. There were also no civil society organizations or political parties to stand up for the rights of the larger society and voice its grievances (Adamu & Basvik, 2018). It was in such contexts that university students emerged as “the most outspoken and visibly the only consolidated opposition group” (Balsvik, 1985, p. xiii) to challenge the system. The students considered themselves the “spokesmen and advocates of the ‘suffering masses’” (Balsvik, 2009, p. 263). Ethnicity also became a political issue when some students of the then Haileselassie I University raised questions of nationalities in the late 1960s. Although this issue was less emphasized during the military regime (1974–1991), it became a major topic of discussion in the early 1990s when the Ethiopian People Revolutionary Democratic Front (EPRDF) took power and introduced an ethnic-based federal and political system.
Since the introduction of the ethnic-based federal and political system, it is becoming more and more evident that many people’s first allegiance is to their ethnic group, and “their country is a poor second” (Milkias, 2011, p. 58). EPRDF also amplified historical grievances and the discourse of oppressor and of oppressed as part of the divide and rule strategy, and this has created hatred and animosity among the society and students from different ethnic backgrounds. This strategy instilled distrust and eroded the intergroup relationship between different ethnic groups. Current university students were born after the EPRDF came to power, and they learned both prejudice and stereotypes associated with different ethnic groups. In a nutshell, it can be argued that the government’s ethnic-based administration and political system has influenced intergroup relations among the society and students from different ethnic backgrounds (Adamu, 2015) by creating a fertile ground for the formation of social stereotypes and prejudice (Bigler & Liben, 2007).
Both the ruling and opposition political parties consider higher education institutions as fertile grounds for executing their political agenda because of (1) university students’ historical contribution to political changes in Ethiopia (Adamu & Balsvik, 2018) and (2) the diverse student body they have from every corner of the country which is considered as a microcosm of the Ethiopian society. Although there were minor ethnic tensions and conflicts among students at least over the past three decades, this has increased since the 2005 general election, where opposition parties rejected the provisional results of the election and university students mounted public protests over the election resonating opposition parties’ complaints (Adamu, 2014). Public universities (e.g., Addis Ababa, Bahir Dar, Dire Dawa, Ambo, Adigrat, Hawassa, Woldia, and Haramaya universities) were the scene of recurring ethnic conflicts. The magnitude of ethnic tension and conflict in public universities has particularly soared (both in frequency and impact) in the past few years, and more so since 2018 (Ashine, 2019). During 2018–2019, campuses were temporarily closed, many students left universities, and some students were killed (Adamu, 2019). In early 2020, universities were so shaken by a series of escalated ethnic conflicts and the government hinted at the possible remedy of closing universities to take time to address the issue (Woldegiyorgis & Adamu, 2022). As a blessing in disguise, COVID-19 forced the closure of universities, and this was a big relief for the government and society.
Although public universities are answerable to the federal government, they have been behaving and acting like local universities serving the society of the region they are located. Instead of addressing this fundamental problem, which is against the wisdom of a university, the government promoted ethnic-based regionalization of universities through ethnic-based senior leadership appointments and university board membership of regional authorities. This coupled with the national-level politicization of ethnicity has impeded positive intergroup relations and intensified tensions and conflicts among ethnically diverse students.
In relation to this, universities have a special role to play in ensuring that a diverse student body is considered as an opportunity rather than a problem. In order to achieve this, universities need to plan and design different approaches, and thus, managing diversity becomes one of their important duties (Liu, 1998). However, ethnic diversity in higher education is considered as the most pronounced aspect of diversity that needs urgent diversity management interventions because of the ever-increasing ethnic tensions and conflicts in public universities (Adamu, 2019; Amnesty International, 2020; Ashine, 2019) which resulted in physical injury and death of students, damage to university and personal properties, interruption of teaching–learning process and temporary closure of universities. Accordingly, the purpose of this study is to examine the practice, challenges, and way forward of ethnic diversity management in Ethiopian public universities.
Method
The study used phenomenological research design to better understand participants’ experiences regarding ethnic diversity management in Ethiopian universities. In Ethiopia, public universities are categorized into four generations based on their year of establishment. Initially, it was planned to collect data from senior and middle-level leaders. However, because of the COVID-19 pandemic, it was not advisable to travel to different parts of the country. Moreover, senior leaders were busy addressing institutional and national challenges faced as a result of the pandemic. Therefore, the necessary data were generated from 48 deans and associate deans who were participating in the Ethiopian higher education leadership training program. The training requires deans to attend three seminars which are three days each. This made the data generation process easy, fast, and effective. The participants were purposefully selected from 19 universities to obtain different perspectives and views regarding the issue under study from the four generations of universities in the country. A semi-structured in-depth one-to-one interview was used to generate data from 12 participants and group interviews to generate data from six groups consisting of five to eight participants. Participants were selected for a one-to-one or group interview based on their availability and preference.
Document reviews were also used to generate data from official documents including the higher education proclamation (Ministry of Science & Higher Education, 2020), higher education policy and strategy (Federal Democratic Republic of Ethiopia [FDRE], 2019), and senate legislations. Data obtained through document review were used to corroborate data generated through interviews. The interview and document review data were analyzed using detailed and context-specific thematic analysis. In order to keep confidentiality, in direct quotations, participants were addressed using abbreviations followed by numbers (e.g., Dean of College of Natural and Computational Sciences = DCNCS1, DCNCS2; Dean of School of Law = DSL1, DSL2…).
Results
This section presents the practices, challenges, and prospects of diversity management in Ethiopian universities.
Practices
In the last two decades, most Ethiopian universities have come to embrace the notion of diversity, and this is often reflected in their institutional core values. Participants indicated that managing ethnic diversity should be given the highest priority in the plan and process of addressing challenges facing Ethiopian public universities. However, the higher education proclamation, the higher education policy and strategy, and the 10-year plan of the ministry of science and higher education are not focused on diversity and how to manage diversity. In universities, there is also a clear lack of policies and practices that consider diversity as one of the guiding values that universities promote and uphold in pursuance of their mission.
Participants also indicated that the absence of a clear policy on diversity management in higher education and the politicization of ethnicity are some of the factors for poor diversity management in higher education. In the words of a participant,If you look at the [higher education] proclamation and the senate legislation of universities, they barely allude to diversity. So, the diversity issue has not received the attention it deserves from the university leaders and the universities community as a whole. (DCSS1, 9 March 2021)
Although the ultimate goal of diversity management in higher education is to benefit from diversity, participants indicated that creating a positive campus climate for diversity, enhancing positive interactions between students from different ethnic groups, and ensuring peaceful co-existence should be the primary goal of diversity management in Ethiopian universities. University leaders and managers are well aware of the individual, institutional, and national benefits of diversity. However, practical measures taken to benefit from diversity in higher education seemed not to be significant. In relation to this, a participant said that,Senior government officials, political leaders, university leaders, and we too say diversity is a beauty. But this must not only be said but also practiced. However, this is not happening and that is why every year there is an ethnic conflict between students in different universities. … I understand that this [diversity] is not what universities alone address, but they have to try their level best. (DCA2, 9 March 2021)
Diversity in higher education is often regarded as a value based on the potential opportunities it provides. In reality, “diversity is not always a value” (Macedo, 2000, p. x) and benefiting from diversity requires good management. That is why diversity management has become a common practice in higher education (Bledsoe et al., 2010) and universities worldwide are adapting and implementing different diversity policies, strategies, and approaches. Although Ethiopian public universities have a diverse student body more than any other setting in the country, they do not have clear diversity management policies and strategies. There are a few activities that contribute to promoting diversity in higher education, but these activities are often underfunded and carried out on an ad hoc and voluntary basis.
University leaders and administrators often do not pay close attention to ethnic diversity until there is tension or conflict that involves different groups. Participants thought that the university could have lessened the impact of ethnic tension and conflict if it had proactive measures and made the necessary follow-up. Studies also indicated the importance of monitoring the campus climate for diversity and taking timely measures to avoid negative consequences (2007; Roksa et al., 2017).
Proactive diversity management strategy which includes creating a positive campus climate for diversity, curricular diversity (diversity-related courses and pedagogical approaches that promote diversity), and cocurricular diversity (diversity training, workshops, seminars, and campus wide events) should be prioritized because it helps to benefit from diversity by promoting diversity and avoiding diversity-related challenges and problems. However, participants indicated that most universities are engaged in reactive measures which only helps to lessen the negative impacts of diversity-related problems. In the words of a participant,There are almost no clear proactive measures that universities are using to deal with issues of diversity. So, challenges and problems are inevitable, and as a result, whether it is good or bad, they will take reactive measures to address diversity-related problems. This is often in consultation with government bodies. (DCHS2, 19 March 2021)
Some of the recent major reactive measures taken include (1) a tripartite agreement signed by parents or guardians, students, and government representatives to prevent the involvement of students in ethnic conflict and other major problems on campus, (2) discussions between federal and regional government representatives and students, (3) peacemaking sessions and advice by elderly people and religious and community leaders, (4) severe disciplinary measures (e.g., suspension and expulsion), and (5) involving the military and other security structures when the situation escalates and becomes difficult to manage with university capacity. The ministry also instructed all public universities not to accept any withdrawal request from students in order to oblige students to stay in class. Despite all these measures, universities and the wider Ethiopian society are still suffering from ethnic conflict. This clearly shows that reactive measures are not the only and preferred diversity management strategies.
Challenges
The study identified several factors that negatively affect ethnic diversity management in public universities. The major factors include absence of clear policy and strategy, lack of diversity programs and activities, lack of leadership commitment and skills, decreased ethnic diversity of faculty members, and socialization and politicization of ethnicity.
Absence of clear policy and strategy
Leaders are often guided by and do their job based on national and institutional policies, strategies, and plans. The inclusion of issues of diversity in such official documents will make diversity one of the issues to be addressed with clear focus, responsibility, and accountability. On the contrary, the absence of clear and feasible national and institutional policies and strategies that direct and support ethnic diversity management will leave diversity as a neglected and an act of goodwill which no one is accountable for. That is also what the participants indicated.Diversity issues are not a priority in our institutional plan. Therefore, there is no set goal and no expected performance and results. It is not even considered as one of the cross-cutting issues to be addressed. (DLH1, 17 March 2021)
It is not clear which unit of the university is mainly responsible for promoting and addressing diversity-related issues. This is not clearly indicated in the organizational structure and roles and responsibilities of different units of the university. So, no one is responsible and accountable for issues of diversity. Leaders too are not held accountable. When they are held accountable, it is not because they did not manage diversity properly, but because they did not control the problem as any other problem, or because of weak leadership, they have contributed to making the problem worse. (DCNCS2, 10 March 2021)
Studies also indicated that recognizing and including diversity in the mission of a university should be the first step in the process of managing diversity because the plans and activities of a university emerge from its mission (Hurtado, 2005, 2007; Lumadi, 2008). Moreover, a university needs to have aspirations to improve the campus climate for diversity, and this should be included and reflected in its policies, strategic plans, programs, and activities (Brown, 2004). Including issues of diversity in an institutional strategic plan guides a university to intentionally accomplish diversity (Milem et al., 2005).
Socialization and politicization of ethnicity
Since 1991 ethnicity has become the ideological basis of the government’s political organization and administration. As a result, Ethiopia has become a federal polity with eleven ethnic-based regional states and two chartered cities that constitute the federation. The ethnic federalism seems to have facilitated ethnic polarization and a politically motivated superior/inferior dichotomy, which in turn increased ethnocentric attitudes among some ethnic groups (Adamu, 2014). The ethnic polarization has also facilitated the ethnicization of Ethiopian politics, which generated increasing grievances among ethnic groups (Valfort, 2007).
Students who joined university in the last two decades were born or started formal education after the introduction of the ethnic-based federal system. These students have grown up listening to and exercising politicization of ethnicity and manipulating historical interethnic grievances to evoke resentment, fear, and hatred toward the “other.” Children are the product of the family, the society, and the education and political system. Participants indicated that family, communities, media, and education played a role in the negative socialization process. The family and the society did not tell and properly teach their children about tolerance, equality, and unity. Rather, they grew up listening to the victim-abuser story and narration. The education system did not enable them to develop critical, perspective taking and problem-solving skills, and skills required to learn and live with diverse others. Some of the ethnic issues broadcasted and posted in public and social media are divisive, undermine and decrease intergroup trust and relations, and increase intergroup prejudices (Adamu, 2014).
Participants noted that the ethnic-based political system reinforced ethno-national sentiments and segregation along ethnic lines. A study also indicated that the “ethnic-based federal and political system contributed to students’ misconceived ethnic and political orientation, which resulted in their ethnicizing various issues that affected the intergroup relations among students and also led to conflict” (Adamu, 2014, p.144). These issues make the diversity management process difficult when students come to university. In relation to this, a participant said,What we see in higher education is the result of many factors including the social upbringing and the education system. Students were grown up listening to ethnic hatred, resentment, and revenge. The education system is not helpful in promoting unity and correcting fabrication and falsification of history. …The higher education curriculum emphasizes on the importance of teaching history for freshman students, but as you may know, so far, there is no history because scholars were not able to agree on the content to be taught. (DCBE6, 18 March 2021)
This clearly indicates that the political discourse and narratives influence and impact the provision of academic knowledge and skills that significantly contributes to managing ethnic diversity in higher education. In general, studies indicated that the ethnic federalism in Ethiopia which takes ethnicity as the basis for its political and administration system (Abbink, 2011) has caused more problems than it was intended to solve (Maru, 2010), and one of the major problems is its negative impacts on managing ethnic diversity in higher education.
Ethnic diversity management can be made more effective by creating a better understanding of issues of diversity among the university community through training, discussion, and education. However, universities are reluctant to do so because of the political sensitivity of ethnic-related issues in Ethiopia.I think that better than any other government institutions, universities could have created forums where individuals, groups and organizations could present their ideas and listen to others perspectives regarding issues of diversity. However, they are not doing this out of fear of the political consequences of dealing with ethnic issues. (DSS7, 19 March 2021)
Studies indicated that in a society where ethnic differences are prevalent and inevitable, the issue of diversity also becomes one of the central educational and civic missions of higher education (Hurtado, 2007). “Universities [should] also serve as a forum for developing critical thinking and stimulating the exchange of ideas on sensitive socio-political issues which impact on peace and development” (Govender, 2020, par.2). Failure to do so will result in diversity becoming more a problem than an opportunity. It also puts a lot of pressure on diversity management.
The politicization of ethnicity went to the extent that regional states, the community, and students and teachers who are from the dominant ethnic group where the university is located perceive the university as “their university” and other students and teachers feel that they are learning and teaching in a university that belongs to “others.” This is despite the fact that all public universities are owned and funded by and accountable to the federal government. The ethnic politics and ethnic-affiliated leadership selection and appointment have crippled universities to stand and act like a federal university. The new selection and appointment process of senior leaders is more merit-based than ethnic and political affiliation. Any individual both within and outside the university can apply and compete. However, individuals from “other” ethnic groups are not interested to apply because of the politicization of ethnicity and the ethnic-based political and administrative system which may not make them successful applicants and leaders. As a result, most, if not all senior leaders, in all public universities are coming from the region where the university is located. In some cases, this even goes to the extent of having leaders from a specific location within the region. Participants indicated that students who study in such universities might have a low sense of belonging and perceive public universities as less inclusive and hospitable, and this is a challenge for effective diversity management.
Lack of diversity programs and activities
One of the major strategies in promoting and managing diversity in higher education is providing various opportunities for students that help them acquire knowledge about and experience diversity (Umbach & Kuh, 2006). As indicated above, there is no unit or center for diversity in universities and it is no one’s job per se. Participants indicated that universities do not have programs and activities which are purposefully developed to promote and create awareness of diversity.I have not seen a single initiative developed by any university to promote diversity. In our university and also in other universities, we used to celebrate the nations and nationalities and people’s day. I think the motive for celebrating this event is more of political than cultural and promoting diversity. That is why in the last three years universities were not celebrating this even as usual because of the change in government and politics. (DSL3, 18 March 2021)
Trotting out diversity in higher education, “only during special occasions for public display” (Aguirre & Martinez, 2002, p. 55) can be considered a political strategy. Nevertheless, the political implication of celebrating the nations and nationalities and people’s day is one thing and its contribution to promoting diversity is another thing. Aside from its arguably political implications, the celebration of the nations and nationalities and people’s day at universities seems somehow to contribute to promoting diversity on campus (Adamu, 2014).
Participants also mentioned some activities which have been carried out on an ad hoc basis and contributed to promoting diversity. They also mentioned the “civic and ethical education” course which is provided to all university students. Participants argued that the course could have developed students’ knowledge about diversity but the content of the course is packed more with political and legal than multicultural and diversity issues. Though not directly, as some participants indicated, this course still contributes to promoting diversity through discussions on democracy, equality, and justice which have much to do with issues of diversity. Studies also showed that diversity-related courses have positive effects on students’ complex thinking skills (Hurtado, 2005), cognitive development (Bowman, 2009), and they are effective in reducing biases and developing skills to work with diverse others (Banks, 2001).
Lack of commitment and leadership skills
Leadership is very important for achieving the mission and vision of any organization, and universities are no exception. In the context of higher education institutions, there are several internal and external leadership challenges. One of the greatest challenges that confront leadership in Ethiopian higher education is ethnic diversity. Therefore, leaders need to have good leadership and management skills to address challenges associated with ethnic diversity on campus. Participants noted that most university leaders do not have the skills to effectively manage ethnic diversity.University-level leaders are elected and appointed from among us. Unless they have the knowledge and skills acquired through reading or experience, I don’t think they have the skills to manage diversity properly. That is why they are not able to come up with strategies to prevent ethnic conflict and prevent ethnic tension and conflict from escalating. …This also works for us as well. In this regard, we have no better leadership skills than they do. (DCNCS2, 10 March 2021)
Even though there are some who may have the skills, they are not determined to do so. They often follow instructions or alternative solutions from the ministry or other government bodies. This is mainly because they are not responsible for the consequences of any action proposed by the government. But they may be held accountable if the measures they took by themselves disappoint some ethnic groups or escalate the situation. Participants said that leaders lack courage and find it difficult to deal with issues of diversity. They often refrain from taking measures to address ethnic tensions and conflicts among different student ethnic groups. This is mainly because ethnic issues are complex and have many political implications.
One of the ways to manage diversity is by creating awareness among students and the university community about the importance of diversity and the need for mutual respect and tolerance. The other strategy is promoting diversity through creating forums for discussion and providing training on diversity which also helps to correct misconceptions, prejudices, and stereotypes. Effective implementation of these diversity management strategies requires a strong commitment of relevant stakeholders, but more than anyone else, leaders and managers need to believe in and support diversity programs and initiatives. This is mainly because no initiative intended to promote diversity can be achieved without the strong support and commitment of the higher officials of an institution (Norris, 2000). Participants indicated that in practice, Ethiopian university leaders lack the courage and commitment to promote and raise awareness about diversity, let alone address serious challenges of ethnic tensions and conflicts. A previous study in Ethiopia also indicated that higher education leaders fail to promote diversity mainly because of a lack of diversity management skills and fear of politicization of diversity-related issues (Adamu, 2013).
Decreased ethnic diversity of faculty members
Participants indicated that increasing and maintaining faculty members’ ethnic diversity is an important strategy for managing student ethnic diversity. A study also indicated that students are happy to see faculty members from their ethnic group and this increases students’ sense of belonging and overall satisfaction with their university, which in turn contributes to managing student diversity (Adamu, 2014).
Participants noted that the recruitment process ensures non-discriminatory employment opportunities and working environment. However, research indicates that there is a tendency for teachers from other regions to leave the university and for teachers from the region where the university is geographically located to come to the university (Adamu & Zellelew, 2007). The university leadership has done nothing to address this issue, and thus, faculty members’ ethnic-related transfer over the past 3 years has been intensified. The increase in numbers of transfers is due to the serious ethnic tensions and conflicts in universities and different parts of the country that followed the change of government leadership in 2018. This is because some faculty members did not feel comfortable and secure to working and living away from family and relatives in such a situation. Some faculty members who support ethnic politics and do not want to work at a university which is located in another ethnic groups’ region and those who want to take advantage of this opportunity to work in a university of their best choice also submitted a transfer request and they were successful. This significantly decreased the diversity of faculty members from different ethnic backgrounds and potentially affects creating an inclusive and welcoming campus climate for diversity. Participants noted that instead of developing strategies and taking measures to retain diverse faculty members, universities and the ministry have facilitated faculty members’ ethnic-related transfer requests. The following responses from three deans indicate the ethnic-related transfer requests in different universities:This should not have happened, but almost all the universities have been releasing and accepting teachers in an unusual and unprecedented way. The ministry has ignored this issue, and at times, it even facilitated the transfer. This is one of the factors for students less interested in studying in universities where there are no teachers from their ethnic group, and I think this seriously affects student ethnic diversity management. (DCA1, 9 March 2021)
In my university, there are still about sixty transfer requests from teachers and most of them want to go to a university which is geographically located in a region where their ethnic group is majority. Universities have contributed to this because they have made the transfer easy and even facilitated this within a short period. They don’t behave like a federal university. Some universities even accept teachers without any communication with the university which the teachers want to leave. (DCNCS8, 10 March 2021)
We do not want to let go teachers from different ethnic backgrounds because we have seen their importance in addressing ethnic-related problems. A discussion between university officials and students to address ethnic-related problems is often successful when university leaders involve teachers from students’ ethnic groups. …I am not sure for how long we will keep them, because the new trend and the political wave are against us. How institutions that are supposed to address ethnic-related problems are involved in activities that escalate ethnic problems. In my view, this is a black historical scar for higher education. (DHS1, 20 March 2021)
The above excerpts clearly indicated the contribution of diverse faculty members to managing student diversity in universities in Ethiopia. Research also showed that diverse faculty members contribute to limiting discrimination claims (Collins & Kritsonis, 2006) which are important for managing diversity. The excerpts also indicated that although this has been threatened by increased transfer which decreased the diversity of faculty members from different ethnic backgrounds, there is no measure taken to attract and retain diverse faculty members.
Way forward
So far, presumably, universities in Ethiopia have not properly managed the ethnic diversity in their own context, let alone played a vital role in addressing the diversity-related problems of the larger society. They seem to have ignored issues of diversity despite the fact they faced several ethnic-related diversity challenges (Adamu, 2014). It is important to note that diversity is not something that will go away through time or ignorance (Levine, 1991).” So, universities should be up to the challenges of diversity.
The Ministry of Education and higher education institutions need to formulate diversity-related national and institutional policies and strategies which help to strategically manage ethnic diversity. Participants noted that national and institutional ethnic diversity policies and strategies should be discussed, accepted, and embraced by relevant stakeholders. They indicated that one of the reasons for poor implementation of policies and strategies in higher education is lack of genuine and purposeful involvement of stakeholders in the policy and strategy development process. They also reminded the government not to adopt any diversity management strategy without contextualizing it. Diversity is a phenomenon that is culturally, socially, and historically formed and reformed (Metcalfe & Woodhams, 2008), and thus, effectively addressing issues of diversity in a given context should take into consideration the specific sociocultural, political, and institutional contexts. The outcomes of ethnic diversity management greatly depend more on the way diversity policies, strategies, and initiatives are implemented than how good the policies, strategies, and initiatives look. Therefore, there should be professional and political commitment of the higher education community and government officials to ensure effective implementation of national and institutional diversity management policies and strategies at different levels.
Universities also need to well understand that effective implementation of national and institutional diversity management policies and strategies require understanding their campus climate for ethnic diversity. This is mainly because the campus climate for diversity is context specific. Hence, universities need to do what needs to be done to understand their campus climate for diversity which may require focused research and discussion with relevant stakeholders (Adamu, 2014).
Managing ethnic diversity in Ethiopia is more of a national issue that requires research/evidence-based political interventions. Therefore, the role of HEIs in this regard is twofold. First, they need to contribute to the national level diversity management endeavors through research, dialogue, and producing active citizens. Second, they need to manage the ethnic diversity in their campus.
Studies showed that the contribution of education in managing ethnic diversity is of paramount importance. Hence, the emphasis given to issues of diversity in the education system needs to be revisited. In the Ethiopian education system, schools are not teaching students about morale, self, and culture. Civic and ethical education courses, which are also provided at primary and secondary schools, are supposed to enhance students’ knowledge about morale and promote unity with diversity. However, the Ethiopian education development roadmap indicates that the courses have not been helpful in achieving this objective in schools (Ministry of Education, 2018).
In recent years, universities put emphasis on graduate employability. Yet, what is needed first is good citizens with a great sense of self, moral integrity, responsible citizenship, etc. Therefore, more emphasis should be given to the process of building responsible and active citizenship at all levels of the education system. One of the participants also emphasized on the importance of including diversity-related courses, and he said,Incorporating diversity-related courses into our education system is a must. If it [diversity-related courses are] is not included in the education system of a country like Ethiopia where ethnic diversity is one of the country’s main features; where ethnicity is the basis for its political and administrative system; and which suffers from ethnic tensions and conflicts, in which country’s education system it should be included then? (DCBE5, 9 March 2021)
Participants commented that it is difficult to manage ethnic diversity in higher education because of different reasons including political sensitivity of ethnic issues. Study also indicated that “managing ethnic diversity stands as a challenging political issue for multi-ethnic societies” (Bélair, 2016, p.1). Yet, universities have to address this challenge in their capacity as socially responsible institutions. They can, among others, raise awareness of diversity, develop diversity management skills of higher education leaders and managers, and enhance and maintain faculty members diversity. In the words of a participant,It is true that there are several challenges in managing diversity in higher education institutions. But we should not externalize everything because the challenges are both internal and external, and we can at least address the internal challenges. In relation to this, the first thing we need to do is depoliticizing ethnicity. I know this is not as easy as I am saying, but it is not impossible. For example, we can try through creating awareness about diversity, developing our management skills and closely working with the government. (DHS2, 18 March 2021)
Implementing diversity management initiatives and activities are difficult (Treven & Treven, 2007), but it is worth investing because the benefit of managing diversity goes beyond higher education and contributes to the political and social stability in the country. This is the point which helps the attention and interest of the government. As indicated in the above excerpt, it is very important to closely work with the government on issues that require its direction and intervention. As this study revealed, the main sources of most of the challenges and problems associated with student ethnic diversity are external to the universities. This also implies the need for close collaboration with government bodies to effectively manage ethnic diversity in higher education. Participants clearly noted that the core solution should be made at national level. They believe that if the government is able to address issues related to ethnic-federalism and politicization of ethnicity at federal, regional, and group levels, universities will be able to easily manage the ethnic diversity in their campuses.
Conclusion
The study revealed that the politics of ethnic diversity in Ethiopia is a complex issue, and this has made managing student ethnic diversity in higher education a hard and difficult process. Although student ethnic diversity is one of the major challenges faced by public universities, the practice of managing ethnic diversity was found to be low. It is an ad hoc exercise not supported by policies and strategies and no accountability and clearly stated responsible unit. This study found major internal factors (decreased ethnic diversity of faculty members, lack of commitment and leadership skills, absence of institutional diversity policy and strategy and lack of diversity programs and activities) and external factors (absence of national diversity policy and strategy, socialization, and politicization of ethnicity) that affect effective implementation of diversity management. Yet, politicization of ethnicity is found to be the mother of all challenges associated with student ethnic diversity management in higher education. Therefore, although effective student ethnic diversity management requires the active participation and commitment of universities, communities, and the government, in the Ethiopian context, addressing issues related to the ethnic political and administrative system is the highest priority. This is a long-term plan which requires political commitment and public consensus. Hence, universities need not wait for this to happen. They should take different measures that help to address ethnic diversity management. In the short-term, instead of establishing units that address only issues related to gender and disability, the ministry of education needs to promote and support universities to establish centers for diversity and inclusion which primarily works to address diversity-related issues including ethnic, gender, and disability, among others. This center should be active and ensure the promotion of diversity in curricular and non-curricular activities, and this enforces universities to pay attention to diversity and its management. Moreover, the ministry and universities need also to take the strongest possible measures against students directly or indirectly involved in intensifying ethnic tensions and conflicts. Finally, this study did not look into the ethnic diversity of leaders which has its own potential influence in managing and/or intensifying diversity-related problems. Therefore, it is important for further research on diversity management to consider this aspect as well.
Acknowledgements
The author would like to thank Addis Ababa University for its generous support of this research.
Funding
This research is generously funded by Addis Ababa University, Ethiopia.
Declarations
Conflict of interest
The author declares no competing interests.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Abbink J Ethnic-based federalism and ethnicity in Ethiopia: Reassessing the experiment after 20 years Journal of Eastern African Studies 2011 5 4 596 618 10.1080/17531055.2011.642516
Adamu AY Pre-service teachers’ awareness about cultural diversity in educational settings and their readiness to teach culturally diverse students: The case of Bahir Dar University Ethiopian Journal of Education 2007 27 2 91 115
Adamu AY Intergroup relations among ethnically diverse university students in Ethiopia Journal of Education and Research 2013 3 2 77 95 10.3126/jer.v3i2.8399
Adamu AY Ethnic and religious diversity in higher education in Ethiopia: The case of Bahir Dar University 2014 University of Tampere Press
Adamu AY Effects of government policies, strategies, and the political system on intergroup relations among university students in Ethiopia The International Journal of Community Diversity 2015 13 4 1 10 10.18848/2327-0004/CGP/v13i04/39952
Adamu AY Zellelew TB Higher education institutions as pavilions of diversity: Opportunities and challenges: The case of Bahir Dar University Ethiopian Journal of Higher Education 2007 4 1 49 68
Adamu, A. Y. & Balsvik, R. R. (2018). Students’ participation in and contribution to political and social change in Ethiopia. In E. Oinas, H. Onodera, & L. Suupaa (Eds.), What politics? In youth and political engagement in Africa (pp. 265–284). 10.1163/9789004356368_017
Adamu, A. Y. (2019). Ethnic violence challenging Ethiopian Universities. Inside Higher ED. https://www.insidehighered.com/blogs/world-view/ethnic-violence-challenging-ethiopian-universities
Aguirre A Martinez R Leadership practices and diversity in higher education: Transitional and transformational frameworks The Journal of Leadership Studies 2002 8 3 53 62 10.1177/107179190200800305
Amnesty International. (2020). Ethiopia: Parents fear for missing Amhara students as universities close over Covid-19. https://www.amnesty.org/en/latest/news/2020/03/ethiopia-parents-fear-for-missing-amhara-students-as-universities-close-over-covid19/
Antonio AL Diversity and the influence of friendship groups in college The Review of Higher Education 2001 25 1 63 89 10.1353/rhe.2001.0013
Ashine, Y. (2019). Universities as contested terrain: making sense of violent conflicts in Ethiopian universities. Working Paper. Center for Advancement of Rights and Democracy.
Balsvik RR Haile Sellassie’s students: The intellectual and social background to revolution, 1952–1977 1985 African Studies Center, Michigan State University
Balsvik, R. R. (2009). “Addis Ababa University in the Shadow of the Derg, 1974–1991.” In, edited by S. Ege, H. Aspen, B. Teferra & S. Bekele (Eds.), Proceedings of the 16th International Conference of Ethiopian Studies: Vol. 3 (pp. 259–272). Trondheim: Norwegian University of Science and Technology.
Banks JA Cultural diversity and education: Foundations, curriculum, and teaching 2001 4 Allyn & Bacon
Bélair J Ethnic federalism and conflicts in Ethiopia Canadian Journal of African Studies 2016 50 2 295 301 10.1080/00083968.2015.1124580
Bigler RS Liben LS Developmental Intergroup Theory Current Directions in Psychological Science 2007 16 3 162 166 10.1111/j.1467-8721.2007.00496.x
Bledsoe MT Oatsvall RJ Condon D Diversity management: Seeking validation Journal of Diversity Management 2010 5 1 23 30 10.19030/jdm.v5i1.802
Bowman NA College diversity courses and cognitive development among students from privileged and marginalized groups Journal of Diversity in Higher Education 2009 2 3 182 194 10.1037/a0016639
Brown LI Diversity: The challenge for higher education Race Ethnicity and Education 2004 7 1 21 34 10.1080/1361332042000187289
Collins CJ Kritsonis WA National viewpoint: The importance of hiring a diverse faculty Doctoral Forum National Journal for Publishing and Mentoring Doctoral Student Research 2006 3 1 1 7
Cox T Creating the multicultural organization: A strategy for capturing the power of diversity 2001 Jossey-Bass
FDRE. (2019). Higher education proclamation (No. 1152/2019). Federal Negarit Gazeta, 78, 11445–11521. Addis Ababa: Author.
Foster C Harris L Easy to say, difficult to do: Diversity management in retail Human Resource Management Journal 2005 15 3 4 17 10.1111/j.1748-8583.2005.tb00150.x
Gottfredson NC Panter AT Daye CE Allen WA Wightman LF Deo ME Does diversity at undergraduate institutions influence student outcomes? Journal of Diversity in Higher Education 2008 1 2 80 94 10.1037/1938-8926.1.2.80
Govender, K. (2020). Panel discussion on the role of academia in the prevention, management, and transformation of conflicts. Retrieved from https://www.accord.org.za/news/accord-hosts-panel-discussion-on-the-role-of-academia-in-the-prevention-management-and-transformation-of-conflicts/
Gurin P Day E Gurin G Hurtado S Diversity and higher education: Theory and impact on educational outcomes Harvard Educational Review 2002 72 3 330 366 10.17763/haer.72.3.01151786u134n051
Hurtado S The next generation of diversity and intergroup relations research Journal of Social Issues 2005 61 3 595 610 10.1111/j.1540-4560.2005.00422.x
Hurtado S Linking diversity with the educational and civic missions of higher education The Review of Higher Education 2007 30 2 185 196 10.1353/rhe.2006.0070
Johnson DW Johnson RT Leicester M Modgil C Cooperative learning, values, and cultural plural classrooms Classroom Issues: Practice, Pedagogy and curriculum 2000 Falmer Press 15 28
Levine A The meaning of diversity Change 1991 23 5 4 5 10.1080/00091383.1991.9939872
Liu G Affirmative action in higher education: The diversity rationale and the compelling interest test Harvard Civil Rights-Civil Liberties Law Review 1998 33 2 381 442
Lumadi MW Managing diversity at higher education and training institutions: A daunting task Journal of Diversity Management 2008 3 4 1 9 10.19030/jdm.v3i4.4996
Macedo S Diversity and distrust: Civic education in a multicultural democracy 2000 Harvard University Press
Maru MT Federalism and conflicts in Ethiopia Conflict Trends 2010 1 36 45
Metcalfe BD Woodhams C Critical perspectives in diversity and equality management Gender in Management: An International Journal 2008 23 6 377 381 10.1108/17542410810897508
Milem JF Chang M Witt D Jones J Hakuta K The educational benefits of diversity: Evidence from multiple sectors Compelling interest: Examining the evidence on racial dynamics in higher education 2003 Stanford University Press 126 169
Milem JF Chang MJ Antonio AL Making diversity work on campus: A research-based perspective 2005 Association of American Colleges and Universities
Milkias, P. (2011). Ethiopia (Africa in Focus). Santa Barbara, CA: ABC-CLIO.
Ministry of Education. (2018). Ethiopian education development road map. Addis Ababa: Author.
Ministry of Science and Higher Education. (2020). Higher education policy and strategy. Addis Ababa: Author.
Norris B Transformation, diversity and organizational change within institutions of higher education South African Journal of Education 2001 21 4 219 222
Pant JJ Vijaya V Challenges in diversity management: A case study of MediHealth systems South Asian Journal of Management 2015 22 1 159 186
Pascarella ET Edison M Nora A Hagedorn LS Terenzini PT Influences on students' openness to diversity and challenge in the first year of college Journal of Higher Education 1996 67 2 174 195 10.1080/00221546.1996.11780255
Pettigrew TF Intergroup contact theory Annual Review of Psychology 1998 49 65 85 10.1146/annurev.psych.49.1.65
Roksa J Kilgo CN Trolian TL Pascarella ET Blaich C Wise KS Engaging with diversity: How positive and negative diversity interactions influence students’ cognitive outcomes The Journal of Higher Education 2017 88 3 297 322 10.1080/00221546.2016.1271690
Shaw, E. J. (2005). Researching the educational benefits of diversity. College Board Research Report No. 2005–4. New York: College Board.
Treven S Treven U Training in diversity management Journal of Diversity Management 2007 2 1 29 36 10.19030/jdm.v2i1.5003
Umbach PD Kuh GD Student experiences with diversity at Liberal Arts Colleges: Another claim for distinctiveness The Journal of Higher Education 2006 77 1 169 192 10.1080/00221546.2006.11778923
Valfort, MA. (2007). Containing ethnic conflicts through ethical voting? Evidence from Ethiopia. Working Paper 35. Households in Conflict Network. Retrieved from http://www.hicn.org/wordpress/wp-content/uploads/2012/06/wp35.pdf
Woldegiyorgis, A. A, & Adamu, A. Y. (2022). Ethiopian Higher Education and the COVID-19 Pandemic. In F. Netswera, A. A. Woldegiyorgis, & T. Hakuta (Eds.). Higher education and the COVID-19 Pandemic: Cross-national perspectives on the challenges and management of higher education in crisis times (pp. 24–41). Brill. 10.1163/9789004520554_003
| 36465487 | PMC9707089 | NO-CC CODE | 2022-12-01 23:20:20 | no | High Educ (Dordr). 2022 Nov 29;:1-16 | utf-8 | High Educ (Dordr) | 2,022 | 10.1007/s10734-022-00974-x | oa_other |
==== Front
J Plant Pathol
Journal of Plant Pathology
2239-7264
Springer International Publishing Cham
1234
10.1007/s42161-022-01234-8
Abstract
Abstracts of presentations at the XXVII Congress of the Italian Phytopathological Society (SIPaV)
September 21-23, 2022. University of Palermo, Palermo, Italy
29 11 2022
174
© The Author(s) under exclusive licence to Società Italiana di Patologia Vegetale (S.I.Pa.V.) 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
==== Body
pmcThe 2022 Journal of Plant Pathology Keynote Address
Exotic Heterobasidion Root Disease in Italy as an unexpected legacy of war
M. Garbelotto 1 , P. Gonthier 2
1 University of California at Berkeley, USA; 2 Dipartimento di Scienze Agrarie, Forestali e Alimentari (DISAFA), Università degli Studi di Torino, Largo Paolo Braccini 2, I-10095 Grugliasco (TO), Italy. E-mail: [email protected]
The presentation will first review those studies that have reconstructed the early establishment and introduction in the Lazio Region of the conifer root pathogen Heterobasidion irregulare by U.S. troops in 1944, during the liberation of Italy in World War II. The main emphasis of the talk, though, will be on research findings that help us predict the future trajectory of this exotic disease in Italy and in the entire European continent. Although multiple studies have shown that H. irregulare has a greater sporulation potential, a faster growth rate in wood and a broader host range than the native Eurasian H. annosum, precise predictions of the future impacts of the exotic disease have been hampered both by the high fragmentation of conifer stands and by the low incidence of the congener H. annosum in the Lazio Region of Italy. Not unexpectedly, population genetics approaches have allowed us to predict with confidence that, as the fragmentation of conifer stands decreases, the spread rate of H. irregulare will increase exponentially. In addition and unexpectedly, we have experimentally shown in the laboratory that the presence of the native H. annosum will accelerate and not slow down the spread of H. irregulare. Both results paint a troublesome scenario for Central and Northern European regions (including coastal Tuscany, Liguria and the Alpine region) characterized by large contiguous conifer woodlands with high incidence of H. annosum. Are these predicted scenarios likely to be true? In order to answer this question, we have studied in depth the interaction between the two pathogens in the Lazio and made several discoveries. Field evidence now proves not only that H. irregulare is truly invasive with a spread rate of about 150 ha per year in a contiguous forest, but also that it is replacing H. annosum. Unexpectedly, we also have discovered that part of the genome of H. annosum is being replaced by genes from H. irregulare, thanks to gene introgression mediated by interspecific hybridization. A recent study has further shown that H. annosum individuals containing H. irregulare genes increase their transmission traits to levels comparable to those of H. irregulare. We conclude that Europe is facing two invasions: one by H. irregulare individuals and one by H. annosum individuals modified by the acquisition of adaptive H. irregulare genes. Both invasions are associated with disease that will spread at a faster rate than that of the disease caused by native H. annosum populations, thus increasing the estimated 800 million Euros per year of damage already caused by native Heterobasidion Root Diseases.
Tree endotherapy with Trichoderma spp. against the agent of chestnut nut rot Gnomoniopsis castaneae
C. Aglietti, A. Benigno, S. Moricca
Department of Agriculture, Food, Environment and Forestry (DAGRI), University of Florence, Piazzale delle Cascine 18, 50144 Florence, Italy. E-mail: [email protected]
An emerging fungal disease caused by the ascomycete fungus Gnomoniopsis castaneae (syn. Gnomoniopsis smitholgivyi) is responsible for nut rot of sweet chestnut. The disease is impacting chestnut production in several Italian regions, with severe economic losses in a marginal cultivation with low income such as the fruit chestnut groves. Since this fungus is endowed with an endophytic lifestyle, surviving at a latent stage in chestnut tissues, it is hard to find effective solutions for its control and management. The considerable size of some chestnut trees and the risks for consumers and the environment arise by chemical pesticides application in chestnut groves (e.g., traditional foliar applications), have prompted the search for alternatives to traditional fungicides as well as to the conventional delivery methods of plant protection products. The exploitation of microbial natural enemies of plant pathogens as biopesticides in plant disease control is an important reality to which the European and Italian institutions strongly aim for a more sustainable and environmentally friendly agriculture. The use of biocontrol agents (BCA) for controlling forest tree diseases is still scarce. The aim of this work was to test the development and optimization of a biological control method against G. castaneae based on trunk injections with Trichoderma species. Trichoderma species and strains were selected in laboratory pre-trials, testing their effectiveness against G. castaneae in an in-vitro dual culture system then injected in fruit chestnut trees stem. Results showed a reduction of G. castaneae infection in treated stands compared to controls (untreated chestnuts), proving endotherapic treatments with BCA to be a promising control strategy against G. castaneae infection.
Pinus radiata – Fusarium circinatum – Phytophthora spp., a model system of a complex host plant – pathogens interaction
F. Aloi 1 , C. Zamora-Ballesteros 2,3 , J. Martín-García 2,3 , J.J. Diez-Casero 2,3 , R. Parlascino 1 , F. La Spada 1 , A. Pane 1 , S.O. Cacciola 1
1 Dipartimento di Agricoltura, Alimentazione e Ambiente, Università degli Studi di Catania, via S. Sofia 100, 95123 Catania, Italy; 2 Department of Vegetal Production and Forest Resources, University of Valladolid, Av. Madrid 44, 34004 Palencia, Spain; 3 Sustainable Forest Management Research Institute, University of Valladolid—INIA, 28040 Madrid, Spain. E-mail: [email protected]
This study investigated the effects of co-infections in Monterrey pine (Pinus radiata) seedlings determined by diverse aggressive pathogens, the fungus Fusarium circinatum, causing Pine Pitch Canker (PPC) disease, and the oomycetes Phytophthora × cambivora and P. parvispora, both causative agents of crown and root rots. Pinus radiata seedlings were wound-inoculated with each single pathogen and with either combinations F. circinatum/P. × cambivora and F. circinatum/P. parvispora. The effects of the co-infections were investigated at 4- and 11-days post inoculation (dpi), in terms of severity of symptoms and modulation of the transcriptomic profile of the pyruvate decarboxylase-(PDC)-encoding gene and three genes encoding pathogenesis- related proteins (PR3, PR5, and PAL) in pine seedlings. Results from plants inoculated singularly with pathogens, highlighted that F. circinatum markedly induced the up-regulation of all four genes mainly at the late stages of infection. Between the two Phytophthora species, only P. cambivora stimulated a significant up-regulations. In seedlings co-inoculated with F. circinatum and P. × cambivora or P. parvispora none of analyzed genes showed a significant up-regulation at 4 dpi. In contrast, at 11 dpi, significant up-regulation was observed for PR5 in the combination F. circinatum/P. × cambivora and PDC in the combination F. circinatum/P. parvispora. In conclusion, two hypotheses were formulated: i. the competition between pathogens could have delayed the infective process by F. circinatum and the plant defense response; ii. co-infection might have repressed the expression of defense-related genes, thus exacerbating the severity of the disease.
Biocontrol of Botrytis cinerea as influenced by grapevine growth stages and environmental conditions
V. Altieri, G. Fedele, V. Rossi
DiProVeS, Università Cattolica del Sacro Cuore, Piacenza, Italy. E-mail: [email protected]
The effective integration of BCAs into a Botrytis bunch rot (BBR) management program should not only include understanding factors, such as berry growth stages and environmental conditions (i.e., temperature, T, and relative humidity, RH), for the pathogen growth and infection, but also for the colonization and the efficacy of the BCA. In this study, four commercial BCAs were evaluated. Turbidimetric assays were conducted to assess the BCA growth at different berry growth stages by inoculating BCA into media mimicking the chemical composition of berries. In a second study, the BCAs colony-forming units (CFUs) were assessed at the ripe berries stage by inoculating the artificial “ripe berries” medium and incubating it for 1 to 13 days at different T/RH conditions. In a third experiment, each BCA was applied to ripe berries and then incubated under different T/RH conditions. After 1 to 13 days, the berries were inoculated with Botrytis cinerea and incubated for 7 days, at which time BBR was assessed. The response of BCA growth to grapevine growth stages and to T/RH conditions, as well as the response of BBR control to T/RH conditions, differed among BCAs. For example, Metschnikowia fructicola grew better on fully ripe berries than in earlier ripening stages; Bacillus amyloliquefaciens showed higher temperature requirements than the other BCAs. The results obtained in this study would assist farmers in selecting the appropriate BCA for application based on the growth stages of grapes and prevailing weather conditions at the time of treatment and later.
Evaluation of the effects of Epicoccum nigrum on the olive fungal pathogens Colletotrichum acutatum and Verticillium dahliae by 1H NMR based metabolic profiling
F. Angilè 1† , M. Rizzo 1† , M. Riolo 1 , L. Riccioni 2 , S.O. Cacciola 3 , F.P. Fanizzi 4 , E. Santilli 1
1 Research Centre for Olive, Fruit and Citrus Crops, Contrada Li Rocchi Vermicelli 83, 87036 Rende, Italy; 2 Research Centre for Plant Protection and Certification, Via C.G. Bertero 22, 00156 Rome, Italy; 3 Department of Agriculture, Food and Environment, University of Catania, 95123 Catania, Italy; 4 Department of Biological and Environmental Sciences and Technologies, University of Salento, Prov.le Lecce-Monteroni, 73100 Lecce, Italy. † These authors have equally contributed to the study. E-mail: [email protected]
Epicoccum is a genus of ubiquitous ascomycetes. Epicoccum species can be found in the air, soil, plants and water. Several species are plant pathogens while some are potential biological control agents (BCAs). Among the latter, Epicoccum nigrum is the most promising and has been shown to reduce the incidence and severity of a wide range of plant diseases. Its antagonistic activity against fungal plant pathogens has been reported in the literature and seems to be due to the production of secondary metabolites with antifungal and antibacterial activity. The aim of this study was to test the ability of E. nigrum to inhibit in vitro the mycelium growth of the fungal pathogens Colletotrichum acutatum and Verticillium dahliae, responsible for anthracnose and Verticillium wilt of olive, respectively. Dual culture assays (E. nigrum + C. acutatum and E. nigrum + V. dahliae) showed that E. nigrum limits the mycelia growth of both C. acutatum and V. dahliae by producing inhibition zones. The metabolomic profiles of mycelium extracts obtained from mono and dual cultures were analyzed by 1H NMR Spectroscopy and Multivariate Statistical Analysis (MVA). For all microorganisms studied, aqueous and lipid extracts were analyzed and showed different metabolic patterns between mono and dual cultures. The first results show differences in relative content of sugars, aminoacids, organic acids and diunsaturated fatty acids. Although further investigations are needed, these preliminary results could represent a starting point for safeguarding the health of olive trees.
This study was funded by SALVAOLIVI Project D.M. MiPAAF n. 33,437 21/12/2017.
Influence of climate change on fungal pathogens of forest plants in Italy
N. Anselmi
University of Tuscia, Department of Biotechnology, Agrofood and Forest Innovation (DIBAF), Via San Camillo De Lellis, snc, Viterbo, Italy. E-mail:[email protected]
Climate change, with rising temperature and recurring and persistent droughts, is causing significant changes in Italy regarding the adaptation, spread and aggressiveness of forest plant pathogens. In some cases there is a reduction in attacks of species whose spread of ‘spores’ and the relative infectious process are favored by rains (e.g. Marssonina, Gloeosporium, Cylindrosporium, Phyllosticta) or disfavoured by too mild springs (e.g. Discella, Discosporium, Pollaccia). Much more often, however, climate changes increase phytopathological events, favouring: (1) the adaptation of newly introduced pathogens; (2) the attacks of species characterized by xero-conidia (e.g. agents of powdery mildew); (3) the resurgence of invasive pathogens, such as Botryosphaeria and Cytospora, and especially Phytophthora. The latter, with an increasing number of specie adapts to ever more numerous hosts and ever more wider ranges (e.g. mountain areas); (4) the suffering of plants with the development of "weakness parasites", which often leads to the notorious complex disease of “forest decline". Among the pathogens that become more aggressive on weakened plants, there is an important expansion of: (a) agents of bark necrosis that stress induces to pass from an asymptomatic endophytic phase to an active pathogenic state (eg Biscogniauxia, Coryneum, Fusarium, Phoma, Phomopsis, Sphaeropsis); (b) agents of root and butt rots (e.g. Armillaria, Ganoderma, Heterobasidion, Rhyzina, Ustulina); (c) wood decay species (e.g. Daedalea, Fistulina, Fomitopsis, Inonotus, Laiteporus, Phellinus). The aforementioned suffering of plants is also often the cause of reductions or involutions of mycorrhizae, often antagonists of phytopathogens, with further depression of the plant defences and vitality.
CRISPR-Cas9 technology for dissecting Pseudopyrenochaeta lycopersici -tomato interaction
M. Aragona 1 , J. Milc 2 , A. Infantino 1 , A. Cuccurullo 3 , A. Nicolia 3
1 Consiglio per la ricerca in agricoltura e l'analisi dell'economia agraria (CREA), Centro di Ricerca Difesa e Certificazione (CREA-DC), Via C.G. Bertero 22, 00156 Roma, Italy; 2 Università degli Studi di Modena e Reggio Emilia, Dipartimento di Scienze della Vita, Via Kennedy 17, Centro Interdipartimentale BIOGEST-SITEIA, Piazzale Europa 1A, 42124 Reggio Emilia, Italy; 3 Consiglio per la ricerca in agricoltura e l'analisi dell'economia agraria (CREA), Centro di ricerca orticoltura e florovivaismo, Via dei Cavalleggeri 25, 84098 Pontecagnano (SA), Italy. E-mail: [email protected]
The soil-borne fungus Pseudopyrenochaeta lycopersici, formerly Pyrenochaeta lycopersici, is the causal agent of corky root rot of tomato. The genome has been sequenced by Illumina and PacBio technologies and the gapless genome assembly of 62.7 Mb, obtained by PacBio technology, showed a large fraction (30% of the total bases) of repetitive sequences, including transposable elements. The most effective tool for controlling the disease is breeding for resistance, however only little is known on the molecular bases of tomato-P. lycopersici interaction. To investigate the underlying genetic bases, a comparative RNA Seq-based transcriptomic analysis was conducted on two tomato cultivars, Mogeor and Moneymaker, resistant and susceptible, respectively, to the pathogen. Overall, transcriptional profilings suggested that susceptibility and resistance share overlapping signalling pathways and responses. In detail, we selected a gene coding for a transcription factor which resulted expressed at a very low level in Mogeor respect to Moneymaker (in control conditions), representing a good candidate as a possible susceptibility gene. Attempts to disrupt this gene via CRISPR-Cas9 technology are in progress, with the aim to analyse its role in resistance to P. lycopersici.
This work was supported by the Programme "Biotecnologie sostenibili per l'agricoltura italiana" BIOTECH, project “Plant pathogen studies for screening of disease resistance” PATHORES, funded by Mipaaf.
A two-year survey on the occurrence of the seedborne barley foliar pathogens Pyrenophora teres and Ramularia collo-cygni in Italy
E. Balducci 1 , L. Roehrig 2 , F. Tini 1 , G. Beccari 1 , N.D. Havis 2 , P. Benincasa 1 , L. Covarelli 1
1 Dipartimento di Scienze Agrarie, Alimentari e Ambientali, Università degli Studi di Perugia, Borgo XX Giugno 74, 06121 Perugia, Italy; 2 Scotland’s Rural College (SRUC), King’s Buildings Campus, West Mains Road Edinburgh EH9 3JG. E-mail: [email protected]
Ramularia leaf spot and net blotch, caused by the fungal pathogens Ramularia collo-cygni (Rcc) and Pyrenophora teres (Pt), respectively, are two of the main foliar fungal diseases of barley causing leaf damages and yield losses. Recently, due to their seedborne attitude, their presence has been recorded constantly in several countries around the world. Fungicide resistance has also been widely recorded in both pathogens in a number of world areas. For this purpose, a two-year survey (2019/2020–2020/2021) was carried out to understand the occurrence and distribution of these two pathogens in the main Italian barley growing areas. Two different plant tissues, kernels and leaves, were analysed with a combination of different isolation methods to obtain fungal strains which were then identified by PCR assays. DNA extracted from barley kernels was also subject to real time qPCR assays to quantify Pt and Rcc DNA. Preliminary results showed that the samples from central Italy showed a higher Pt presence, both in terms of incidence and fungal DNA in the grains. Rcc was principally isolated and quantified in the samples from northern Italy. The incidence of the two pathogens in all the analysed grain samples did not exceed 30%, and, concerning fungal DNA, Pt showed the highest amount with respect to Rcc. This study will allow to better understand Pt and Rcc occurrence in the seed and their distribution in Italy as well as to obtain isolates to be used in population studies, including fungicide resistance tests.
Presence of Gnomoniopsis smithogilvyi in chestnut in Campania Region and sensibility to some active ingredients
V. Battaglia 1 , M. Petriccione 2 , A. Merola 1 , M. Cermola 1 , P. Rega 2 , E. Ferrara 2 , R. Sorrentino 1 , E. Lahoz 1
1 CREA—Consiglio per la ricerca in agricoltura e l'analisi dell'economia agraria Centro di Cerealicoltura e Colture Industriali, Via Torrino 2, 81100 Caserta, Italia; 2 CREA—Consiglio per la ricerca in agricoltura e l'analisi dell'economia agraria Centro di Olivicoltura, Frutticoltura e Agrumicoltura, Via Torrino 2, 81100 Caserta, Italia. E-mail: [email protected]
The production of chestnuts (Castanea sativa) in Italy plays a pivotal role in the nut sector and especially in Campania. In 2005, the presence of Gnomoniopsis smithogilvyi (syn. G. castaneae), agent of the chestnut rot, which causes huge losses of product both at harvest and post-harvest, was reported for the first time in Italy. The extreme environmental conditions play an important role in the epidemiology of the disease. In 2021, the presence of the disease was monitored in different cultivation conditions in the Campania region. Visual samplings were carried out on the fruits which were subsequently placed in a humid chamber to favor the development of the fungi which were isolated on culture medium. All the isolates were obtained in pure culture, to proceed with the morphological and molecular identification by PCR amplification. Eleven isolates obtained from the samples from different areas have been identified as G. smithogilvyi with 99.9% nucleotide sequence identity. Sensitivity tests were conducted on the eleven isolates with 4 commercial formulations, relating to 3 categories of fungicides: “biological” (tribasic copper sulphate; Eugenol + Geraniol + Thymol); succinate dehydrogenase inhibitors (Fluxapyroxad), triazoles (Tetraconazole). In addition, in vitro tests were carried out to evaluate the resistance to the various active ingredients. The results obtained in this work suggest that in addition to good agronomic practices, in the rational management of the orchard, such as the removal of residues and infected organs and the use of effective phytoiatric interventions can contribute to control the pathogen.
Dissecting foliar fitness of a Pseudomonas mediterranea strain nonpathogenic to citrus to understand the interaction with Plenodomus tracheiphilus
M.C. Bazzano 1 , P. Bella 2 , A.F. Catara 1 , G. Scuderi 1 , G. Dimaria 3 , F. Modica 3 , V. Catara 3 , M. Russo 1
1 Agrobiotech soc. coop. ZI Blocco Palma I, Str.le V. Lancia 57, 95121 Catania; 2 Department of Agricultural, Food and Forestry Sciences, University of Palermo, Italy; 3 Department of Agriculture, Food and Environment, University of Catania, Italy. E-mail: [email protected]
Citrus mal secco disease is the most limiting factor of lemon industry. The use of biological control agents to mitigate its symptoms is based on the capability of bacteria to colonize the same ecological niche invaded by the causal agent Plenodomus tracheiphilus. We have investigated several strains of Pseudomonas mediterranea (Pme), a bacterial nonpathogenic to citrus, for the frequent and rich presence in plants’ rhizosphere, the arsenal of secondary metabolites, the plant growth-promoting traits present in the genome, and the direct antagonism against a class of plant pathogens. The strain Pme 3C was tested in greenhouse and field on lemon grafted on sour orange to understand its foliar fitness and capability to reduce the penetration of P. tracheiphilus and to mitigate the symptoms of the disease. Results of dual tests in vitro show that the strain penetrates the leaves through the wounds and reduces the conidia germination and mycelial growth of the pathogen at different levels. Other tests, based on detection of bacterial DNA by real time PCR, evidence Pme survives epiphytically on leaf surfaces, penetrates the leaf wounds, and colonizes the mesophyll veinlets. Field tests confirm a reduction of penetration of the pathogen into the leaves, and a minor extension of the host xylem colonization, associated to a delayed expression of symptoms of the disease. Over all, the pathogen is not killed but is temporarily confined by the bacterium, and it competes for nutrient sources essential for hyphal elongation and host penetration.
This work was supported by the project S.I.R.P.A. funded by PO FESR 2014–2020 Sicilia action 1.1.5.
Climate change, new pathogens and multi-trophic interactions threaten forest ecosystems in the Mediterranean basin
A. Benigno, C. Aglietti, S. Moricca
Department of Agriculture, Food, Environment and Forestry (DAGRI), University of Florence, Piazzale delle Cascine 18, 50144 Florence, Italy. E-mail: [email protected]
In recent decades, forest ecosystems of the Mediterranean basin (a renowned hotspot for climate change) have been severely impacted by climate warming. Anomalous periods of rainfall, with recurrent droughts, have strongly stressed forest trees in a variety of situations (natural/artificial forests), predisposing trees to the attack by various fungal and oomycete pathogens. While, on the one hand, there has been a resurgence of old and known diseases, on the other hand new pathogens have arrived in uncontaminated territories, heavily infecting trees and spreading at extremely rapid rates (invasive species) over forest areas. Due to a lack of host–pathogen coevolution, these alien invaders often cause huge damage, posing a serious threat to tree populations and questions about the future management of the attacked forest ecosystems. Furthermore, a combined attack by immigrant and/or resident pathogens is in some instances observed, with lethal effects on the impaired tree species. This work reports some new and emerging pathogens that are threatening some Italian forest ecosystems. There is evidence that thermophilic or thermotolerant pathogens, which normally live as opportunistic endophytes in the tissues of host plants at an asymptomatic state, are capable to extensively colonize hosts that are impaired by climate anomalies (e.g. water stress). Some examples of multiple attacks by different pathogens on the same tree host are also reported. It becomes increasingly clear that in forest ecosystems we are often dealing not with single host–pathogen interactions but with multi-trophic interactions that can seriously compromise the stability and resilience of tree populations that are already heavily compromised by climate change.
Thousand cankers disease of walnut threatens the walnut groves of the Italian peninsula
A. Benigno 1 , C. Aglietti 1 , E. Falciani 1 , M. Bracalini 1 , D. Rizzo 2 , T. Panzavolta 1 , S. Moricca 1
1 Department of Agriculture, Food, Environment and Forestry (DAGRI), University of Florence, Piazzale delle Cascine 18, 50144 Florence, Italy; 2 Laboratory of Phytopathological Diagnostics and Molecular Biology, Plant Protection Service of Tuscany, Via Ciliegiole 99, 51100 Pistoia, Italy. [email protected]
Thousand cankers disease (TCD) of walnut is an emerging new disease of walnut caused by the ascomycete fungus Geosmithia morbida (Hypocreales, Bionectriaceae) and its vector insect, the walnut twig beetle Pityophthorus juglandis (Coleoptera, Curculionidae, Scolytinae). TCD is a disease of North American origin (USA), which in the last decade has spread to some Italian areas: first reported in the Veneto region in 2013, the disease was subsequently found in various regions of northern Italy and in 2019 it was reported in Tuscany, in the province of Florence. Main host of TCD is the black walnut (Juglans nigra) but in some situations the common walnut (Juglans regia) can also be attacked, e.g. in mixed stands with the two species growing adjacently. There is currently a strong concern that the disease may spread to other regions of our country, because the black walnut has been extensively used in tree plantations since the early nineties of the last century, with the financial support of the European Union (EU Regulation 1992/2080) aimed at boosting the cultivation of valuable hardwoods. But, being Italy the only country in which TCD is currently reported outside the USA, the concern is also high at the European level. Indeed, there is a strong fear that the disease could spread to other EU countries, where black walnut plantations have also been established with similar financial measures. For this reason, both the fungal pathogen and its insect vector have been included in the EPPO A2 List of quarantine pests. Given this alarming situation, molecular diagnostic tools have been developed to be used in phytosanitary surveillance at sensitive sites (ports, airports, etc.) and over the territory (nurseries, plantations, etc.). Such diagnostic protocols revealed the capacity to detect the DNA of the two organisms from a variety of matrices, such as necrotic areas (cankers) around insect holes, infected internal tissues (galleries), adults and insect larvae, insect frass. They therefore constitute a promising tool to try to counter the epidemic spread of the disease.
Survey of Kiwifruit Vine Decline Syndrome in Lazio region
V. Bergamaschi 1,2 , L. Pirone 2 , M.T. Valente 2 , S. Vitale 2 , L. Luongo 2 , A. Grottoli 2 , F. Marocchi 3 , L. Riccioni 2
1 Department of Environmental Biology, Sapienza University of Rome, P. le Aldo Moro 5, 00185 Rome, Italy; 2 Council for Agricultural Research and Economics, Research Centre for Plant Protection and Certification, Via C. G. Bertero 22, 00156 Rome, Italy; 3 Apofruit Italia s.c.a. Technical Office and Plant, Via Guardapasso snc Aprilia-Latina. E-mail: [email protected]
Kiwifruit Vine Decline Syndrome (KVDS) is a disease that affects Actinidia chinensis (kiwifruit) causing plant decline. Symptoms appear as root rot and branch wilting. The disease spread out in Lazio (central Italy) in 2017, while in northern Italy appeared as early as 2012. The disease probably is caused by abiotic and biotic factors. In this study, we aim to investigate the aetiological agents causing the symptoms, using both traditional, molecular and Next Generation Sequencing (NGS) approaches. Root and rhizosphere samples were collected from 20 kiwifruit orchards localized in Lazio region. Isolation from symptomatic roots revealed the presence of Cylindrocarpon sp. and oomycetes such as Phytophytium sp. In order to understand whether there is a clear difference in microbiome between kiwifruit plants affected or not by KVDS, we set up a DNA extraction method and a metabarcoding analysis using ITS2 for fungi and oomycetes. Preliminary results show that the DNA extraction method from root and soil samples works properly despite the high polysaccharides content in roots. Furthermore, NGS sequencing confirms the differences in the microbial community of asymptomatic and symptomatic samples. The next step is to analyse the other soil and root samples to confirm the difference between the symptomatic and asymptomatic plants to associate a pool of potential pathogens to the disease and a pool of microorganism to the healthy root and rhizosphere.
This work was founded by Programma Operativo Nazionale (PON) “Ricerca e Innovazione” 2014–2020—Azione IV.5 “Dottorati su tematiche green”.
Spread of Olea europaea geminivirus (OEGV) in olive trees in Sicily
S. Bertacca 1 , A.G. Caruso 1 , S. Matic 2 , E. Noris 2 , S. Panno 1 , S. Davino 1
1 Università degli Studi di Palermo, Dipartimento di Scienze Agrarie, Alimentari e Forestali, Viale delle Scienze, Edificio 5, 90133 Palermo, Italy; 2 Istituto per la Protezione Sostenibile delle Piante, Consiglio Nazionale delle Ricerche, Strada delle Cacce, 73, 10135 Torino, Italy. E-mail: [email protected]
Olea europaea geminivirus (OEGV), a putative member of the Geminiviridae family, has been recently identified in olive trees in several cultivation areas worldwide. For this reason, we aimed to investigate the presence and spread of OEGV in Sicily. Several surveys were conducted, focusing on olive production sites in the provinces of Agrigento, Caltanissetta, Ragusa and Trapani. Specifically, 70 sample batches, each consisting of eight different trees (560 sampled trees in total) of different cultivars were collected and geo-referenced with Planthology mobile application. Positive batches were re-sampled and each plant was analysed individually. All samples were prepared with a rapid extraction method, avoiding the use of DNA extraction with commercial kits and analysed using a real-time LAMP . These analyses showed that 30 out of the 70 cultivars tested were positive for OEGV (~ 43%), indicating a relatively high incidence and prevalence of OEGV across the sampling locations and the cultivars. In addition, the majority of plants (235 out of 240) within positive batches resulted OEGV positive, except the ‘Calatina’ cv. for which only three out of eight plants tested positive. Our survey revealed a considerable presence of OEGV in olive trees in Sicily, probably due to the inadvertent movement of clonally propagated infected but asymptomatic plant material.
Detection and management of “huanglongbing” in some Caribbean countries
A. Bertaccini 1 , F. Pacini 1 , N. Contaldo 1 , M. Nicolaisen 2 , R. Morillon 3 , Y. Uneau 4 , M. Luis Pantoja 5 , W. Myrie 6
1Alma Mater Studiorum – University of Bologna, Department of Agricultural and Food Sciences, Bologna, Italy; 2Department of Agroecology, Aarhus University, Denmark; 3CIRAD, Station de Roujol, Petit Bourg, Guadeloupe, France; 4ASSOWFI, Vieux-Habitants, Guadeloupe, France; 5Research Institute on Tropical Fruit Growing, La Habana, Cuba; 6Coconut Industry Board, Kingston, Jamaica. E-mail: [email protected]
“Huanglongbing” (HLB) is a bacterial citrus disease associated with huge economical damages whose most distinctive symptom is diffuse and asymmetrical mottling on the leaves. The HLB presence and management were monitored in Cuba, Guadeloupe and Jamaica during 2017–2022 where the presence of ‘Candidatus Liberibacter asiaticus’ was known since several years. This bacterium was present very often in mixed infection with diverse phytoplasmas, that in some cases were the only bacteria detected in plants showing HLB. These pathogens presence was confirmed by using a next generation sequencing method specifically adapted to detect these bacteria. In Cuba the elimination of symptomatic trees at regional scale resulted in the best management strategy to reduce the disease impact. In the orchards without eradication the disease incidence ranged between 13 and 18%. After two years, the percentage of symptomatic trees was 20% lower in the area with eradication program. The use of kaolin against the vector Diaphorina citri was reducing the infestation at very low percentages. In Guadeloupe, under very high HLB levels and mortality, different combinations of rootstocks / varieties were selected to study the impact of polyploidy on disease tolerance. In phloem petiole observed under scanning electron microscopy, a more limited callose deposition into sieve pores was observed in triploid compared to diploid genotypes; moreover, a more limited oxidative stress was observed in triploid petiole and tetraploid root samples compared to diploids. The impact of cultivation practices as adapted nutrition and fertilization, seems to be crucial in maintaining the trees productive under HLB.
Research founded from EU H2020 R & I programme under the project “Insect-borne prokaryote-associated diseases in tropical and sub-tropical perennial crops—TROPICSAFE” Grant Agreement No 727459.
Reverse transcription-droplet digital PCR for plum pox virus detection and quantification
G. Bertinelli, L. Tizzani, M. Luigi, V. Ilardi
Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di Ricerca Difesa e Certificazione sede di Roma (CREA-DC), Via C. G. Bertero 22, 00156 Roma, Italia E-mail: [email protected]
Plum pox virus (PPV) is the causal agent of sharka, one of the most detrimental diseases of stone fruits worldwide. PPV is one of the "top ten" viruses in molecular plant pathology. Currently most sensitive detection methods use reverse transcription real-time PCR (RT-qPCR), which needs a known standard to quantify the sample. Droplet digital PCR (ddPCR) is the third-generation PCR technique that overcomes this limitation, giving an absolute quantification. In this study, a validated one step RT-qPCR system was optimized and transferred to one step RT-ddPCR. Both methods were validated according to the European and Mediterranean Plant Protection Organization (EPPO) recommendations on sensitivity, specificity, selectivity, repeatability and reproducibility. PPV was detected and quantified using plant-infected RNA. Interestingly, not only RT-qPCR but also RT-ddPCR was suitable for plant-PPV-infected crude extract, suggesting a possible faster, cheaper detection. The methods succeeded in detecting 5 different PPV strains (Marcus, Dideron, Rec, El Amar, and Sweet Cherry) coming from different countries and four Prunus species. The exclusivity tests confirmed there were not signals when using Prunus-related viruses/viroids. Comparisons between the methods demonstrated that RT-ddPCR is 10 times more sensitive than the RT-qPCR when using total RNA from infected plants, while showed the same detection limit when using crude extracts. The obtained results shed light on this useful tool for the early detection and absolute quantification of PPV strains in Prunus species.
This work was supported by the project BIOTECH-BIOSOSFRU funded by MIPAAF (Ministero delle Politiche Agricole, Alimentari e Forestali).
Validation of high-throughput sequencing (HTS) for routine plant virus diagnostics
R. Bester 1,2 , G. Cook 3 , J.H.J. Breytenbach 3 , C. Steyn 3,4 , R. De Bruyn 1,2 , H.J. Maree 1,2
1 Citrus Research International, P.O. Box 2201, Matieland, 7602, South Africa; 2 Department of Genetics, Stellenbosch University, Private Bag X1, Matieland, 7602, South Africa; 3 Citrus Research International, P.O. Box 28, Nelspruit, 1200, South Africa; 4 Department of Plant Pathology, Stellenbosch University, Private Bag X1, Matieland, 7602, South Africa. E-mail: [email protected]
The credibility of a pathogen detection assay is measured using specific parameters including repeatability, specificity, sensitivity, and reproducibility. The use of high-throughput sequencing (HTS) as a routine detection assay for viruses and viroids in citrus was evaluated. Plant infections were established by graft inoculating healthy seedlings with a suite of viruses and viroids. Plants were sampled in triplicate and total RNA was extracted using two different methods and sent for HTS on both the Illumina and Ion Torrent platforms. The data were evaluated for biological and technical variation focussing on RNA extraction method, platform used and bioinformatic analyses. To evaluate the reproducibility of HTS, the same plants were evaluated again, one year later. The sensitivity of the HTS assay was compared to routinely used RT-PCR assays in a time course experiment. Both extraction method and sequencing platform resulted in significant differences between the data sets. The expression profiles of reference genes were also investigated to assess the suitability of these genes as internal controls to allow for the comparison between samples across different protocols. Even though the limit of detection of HTS was influenced by pathogen concentration, sample processing method and sequencing depth, HTS detection in this study was found to be either equivalent or more sensitive than RT-PCR. HTS is more comprehensive than any other assay and with the necessary validations the implementation of HTS as part of routine pathogen screening practices is possible.
This research was funded by Citrus Research International (CRI) and Research for Citrus Exports Fund of the Department of Science and Innovation (DSI), South African Government, grant number 1241.
Strawberry angular leafspot: efficacy of biocontrol agents and acibenzolar-S-methyl towards Xanthomonas fragariae
E. Biondi 1 , N. Biondo 1 , P. Lucchi 2 , M. Vibio 2 , L. Tazzari 1 , M. Cardoni 2 , P. Minardi 1
1 Department of Agricultural and Food Sciences (DISTAL), Alma Mater Studiorum—University of Bologna, V. le G. Fanin 42, 40127 Bologna, Italy; 2 Centro Attività Vivaistiche (CAV), Via Tebano, 45 – 48018 Faenza, Italy. E-mail: [email protected]
The angular leafspot of strawberry (ALS) is present in all areas where strawberry is cultivated, and in nurseries it represents a significant problem for the production of pathogen-free plant material. The control of Xanthomonas fragariae (Xf) is mainly achieved through preventive treatments based on copper compounds or antibiotics, where they are allowed. The efficacy of three bacterial antagonists (Bacillus sp.; D747 and QST713; Pseudomonas fluorescens: BB20), two essential oil (EO)-based products (Vitibiosap Plus® 458 and Vitibiosap Plus® 1R45) and the resistance inducer Bion® was tested against Xf in vitro and in vivo. All the biocontrol agents were able to directly reduce the pathogen growth in vitro, whereas Bion®, as expected, resulted ineffective. Under climatic chamber conditions, the three antagonists and the resistance inducer reduced the ALS severity and provided approx. 50% and 38% relative protection, respectively; besides, the two EO-based products provided approx. 40% relative protection. In addition, during the two years field trial, the EO-based products were able to significantly reduce ALS severity with a relative protection ranging to 60 to 80%, compared to the conventional treatments with copper compounds. Indeed, biological agents, such as bacterial antagonists or EO, and resistance inducers could represent an effective alternative in the framework of integrated control strategies.
An emerging disease challenges the use of cherry laurel ( Prunus laurocerasus L.) as a hedge plant in urban and periurban landscapes
D. Biondi 1 , E. Scali 1,2 , P. Capretti 1 , F. Fratini 1 , M. Benedetti 1 , C. Aglietti 1 , G. Marchi 1 , L. Ghelardini 1,3
1 Università degli Studi di Firenze, Dipartimento di Scienze e Tecnologie Agrarie, Alimentari, Ambientali e Forestali, Piazzale delle Cascine 18, 50144 Firenze, Italy; 2 University of California, Department of Environmental Science, Policy & Management, 137 Mulford Hall, Berkeley, CA 94720, USA; 3 Consiglio Nazionale delle Ricerche, Istituto per la Protezione Sostenibile delle Piante, Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy. E-mail: [email protected]
In the past five years extensive dieback of cherry laurel (Prunus laurocerasus L.) hedges has been observed with increasing frequency in the metropolitan area of Florence, Italy. Affected plants show bark necrosis and cankers of the lignified crown down to the collar with mucilage bleeding and gummosis. In the early disease stages the hedge appears healthy from the outside, although on closer inspection the first symptoms are present on the woody crown organs. In the more advanced stages internerval leaf chlorosis is followed by marginal orange spots progressing towards the centre of the leaf blade. The leaves turn rust-coloured, dry out and remain attached to the branches. Over time, the crown dries out and the plant dies. Isolations from symptomatic branches at numerous sites over the years revealed the constant presence of Diplodia seriata and other associated fungal pathogens with varying frequency. Disease monitoring and assessment (over 250 hedge plants scattered in the territory) revealed the ubiquitous presence of the disease (100% of the sampled sites affected), and extensive crown damage (85% of sampled plants affected) often without apparent foliar symptoms, since they usually become evident in later disease stages. Data analysis and inoculation tests suggest multiple disease drivers including rising temperatures that may favour D. seriata and alter host defenses by increasing water shortage. Results suggest that greater damage will occur in the coming years. The spread of the disease and the severity of the damage call into question the customary use of cherry laurel as a hedge plant.
Peptide analogs of a Trichoderma peptaibol effectively control downy mildew in leaf disc assay and vineyard
A. Bolzonello 1 , L. Sella 1 , S. Tundo 1 , L. Morbiato 2 , I. Baccelli 3 , R. Musetti 1 , M. De Zotti 2 , F. Favaron 1
1 Department of Land, Environment, Agriculture and Forestry, University of Padova, 35020 Legnaro, Italy; 2 Department of Chemistry, University of Padova, 35100 Padova, Italy; Institute for Sustainable Plant Protection, National Research Council, 50019 Sesto Fiorentino, Italy. E-mail: [email protected]
Grapevine (Vitis vinifera L.) is one of the main crops worldwide. Plasmopara viticola, the causal agent of downy mildew, causes enormous economic damage in terms of yield, quality and productivity to this crop. Although the disease control is largely based on the use of synthetic fungicides, the European Union (UE) policies promote the reduction of the reliance on synthetic plant protection products (PPPs) and the implementation of integrated approaches to crop protection. Biocontrol agents (BCA) constitute a great resource for the development of biopesticides. The fungal genus Trichoderma includes different species commonly used as BCAs against several crop pathogens and among their antifungal strategies, there are secondary metabolites such as peptaibols. The promise of peptaibols as agrochemicals is, however, hampered by their poor water solubility, preventing an efficient delivery for practical use in crop protection. Some water-soluble trichogin analogs with Gly to Lys substitutions were synthesized and analyzed for their antimicrobial activity against P. viticola. Among the peptides that proved effective in inhibiting the oomycete sporulation on leaf discs assay, the peptide 4r was selected for a two-year field trial experiment. In comparison with the untreated control, the protection level was comparable to that obtained with a cupric fungicide. The peptide did not show any phytotoxic effect and reduced downy mildew incidence and severity on leaves and bunches. Since it was demonstrated active also against the fungal pathogen Botrytis cinerea, the peptide 4r may be a candidate broad-spectrum fungicide whose biological properties deserve further investigations.
This research is funded by the Italian Ministry for University and Research, PRIN 20173LBZM2.
The high complexity of soybean anthracnose revealed by comparative genomics and transcriptomic analyses
T.R. Boufleur 1,2 , N.S. Massola Jr. 1 , S. Becerra 2 , Í. Tikami 1 , E. Baraldi 3 , S.A. Sukno 2 , M.R. Thon 2 , R. Baroncelli 2,3
1 Luiz de Queiroz College of Agriculture (ESALQ), Department of Plant Pathology and Nematology, University of São Paulo (USP) Piracicaba, Brazil; 2 Institute for Agribiotechnology Research (CIALE), Department of Microbiology and Genetics, University of Salamanca (USAL), Villamayor, Spain; 3 Department of Agricultural and Food Sciences (DISTAL), University of Bologna, Bologna, Italy. E-mail: [email protected]
Soybean (Glycine max) is one of the most important crops worldwide. A major limiting factor of soybean production is anthracnose, a disease caused by fungi belonging to the genus Colletotrichum. We performed a survey based on publicly available genomic data and showed that at least 12 Colletotrichum lineages are associated with soybean, with the C. truncatum species complex (s.c.) and C. orchidearum s.c. having the greatest impact and the broadest worldwide distribution. We used a combined approach of biological experiments with ‘omic’ technologies to gain a better understanding into the evolution of Colletotrichum species associated to soybean and the molecular basis of soybean resistance. The genomes of C. truncatum, C. musicola, C. plurivorum and C. sojae were sequenced and compared to eight additional Colletotrichum species not pathogenic to soybean. Our results show that C. truncatum and the three species belonging to the C. orchidearum s.c. did not evolve the capability to infect soybean from a common ancestor, due to the absence of effector candidates shared only among these four species. Therefore, we selected C. truncatum as a model system to study the molecular interaction with soybean. Pathogenicity assays revealed that the interaction between C. truncatum and soybean is genotype dependent, meaning that a soybean cultivar can present contrasting levels of resistance to different strains of C. truncatum, and a strain of C. truncatum can have different levels of aggressiveness in different soybean cultivars. Two soybean cultivars and two strains of C. truncatum were selected for cross inoculation assays. The transcriptomes of the four interactions were sequenced and analyzed in a time course experiment, revealing a coordinated pattern of gene expression over time for the resistant interactions, including the over expression of genes involved in pathogen recognition, signaling and defense responses. Our results show a strong correlation between plant/pathogen recognition and the activation of defense response opening new perspectives in the complexity of this pathosystem.
Exploiting the seed associated endophytes in Brassica oleracea genotypes as a potential source for plant growth promoting bacteria and biological control agents
N. Bova 1 , S. Conti 1 , A. Miceli 1 , A. Moncada 1 , F. Vetrano 1 , S. Bertacca 1 , A.G. Caruso 1 , S. Panno 1 , S. Davino 1 , G. Mirabile 1 , L. Torta 1 , V. Catara 2 , P. Bella 1
1 Department of Agricultural, Food and Forest Sciences, University of Palermo, Viale delle Scienze Ed. 4, 90128 Palermo, Italy; 2 Department of Agriculture, Food and Environment, University of Catania, Via Santa Sofia 100, 95123 Catania, Italy. E-mail: [email protected]
Plant seeds are characterised by a wide diversity of microbial communities which interact with the plant during its development. Seed associated microbial communities are influenced by genotype, environmental condition and crop management practices and may promote plant growth and act as biological control agents. Studies on seed microbiome associated with brassica seeds were mainly focused on canola, whereas there is a paucity of information on seed microbial signature in other Brassica genotypes. In order to explore the seed bacterial community structure, seed batches from Brassica oleracea groups were analysed for seed-originating endophytic bacteria with the aim to select beneficial microorganisms with potential application in suitable crop protection. Seed samples were analysed by direct isolation techniques after seed sterilisation and bacterial isolates were identified based on the PCR amplification of partial 16S rDNA gene sequences. Screening for plant growth-promoting activities revealed that most of the isolates showed PGPR traits and produced ACC deaminase, siderophores and were able to solubilize phosphate and to grow in presence of 4% and/or 8% NaCl. In vitro tests showed that some bacterial isolates inhibited the growth of both fungal and bacterial strains including the seed-borne brassica pathogens Xanthomonas campestris pv. campestris, X. campestris pv. raphani, Leptosphaeria maculans and Alternaria brassicicola. Bacterial isolates sharing PGPR traits and direct antagonistic activity against all tested pathogens were also identified. Preliminary results indicated that seeds of Brassica oleracea genotypes store their own beneficial inoculum which may be involved in maintaining plant health. Further studies are ongoing to analyse the in vivo bacterial endophyte-plant interaction after seed treatments.
In vitro exploitation of bioactivity of brassica extract on the growth of plant pathogenic bacteria and fungi
N. Bova 1 , S. Conti 1 , G.M. Infurna 2 , S. Nigro 2 , G.F. Rizzo 3 , F. Branca 3 , P. Bella 1
1Department of Agricultural, Food and Forest Sciences, University of Palermo, Viale delle Scienze Ed. 4, 90128 Palermo, Italy; 2Itaka srl, Via Montenapoleone 8, 20121 Milano, Italy; 3Department of Agriculture, Food and Environment, University of Catania, Via Santa Sofia 100, 95123 Catania, Italy. E-mail: [email protected]
In recent years, there has been a growing interest in exploitation and developing new environmentally sustainable solutions to counteract plant disease. Glucosinolates are secondary metabolites found almost exclusively in Brassicaceae family and are important contributors to the health benefits of these plants. In this study, the activity of a natural brassica extract coupled with propolis was evaluated in vitro against plant pathogenic bacteria and fungi at three different concentrations 1.0, 1.5 and 3.0 mL/L according to the recommendations of the company who provided them. Antimicrobial activity was tested against the fungal pathogens Alternaria brassicicola, Neocosmospora phaseoli, Fusarium oxysporum f. sp. lycopersici and Sclerotinia sclerotiorum by the poisoned food method. The natural compound induced a reduction of mycelial growth of all tested fungal isolates with significant differences compared to the control. Inhibitory effect was dose dependent, varied according to the fungal species and decreased with the time. All the extracts at 3.0 mL/L concentration reduced the fungal growth by 70–100% up to about seven days after inoculation. No activity was observed by well diffusion assay against the plant pathogenic bacteria Pseudomonas savastanoi pv. phaseolicola, P. syringae pv. tomato, Xanthomonas vesicatoria, X. campestris pv. campestris and Clavibacter michiganensis subsp. michiganensis. The interference of the natural compound with the kinetic growth of all bacterial strains was further evaluated by an automated turbidimetry-based system and the AUC (area under curve) was calculated. All concentrations of the natural compound induced a growth reduction of all strains with a higher AUC reduction in presence of the higher concentration.
This work was supported by the European project H2020 BRESOV “Breeding for Resilient, Efficient and Sustainable Organic Vegetable production”.
Phytophthora : an ancient, historic, biologically and structurally cohesive and evolutionarily successful generic concept in need of preservation
C. Brasier 1 , B. Scanu 2 , D. Cooke 3 , T. Jung 4
1 Forest Research, Alice Holt Lodge, Farnham, Surrey GU10 4LH, UK; 2 Department of Agricultural Sciences, University of Sassari, Viale Italia 39A, 07100 Sassari, Italy; 3 The James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK; 4 Department of Forest Protection and Wildlife Management, Phytophthora Research Centre, Mendel University in Brno, 613 00 Brno, Czech Republic. E-mail: [email protected]
In response to evidence that the 20 genera of the obligately biotrophic, angiosperm-foliage specialised downy mildews (DMs) evolved from Phytophthora at least twice via convergent evolution, making the DMs as a group polyphyletic and Phytophthora paraphyletic in cladistic terms, a proposal has been made to split the genus into multiple new genera. We have reviewed the status of Phytophthora and its relationship to the DMs. Currently some 200 species are distributed across twelve major phylogenetic clades that, based on our assessment for twenty morphological and behavioural criteria, show good biological cohesion. Saprotrophy, necrotrophy and hemi-biotrophy of woody and non-woody roots, stems and foliage occurs across the clades. Phylogenetically less related clades often show strong phenotypic and behavioural similarities and no one clade or group of clades shows the synapomorphies that might justify a unique generic status. The proposal to divide Phytophthora appears more a device to address the issue of the convergent evolution of the DMs than the structure of Phytophthora per se. We consider it non-Darwinian, putting the emphasis on the emergent groups (the DMs) rather than the progenitor (Phytophthora) and ignoring the evolutionary processes that gave rise to the divergence. Considering the biological and structural cohesion of Phytophthora, its historic and social impacts and its importance in scientific communication and biosecurity protocol, we recommend that the current broad generic concept is retained by the scientific community.
Emerging Phytophthora -related diseases in the subalpine European vegetation
C. Bregant, G. Rossetto, L. Montecchio, N. Sasso, B.T. Linaldeddu
Dipartimento Territorio e Sistemi Agro-Forestali, Università degli Studi di Padova, Viale dell’Università 16, 35020 Legnaro, Italy. E-mail: [email protected]
Extensive decline and mortality of shrubs and trees typical of alpine areas has recently been observed in Southern Europe. Decline phenomena affecting different ecological niches such as riparian systems, forest margins and Alpine heathlands were characterized by a complex symptomatology including foliar necroses, shoot blight and bleeding cankers. Infections resulted in progressively widening patches of dying vegetation, suggesting the involvement of airborne Phytophthora species in the aetiology. Between 2019 and 2021, an in-depth study aimed to determine the impact and the diversity of Phytophthora species involved in these phenomena, was carried out on 47 different sites distributed in the mountainous areas of Sardinia, Tuscany Apennines and North-eastern Alpine regions among Italy, Austria, and Slovenia. A total of 286 Phytophthora isolates were obtained from 322 symptomatic samples collected from 26 plant species. Isolates have been identified as Phytophthora pseudosyringae (169 isolates), P. plurivora (39), P. ilicis (20), P. acerina (13), P. alpina (6), P. cactorum (6), P. gonapodyides (6), P. idaei (4), P. cambivora (3), P. psychrophila (3), P. pseudocryptogea (2), P. bilorbang (2), P. gregata (2), P. hedraiandra (2) and P. kelmanii (1). The results highlighted a wide diversity of Phytophthora belonging to the clade 1 and 3, species able to survive under cold conditions and to cause aerial infections, due to the production of caducous sporangia. Among the different species, Phytophthora pseudosyringae was identified as the key pathogen in these habitats, being isolated from 22 out of 26 investigated hosts.
The fungal and bacterial bunch microbiome associated with sour rot in vineyards
C. Brischetto, V. Rossi, G. Fedele
Università Cattolica del Sacro Cuore di Piacenza, Dipartimento di Scienze delle produzioni vegetali sostenibili, Via Emilia Parmense 84, 29122 Piacenza, Italy. E-mail: [email protected]
Sour rot (SR) is a grapevine disease complex that is not completely understood for both etiology and epidemiology. In recent years, SR has received special attention due to its increasing economic importance due to the heavy crop losses and reduced wine quality it can cause. In this study, the bacterial and fungal microbiota of healthy and affected (i.e., showing SR symptoms) ripe bunches were characterized across 47 epidemics (38 vineyards in 5 grape-growing areas) over a 3-year period. The 16S rRNA gene and ITS high-throughput amplicon sequencing, as well as quantitative PCR (qPCR), were used to assess the relative abundance and dynamic changes of microorganisms associated with SR. The estimators of genera richness within samples indicated that diversity was not significantly different between healthy and affected bunches. When the community composition between samples was evaluated, the bunch status (i.e., healthy and affected) was a significant source of diversity (p < 0.05), indicating that microbiome composition varied between healthy and affected bunches. In particular, these two categories shared 43.1% and 56.5% of fungal and bacterial genera, respectively; 31.6% (fungal) and 26.1% (bacterial) genera were associated with affected bunches only. Based on a linear discriminant analysis, the following bacteria: Orbus and Gluconobacter spp., and fungi Zygosaccharomyces, Pichia, Issatchenkia, and Zygoascus spp. were closely associated with bunches showing SR symptoms. Understanding the diversity of the microflora associated with SR affected bunches is the first step for the development of effective protection strategies.
Solid-State fermentation of organic wastes: a source of bioactive molecules
D. Bulgari 1 , C. Alias 1 , G. Peron 2 , G. Ribaudo 2 , A. Gianoncelli 2 , E. Gobbi 1
1Agri-Food and Environmental Microbiology Platform (PiMiAA), Department of Molecular and Translational Medicine, University of Brescia, Via Branze 45, 25123 Brescia, Italy; 2Proteomic Platform, AgroFood Lab, Department of Molecular and Translational Medicine, University of Brescia, Viale Europa, 11, 25123 Brescia, Italy. E-mail: [email protected]
Fungi used in biocontrol often exert their activity through the production of a wide-range of bioactive secondary metabolites, which could be exploited for the development of bio-pesticides, bio-stimulants and medicines. In this work, we evaluated the possibility of using digestate as a source for the inoculation in the open field of beneficial microorganisms, including the associated molecules with high added value, such as organic acids and enzymes. Digestate is a by-product of anaerobic digestion mainly used as a fertilizer in the agricultural field. Whole digestate, suitably mixed with agro-food waste, was used as a substrate for Solid State fermentation (SSF) by different fungal genera. The digestate-based substrates have allowed the growth of fungi belonging to the genera Aspergillus, Penicillium, Cryphonectria and Trichoderma. The aqueous crude extracts obtained from SSF of four Trichoderma sp. were tested for enzymes production, organic acids and bioactive molecules. Enzymatic assay revealed cellulase and esterase activities. Citric acid and other secondary metabolites produced during fermentation were detected by UPLC-QTOF-MS analysis. Moreover, experimental evaluation of germination parameters (germination index) highlighted a strong promotion of tomato seed germination and root elongation induced by T. harzianum and T. atroviridae crude extracts from SSF. This study suggests an innovative use of the whole digestate mixed with agro-food waste as a valuable substrate for growing plant growth-promoting fungi and for producing bioactive molecules. We are currently evaluating metabolites with biocontrol potential from the same extracts.
A phytocomplex obtained from a Salvia officinalis cell culture effectively controls the grapevine downy mildew pathogen Plasmopara viticola
I. Busato 1 , O. Bertaiola 2 , S. Tundo 1 , C. Guarnerio 2 , M. Lucchetta 3 , L. Sella 1 , G. Pressi 2 , F. Favaron 1
1 >Department of Land, Environment, Agriculture and Forestry, University of Padova, Legnaro, Italy; 2 Aethera Biotech s.r.l., 36043 Camisano Vicentino (VI), Italy; 3 Coccitech s.r.l., 3105 Conegliano (TV). E-mail: [email protected]
The impact of conventional fungicides on environmental and human health has increased the interest in safer alternatives such as plant secondary metabolites, generally having a better toxicological profile. However, plant genetics, agronomic practices, climatic conditions and extraction methods strongly affect the quality and quantity of secondary metabolites obtained from field grown plants. These factors cause limitations to the standardization needed for industrial production. Plant cell culture technology can meet this need: totipotent cells grown in controlled conditions can provide a highly homogeneous biomass with specific chemical characteristics. A phytocomplex, with standardized rosmarinic acid content, was obtained from a selected cell line of Salvia officinalis. The Salvia officinalis phytocomplex (SOP) was tested against the grapevine downy mildew pathogen, Plasmopara viticola. Grapevine leaf disks were sprayed with SOP and with fresh P. viticola sporangia. Sporulation level on each disk was assessed after 7 days with an image processing software. SOP at 5 g/l reduced by 95% the sporulation level compared to the control treatment. SOP was significantly more effective than rosmarinic acid alone, tested at the concentration found in SOP. Persistence of the phytocomplex was also assessed: leaves were sprayed with SOP and, after a few days, they were detached and inoculated. SOP applied 5 days before inoculation reduced by 90% the sporulation level compared to the control. These results highlight the possibility for plant protection industry to take advantage of cell culture techniques to produce safer pesticides with high quality standards.
The stress related – calcium dependent – transcription factor CRZ1 is a pivotal factor in synchronising host perception and fumonisins biosynthesis in the Fusarium verticillioides- Zea mays interaction
A. Cacciotti 1 , M. Beccaccioli 1 , S. Vitale 2 , A. Di Pietro 2 , D. Turrà 2 , V. Scala 3 , M. Reverberi 1
1 Department of Enviromental Biology, “Sapienza” University of Rome, P.le Aldo Moro 5, 00185 Rome, Italy; 2 Department of Genetics, Campus de Excelencia Internacional Agroalimentario ceiA3, Universidad de Córdoba, Córdoba, Spain; 3 CREA-DC, Via Carlo Giuseppe Bertero 22, 00156 Roma, Italy. E-mail: andrea.cacciotti @uniroma1.it
Calmodulin (CaM), the main calcium binding protein in eukaryotes, acts as a part of calcium signal transduction pathway, involved also in colony growth, stress response and pathogenicity regulation in fungi. Here we investigated the role of the fungal protein Crz1, a downstream transcription factor of the CaM pathway, in the fumonisin (FUM)-producing fungus Fusarium verticillioides. Previous studies have shown that the trigger for the onset of FUM synthesis in maize by Fusarium is related to some specific fatty acids (FAs) and oxylipins (e.g., oxidised FAs) that can play, during the infection, as intra-extra cellular signals exchanged by the pathogen and its host. In this study, we suggest the role of Crz1 as regulator and “synchronizer” of lipid metabolism and FUM biosynthesis during the F. verticillioides-Zea mays interaction. To investigate the possible role of the transcription factor during the infection, we found through a mass spectrometry and transcriptomic approach that the deletion of crz1 was consistently associated with an overall reduction in oxylipin, FA and FUM expression and amount in the infected kernels, whilst this deletion did not affect the fungal lipid metabolism (under in vitro growth); this evidence suggests a key role for Crz1 in regulating this complex and interconnected pathways during the host–pathogen interaction. We postulate that Crz1 controls the production of secreted signals during the interaction with the host that can trigger the oxylipin pathways in maize. In turn, the oxylipins (e.g. 9-HPODE) generated by maize would provide feedback to the fungus to switch on the biosynthesis of FUM as already stated elsewhere.
Monitoring and characterisation of Colletotrichum species associated with bitter rot isolated from diseased apple in the Emilia-Romagna region orchards
M. Calì 1 , G. Amaral Carneiro 1 , R. Baroncelli 1 , E. Cappelletti 1 , R. Bugiani 2 , A. Prodi 1
1 Università di Bologna, Dipartimento di Scienze e Tecnologie Agro-alimentari, Viale Fanin 40, 40127 Bologna, Italy; 2 Servizio Fitosanitario Regione Emilia-Romagna, via Andrea da Formigine 3, 40128 Bologna, Italy. E-mail: [email protected]
Italy is one of the major apple producers in Europe, ranking second after Poland. An important disease that affects apples both pre-harvest and post-harvest is bitter rot. In Europe, the species causing the most significant losses in the orchard during the vegetative period belong to the Colletotrichum acutatum species complex. Over the past few years, however, species of the Colletotrichum gloeosporioides complex are increasingly emerging as bitter rot pathogens. In Italy in 2019, a serious outbreak of fruit with bitter rot symptoms was observed in commercial apple orchards of the 'Pink Lady' variety in the Emilia- Romagna region. One of the peculiarities of Colletotrichum species is to infect apple fruits through lenticels in pre-harvest, remain latent and develop symptoms after months of cold storage under controlled atmosphere, leading to economic losses of the product. Therefore, the aim of this work is monitoring the incidence of bitter rot in apple orchards in the Emilia-Romagna region and bordering areas. Moreover, special attention will be given to the phenotypic and molecular characterisation, with the application of multilocus sequencing typing, of emerging species belonging to the C. gloeosporioides complex isolated from leaves and fruits with symptoms related to bitter rot. Knowledge of the incidence of bitter rot in apple orchards in Emilia Romagna, as well as accurate identification of the pathogenic species will be of paramount interest in order to implement targeted disease control systems in pre-harvest. Furthermore, to date no post-harvest treatment is able to eradicate latent infection of apple fruits during cold storage, which means that monitoring of these pathogens in pre-harvest is essential to prevent post-harvest losses.
Development and validation of a mechanistic model for the prediction of hazelnut defects caused by Diaporthe eres
M. Camardo Leggieri 1 , R. Arciuolo 1 , G. Castello 2 , N. Spigolon 2 , P. Battilani 1
1 Department of Sustainable Crop Production, Università Cattolica del Sacro Cuore, Via Emilia Parmense 84, Piacenza, Italy; 2 SOREMARTEC ITALIA S.r.l., Piazzale Pietro Ferrero 1, Alba (CN), Italy. E-mail: [email protected]
Hazelnuts are high-value products that suffer defects sometimes associated with off flavours. They cause yield losses due to non-compliance with the required market standards. The browning of the internal tissues of kernels, which is visible when nuts are cut in half, discolouration and brown spots on the kernels surface are important defects mainly attributed to Diaporthe eres. Timely indications on the expected incidence of defective hazelnuts would be essential both for buyer supplies, to identify risk areas for uncompliant products and for farmers to have timely warnings of D. eres infection and defect outbreak. Knowledge regarding this fungus and its interaction with hazelnut is poor. Nevertheless, a mechanistic model was developed using information regarding Diaporthe infection cycle and D. eres ecological needs. The predictive model inputs hourly meteorological data (air temperature, relative humidity and rainfall). The output is the risk of hazelnut defects at the end of the growing season. Georeferenced data on the occurrence of hazelnut defects from 2014 to 2019 were collected from orchards in the Caucasus region and Turkey to validate the model. Hourly meteorological data from weather stations close to the hazelnut orchards selected were used to run the model. Then, using appropriate statistical techniques, predictions were compared with observed data (480 records). Overall, correct predictions were > 80%, with < 16% of false negative. The results confirmed the developed model's accuracy in predicting hazelnut defects caused by D. eres.
Scientific community actions and reactions to the predicted impact of climate change on aflatoxin
M. Camardo Leggieri 1 , P. Toscano 2 , P. Battilani 1
1 Department of Sustainable Crop Production, Università Cattolica del Sacro Cuore, Via Emilia Parmense 84, Piacenza, Italy; 2 IBE-CNR, Institute of BioEconomy-National Research Council, Via Giovanni Caproni 8, 50145 Firenze, Italia. E.mail: [email protected]
Climate change (CC) is confirmed as a driver for emerging food/feed safety issues worldwide, and its expected impact on the presence of mycotoxins in commodities is of great concern. Aflatoxins (AF) have the highest toxicity of all mycotoxins and are regulated in food/feed worldwide. As highlighted in 2011 by a project supported by EFSA, CC is predicted to increase the risk of AF contamination in maize. In 2021 a comprehensive literature search was performed using the Scopus search engine to extract peer-reviewed studies related to this prediction. A total of 224 papers were identified after step I filtering, while step II filtering identified 25 papers for quantitative analysis. The unselected papers (199) were categorized as “actions” because they provided a sounding board for the expected impact of CC on AFB1 contamination without adding new data on the topic. The remaining papers were considered “reactions” of the scientific community because they went further in data and ideas reporting. Interesting statements taken from the “reactions” could be summarized with keywords: chain and multi-actor approach, intersectoral and multidisciplinary, resilience, human and animal health, and global vision. In addition, fields deserving increased research efforts were summarized as: i) improvement of predictive modelling, ii) extension to different crops and geographic areas, and ii) impact of CC on fungi and mycotoxin co-occurrence, both in crops and their value chains, up to the consumers. Ten-years of work after the prediction release indicated improved approaches but also emphasized that relevant actions are still requested.
Phloem-restricted phytoplasmas impair carbon fixation in tomato
F. Cantao 1 , S. Santi 1 , F. Marroni 1 , M. Guazzini 1 , R. Musetti 2
1 University of Udine, Department of Agricultural, Food, Environmental and Animal Sciences, Via delle Scienze, 206—33100 Udine; 2 Department of Land, Environment, Agriculture and Forestry (TESAF), University of Padova Agripolis, Viale dell’Università 16 – 35020 Legnaro, Padova, Italy. E-mail: [email protected]
Phytoplasmas are prokaryotic obligated parasites of plants that colonize the sieve elements of their host plant, causing alterations in phloem function such as occlusion, nutrients uptake and utilization, impairment of photosynthesis and photo assimilates translocation. Symptoms have been well described and include inhibition and decline of photosynthesis efficiency and alterations in sugar metabolism. Our research group has been investigating the effects of ‘Candidatus Phytoplasma solani’ infection on tomato plants (Solanum lycopersicum cv. Micro-Tom) by RNA-sequencing. Analyzing tomato plants grown under different Fe regimes, we demonstrated that infection changes Fe distribution in leaves, affects photosynthetic machinery and perturbs shoot-to-root communication. Moreover, infection impacts mineral nutrient fluxes and alters ion homeostasis of K, Ca, Mg, Fe and Mn. Here, we focus our analysis on the transcriptional regulation of genes involved on carbon fixation and carbon metabolism. A gene co-expression network was investigated with a weighted correlation approach implemented in the R package WGCNA. Gene expression and protein analysis highlighted key genes of carbon fixation pathway and confirmed impairment of carbon metabolism.
Isolation and molecular characterization of endophyte bacteria in three different durum wheat cultivars susceptible to Fusarium Foot and Crown Rot
E. Cappelletti, C. Braglia, D. Alberoni, M. Calì, G. Amaral Carneiro, D. Di Gioia, A. Prodi
Università di Bologna, Dipartimento di Scienze e Tecnologie Agro-alimentari, Viale Fanin 40, 40127 Bologna, Italy. E-mail: [email protected]
Triticum durum is one of the world’s most important food crops and Italy is one of the main producers and consumers of this cereal, that unfortunately is susceptible to the attack of microorganisms responsible for different diseases causing both quantitative and qualitative damages. Specifically, two main diseases that affect durum wheat in the early stages of development are Fusarium Foot Rot (FFR) and Fusarium Crown Rot (FCR). The main fungal species involved in the process belong to the genus Fusarium, specifically F. culmorum, F. graminearum and F. pseudograminearum. These species can act alone or in combination, affecting the roots and the internodal portions of the culm, and are responsible for the production of high levels of a wide variety of mycotoxins. The use of organic seed coating technique with endophytic bacteria, that colonize healthy plant tissue without causing obvious disease symptoms in host plant, may represent a reliable and safe strategy able to control soil-borne pathogens. The present investigation was carried out to analyze the microbiome of three durum wheat varieties viz. Claudio, Marco Aurelio and Odisseo to identify potential biocontrol agents. A total of 64 bacterial endophytes were isolated from crown and root tissues. Molecular characterization of all the strains was carried out by partial PCR amplification of 16S rRNA gene using the primers 27F and 1492R. Isolated endophytes are closely related phylogenetically to Pseudomonas, Bacillus, Stenotrophomonas, Microbacterium, Achromobacter, Ochrobactrum, Peribacillus, Arthrobacter, and Rhizobium by sequence analysis.
Pomegranate and citrus fruit rots caused by Alternaria species in Albania
M. Cara 1 , M. Toska 2 , D. Frasheri 2 , R. Baroncelli 3 , S.M. Sanzani 4
1 Department of Plant Protection, Faculty of Agriculture and Environment, Agricultural University of Tirana, Tirana, Albania; 2 CIHEAM BARI, Via Ceglie 9, 70010 Valenzano (BA), Italy; 3 Department of Agricultural and Food Sciences, University of Bologna, Viale Fanin 46, 40127 Bologna, Italy; 4 Department of Soil, Plant and Food Sciences, University of Bari “Aldo Moro”, Via Amendola 165/A, 70126 Bari, Italy. E-mail: [email protected]
The genus Alternaria is a relevant fungal pathogen for several commodities including citrus and pomegranate fruits. On citrus, it mostly causes brown spots on fruits and leaves, whereas, on pomegranate, it primarily causes a fruit heart rot. In the present study the Alternaria rots on citrus and pomegranate fruits cultivated in Albania was assessed. Representative fruits were collected from different regions. Nineteen and thirteen Alternaria isolates were obtained from pomegranate and citrus samples, respectively. The isolates were identified at species and morphotype level. Micro- and macroscopic features separated isolates into four morphotypes. BLAST and phylogenetic analysis using the SCAR Marker OPA1-3 confirmed the isolate identity. All 32 isolates proved to be Alternaria alternata and to mostly belong to morphotype alternata, followed by limoniasperae and tenuissima. All Alternaria strains proved to possess the pksI gene of the mycotoxin alternariol biosynthesis. Citrus isolates were tested for the presence of genes of the biosynthesis of the phytotoxins ACT and ACR, but none of them proved to possess them. Concluding, Alternaria spp. might represent a treat to pomegranate and citrus production in Albania and commercialised abroad, and thus effective diagnostic and control means are needed.
Cortical necrosis of grapevine (known as “excoriose”) and Diaporthe dieback: an undervalued damage to grape production
G. Carella 1 , A. Paolini 1 , A. Risaliti 1 , L. Mugnai 1
1 Dipartimento di Scienze e Tecnologie Agrarie, Alimentari, Ambientali e Forestali (DAGRI), Sez. Patologia vegetale ed Entomologia, Università di Firenze, P.le delle Cascine 28, 50144 FIRENZE, Italy. E-mail: [email protected]
Damages on grapevine caused by Diaporthe species are increasingly reported both on wine grape and table grape. Several species are detected, the most common of which is Diaporthe ampelina, the currently correct name of the anamorph Phomopsis viticola, the agent of cortical necrosis (often erroneously referred to as excoriose). The increase in the incidence of the disease in recent years is due to a series of causes such as scarce or not careful monitoring in the vineyard due to the limited importance given to fungi of this genus, considered minor pathogens compared to others associated with wood diseases of the grapevine (Grapevine Trunk Diseases, or GTDs). In Italy the main GTD is surely the well know Esca complex. In a recent survey in a vineyard where heavy damages by esca were reported a high incidence of spur dieback was noticed that was considered as one of the symptoms caused by the esca agents. Sectorial necrosis or total necrosis of the spur could be observed when cross cutting the spurs, cordon or trunk. A careful monitoring revealed a high presence of cortical necrosis symptoms in the basal internodes of the canes. Field survey and sampling proved the spur dieback to be actually strongly associated to Diaporthe species colonizing the wood starting from pruning wounds infections, up to causing the death of the whole cordon, that could be described as a Diaporthe dieback. The relevance of detecting the wood disease actually present in the vineyard is discussed in view of setting up efficient preventive strategies.
Complete genomes of several ‘ Candidatus Phytoplasma solani’ strains causing different symptoms in experimental host Solanum lycopersicum
G. Carminati 1 , G. Firrao 1 , P. Ermacora 1 , F. Pavan 1 , M. Kube 2 , M. Martini 1
1 Department of Agricultural, Food, Environmental and Animal Sciences (DI4A), University of Udine, Via delle Scienze 206, 33100 Udine, Italy; 2 University of Hohenheim, Integrative Infection Biology Crops-Livestock, Garbenstr. 30, 70599 Stuttgart, Germany. E-mail: [email protected]
‘Candidatus Phytoplasma (Ca. P.) solani’ (16SrXII-A) is associated with bois noir (BN) of grapevine and stolbur disease of herbaceous plants and is primarily transmitted by Hyalesthes obsoletus Signoret. The aim of our study is to investigate the interaction of different ‘Ca. P. solani’ strains with tomato (Solanum lycopersicum L. cv. Micro-Tom). Strains of ‘Ca. P. solani’ were acquired by individuals of H. obsoletus captured on bindweed and stinging nettle in vineyards of northeastern Italy with high incidence of BN and forced to feed on tomato plants. Strains originating in stinging nettle and bindweed belonged to tuf-a and tuf-b genotypes, respectively, as expected. The genomes of four strains were shotgun sequenced by Oxford Nanopore Technology (ONT) and NovaSeq (Illumina). The complete and circular genomes of three ‘Ca. P. solani’ tuf-b and one tuf-a strains with a size between 751,188 bp and 973,640 bp, were reconstructed by hybrid assembly. The number of automatically annotated CDS varied between 725 and 1064. The four genome sequences were mined for the presence of candidate effector genes by signal peptide (SP) and transmembrane signal peptide (TMSP) predictions. The protein predicted as SP from Phobius ranged between 23 and 40; in particular, all strains lacked TENGU-like proteins, whereas proteins like SAP11, SAP54/PHIL1 and SAP05 were found. Comparisons with already known phytoplasma effector proteins and putative secreted proteins evidenced strain-specific sets of effectors that may explain the quite different symptoms induced on tomato.
Insight of tomato brown rugose fruit virus (ToBRFV) dispersion applying remote RT-qPCR procedures in tomato production environment
A.G. Caruso, S. Davino, S. Bertacca, A. Ragona, S. Panno
Università degli Studi di Palermo, Dipartimento di Scienze Agrarie, Alimentari e Forestali, Viale delle Scienze, Edificio 5, 90133 Palermo, Italy. E-mail: [email protected]
Tomato brown rugose fruit virus (ToBRFV) causes severe disease in tomato crops. After its outbreak in Sicily (2018), it still represents a serious threat for tomato production.
In the present work, during 2021, to assess the ToBRFV dispersion in the main tomato production Sicilian areas, five tomato commercial greenhouses with 600 tomato plants each (3 in Ragusa province, 1 in Syracuse province, 1 in Agrigento province) were monitored with the bCUBE® portable system (Hyris Ltd). The ToBRFV dispersion was studied using a network of mini-laboratories for early in-field ToBRFV detection, established and monitored by the Plant Virology Laboratory at the University of Palermo for remote network management, data visualization, and analysis confirmation. For each greenhouse, three samples per month were collected, for eight months. Each sample, consisting of 200 plants pool (⁓0.5 cm2 leaf portion from each plant), was tested by RT-qPCR developed by Panno and co-workers (2019) using the bCUBE® system. To confirm the bCUBE® analyses, 20% of the samples were also analyzed at the Plant Virology laboratory. The 25% of the samples resulted positive to ToBRFV with the bCUBE® system, confirmed by the laboratory analyses. The positive samples were sequenced in both directions, showing a 100% identity with the previously characterized ToB-SIC01/19 Sicilian isolate. The developed mini-laboratories network allowed continuous ToBRFV monitoring throughout the tomato production. Compared to previous years, a decrease of ToBRFV dispersion was observed, thanks to early in-field diagnosis using the bCUBE® diagnostic network, continuous monitoring by phytosanitary services and the prophylactic measures implementation.
Survey of main grapevine cultivars for grapevine fanleaf virus dispersion in Sicily
A.G. Caruso 1 , S. Panno 1 , S. Bertacca 1 , A. Ragona 1 , N. Bova 1 , G. Mirabile 1 , L. Torta 1 , P. Bella 1 , S. Matić 2 , S. Davino 1
1Università degli Studi di Palermo, Dipartimento di Scienze Agrarie, Alimentari e Forestali, Viale delle Scienze, Edificio 5, 90133 Palermo, Italy; 2Institute for Sustainable Plant Protection, National Research Council (IPSP-CNR), Strada delle Cacce 73, 10135 Turin, Italy. E-mail: [email protected]
Grapevine (Vitis vinifera L.) is one of the most important crops worldwide. Grapevine cultivation is undermined by several viral diseases, such as grapevine fanleaf degeneration disease caused by grapevine fanleaf virus (GFLV) (family Secoviridae). In this survey, twenty commercial vineyards in four Sicilian provinces (Italy) were investigated for GFLV presence, studying its genetic structure and molecular variability. A total of 800 grapevine samples from eleven autochthonous Sicilian cultivars (cvs) were collected and analyzed by end-point RT-PCR using specific primers for the CP gene. Forty-eight out of 800 tested samples resulted positive to GFLV, with 6% of the overall infection rate. The highest percentage of positive samples was recorded in cvs “Nerello Cappuccio”, “Nero d’Avola” and “Catarratto”, with an incidence of 26.8%, 12.3%, and 9.6%, respectively, while in cvs “Grillo” and “Moscato” no infection was found, possibly related to the different agronomic cultivation practices. Phylogenetic analyses showed that the 18 selected Sicilian GFLV-CP sequences were grouped in five sub-clades inside the same cluster with sequences from Italy and other countries. Recombination analyses detected only putative recombination events among Sicilian isolates, while nucleotide diversity analysis showed a very low differentiation within Italian isolates (0.1258 ± 0.008) and between isolates from Italy and from other countries. The hypothesis of negative selection was confirmed by the dN/dS ratio of 0.249. The results obtained highlighted a certain degree of variability within the sequences obtained, suggesting a different origin, probably due to the continuous interchange of plant propagation material with other Italian regions or European countries.
Genetic structure, molecular variability and spread of grapevine leafroll‐associated virus 1 and 3 in Sicilian autochthonous grapevine cultivars
A.G. Caruso 1 , S. Davino 1 , S. Bertacca 1 , A. Ragona 1 , N. Bova 1 , G. Mirabile 1 , L. Torta 1 , P. Bella 1 , S. Matić 2 , S. Panno 1
1Università degli Studi di Palermo, Dipartimento di Scienze Agrarie, Alimentari e Forestali, Viale delle Scienze, Edificio 5, 90133 Palermo, Italy; 2Institute for Sustainable Plant Protection, National Research Council (IPSP-CNR), Strada delle Cacce 73, 10135 Turin, Italy. E-mail: [email protected]
Grapevine leafroll disease is one of the most important viral diseases caused by several viruses of the Closteroviridae family, including grapevine leafroll‐associated viruses (GLRaV) 1 and 3. In this study, we investigated the genetic structure, molecular diversity and spread of GLRaV‐1 and GLRaV‐3 in 800 grapevine samples, collected from eleven autochthonous cultivars in four Sicilian provinces. Serological analyses were performed using GLRaV‐1/GLRaV‐3 polyclonal antibodies by DAS‐ELISA; molecular analyses were conducted to discriminate single and/or mixed infections, using specific primers for the CP gene of GLRaV-1 and 3. Thirty-three and 138 samples resulted positive to GLRaV‐1 and GLRaV‐3, with a 4.1% and 17.3% infection rate, respectively; only 15 samples showed a mixed infection, with a 1.9% incidence. Both viruses were not detected in ‘Grillo’ and ‘Moscato’ cultivars. Phylogenetic analyses conducted on 12 GLRaV‐1 selected CP sequences showed a close relationship with European isolates. The analyses conducted on 31 GLRaV‐3 selected CP sequences demonstrate a close relationship between Sicilian and isolates from different countries. Regarding GLRaV-1, the discrete nucleotide differentiation and positive selection (dN/dS: 3.147) could indicate a current increase in population fitness; a certain stability of GLRaV‐3 in Sicilian cultivars is suggested by the high nucleotide differentiation and negative selection (dN/dS: 0.2519). A different incidence of both viruses was detected in the main Sicilian cultivars, and it was the highest in ‘Carricante’ for GLRaV-1 and ‘Nero d’Avola’ for GLRaV-3. The absence of GLRaV-1 in ‘Grillo’, ‘Nerello Cappuccio’, and ‘Moscato’ could be due to a low prevalence of infected propagation material and a possible low presence of GLRaV-1 vectors.
Insights into the grapevine virus A spread in Sicily: epidemiological and evolutionary analysis
A.G. Caruso 1 , S. Panno 1 , S. Bertacca 1 , A. Ragona 1 , N. Bova 1 , G. Mirabile 1 , L. Torta 1 , P. Bella 1 , S. Matić 2 , S. Davino 1
1Università degli Studi di Palermo, Dipartimento di Scienze Agrarie, Alimentari e Forestali, Viale delle Scienze, Edificio 5, 90133 Palermo, Italy; 2Institute for Sustainable Plant Protection, National Research Council (IPSP-CNR), Strada delle Cacce 73, 10135 Turin, Italy. E-mail: [email protected]
Sicily is one of the most important Italian regions for the grapevine industry. Grapevines are threatened by several pests and pathogens including different viral diseases. Among these, phloem-restricted grapevine virus A (GVA) of the family Betaflexiviridae causes losses of 5–22%. In this study, the GVA epidemiology, genetic structure and molecular variability were ascertained in twenty commercial vineyards of four Sicilian provinces. Eleven autochthonous cultivars were investigated and a total of 800 samples were analyzed by end-point RT-PCR assay, using specific primers for the CP gene. Only forty-nine positive samples were detected, with an infection rate of 6.1%. Ten out of 11 cultivars analyzed resulted positive to GVA infection; a higher incidence was observed in ‘Nerello Mascalese’, ‘Carricante’, ‘Perricone’ and ‘Nero d’Avola’, with an infection rate of 30%, 12.8%, 7.7% and 4.8%, respectively, while in ‘Moscato’ cultivar no infection was found. Evolutionary relationships showed that the sixteen GVA-CP selected sequences were separated into two clusters. All Sicilian isolates were grouped into the same sub-clade with the other three Italian GVA isolates, showing a low variability. Recombination analyses identified putative recombination events among the Sicilian isolates analyzed, while the analysis of the nucleotide diversity showed a low differentiation within Italian isolates (0.0681 ± 0.015). The dN/dS ratio of 0.099 confirmed the hypothesis of negative selection. This study revealed a low variability degree among the Sicilian isolates, and the absence of candidate recombination events, suggesting a common origin, probably due to the exchange of infected propagation material within the Italian territory.
Inoculation of Citrus relatives ornamental rutaceous and rootstocks highlights a different replication rate of local CTV isolates
P. Caruso 1 , G.E. Massimino Cocuzza 2 , S. Di Silvestro 1 , C. Licciardello 1 , D. Puglisi 1 , M.C. Bazzano 3 , G. Scuderi 3 , A.F. Catara 4 , G. Licciardello 1
1 CREA—Council for Agricultural Research and Economics, Research Centre for Olive, Citrus and Tree fruit. 95127 Acireale (CT), Italy; 2 Dipartimento di Agricoltura, Alimentazione e Ambiente, Università degli Studi di Catania, via Santa Sofia 100, 95123 Catania, Italy; 3 Agrobiotech Soc. Coop. z.i. Blocco Palma I, Stradale Lancia 57, 95121 Catania, Italy; 4 Formerly, Department of Phytosanitary Science and Technologies, University of Catania, 95123 Catania, Italia. E-mail: [email protected]
The study of the effects of Citrus tristeza virus (CTV) on different host plants has been attempted in many citrus countries worldwide. Within the project SIRPA several ornamental rutaceous grafted onto seedling rootstocks were bark patch inoculated in greenhouse with a severe SY-VT isolate (P7/2) and its homologous and asymptomatic variant, cross protective on sour orange (M39D). Young plantlets included seven ornamentals rutaceous (Afraegle paniculata, Atalantia ceylanica, Clausena excavata, Eremocitrus glauca, Microcitrus australis, M. papuana, Severinia buxifolia, S. disticha), three species of Fortunella (F. margarita, F. obovata, and F. polyandra) and four species of Citrus (C. micrantha var. microcarpa, C. hystrix, C. madurensis and C. jambhiri). All the plants were monitored for 6 – 12 months and tested by ELISA and/or RT-real time PCR. ELISA tests of young new flushes taken at different times showed a fast and high replication in A. ceylanica, M. australis and M. papuana, Fortunella and Citrus species, whereas no replication was detected in A. paniculata, E. glauca, S. buxifolia. and S. disticha. Additional inoculations carried out on Troyer citrange and Poncirus trifoliata seedlings discovered a different replication rate of a subset of representative CTV isolates belonging to VT and T30 genotypes. RT-real time PCR tests, specific for detection of resistance breaking isolates, were negative. The study highlights the risk of aphid transmission in a system which includes many potential host reservoirs of the virus.
This research is part of the project S.I.R.P.A (n. 08CT7211000254) funded by PO FESR Sicilia 2014–2020, action 1.1.5.
Microbiome-informed selection of biocontrol bacteria from the tomato endophytome
V. Catara 1 , D. Nicotra 1 , A. Mosca 2 , A. Anzalone 1 , G. Dimaria 1 , G.F. Privitera 2 , A. Pulvirenti 3
1Department of Agriculture, Food and Environment, University of Catania, Catania, Italy;2Department of Physics and Astronomy University of Catania, Catania, Italy; 3Department of Clinical and Experimental Medicine, University of Catania, Catania, Italy. E-mail: [email protected]
Metagenomics by next generation sequencing (NGS) provided a window into the microbial community of cultivated plants and changed our belief of the complex interactions in the microbiome. Keystone operational taxonomic units (OTUs) in this context represent highly connected taxa, co-expressed in the same conditions, which exert a considerable influence on microbiome structure and could be identified by microbial co-occurrence network analysis. Here we present the data analysis of tomato root bacterial communities obtained by amplicon-based metagenomics and the identification of core taxa in two different growing conditions, i.e. either in soil or soilless cultivation. The properties of the network structure were evaluated. Microbial networks were compartmentalized into several “network modules” and the keystone OTUs identified. The next step was to align the sequences of the OTUs to those from a cultivable bacterial collection obtained from the tomato endosphere to find their role in the network. Purpose of the third step was to select a subset of bacterial endophytes based on bacterial community data such as abundance (either highly abundant or rare taxa), number or type of connections in the co-occurrence network as well as in vitro plant growth promotion (ACC deaminase, siderophores, phosphate solubilization, etc.) and biocontrol activities against tomato pathogens (bacteria and fungi). Microbial consortia to be used as bioinoculants were formed from these data after the evaluation of the in vitro growth compatibility between the selected bacteria. Individual and sum effects of microbial consortia are being evaluated in vivo for biotic and abiotic stresses.
Temperature plays a decisive role in the ability of Pseudomonas syringae strains carrying a recognized avirulence gene to induce a hypersensitive response in Arabidopsis thaliana
E. Caullireau 1,2 , D. Danzi 1 , V.M. Tempo 1 , C.E. Morris 2 , E. Vandelle 1
1 Department of Biotechnology, University of Verona, Strada Le Grazie 15, 37134 Verona, Italy; 2 Plant Pathology Research Unit, INRAE UR407, 84140 Montfavet, France. E-mail: [email protected]
The phytopathogenic species complex P. syringae has a very important economic impact, causing diseases on many important crops and woody plants. The P. syringae complex is currently divided into phylogenetic groups, and according to a certain recognized host specificity, they have been divided into more than 50 pathovars. A considerable research effort is invested in understanding how effector repertoires could determine the range of plants that a given strain can infect. Recently, we showed that the incapacity of P. syringae pv. actinidiae to induce the hypersensitive response (HR) in A. thaliana is due to its incapability to inject effectors rather than to the absence of a recognized effector. In this context, we report here results from the comparison of different P. syringae strains, belonging to different phylogroups and carrying the same plasmid-borne avirulence gene, for their capacity to induce an HR in A. thaliana Col-0, at different temperatures. Pto DC3000 and Pma M6 consistently trigger a strong hypersensitive cell death, while the other strains induce an HR, at different intensities, significantly dependent on temperature. Surprisingly, differences were also observed among quasi-clonal strains. These results i) highlight the necessity to study bacterial virulence in a broader set of strains and reveal that Pto DC3000 is a reliable model strain but not representative of the Pseudomonas complex, ii) support the notion that the presence/absence of effectors is not sufficient to predict the outcome of plant-bacteria interactions, and iii) indicate that temperature may play a crucial role in regulating effector injection.
Effect of different light spectra on Zymoseptoria tritici wheat colonization
M. Cerón-Bustamante, F.Tini, G. Beccari, P. Benincasa, L. Covarelli
1 Dipartimento di Scienze Agrarie, Alimentari e Ambientali, Università degli Studi di Perugia, Borgo XX Giugno 74, 06121 Perugia. E-mail: [email protected]
Light plays a crucial role in the growth and development of fungi. Zymoseptoria tritici, the causal agent of Septoria tritici blotch, is capable to sense and respond to specific wavelengths by modulating gene expression. To study the effect of light on Z. tritici infection, bread wheat plants of the susceptible variety A416 were inoculated with a conidial suspension of 107 ml–1 and grown under 100% red (627 nm) (R100), 100% blue (470 nm) (B100), blue:red (B50:R50), and white lights (W), with a photoperiod of 12 h, as well as in the dark (D). The overall photon flux density in all light treatments was 200 micromoles m−2 s−1. Fungal DNA was detected by qPCR at seven days after inoculation in leaf samples from all treatments. To investigate if light directly affected Z. tritici, fungal cultures were grown on PDA under the same light conditions and measured after 14 dpi. The highest disease severity was observed in B100 plants, followed by B50:R50 and W ones, while no symptoms were observed on R100 plants. In agreement with symptom expression, a steady increase of Z. tritici DNA amount was recorded in B100 and B50:R50 plants up to 21 dpi, while the lowest fungal DNA concentration was detected in R100 plants. In contrast to the in planta results, the B100 treatment reduced the in vitro growth of the pathogen, while no significant differences were observed among R100, W and B50:R50 treatments. These results suggest that the reduced infection levels obtained under the red wavelength is probably related to the host response, as already observed for other foliar pathogens on rice plants.
LIFE MICROFIGHTERS: an EU funded project for the implementation and use of innovative Zeo-biopesticides, based on beneficial microorganisms, as an alternative to the use of copper-based products
A. Cerrato 1 , D. Giovanardi 2 , M. Coltorti 3 , E. Stefani 2
1Symbiagro srl, via dell’Artigianato 1/A, 25030 Roncadelle, Brescia, Italy; 2Biogest-Siteia, University of Modena and Reggio Emilia, via Amendola 2, 42122 Reggio Emilia, Italy; 3Department of Physics and Earth Sciences, via Saragat 1, 44122 Ferrara, Italy. E-mail: [email protected]
Copper, as a pesticide, has been used in agriculture for more than 150 years. During the last three decades, limitations in its use have been imposed by the European Commission, in order to avoid its accumulation in agricultural ecosystems. Meanwhile, alternatives to copper–based chemicals have been suggested, e.g. by using microbial biocontrol agents or by applying natural zeolites to the crops. The LIFE MICROFIGHTER project was conceived to demonstrate in open field, and on different crops, the effectiveness of innovative pesticide formulations based on the intimate association of bacterial antagonists with zeolites micro-particles, which will support bacterial viability and efficacy after their spray on the crop canopy. The pathogen-crop association considered by the project are: Plasmopara viticola and grapevine; Pseudomonas syringae pv. tomato and Xanthomonas vesicatoria on tomato; Pseudomonas savastanoi pv. savastanoi and olive. The Zeo-biopesticides will be used to control pathogens in vineyards, tomato fields and olive groves located in Italy, Croatia and Spain, and results will be compared to those obtained by using copper compounds according the local IPM. Copper dynamics, as an environmental pollutant, will be studied in all fields where the experiments are conducted, in order to confirm its reduction in the agricultural environments. The project has a duration of 42 months and 9 partners are participating: Symbiagro (coordinator, Roncadelle, Italy), UNIMORE (Reggio Emilia, Italy), UNIFE (Ferrara, Italy), UNIZD (Zadar, Croatia); Astra Innovazione e Sviluppo (Faenza, Italy), Consorzio Agrario di Ravenna (Cotignola, Italy), Ibe-CNR (Bologna, Italy), Consorcio Cooperatives Agroalimentaries (Valencia, Spain); CO&SO (Firenze, Italy).
This research is funded by the European Commission, under the LIFE-2021-SAP-ENV Call (Circular Economy and Quality of Life—Standard Action Projects).
In vitro efficacy of natural compounds on plant pathogenic bacteria affecting kiwi and olive crops
E. Cesari 1 , F. Costantini 2 , G. Tatulli 2 , V. Scala 2 , S. Loreti 2 , M. Scortichini 1 , N. Pucci 2
1 Council for Agricultural Research and Economics – Research Centre for Olive, Fruit and Citrus Crops, Rome, Italy; 2 Council for Agricultural Research and Economics – Research Centre for Plant Protection and Certification, Rome, Italy. E-mail: [email protected]
Plant bacterial diseases have always led to major economic losses for producers worldwide. In Italy, two relevant outbreaks are still affecting Actinidia crops: kiwifruit canker and blight disease, caused by Pseudomonas syringae pv. actinidiae (Psa) and Pseudomonas viridiflava (Pv), respectively. Another severe epidemic is represented by the ongoing Olive Quick Decline Disease (OQDS) and the control of his causal agent, Xylella fastidiosa (Xf), continues to be a priority in the management of Apulian olive crops. Nowadays, the use of copper-based formulations and an early disease diagnosis are the only two strategies for a preventive control. It is therefore important to choose a more eco-compatible approach. For this purpose, we investigated the in vitro antibacterial effect of natural molecules and resistance inductors (BABAⓇ, Atholio BioⓇ, Trametes versicolor extract, chitosan, algae and essential oil based formulations, clove oil, free-fatty acids, diacylglycerides and oxylipins) and synthetic compounds with low environmental impact (BiozonⓇ, DentametⓇ and propionic acid). A preliminary screening of different methodologies was carried out to evaluate the best one for Xf, a slow-growing bacterium difficult to cultivate. Among the several assessed methods (broth macrodilution, broth microdilution, agar dilution assay, agar disk diffusion and a novel bacterial culture method based on microfluidic channels), broth macrodilution is the most accurate for Xf. On the other hand, microfluidic channels permit a real-time observation of the bacterial behaviour under continuous-flow conditions. The overall results highlight that most of the tested products show a good antimicrobial efficacy, suggesting a possible future open field application.
This work was supported by the project SALVAOLIVI—MIPAAF (D.M. 33,437 del 21–12-2018).
Viroid classification: Reassessment of the species demarcation criteria
M. Chiumenti, B. Navarro, F. Di Serio
Istituto per la Protezione Sostenibile delle Piante, Consiglio Nazionale delle Ricerche, Via Amendola 122/D , Bari, Italy. E-mail: [email protected]
Viroids are small (240–430 nt), circular, single-stranded, non-coding RNAs that can infect plants and induce symptoms. Since their discovery over 50 years ago, more than forty viroids have been identified. Similarly to viruses, viroid populations in a single host exist as a group of variants closely related but slightly differing from each other, thus behaving as a quasi-species. To take into account the intraspecies variability, the early demarcation threshold for viroid species was set at 90% of sequence identity on the overall genome in addition to divergent biological evidence (i.e., host range, symptoms) compared to the closest related viroid. High-throughput sequencing technology allowed the identification of many new viroids not eliciting symptoms in their hosts with narrow experimental host range, thus making difficult the fulfilment of the biological criteria needed for the establishment of a new viroid species. To overcome this problem, pairwise identity scores (PWISs) thresholds at genus level were tested as potential species demarcation criteria. We demonstrated that applying these thresholds as a major species demarcation criterion did not modify the current classification and allowed the assignment of the unclassified viroids to a known or a new species. Starting from 2021, the proposed thresholds have been adopted by the International Committee on Taxonomy of Viruses (ICTV). The adoption of these demarcation criteria allowed the establishment of ten new viroid species. Such approach based on pairwise identity matrices will facilitate the classification of new viroids, limiting the need of providing biological evidence only to the most complicate cases.
This project has been partially funded from the European Union’s Horizon 2020 Research and InnovationProgram under the Marie Skłodowska-Curie grant agreement no. 734736. This article reflects only the author’s view, and the agency is not responsible for any use that may be made of the information it contains.
Molecular characterization and effectiveness evaluation of Aureobasidium spp. strains against brown rot of stone fruit
R. Cignola, A. Sadallah, M. Martini, A. Di Francesco
Department of Agriculture, Food, Environmental and Animal Sciences, University of Udine, Via delle Scienze 206, 33100 Udine, Italy. E-mail: [email protected]
Given the need to reduce the use of chemicals during the postharvest phase, the study aimed to exploit the natural diversity of Aureobasidium spp. strains, identifying new Biocontrol Agents (BCAs) as more suitable or simply alternative or complementary to those already available. For these reasons, an alternative defence strategy was investigated. A population of 40 epiphytic and endophytic strains of Aureobasidium spp., isolated from common plants (e.g., olive, laurel, plum tree) during the winter season at low temperatures, was molecularly characterized by multi-locus sequence analysis (ITS, ELO and EF1α). The most representative strains, belonging to every single cluster and different subspecies, were tested as active BCAs by in vitro and in vivo assays against Monilinia spp. of stone fruit. Molecular results displayed, in a population consisting mainly of A. pullulans, the presence of an unusual species: A. namibiae (UC14 strain). According to the results of in vitro and in vivo assays, the selected strains showed a certain ability to control brown rot diseases at different levels of effectiveness. In fact, our results demonstrated that Aureobasidium spp., isolated during the winter season, could be more capable of withstanding low temperatures, a key feature for the storage. In addition, from the unusual source of isolation, we detected a greater and more interesting potential of species diversification.
PvDCL1/2 dsRNA as an eco-friendly approach for downy mildew control in grapevine
A. Ciofini 1 , Z.M. Haile 1,2 , D.E. Gebremichael 1 , F. Negrini 1 , L. Capriotti 3 , B. Molesini 4 , F. Negrini 1 , M. Collina 1 , S. Sabbadini 3 , B. Mezzetti 3,5 , E. Baraldi 1
1 Department of Agricultural and Food Sciences (DISTAL), University of Bologna, Bologna, Italy; 2 Ethiopian Institute of Agricultural Research (EIAR), Addis Ababa, Ethiopia; 3 Department of Agricultural, Food and Environmental Sciences, Polytechnic University of Marche, Ancona, Italy; 4 Department of Biotechnology, University of Verona, Verona, Italy; 5 Research Group on Food, Nutritional Biochemistry and Health, Universidad Europea del Atlántico, Santander, Spain. E-mail: [email protected]
Downy mildew caused by Plasmopara viticola is the most serious fungal disease of grapevine and its control requires the use of a large amount of pesticides throughout the vegetative season with a consequent high environmental impact. During the last decades, the RNA interference (RNAi) mechanism was exploited for the development of eco-friendly approaches for pest management in crops paving the way toward innovative and sustainable alternatives to traditional toxic Plant Protection Products (PPP). In this work, we show the potential of dsRNA targeting P. viticola Dicer-like (DCL) homolog genes for grapevine protection against downy mildew. The topical application of the chimeric dsRNA construct PvDCL1/2 at different concentrations (75, 100, and 125 ng/µl) was correlated with a significant reduction of the pathogen virulence. In particular, grapevine leaves treated with PvDCL1/2 and subsequently inoculated with a P. viticola suspension (1*105 sporangia/ml) exhibited lower downy mildew incidence and severity compared to those subjected to treatments with water (negative control) or unspecific dsRNA (BcDCL1/2, positive control). The RNAi mechanism exerted by PvDCL1/2 dsRNA was confirmed by the significant reduction of PvDCL1 and PvDCL2 transcript levels in infected tissues previously treated with these molecules compared to control samples. Besides the preventive effect, PvDCL1/2 dsRNA also exhibited a curative action by reducing downy mildew progression of already established infections. Therefore, our results suggest that PvDCL1/2 dsRNA is efficient in controlling P. viticola infections and can be considered as a promising candidate to substitute toxic pesticides.
Use of RNAi-based protection strategies against fungal pathogens
A. Ciofini 1 , F. Negrini 1 , Z.M. Haile 1,2 , D.E. Gebremichael 1 , L. Capriotti 3 , F. Negrini 1 , M. Collina 1 , S. Sabbadini 3 , B. Mezzetti 3 , E. Baraldi 1
1 Department of Agricultural and Food Sciences (DISTAL), University of Bologna, Bologna, Italy; 2 Ethiopian Institute of Agricultural Research (EIAR), Addis Ababa, Ethiopia; 3 Department of Agricultural, Food and Environmental Sciences, Polytechnic University of Marche, Ancona, Italy; E-mail: [email protected]
Plant pathogenic fungi represent a serious and constant threat to the agri-food sector worldwide and traditionally their control has been approached with chemical fungicides, harmful to both the environment and human health. During the last decades, the urgency of their replacement with eco-friendly tools led to the investigation of control strategies based on the RNA interference (RNAi), a post-transcriptional silencing mechanism that, acting in a very specific manner, allows to downregulate the expression of key genes and consequently arrests the infective process of plant pathogens. In particular, two main approaches have been explored: Host-Induced Gene Silencing (HIGS) and Spray-Induced Gene Silencing (SIGS). HIGS relies on the development of Genetically Modified (GM) crop plants expressing siRNAs that target pathogen key genes, while SIGS is based on the topical application of interfering dsRNA molecules in replacement of toxic fungicides. Of course, the high specificity of the RNAi mechanism requires the case-by-case identification of genes and gene regions efficient to properly arrest the infective process. Our studies focus on the detection of the most efficacy sequences to be used to develop RNAi-based strategies to control main pathogens (Plasmopara viticola, Botrytis cinerea, Phytophthora infestans, Stemphylium vesicarium, Colletotrichum spp., and Podosphaera aphanis) of economically strategic crops for our country, such as grapevine, tomato, pear, and strawberry. Results suggest that this approach represents a promising candidate for the implementation of sustainable pathogen management strategies, although several issues remain to be addressed for their field application.
Microbial BCA-PGP inoculants on food crops in Kenya and evaluation of rhizosphere microbiota
E. Comite 1 , S. Kamau 2 , F. De Filippis 1 , G. Sequino 1 , A. Bottiglieri 1 , N.N. Karanja 2 , S.L. Woo 3
1 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 2 Department of Land Resource Management and Agricultural Technology (LARMAT), College of Agriculture and Veterinary Sciences, University of Nairobi, P.O. Box 29053, Nairobi 00625; 3 Department of Pharmacy, University of Naples “Federico II”, 80131 Naples, Italy. E-mail : [email protected]
In Kenya, agronomic techniques used in small farms are manual and fertilizers are manures applied to soils, the availability of water is strongly related to the two rainfall seasons. Microbial inoculants may improve the assimilation of nutrients/water in conditions of abiotic stress, enhance plant fitness and disease control. In order to test small-scale agriculture systems operating in limited-water conditions, the seed of two food crops (green gram and sorghum) were treated with fungal spore suspensions of Trichoderma afroharzianum strain T22 and T. asperellum, known biocontrol agents (BCA) and plant growth promoters (PGP). The aim of this study was to characterize the rhizosphere microbiota of plants treated with manure and Trichoderma after short or long rain seasons in Kaliku-Kenya. DNA was extracted from soil rhizosphere samples, bacterial and fungal diversity were assessed by Illumina amplicon sequencing, then analyzed by Qiime software. Actinobacteria were the most abundant bacterial phylum in both crops, followed by Firmicutes and Proteobacteria; in particular, all these phyla and Chloroflexi were reduced during the long-rain season, while Acidobacteria, Planctomyces and Gemmatimonadetes increased. Ascomycota was the most abundant fungal phylum, with Dothideomycetes and Sordariomycetes demonstrating a seasonal pattern in the rhizosphere of green grams, with a negative correlation to the rainfall lengths. Trichoderma abundance increased in long-rains and there was a reduction in the relative abundance of the soilborne phytopathogen Rhizoctonia, which was absent in sorghum and T22-treated green gram. Thus, there was a variation of rhizospheric microbiota due to rain seasonality and microbial inoculants.
Molecular markers for differentiation of a ‘ Candidatus Phytoplasma solani’ strain associated with “bois noir” disease in Sicily
N. Contaldo, G. Tranchina, A. Bertaccini
Alma Mater Studiorum – University of Bologna, Department of Agricultural and Food Sciences, Bologna, Italy; E-mail: [email protected]
“Bois noir” (BN) associated with the presence of ‘Candidatus Phytoplasma solani’ is one of the most important and widespread grapevine yellows diseases in the Euro-Mediterranean region. Its epidemiology is complex since it is related to interactions of the pathogen with host plants and insect vectors and its spread through propagation material. During monitoring in a vineyard located in Palermo province (Sicily) symptomatic grapevine plants sampled tested positive for 'Ca. P. solani '. The detected phytoplasmas were then subjected to a multigenic characterization on the tuf, stamp, vmp1 and secY genes to study their molecular variability in relationship with possible environmental and epidemiological aspects. The vmp1 gene provided 3 restriction profiles (V4-V11-V12) confirming results reported in Sicily, Sardinia and central western Italy. No variability was present on the 16S rRNA and tuf genes (tuf type b1). On stamp and secY sequences it was possible to verify the presence of a molecular variant having two SNPs to the reported St4 and 99.75% identical to strain P-TV on secY gene. The stamp gene encodes a membrane protein involved in phytoplasma adaptation to different vector species, and the Sicilian variant is different from those identified in various Italian regions. The BN vector Hyalesthes obsoletus is not common in Sicily and this variant might suggest phytoplasma adaptation to insect vectors other than H. obsoletus. This stamp and secY variant is therefore an excellent molecular marker for epidemiological studies that are the basis to improve BN prevention and control in Sicily.
Research founded from EU H2020 R & I programme under the project “Insect-borne prokaryote-associated diseases in tropical and sub-tropical perennial crops—TROPICSAFE” Grant Agreement No 727459.
Characterization of micotoxins produced by Fusariun proliferatum and F. sacchari
S. Conti Taguali 1,2,3 , M. Riolo 1,2,3 , F. Aloi 1 , R. Parlascino 1,3 , A. Pane 1 , E. Santilli 3 , J.M. Quiles 4 , G. Meca 4 , S.O. Cacciola 1
1 Department of Agriculture, Food and Environment (Di3A), University of Catania, Via Santa Sofia 100, 95123 Catania, Italy; 2 Department of Agricultural Science, Mediterranean University of Reggio Calabria, Localitá Feo di Vito, 89122 Reggio Calabria, Italy; 3 Council for Agricultural Research and Agricultural Economy Analysis, Research Centre for Olive, Citrus and Tree Fruit- Rende CS (CREA-OFA), 87036 Rende, Italy; 4 Department of Preventive Medicine, University of Valencia, Av. Vicent Andrés Estellés s/n, 46100 Valencia, Spain. E-mail: [email protected]
The genus Fusarium is responsible for many serious pre- and post-harvest diseases of agricultural crops. In this respect, product losses along the post-harvest supply chains (storage, transport, and final distribution) also result in severe consequences for product quality. Harvesting and handling of the fruit is, therefore, an extremely important process, due to the risk that lesions may promote the development of fungal agents that rapidly deteriorate the fruit by also producing mycotoxins. In an open-air market in Catania (southern Italy), lady finger banana fruit showed dark circular spots from which Fusarium sacchari and F. proliferatum were isolated. The aim of this study was to test the pathogenicity of F. sacchari and F. proliferatum, and to characterize the mycotoxins produced by both species. The characterization of the mycotoxins was carried out by individually inoculating with either species, F. sacchari and F. proliferatum, corn and barley. Seven mycotoxins were detected from the samples inoculated with F. sacchari (ENB2, ENB3, ENB4, FA1, FB2, HFB1 and FC), of which 2 (ENB2 and FC) in both matrices, while FA1 only in maize and ENB3, ENB4, FB2 and HFB1 only in barley. On the other hand, seven mycotoxins (FUS, FA1, FA2, FB1, HFB1, FC and MON) were detected from the samples inoculated with F. proliferatum, of which FHB1 and MON only from barley.
Genome-barcoding (NGS) of quarantine bacterial pathogens relevant to the European Union: development and application with Nanopore methodology
V. Crosara 1 , V. Scala 2 , M. Reverberi 1 , L. Faino 1 , S. Loreti 2
1 Department of Environmental Biology, Sapienza University of Rome, 00185 Rome, Italy; 2 Council for Agricultural Research and Economics, Research Centre for Plant Protection and Certification, Via C.G. Bertero 22, 00156 Rome, Italy. E-mail: [email protected]
Globalization and climate change are accelerating the spread of plant pathogens in new areas. Keeping the expansion of these pathogens under control is fundamental because they impact both the conservation of ecosystems and nations’ economy. Their spread often comes from asymptomatic plant materials so, to prevent it, it is necessary to improve detection and identification methodologies. In this study, quarantine relevant pathogens Xylella fastidiosa, Xanthomonas citri and Pantoea stewartii have been considered. It is key to identify these pathogens at sequence-type, pathovar and subspecies levels, as these provide insight on their host specificity as well as potential damage. Therefore, this work aims at developing a sensitive and specific diagnostic system for a precise identification even in asymptomatic conditions. We decided to rely on MinION from Oxford Nanopore Technology, a portable, fast, and easy to use device, paired with an ad hoc bioinformatics pipeline. An amplicon-Nanopore sequencing, based on selected housekeeping genes (e.g. MLST), has been developed. These selected genes allow the distinction of the pathovar, subspecies or ST of the different tested bacteria. Several plant species have been spiked with different bacterial suspensions at known concentrations. Then, a multiplex- (for Xylella fastidiosa) and duplex- (Xanthomonas citri) PCR amplification of housekeeping genes has been developed. The prepared samples were finally sequenced with Nanopore to test the workflow. Preliminary results indicate that this approach is promising for the detection and identification of these priority pathogens. Further study will be done to determine the limit of detection and specificity.
LAMP assay for in-field detection of Flavescence dorée phytoplasma in Scaphoideus titanus vector
C. D’Errico 1 , V. Candian 2 , R. Pierro 1 , S. Panno 3 , S. Davino 3 , E. Noris 1 , R. Tedeschi 2 , S. Matić 1
1 Institute for Sustainable Plant Protection, National Research Council, Strada delle Cacce 73, 10135 Turin, Italy; 2 Department of Agriculture, Forestry and Food Science DISAFA—Turin University, Largo Braccini 2, 10095 Grugliasco, Italy; 3 Department of Agricultural, Food and Forest Sciences, University of Palermo, Viale delle Scienze, Ed. 4, 90128 Palermo, Italy. E-mail: [email protected]
Flavescence Dorée is one the most notorious diseases of grapevine, caused by EPPO quarantine Flavescence Dorée phytoplasma (FDp) belonging to the 16Sr-V group. Although many molecular techniques such as Loop-Mediated Isothermal Amplification (LAMP) have frequent use in the rapid detection of FDp in infected grapevine plants, there is still a lack of developed isothermal amplification protocol for FDp detection in the insect vectors. In this study, a simple in-field real-time LAMP assay was developed for the detection of FDp in the Scaphoideus titanus vector. The LAMP assay was optimized to work with crude insect extracts obtained with 5 min-manual shaking of a single insect in the buffer. An easy, sensitive, specific, low-cost and point-of-care LAMP assay was able to detect FDp in S. titanus in less than half an hour directly in the infected Piedmont vineyards. These results enable new possibilities for FDp detection in both, plants and vectors, directly in the field without the requirement of highly specialized personnel, at relatively low operating costs.
This study was funded by Fondazione Cassa di Risparmio di Torino, Project FD_LAMPvett, Ref. No. 2020.2066.
Psa3 virulence reduction by natural molecules: insights on subtle regulation mechanisms
D. Danzi 1 , M.R. Puttilli 1 , J. Brandi 1 , D. Cecconi 1 , M. Manfredi 2 , E. Marengo 3 , A. Polverari 1 , E. Vandelle 1
1 Department of Biotechnology, University of Verona, Strada Le Grazie, 15, 37134 Verona, Italy; 2 Department of Translational Medicine – Center for Translational Research on Autoimmune & Allergic Diseases – CAAD, University of Piemonte Orientale, Corso Trieste 15, 28100 Novara, Italy; 3 Department of Science and Technological Innovation – University of Piemonte Orientale, Viale T. Michel 1511, 15121 Alessandria, Italy. E-mail: [email protected]
Pseudomonas syringae pv. actinidiae (Psa) is the causal agent of kiwifruit bacterial canker disease, and currently, the disease management relies on copper-based compounds, which could force the insurgence of copper-resistance Psa strains, as already observed with other plant pathogens. Thus, new control strategies should target bacterial virulence mechanisms to weaken pathogens rather than prevent their growth in planta. One of the most important factors of virulence is the ability of the bacterium to inject effectors into the plant cells through the type III secretion system (T3SS). Thus, exploiting a reporter system based on the GFP expression driven by the promoter region of the hrpA1 gene, which encodes one of the major components of the T3SS, a chemical library composed of 502 natural molecules was screened looking for those able to reduce the T3SS induction in the most aggressive Psa biovar 3, without affecting its growth. Among selected candidates, dicoumarol showed the ability to reduce the hypersensitive response (HR) in Arabidopsis thaliana, further supporting the hypothesis of T3SS inhibition. Moreover, a proteomic analysis revealed that dicoumarol reduces the secretion of proteins by Psa3 cultured in minimal medium, in particular the injectisome proteins HrpZ and proteins of the flagellum-related T3SS, thus suggesting the inhibition of a common mechanism involved in the regulation of both secretion systems. The possible target of dicoumarol will be discussed.
eCPMV nanoparticles: the potential of a bio-inspired strategy for plant protection
D. Danzi, R. Zampieri, A. Polverari, L. Avesani, E. Vandelle
Department of Biotechnology, University of Verona, Strada le Grazie 15, 37134 Verona, Italy. E-mail: [email protected]
Nano-agrochemicals are an unprecedent opportunity for crop disease management since they can provide a valuable alternative to common pesticides, due to a targeted and enhanced efficacy, resulting in a lower dosage of active substances and a reduction of possible off-target effects. Plant viruses can play an important role in this context since they are naturally occurring proteinaceous nanoparticles able to carry, protect, and deliver a cargo. Moreover, the biosafety concerns about the use of plant viruses for crop protection have been exceeded by the development of nucleic acid-free virus nanoparticles, named empty virus-like particles (eVLPs). (CPMV) can be produced as eVLPs (eCPMV) at high titres through transient expression in Nicotiana benthamiana plants. Additionally, eCPMV is prone to a vast array of functionalization by chemical and genetic engineering. In this work, eCPMV nanoparticles have been exploited in an attempt to develop new products for crop protection, combining the physical stability of the eCPMV capsid with natural bioactive peptides and molecules. Particularly eCPMV has been exploited for three purposes: (i) exposure of antimicrobial peptides (AMP); (ii), as a plant immunity-triggering nanoparticle, and (iii) eCPMV as a functional nanocarrier. Overall, the data enforce to reconsider the paradigms regarding eCPMV functionalization, in particular in terms of peptide features required for genetical modification. Moreover, this work supports the idea of eCPMV as a promising tool to develop new nano biopesticides applicable in agriculture.
Evaluation of copper, zinc and organic fertilizers for their side-effect against Xylella fastidiosa subsp. pauca in vitro and on naturally infected olive plants
C. Del Grosso 1,2 , M. Saponari 2 , R. Abou Kubaa 2 , G. Lima 1
1 Department of Agricultural, Environmental and Food Sciences, University of Molise, Via F. De Sanctis, 86100 Campobasso, Italy; 2 Institute for Sustainable Plant Protection, National Research Council, via Amendola 122/D, 70126 Bari, Italy. E-mail: [email protected]
Xylella fastidiosa (Xf), a dangerous xylem-limited bacterium, can infect more than 600 plant species. The bacterium was first discovered in Europe in 2013 in the Apulia region (southern Italy) where the outbreak of X. fastidiosa subsp. pauca (Xfp) raised major concerns due to the deadly disease caused on infected olive trees (Olea europaea). The lack of effective therapeutic tools to cure infected plants is also a key factor boosting research for practical solutions to contain the infections. In this framework, the screening for new antibacterial products such as zinc and copper phosphites, and an organic fertilizer based on a balanced mixture of seaweeds and vegetal polyphenols, evidenced a broad-spectrum antibacterial activity in vitro against various species of phytopathogenic bacteria. Through the optimization of specific protocols, the selected products were then tested in vitro also for their antimicrobial activity against Xf strains belonging to different subspecies. Although with some differences related to Xf strain or subspecies, results confirmed the broad antibacterial activity of the tested compounds. Subsequently, in the framework of an integrated control approach, these products were tested in the field in naturally Xfp-infected olive orchards in the Salento Peninsula. Upon two years of field assessments, positive effects of the application were recorded in terms of appreciable improvement of vegetative, productive and phytosanitary conditions of the treated plants. However, further observations are needed to assess the long-term effect and sustainability of selected products application, while additional researches are needed to optimize their formulation/application and understanding the mechanism of action.
This research was/is supported by the following projects: OLIDIXIIT, “Olivicoltura e difesa da Xylella fastidiosa e da insetti vettori in italia" – MiPAAF—DM n. 23,773/2017; DEMETRA, “Ideazione e validazione di sistemi produttivi multifunzionali e diversificati basati sull’integrazione tra produzioni vegetali e animali nelle aree marginali dell’Italia centro-meridionale. MiPAAF, DM 27 settembre 2018 n. 67,374; XF-ACTORS “Xylella fastidiosa Active Containment Through a multidisciplinary-Oriented Research Strategy” H2020 GA 727,987.
A rapid lab screening to select biological and integrated tools to be used against Phytophthora cinnamomi root rot of holm oak and other woody plants
C. Del Grosso 1,2 , D. Palmieri 1 , L. Marchese 1 , F. De Curtis 1 , G. Lima 1
1 Dipartimento Agricoltura, Ambiente e Alimenti, Università del Molise, Via F. De Sanctis, 86100 Campobasso, Italy; 2 Istituto per la Protezione Sostenibile delle Piante, Consiglio Nazionale delle Ricerche, via Amendola 122/D, 70126 Bari, Italy. E-mail: [email protected]
Phytophthora species, as well as other telluric pathogens, are harmful microorganisms very difficult to control with conventional means. The low effectiveness of chemical formulates is principally due to the development of strains resistant to fungicides. Furthermore, the use of resistant plants has several limitations due to the extreme ability of Phytophthora species to mutate and overcome host resistance. These factors together with the risks and negative impact on human health and the environment underline the need to develop new screening methods and sustainable control strategies. In this work, antagonistic bacterial strains and fertilizers with collateral antimicrobial activity, already selected for their antagonistic/inhibitory activity against fungal and/or bacterial pathogens, were evaluated against P. cinnamomi. Assays were carried out in vitro, on agarized media as well as in vivo on pot grown young holm oak plants. Moreover, as a new rapid lab screen in vivo we proposed the use of apple fruit. More specifically, surface sterilized apples were artificially wounded, treated with biocontrol or chemical products and inoculated before (curative tests) or after (preventive test) with the pathogen. Several experiments have shown a good agreement between the rapid fruit screening and the classic plant test method based on treated and artificially inoculated olm oak plants. Therefore, the fruit method could be used as a new alternative model in vivo to pre-select/evaluate new biological or integrated tools/strategy.
First report of Fusarium pseudograminearum causing root and crown rot in the halophyte Salicornia europaea
E. Delli Compagni, A. Pardossi, S. Pecchia
Department of Agriculture, Food and Environment, University of Pisa, Via Del Borghetto 80, 56124 Pisa, Italy. E-mail: [email protected]
Salicornia europaea L. (glasswort), a euhalophyte in the Amaranthaceae family, is a valuable green vegetable. In July 2021, an outbreak of root and crown rot disease occurred on S. europaea grown in peat-filled pots under greenhouse. Symptoms appeared on 20–25% of 6-month-old plants. The fungus was identified as F. pseudograminearum by means of morphological observations and molecular sequence analysis based on tef-1α gene (EF-1/EF-2) and using species-specific PCR primers (Fp1-1/Fp1-2). This pathogen is known as the causal agent of crown rot in cereals and has sporadically been reported on wheat plants and seeds of soybean and vetch in Europe. A pathogenicity test was then conducted in a growth chamber to fulfill Koch’s postulates. Forty-eight seedlings (57 days after sowing) were grown in aerated non-saline nutrient solution in which a suspension of F. pseudograminearum macroconidia had been poured (final concentration 105 ml−1). Other 48 plants (controls) were grown hydroponically in a separate growth chamber and inoculated with sterile distilled water. Twenty-four days after inoculation (dpi), half of control and inoculated plants was transferred into a new sterile nutrient solution while the other half was transplanted into pots filled with sterilized peat. After 80 dpi, 100% of pot-grown plants showed root and crown rot symptoms whereas only 70% of infected hydroponically-grown plants developed symptoms. No evidence of disease was observed in the controls. F. pseudograminearum was consistently re-isolated from diseased plants in both cultivation systems (64.5–83.0%). Further investigations are in progress on this new pathosystem in saltwater hydroponics.
Advances in plant disease detection: pulse thermography as new tool to predict Botrytis cinerea infection in pepper and tomato
C. Del Regno 1 , A. De Rosa 1 , R. Curcio 2 , A. Russo 3 , F. Mele 4 , E. Cappetta 1 , P. Mormile 2 , M. Rippa 2 , M. Ruocco 4 , A. Ambrosone 1 , A. Leone 1
1 Dipartimento di Farmacia, Università di Salerno, Fisciano (SA); 2 Istituto di Scienze Applicate e Sistemi Intelligenti, Pozzuoli (NA), (CNR-ISASI); 3 Dipartimento di Agraria Università di Napoli “Federico II”, Portici (NA); 4 Istituto per la Protezione Sostenibile delle Piante SS Portici (CNR-IPSP), Piazzale Enrico Fermi I, 80055 Portici (NA). E-mail: [email protected]
Botrytis cinerea is a necrotrophic pathogen characterized by short life cycle, high reproduction and genetic variation. Infrared thermal imaging is a non-destructive and fast technique, which holds great promise for the detection of pathogen attacks in plants. Herein, we evaluated and predicted the development of gray mold in pepper (Capsicum annum) and tomato (Solanum lycopersicum) plants by pulse thermography of the leaves in the pre-symptomatic phase. For this purpose, pepper and tomato plants were inoculated with different concentrations of B. cinerea or Trichoderma harzianum spores, a beneficial fungus employed as non-pathogenic control. Thermographic measurements, carried out during seven days after infection, revealed specific thermal patterns in the infected leaves after a few hours (6 – 48 h) and earlier than the appearance of the characteristic lesions caused by B. cinerea. Diagnostic parameters, such as specificity, sensitivity, positive and negative predictive thermographic values, confirmed a good reliability of the pulse thermography technique in the early detection of B. cinerea infections. In order to better understand the mechanisms underlying the thermographic patterns caused by the gray mold, stomatal opening and conductance as well as expression of genes typically involved in plant-pathogen interactions were evaluated. Altogether, our data, supported by physiological, cellular and molecular evidence, demonstrate that pulse thermography imaging is a valid and reliable diagnostic tool for the rapid detection of B. cinerea infection and, possibly, to other necrotrophic fungi, to be applied for a rapid plant phenotyping of resistance/tolerance as well as for a more automatized, and sustainable plant disease control.
Yet another valuable trait of the biological control agent, Aurebasidium subglaciale
A. Di Francesco 1 , J. Zajc 2 , G. Firrao 1 , R. Cignola 1 , H. Brötz-Oesterhelt 3 , C. Hughes 3
1 University of Udine, Department of Agriculture, Food, Environmental and Animal Sciences, Via delle Scienze 206, 33100 Udine, Italy; 2 Agricultural Institute of Slovenia, Plant Protection Department, Hacquetova ulica 17, 1000, Ljubljana, Slovenia; 3 University of Tübingen, Interfaculty Institute for Microbiology and Infection Medicine, 72076, Tübingen, Germany. E-mail: [email protected]
Aureobasidium subglaciale is a psychrotolerant yeast isolated from subglacial ice that is able to grow at low temperatures. In fact, due to such feature it is an excellent biocontrol candidate that can be applied to long postharvest storage. The potential of A. subglaciale (strain AS13) against different fungal pathogens was previously verified by the assessment of the production of various metabolites, including both VOCs (volatile organic compounds) and non-VOCs. Furthermore, a more in-depth look at the bioactive compounds produced by strain AS13 was investigated by using preparative high-performance liquid chromatography (HPLC) in combination with nuclear magnetic resonance (NMR) spectroscopy. The biological activity of the strain AS13 and the fractionated extracts were tested in in vitro and in vivo assays against Penicillium spp., confirming the strain’s potential as a biocontrol agent against green and blue molds. Furthermore, the most potent purified fraction consisted of an unsaturated triglyceride similar to glyceryl tripalmitoleate (C51H92O6), which is known to be physiologically active and able to bind to the lipid bilayer and other components of the microbial cell membrane. With new findings on the type of compounds produced by AS13 further applications of the strains or derived compounds can be envisioned, providing a framework to develop environmentally-friendly and sustainable alternatives to current commercial pesticides.
This work was supported by DAAD-Research Stays for University Academics and Scientists, 2021
Identification and characterization of Xanthomonas arboricola pv. pruni strains isolated from almond in Sicily
G. Dimaria 1 , A. Mosca 2 , O. Viscardo 1 , M. Russo 3 , V. Catara 1
1 Department of Agriculture, Food and Environment, University of Catania, Catania, Italy; 2 Department of Physics and Astronomy University of Catania, Catania, Italy; 3 Agrobiotech Soc Coop S.r.l, Catania, Italy. E-mail: [email protected]
Xanthomonas arboricola pv. pruni (Xap) is the causal agent of bacterial spot disease of stone fruits, severely affecting health and productivity of many fruit and ornamental Prunus species. In the EU, the pathogen is regarded as a Regulated Non-Quarantine Pest (RNQP). In Italy, severe outbreaks have been reported on plum and peach in the north-east of the country. On almond it has been recently described in Apulia. Samples consisting of young sprouts and fruits of almonds from a commercial almond orchard near Agrigento (Sicily) were delivered to the laboratory of phytobacteriology of the University of Catania, Italy. Symptoms consisted in leaf and fruit spots. The symptoms on almond’s fruits were very characteristic, with sunken, corky lesions and oozing gum. Colonies with typical Xap morphology were isolated both from leaf and fruit samples. Xap identification was achieved by duplex PCR and supported by gyrB and rpoD gene sequence analysis. The pathogenicity of representative isolates was confirmed by prick-inoculation of young almond stems and by a detached-leaf assay. This latter assay also suggested differences in resistance amongst six Sicilian almond cultivars, with no immunity. The draft genome of X. arboricola pv. pruni strain PVCT262.1 was obtained. Genome-wide average nucleotide identity with other Xap from different hosts and countries was above 99%. However, it clustered more tightly with Xap almond strain CITA99 than CITA33 both isolated in Spain, in 2006 and 2009, respectively. No other genomes from Italian strains are available to date. Pathogenicity and virulence genes genome mining results will be discussed.
Virulence and host-specificity in Fusarium oxysporum ff. spp. interactions
A. Doddi 1 , G. Lancia 1 , W. Groot 1 , M. Amadei 2 , M.C. Bonaccorsi 2 , M. Rep 3 , M. Reverberi 1 , L. Faino 1
1 Department of Environmental Biology, University of Rome “La Sapienza”, Piazzale Aldo Moro 5, 00185 Roma; 2 Department of Biochemistry, University of Rome “La Sapienza”, Piazzale Aldo Moro 5, 00185 Roma; 3 Molecular Plant Pathology, Swammerdam Institute for Life Science, Universiteit van Amsterdam, The Netherlands, E-mail: [email protected]; [email protected]
Pathogenic fungi use effectors to colonize plants. These molecules regulate the interaction between pathogen and host, manipulating the ability of the pathogen to induce disease on plants. For example, Verticillium dahliae (Vd), a well-known pathogenic fungus, can infect a wide variety of plants, while Fusarium oxysporum (Fo), is able to infect only one or a few related plant species and therefore each strain is subcategorized in formae speciales (ff. spp.). Recently, an essential effector protein involved in the interaction Vd-cotton was identified in Vd isolates and named d-Vd. Additionally, we found that the d-Vd gene has many homologs in different ff.spp. of Fo. We were able to express the d-homolog protein in a bacterial heterologous system for the version from Fo f. sp. vasinfectum that infect cotton, for Fo f. sp. radices-cucumerinum that infect cucurbits and for Vd non-cotton infecting strains. Subsequently, we set an assay using plants growing in a protein solution and check the effect of the different homologs. The results showed that the d-Vd and the d-Fov alone can induce wilting on cotton plants while plants grown in a solution containing the nd-Vd or d-Forc did not show any symptoms. Additionally, we generate a deletion mutant for the d-Forc homolog and test the virulence of the mutant on different plants. Interestingly, deletion mutants of d-Forc showed a reduction of virulence on cucumber and watermelon after infection while no difference was observed on melon plants. All the obtained results show that the d-homologs may have different effect in different interactions rather than a common target.
Characterization of fungi associated with olive fruit rot in central Italy
M.I. Drais, F. Brugneti, S. Turco, S. Spagnuolo, A. Mazzaglia
Dipartimento di Scienze Agrarie e Forestali, Università degli Studi della Tuscia, via S. Camillo de Lellis snc, 01100 Viterbo, Italy. E-mail: [email protected]
Olive (Olea europaea L.) is one of the most cultivated tree species in Mediterranean countries with a major social, economic and environmental importance. Olive is commonly affected by diverse fungal pathogens causing fruit rot and thus, diminishing productivity and oil quantity and quality in oil-producing varieties. In Autumn 2021, a survey was carried out in an oil mill in Tuscia area (Viterbo Province). Fruit rot symptoms on olive fruits were observed with 13% prevalence. Depressed, round, and ochre or brown lesions leading to fruit rot were observed on the Italian olive cultivars Leccino, Canino and Frantoio collected from four different municipalities (Viterbo, Carbognano, Caprarola and Sutri). Visual, stereoscopic, and microscopic observations were carried out. Fungal strains isolated from symptomatic fruits were characterized by morphological and molecular analyses. Fungal species were identified belonging to seven different genera: Didymella, Botryosphaeria, Colletotrichum, Diaporthe, Epicoccum, Alternaria and Fusarium by the analysis of their internal transcribed spacer regions of ribosomal DNA region. Pathogenicity tests on punctured and non-punctured fruits were carried out for selected isolates. Further molecular analyses are ongoing on additional housekeeping genes for an accurate species-level identification of the isolated fungi. This research allowed to define a list of the most frequent fungal pathogens affecting olive drupes in Lazio region in central Italy.
Evaluation of alternative seed treatments for the management of seed-borne diseases
E. Emmanuello 1 , N. Bova 2 , G.F. Rizzo 1 , S. Conti 2 , S. Nigro 3 , V. Lefebvre du Prey 3 , P. Bella 1 , F. Branca 1 , V. Catara 1
1Department of Agriculture, Food and Environment, University of Catania, Via Santa Sofia 100, 95123 Catania, Italy; 2Department of Agricultural, Food and Forest Sciences, University of Palermo, Viale delle Scienze Ed.4, 90128 Palermo, Italy; 3Itaka srl, via Montenapoleone 8, 20121 Milano, Italy. E-mail: [email protected]
The European project H2020 BRESOV “Breeding for Resilient, Efficient and Sustainable Organic Vegetable production” aims to improve the competitiveness of three vegetable crops (tomato, broccoli and snap beans) when grown in an organic production system. In particular, workpackage 4 aims at developing protocols and tools to maximize yield and ensuring high sanitary quality and high genetic purity of organic seeds. In this framework, alternative seed treatments on different pathosystems are being evaluated. Here we report on the efficacy of seed treatments with three microbial consortia (Azotobacter, Bacillus, Pseudomonas, Streptomyces, Glomus, Trichoderma) and two natural compounds (chitosan and glucosinolates based). Organic seeds of two cultivars were inoculated in each trial. Snap beans and broccoli seeds were artificially inoculated with Neocosmospora phaseoli (106 conidia ml−1) and Alternaria brassicicola (105 conidia ml−1), respectively, before the treatment. Only one of the microbial consortia and the chitosan based product reduced significantly symptoms caused by A. brassicicola on broccoli cotyledons and brown discoloration of hypocotyl of snap beans induced by N. phaseoli. Trials on the tomato-Xanthomonas euvesicatoria pv. perforans pathosystem aimed at evaluating a possible biopriming effect of the seed treatment. To this aim 1-month-old tomato seedlings were spray inoculated with a bacterial suspension of the pathogen (108 cfu ml−1) and symptoms evaluated two weeks after inoculation. Results showed a reduction of bacterial disease severity with some of the treatments with MCs and NCs although the chitosan was the most effective.
This work was supported by the European project H2020 BRESOV “Breeding for Resilient, Efficient and Sustainable Organic Vegetable production”.
Pigmented wheat genotypes as innovative tools against Fusarium Head Blight disease
L. Felici 1 , C. Miccoli 1 , S. Francesconi 1 , F. Sestili 1 , M. Vitali 2 , G.M. Balestra 1
1 Università degli Studi della Tuscia, Dipartimento di Scienza Agrarie e Forestali, Via San Camillo De Lellis snc, 01100 Viterbo, Italy; 2 Università di Roma La Sapienza, Dipartimento di Sanità Pubblica e Malattie Infettive, Piazzale Aldo Moro, 500185 Roma Italy. E-mail: [email protected]
Fusarium head blight (FHB) is one of the main severe cereal diseases, including bread wheat (Triticum aestivum). To date, few fungicides to counteract this disease are available, thus there is a high priority need to find new protective compounds. Polyphenols are secondary metabolites that act as stress protecting agents, playing a significant role in plant resistance against plant pathogens. Pigmented wheat genotypes have flavonoid-rich kernels due to the presence of anthocyanins in the external layers (pericarp and aleurone). The aim of the present work was to evaluate whether the pigmented wheat genotypes show an increased resistance to FHB and if the development of new crop protection products based on nanomaterials obtained from pigmented wheat wastes (such as bran) can be successfully used to counteract FHB. The work was articulated in several steps: 1) evaluation of the resistance level of five pigmented wheat genotypes (two blue-aleurone genotypes, Purendo and Skorpion and three purple-pericarp genotypes, Rosso, Vanilnoir, and Indigo) compared to a resistant genotype (Sumai3) and to a control non pigmented cultivar (Rebelde); 2) synthetize cellulose nanocrystals (CNCs) from cellulose of wheat bran to be used as crop protection products. Results highlighted that blue aleurone genotypes showed high susceptibility to FHB, purple pericarp genotypes are less susceptible and Vanilnoir showed a resistance level comparable to Sumai3. CNCs was successfully obtained by pigmented wheat bran and further analysis to verify its antimicrobial activity (both in vitro and in vivo) has already been planned.
This work was supported by the project Valorizzazione Multifunzionale di Frumenti Pigmentati per la realizzazione di Alimenti Funzionali e di NanoAgrofarmaci Green per la Protezione del Frumento acronym MULTIFRU funded by POR FESR Lazio 2014–2020 action 1.2.1
Endophytes isolated from maize Lombard landraces: new perspectives for the control of Fusarium verticillioides
A. Follador 1 , A. Passera 1 , N. Miotti 1 , M. Ghidoli 1 , G. Consonni 1 , C. Balconi 2 , R. Pilu 1 , P. Casati 1
1 Department of Agricultural and Environmental Sciences–Production, Landscape, Agroenergy, Università degli Studi di Milano, Via Celoria 2, 20133 Milan, Italy. E-mail: [email protected]; 2 CREA-Centro di ricerca Cerealicoltura e Colture Industriali/Research Centre for Cereal and Industrial Crops— Via Stezzano 24—24126 Bergamo, Italy. E-mail: [email protected]
Maize is one of the most important crops in our country. However, in the past years yields have gradually decreased due to reduced cultivated areas and climate change. GEMMA project aims to study the endophytic bacterial community of landraces and their relationship with Fusarium verticillioides, one of the most widespread toxicogenic fungal pathogens in Lombardy region. Recent studies show how endophytes have abilities to inhibit the development of pathogens. Over the past three years four traditional landraces (Nero Spinoso Valcamonica, Spinato Gandino, Rostrato Rosso Rovetta, Fiorine Clusone, preserved at CREA Bergamo Genebank) and a pure line (B73) were grown in different locations (Landriano, Bergamo, Verderio, Carvico) with low-input farming techniques. Embryo of seeds harvested in 2019 allowed the isolation of 96 bacterial strains, mostly Firmicutes. In vitro antifungal assays against F. verticillioides allowed the selection of 12 strains with inhibition percentage higher than 50% for in vivo test. Only 2 isolates (Bacillus and Psychrobacillus) significantly reduced fungal infection in vivo. In 2021, experimental inoculations with F. verticillioides conidia were carried out in field during flowering. Values of infected seeds/ear and % infected area/ear show that Spinato di Gandino is the most susceptible variety while Nero Spinoso is confirmed as being the least susceptible. The study of microbiota and resistance traits from landraces shows potential for future organic maize production where biocontrol agents and resistant genotypes can contribute to a better, and more sustainable, response to stresses that will allow a higher maize yield even in the scenario of climate change.
Basalt-based novel agrochemicals demonstrated in vitro antimicrobial activities against fungal and bacterial plant pathogens
S. Francesconi 1 , A. Riccini 2 , F. Primavera 2 , G. Pizzuti 2 , G.M. Balestra 1
1 Dipartimento di Scienze Agrarie e Forestali (DAFNE), Università della Tuscia, Via San Camillo de Lellis snc, 01100 Viterbo, Italy; 2 Basalti Orvieto s.r.l., Località il Cornale snc, 05014 Castel Viscardo, Italy. E-mail: [email protected]
Basalt-based flours are natural products of volcanic origin, admitted in organic agriculture, obtained by a grinding process and rich in silicon, potassium, iron, calcium, and magnesium, which restore nutrients for soil fertility and can also potentially protect plants directly and indirectly as a physical barrier against pathogens and thanks to their biostimulant and nutritional properties. The aim of this study was to test in vitro the antimicrobial potentiality of three different particle size basalt-based flours, denominated as Farina di Basalto® XF, Farina di Basalto® F and Farina di Basalto® XM, against three impacting phytopathogen: Fusarium graminearum (Fg) causing Fusarium head blight (FHB) on wheat, Botrytis cinerea (Bc) causing grey mould on grapevine, and Pseudomonas savastanoi pv. savastanoi (Psav) causing olive knot. The three basalt-based products were assayed at five concentrations (10–15-20–25-30 g/L). The direct antimicrobial properties have been assayed by broth and agar assays and the highest concentration demonstrated high antimicrobial activity compared to the mock controls. The indirect antimicrobial properties were assayed by evaluating the ability to form a barrier against the pathogens and the physical interaction between the basalt particles and the pathogen cells. In particular, the basalt-based flours formed an effective barrier against Fg and Psav, inhibited the biofilm biosynthesis in Fg and Psav, inhibited the adhesion of cells and favoured their aggregation, thus confirming that such basalt-based particles can mechanically interact with the pathogen cells to reduce their contact surface with the host. Further investigations will be focused on validating such evidences by in vivo experiments.
Potato late blight: detached leaf and in planta assays to evaluate Phytophthora infestans colonization in several potato varieties
R.S. Fratini, M. Beccaccioli, M. Reverberi
Department of Environmental Biology, Sapienza University of Rome, 00185 Rome, Italy. E-mail: [email protected]
Late blight caused by the oomycete Phytophthora infestans is one of the most devastating diseases affecting potato (Solanum tuberosum L.). Under favorable conditions, P. infestans can infect leaves and tubers and severely spread in field. The use of resistant cultivars, when available, is considered a sustainable option to control the disease. However, the main strategy is still the oomycide application. Different potato varieties, all cultivated in central Italy, were tested in these experiments. The first aim of this work was to develop a reliable method to phenotypically evaluate the susceptibility of potato varieties to late blight with a detached leaf assay, leaves from adult plants were inoculated with a sporangial suspension of P. infestans and after 3 days foliar lesions were measured. The second aim was to evaluate the development of the infection caused by P. infestans in planta. Molecular approach (qPCR for pathogen quantification) coupled with an analytical approach (HPLC–MS/MS for plant hormones detection) was used. For the in planta assay, 2-week-old seedlings were inoculated with sporangia and kept under controlled conditions in a phytotron. The infection was monitored for 10 days and leaves samples were collected at early, middle and late stages for DNA and plant hormones extraction. The molecular analysis via qPCR was performed in order to estimate P. infestans colonization based on the infection stage and presence/absence of symptoms, while the quantification of phytohormones (salicylic acid and jasmonate) was conducted for a better comprehension of plant response to the infection of hemibiotrophic oomycete P. infestans.
This work was supported and funded by the project SOS TATA SVILUPPO DI NUOVE TECNOLOGIE NELL'AGRICOLTURA DI PRECISIONE PER LA PRODUZIONE SOSTENIBILE DI GENOTIPI DI PATATA CON ELEVATE QUALITÀ NUTRIZIONALI, n. F/200088/01–03/X45, Asse 1, azione 1.1.3. del Programma Operativo Nazionale «Imprese e Competitività» 2014–2020 FESR.
The evolution of the Plant Health Service diagnostic activity: the role of an Official Laboratory
F. Gaffuri 1 , L. Giordano 1 , M. Zugno 1 , M. Calvi 1 , R. Croci 1 , S. Sacchi 1 , B. Cavagna 1
1 Laboratory of Plant Health Service of Lombardy Region, c/o Fondazione Minoprio, Viale Raimondi 54, 22070 Como, Italy. E-mail: [email protected]
In December 2019, the Regulation (EU) 2016/2031 on protective measures against plant pathogens and pests entered into force and became applicable. The Regulation introduced the new plant health regime to reduce the risk of introduction into and spread within the territory of the European Union of invasive organisms, that can be responsible of substantial economic losses in both agricultural and forestry sectors. The Legislative Decree No. 19/2021 redefined the new national phytosanitary system and established a Network of Laboratories including the National Reference Laboratory, the Official Laboratories and others operating on the national territory in the plant protection sector. The Official Laboratories perform analyses, tests and diagnoses in the context of official controls and other official activities. To ensure highest standard results, these Laboratories must be accredited according to standard EN ISO/IEC 17,025. In this context, the Official Laboratory of Plant Health Service of Lombardy Region performs analyses for detection, identification and characterization of bacteria, fungi and oomycetes, insects and mites, nematodes, viruses, viroids and phytoplasmas. The diagnostic activity encompasses from traditional to advanced molecular techniques. In the era of climate change and globalization, the early detection and the accurate identification of pathogens and pests represents one of the most important goals of Plant Health Service. Therefore, we present our experience as Official Laboratory including the accreditation process, the contribution to development of new standard diagnostic protocols, and the collaboration with national and international research centers.
Development of a new LAMP assay for the fast diagnosis of Gnomoniopsis castaneae
L. Ghelardini 1 , C. Aglietti 1 , A. Benigno 1 , P. Capretti 1 , P. Gonthier 2 , N. Luchi 3 , S. Moricca 1
1 Department of Agriculture, Food, Environment and Forestry (DAGRI), University of Florence, Piazzale delle Cascine 18, 50144 Florence, Italy; 2 Department of Exploitation and Protection of Agricultural and Forest Resources (Di. Va. P. R. A.) – Plant Pathology, University of Torino, Via L. da Vinci 44, I-10095 Grugliasco, Italy; 3 Institute for Sustainable Plant Protection, National Research Council (IPSP-CNR), Via Madonna del Piano 10, Sesto Fiorentino 50019, Firenze, Italy
Gnomoniopsis castaneae (syn. Gnomoniopsis smitholgivyi) is an emerging fungal pathogen causing nut rot in sweet chestnut (Castanea sativa) but also reported as agent of cankers on twigs as well as of necroses on galls and leaves. Nut rot disease was observed in northern Italy since the second half of the nineteenth century, but only recently its incidence and severity was associated with outbreaks of G. castaneae. The little knowledge about the disease epidemiology, its almost ubiquitous distribution in chestnut groves and its endophytic occurrence in chestnut organs make it difficult to find valid solutions for its management and control. The possibility of a fast and accurate diagnosis could represent a valuable alternative to monitor pathogen occurrence both in pre- and in post-harvest conditions. Among the strategies today available to counter the disease, molecular detection tools are the most effective for disease diagnosis with high sensitivity and specificity. However, molecular detection often requires a well-equipped laboratory to be applied, limiting the speed and user-friendliness of the diagnosis. The aim of this work was to develop a new LAMP-based tool to be applied directly on site for the early detection of G. castaneae. The assay, optimized on the portable instrument Genie III (Optigene, UK) and based on the Ef1-α target region, can recognize the pathogen with a high level of specificity and sensitivity in about 20 min. Application of this method in chestnut orchards and in the subsequent processing steps of the fruit and derived products (e.g. chestnut flour) might provide a new, intriguing perspective for disease prevention and control directly in the field and in the chestnut processing chain.
This work was supported by the project Innovation and sustainable recovery in some piedmont agri-food chains of the Tuscan Apennines”—Fondazione Cassa di Risparmio di Firenze and by the project INGECA (Innovative low-impact strategies for the management of adversity of Chestnut orchards)- PSR—FEASR 2014–2020 Regione Toscana-PS-GO 2017.
Trichoderma application for reducing copper-based fungicides use and controlling phytopathogens
L. Gioia 1 , M. Michelino 1 , S. Lanzuise 1 , G. Manganiello 1 , R. Barone 1 , A. Raucci 2 , A. Pappalardo 2 , S. Cinti 2,3 , S.L. Woo 2,3
1 University of Naples “Federico II”, Department of Agricultural Sciences, Via Università 133, 80055 Portici (NA), Italy; 2 University of Naples “Federico II”, Department of Pharmacy, Via D. Montesano 48, 80131 Naples, Italy; 3 BAT Center—Interuniversity Center for Studies on Bioinspired Agro-Environmental Technology, University of Naples “Federico II”, Via Università 100, 80055 Portici, Italy. E-mail: [email protected]
Copper (Cu)-based agricultural products are widely used as antimicrobial compounds in different pathosystems. However, their frequent applications can cause indiscriminate Cu accumulation in soil and water sources, negatively impacting micro- and macro-organisms. On 22th June 2022, European Commission proposed a 50% reduction in Cu pesticide applications by 2030, as part of its mandate for sustainability and biodiversity preservation. The development of effective alternative products which can contribute in minimizing pollutant release in the agroecosystem is needed. The present research was aimed at testing the efficacy of Trichoderma sp. in combination with reduced doses of copper-based fungicide (CuSO4, Cupravit Bioadvance®, Bayer) for controlling Botrytis cinerea and Alternaria alternata on tomato. In order to select Trichoderma Cu tolerant strains, the growth of 4 isolates was monitored on PDA enriched with increasing concentrations of fungicide, ranging from 0.18 to 1.8 g/L. T. afroharzianum T22 and T. asperellum T25 were able to growth on media contained up to 90 mg/L of Cu (metal), resulting as the best candidates for in vivo biocontrol assays. A consistent disease severity reduction was observed on plants treated with one-half and one-quarter doses of Cupravit® (0.45 g/L and 0.9 g/L) when combined with either Trichoderma strains. T25 applied by soil drenching with 0.45 g/L CuSO4 was more effective than the full fungicide dose in reducing Alternaria diseases symptoms. The data suggested that Cu tolerant Trichoderma strains can be used in combination with Cu-based fungicides for the effective management of phytopathogens, reducing drastically the Cu released in the environment.
Pathogenicity assay and genome assembly of Fusarium oxysporum f. sp. melongenae that causes Fusarium wilt in eggplant
A. Grottoli 1,2 , A. Infantino 2 , S. Vitale 2 , L. Luongo 2 , A. Haegi 2
1 SARA EnviMob s.r.l., Rome, Italy; 2 CREA-DC, Research Centre for Plant Protection and Certification, Via C.G. Bertero 22, 00156 Roma, Italy. E-mail: [email protected]
Fusarium oxysporum f. sp. melongenae (FOMG) is the causal agent of Fusarium wilt in eggplant (Solanum melongena). FOMG is a tracheomycotic fungus that penetrates roots and invades plants systematically, causing leaf chlorosis and dropping of leaves, followed by wilt and finally plant death. FOMG is a soil-borne plant pathogen that can spread widely, it can persist for many years in the soil causing severe diseases on eggplant and seriously affecting the production. To better understand FOMG pathogenicity in FOMG-eggplant interaction, we set up a fast and easy pathogenicity in vitro assay. The test was used in order to rapidly identify pathogenic strains among Fusarium isolates from symptomatic eggplants; moreover, the test was used for molecular analyses on in vitro inoculated plants. We performed whole genome assembly of an Italian FOMG strain (FOMG-2267) and RNA-seq analysis in order to identify fungal genes involved in FOMG—eggplant interaction. Future reverse genetics analyses will be performed in order to assess the role of these genes in the interaction.
This work was founded by the Italian Ministry of Agriculture (MiPAAF) within the research project QUALIMEC ‘Miglioramento delle proprietà qualitative in melanzana e carciofo mediante approcci di genome editing e cisgenesi’.
The introduction of new cultivar and the risk of new virus: the case study of cherry necrotic rusty mottle virus (CNRMV) in northern Italy
V. Gualandri 1,2 , A. Gottardello 1 , T. Pantezzi, G. Angeli 1
1 Fondazione Edmund Mach, Center for Technology Transfer, Via E. Mach, 1—38010 S. Michele all’Adige, Trento; 2 Center of Agriculture, Food and Environment (C3A), University of Trento, I-38010 San Michele all’Adige, Italy. E-mail: [email protected]
Sweet cherry trees (Prunus avium L.) can be infected with various viruses, the presence of which can lead to production losses and orchard decline. In Trentino region in northern Italy the most important cultivars are Kordia and Regina but from 2016 for commercial purposes it was introduced over a large area a cv. “Giant Red (Mariant)” on Gisela 5 rootstocks. During spring in 2019 a high number of trees displaying virus-like disease symptoms such as brown angular necrotic spots on leaves of which can drop out giving a shot-hole appearance, yellow alone along the veins and necrosis on the margin, on the bark sometimes necrosis with gum spill on. Symptomatic and asymptomatic plants of sweet cherry cv. Giant Red planted in the previous 1–2 years were collected in several orchard in 2019. Total RNAs were extracted and analyzed by RT-PCR for detection of prunus necrotic ringspot virus (PNRV), prune dwarf virus (PDV), apple chlorotic leaf spot virus (ACLSV), cherry virus A (CVA), cherry leaf roll virus (CRLV) and cherry necrotic rusty mottle virus (CNRMV) with diagnostic method already published. Only CVA and CNRMV were identified, and relative amplicons obtained were purified and sequenced in both directions. The sequences were analyzed using blast. The analysis confirmed the only presence of CVA and CNRMV. While CVA is generally considered a latent virus, this result suggests that CNRMV is the cause of observed symptoms. This finding underscores the importance of virus regulation that should be highlighted and points again the risk for virus distribution utilizing infected propagation material. To our knowledge, this is the first report of CNRMV causing necrotic rusty mottle of cherry in Italy.
Paraconiothyrium fuckelii, Diaporthe eres and Neocosmospora parceramosa causing cane blight of red raspberry in Northern Italy
V. Guarnaccia 1,2 , I. Martino 1 , L. Brondino 3 , M.L. Gullino 1
1 Centre for Innovation in the Agro-Environmental Sector, AGROINNOVA, University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy; 2 Department of Agricultural, Forest and Food Sciences (DISAFA), University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy; 3 Ortofruit Italia Soc. Agr. Coop. O.P., Via Colombaro dei Rossi 16/bis, 12037 Saluzzo (CN), Italy. E-mail: [email protected]
Red raspberry (Rubus idaeus) is considered a crop of global significance and its production has largely increased during the last decade. In Piedmont, the increasing interest in raspberry cultivation has led to a progressive intensification of the production systems and to the adoption of new cultivars to supply the consumer demand. However, the consequent consistent movement of plant materials and fruit, along with the presence of favourable climate conditions, resulted in the emergence of diseases previously not reported in this area. Thus, a monitoring activity was conducted in red raspberry orchards over a three years period (2019–2021) to investigate the etiology and pathogen diversity in association with cane blight, the most common symptom observed and one of the most diffused and serious fungal diseases of raspberry. Isolates were collected from symptomatic plants of the cultivars ‘Diamond Jubilee’ and ‘Grandeur’. Three fungal species were identified: Paraconiothyrium fuckelii, Diaporthe eres and Neocosmospora parceramosa. The identification was achieved through morphological features assessment and multi-locus phylogenetic analyses on four different genomic loci (ITS, tef1, tub2 and rpb2). All the species found were confirmed as pathogenic and P. fuckelii was the most aggressive. This study provides the first insight on raspberry cane blight in Italy. This preliminary knowledge can be applied to further epidemiological and diagnostic studies in order to adopt effective integrated strategies to control and prevent the disease spread.
Kiwifruit Vine Decline Syndrome: the roles of microbiome and abiotic stress in plant health
M. Guaschino 1,2 , M. Garello 1,2 , L. Nari 3 , D. Spadaro 1,2
1 Department of Agricultural, Forest and Food Sciences (DISAFA), University of Torino, Largo Paolo Braccini 2, 10095 Grugliasco (TO), Italy; 2 AGROINNOVA, University of Torino, Largo Paolo Braccini 2, 10095 Grugliasco (TO), Italy; 3 Agrion, Fondazione per la Ricerca, l’Innovazione e lo Sviluppo Tecnologico dell’Agricoltura Piemontese, 12030 Manta (CN), Italy. E-mail: [email protected]
Kiwifruit vine decline syndrome (KVDS) has been threatening kiwifruit cultivation in Italy for the last decade. Symptoms lead to severe root decay leading to irreversible wilting that develops with the high temperatures of mid-late summer when kiwi vines have the highest transpiration rates. Root symptoms include disappearance of root hairs, breakdown of root cortex, blocked vessels, hypertrophy, and phloem detachment from the cortex. Among the microorganisms involved in the disease development, there are soilborne oomycetes belonging to the genus Phytopythium. High summer temperatures and flooding play a key role in the development of the disease. Trials performed in controlled conditions revealed a high virulence for the species P. helicoides, which showed a higher temperature tolerance even if the most isolated species is P. vexans. The use of a naturally infected soil to reproduce the symptoms revealed higher disease severity compared to the soil inoculation of a single species. For this reason, the soil, rhizosphere, and root microbiota of eight kiwifruit orchards, both healthy and KVDS-affected. Total DNA was extracted, and the population dynamics of fungi, bacteria and oomycetes was analyzed through metabarcoding. Defining the interplay between the factors involved will contribute to the understanding of multifactorial diseases where biotic and abiotic components simultaneously or sequentially affect plant health.
Work supported by the project KIRIS (La moria del kiwi – Approfondimento sull’eziologia e strumenti di prevenzione e difesa), funded by Piedmont Region.
A gnotobiotic approach to investigate microbiome functions in grapevine and lettuce
M. Guaschino 1 , G. Schiavon 1 , F. Mapelli 2 , S. Stassi 3 , G. Gambino 4 , L. Schena 5 , D. Spadaro 1
1 AGROINNOVA – Centre of Competence, and Department of Agricultural, Forest and Food Sciences (DISAFA), University of Torino, Largo Paolo Braccini 2, 10095 Grugliasco (TO), Italy; 2 Department of Food, Environmental, and Nutritional Sciences, Università degli Studi di Milano, Milan, Italy; 3 Politecnico di Torino, Dipartimento di Scienza Applicata e Tecnologia (DISAT), Corso Duca Degli Abruzzi 24, 10129 Torino, Italy. 4 Institute for Sustainable Plant Protection, National Research Council (IPSP-CNR), Torino, Strada delle Cacce 73, 10135 Torino, Italy; 5 Dipartimento di Agraria, Università degli Studi “Mediterranea” di Reggio Calabria, Reggio Calabria, Italy. E-mail: [email protected]
Plant microbiome plays a key role on impacting plant health, however knowledge on the exploitation of certain beneficial microorganisms to improve crops is still partial. The aim of the project is to investigate timing and patterns of the plant microbial colonization and the effect of microorganisms on stress tolerance, fungal pathogenesis and plant growth by using gnotobiotic plants. Gnotobiotic plants, defined as plants with a characterized microbiome, could represent a new system to study microbial interactions and their benefits against abiotic and biotic stresses. For this study, grapevine (Vitis vinifera cv. Chardonnay) and lettuce (Lactuca sativa) were selected as representative crops for woody and herbaceous systems. Firstly, colonization from selected Biome Agents (BAs) and their efficacy in biocontrol against fungal pathogens were evaluated in controlled conditions. We tested bacterial strains isolated from lettuce leaf endosphere against biotic stress using lettuce plants inoculated with two races of the wilting agent: Fusarium oxysporum f. sp. lactucae. The efficacy assay was performed in vivo by assigning the disease index at 12-, 21- and 28-days post inoculation. The average biomass weight was also evaluated for in vivo trials. The identified best performing agents will be evaluated using gnotobiotic conditions. Gnotobiotic plants of both species were obtained by testing different sterilization methods. The gnotobiotic conditions are maintained by using smart boxes able to completely isolate the plants from the external environment while monitoring humidity and plant eco-physiological parameters in real time.
Work supported by the NATURE project funded in the framework of the PRIN 2017 Programme (Italian Ministry of University and Research).
The GPI-anchored protein HAM-7 regulates root adhesion in Fusarium oxysporum
V.M. Guastaferro 1 , S. Vitale 2 , E. Russo 1 , D. Turrà 1
1 Department of Agricultural Sciences, University of Naples “Federico II”, Via Università 100, 80055 Portici (Na), Italy; 2 Institute for Sustainable Plant Protection, National Research Council of Italy, 80055 Portici, Italy. E-mail: [email protected]
The fungal cell wall is a dynamic structure protecting cells from environmental stresses, providing cell type-specific morphology and functioning as a physio-chemical rheostat for the transmission of extracellular signals through a large set of cell-wall-anchored proteins. HAM-7 is a highly conserved protein present in all filamentous ascomycetes, and it is known to form a sensor complex at the cell wall/plasma membrane interface for the activation of the MAK-1 cell wall integrity mitogen-activated protein kinase (MAPK) pathway in Neurospora crassa. Additionally, absence of the GPI-anchored cell wall protein HAM-7 in N. crassa leads to severe defects in cell-to-cell fusion and sexual development. BlastP searches using the N. crassa OR74A HAM-7 protein as a bait identified a single 233-amino-acid long orthologue with 64.19% identity in the Fusarium oxysporum f. sp. lycopersici (Fol) genome. Here, we genetically dissected the contribution of Fol ham-7 gene in the regulation of stress response, vegetative hyphal fusion, hyphal agglutination, plant root adhesion and virulence. Similarly to N. crassa, Fol ham-7Δ mutants are severely impaired in vegetative hyphal fusion, but not in vegetative growth under stress conditions (i.e. cell wall, hyperosmotic and heat stress). Importantly, despite being unable to undergo hyphal agglutination and plant root adhesion ham-7Δ mutants only showed minor however not significant defects in plant virulence.
BNYVV genomic formula changes during host infection and vector transmission
Y. Guo 1 , M. Dall’Ara 1 , D. Gilmer 2 , C. Ratti 1
1 University of Bologna, Department of Agricultural and Food Sciences, Viale G. Fanin, 40127 Bologna, Italy; 2 Institut de Biologie Moléculaire des Plantes (IBMP), 12 rue du Général Zimmer, 67000 Strasbourg, France. E-mail: [email protected]
Multipartite viruses possess genomes split in more than two genomic segments each one packaged into independent particles. One of the benefits of being a multipartite virus is the gene expression modulation via changes of the segment copy number. The soil-borne beet necrotic yellow vein virus (BNYVV) may be considered a model of multipartitism since, with its 4 to 5 genomic RNAs, it has the highest number of genomic segments among RNA viruses infecting plants. In this work, we investigate the ratio of the four genomic segments of BNYVV type B in different host types analyzing tissues from infected roots and leaves by a validated protocol of dual step reverse transcriptase droplet digital (RT-dd)-PCR. BNYVV genome formula was also calculated within the vector Polymyxa betae after zoospore purification from infected Beta vulgaris roots evaluating the plant rate of contamination. Results showed that some viral gene segments accumulate at low frequency, whereas others dominate. BNYVV segment copy numbers change according to the type of host and organ infected, moreover the virus seems to reach a dedicated set-point genome formula also within its vector. These data together with the biology of this virus raise questions about the genome integrity preservation of BNYVV during the host infection and transmission by the vector.
Current status of Botryosphaeriaceae : fourteen years surveys among nursery, urban and agro-ecosystems in Sicily
G. Gusella, A. Fiorenza, M.B. Costanzo, G.R. Leonardi, D. Aiello, A. Vitale, G. Polizzi
Università degli Studi di Catania, Dipartimento di Agricoltura, Alimentazione e Ambiente, Via Santa Sofia 100, 95123 Catania. E-mail: [email protected]
Botryosphaeriaceae represents a wide fungal family of which most members are known to be serious plant pathogens worldwide. During 2007–2022 numerous investigations have been conducted in Sicily to ascertain the etiology of many diseases of different crops. Specifically, ornamental crops, nut trees, fruit trees, and urban trees showing symptoms of dieback, cankers, shoot blight, fruit rot, and leaf spot, were observed, collected and furtherly investigated. Fungal species isolated from symptomatic samples were fully characterized morphologically and molecularly. Sequencing of the internal transcriber spacer region (ITS) of the nuclear ribosomal RNA cluster, part of the translation elongation factor 1α gene (tef1-α), and partial β-tubulin gene (tub2) was performed to infer phylogenetic analyses of concatenated gene datasets (multi-locus approach) for a reliable and robust characterization. Phylogenetic analyses, based on Maximum Parsimony (MP), Maximum Likelihood (ML) and Bayesian Inference (BI), led us to identify different genera and species belonging to Botryosphaeriaceae including Botryosphaeria dothidea, Lasiodiplodia spp., Neofusicoccum spp., Neoscytalidium dimidiatum. Pathogenicity tests, under greenhouses and field conditions, were always performed for all the species characterized among the years to ascertain their role as plant pathogens on the target hosts. Presence of Botryosphaeriaceae in Sicilian ecosystems represents an important epidemiological information since these group of fungi are known to be latent pathogens, cosmopolitan and having neutral host behavior. Monitoring of Botryosphaeriaceae among different hosts is a crucial step in order to assess effective control strategies.
Preliminary metabarcoding analysis of potential plant pathogens transmitted by hazelnut pollen
A. Haegi 1 , M. Petrucci 3 , M. Moizio 3 , G. Schreiber 3 , M. Maspero 2 , G. Castello 2 , T. De Gregorio 2 , L. Luongo 1 , S. Vitale 1
1 CREA DC Research Centre for Plant Protection and Certification, Via C.G. Bertero 22, 00156 Roma, Italy; 2 Hazelnuts Company division of Ferrero Group (Luxemburg). 3 SAGEA srl, Castagnito d’Alba (Italy). E-mail: [email protected]
Pollination is a vital process in the reproduction of most spermatophytes and requires the transfer of pollen to the plant carpel, which contains the ovule, in order that fertilisation can occur. It is well known that many insects are vectors for plant viruses by contamination of pollen with these pathogens when feeding, while there are a few reports of fungi associated with pollen. However, pollen grains are nutritionally rich and represent an important source of food for fungi. Pathogenic fungus can be spread with pollen as it has been reported for Citrus sinensis pollen that may, in fact, play a role in the spread of Colletotrichum acutatum in citrus orchards. To clarify if fungi causal agents of rotten hazelnut can be transmitted with pollen, we analyzed the fungal populations present on pollen by metabarcoding. To this aim, we extracted DNA from microorganisms present on pollen of two commercial hazelnut “Tonda Gentile Trilobata” and “Tonda di Giffoni”, performing a mild extraction to avoid pollen breakage and plant DNA extraction. Metabarcoding was performed by amplification of the Internal Transcribed Spacer (ITS) region of rDNA followed by short-reads Illumina sequencing. Preliminary analysis showed a high number of fungal microorganisms, and the most represented fungus is Colletotrichum lagenaria both in GF (26.73%) than in TGT (19.41%), but also different species of Fusarium sp. including F. lateritium, Alternaria sp., Cladosporium sp., Ramularia sp., (especially TGT) are present, in addition to different epiphyte/saprophyte fungi.
Rapid detection of Eremothecium coryli from kernel hazelnut and Halyomorpha halys
A. Haegi 1 , G. Sabbatini Peverieri 2 , T. De Gregorio 3 , M. Maspero 3 , G. Castello 3 , M. Petrucci 4 , L. Marianelli 2 , L. Luongo 1 , P.F. Roversi 2 , S. Vitale 1
1 CREA DC Research Centre for Plant Protection and Certification, Via C. G. Bertero 22, 00156 Roma, Italy; 2 CREA DC Research Centre for Plant Protection and Certification, Via Lanciola 12A—50125 Cascine del Riccio, Impruneta, Florence (Italy); 3 Hazelnuts Company division of Ferrero Group (Luxemburg). 4 SAGEA srl, Castagnito d’Alba (Italy). E-mail: [email protected]
Eremothecium coryli (Nematospora coryli), a yeast pathogen of some fruit trees, cotton, and vegetables (especially tomatoes and Lima beans) was first isolated from Corylus avellana by Peglion in 1897. Since 2017, an intense survey on Italian hazelnut orchards has been carried out to investigate defects of hazelnut kernel. In these studies, E. coryli was isolated in Piedmont and Campania regions, from kernels that presented the typical damage caused by Pentatomids bug (e.g., Halyomorpha halys) and showing an altered area with symptoms of dry rot. To study the presence, transmission and diffusion of E. coryli on hazelnut, we developed a simple, fast, reliable and sensitive method for detection and quantification of the pathogen. We set up a real-time quantitative PCR (qPCR) on total DNA extracted directly from kernel hazelnut or from insect, with primers ERM-F1/ERM-R1 developed on the ITS region. Specificity analysis of this primer pair demonstrated that fungi that usually are isolated from hazelnuts, such as Alternaria spp., Colletotrichum spp., Botryosphaeria spp., Phomopsis spp., and Didymella corilicola, did not give cross reaction. Conversely, the phylogenetically close yeast Saccaromyces cerevisiae gave cross reaction. For this reason, qPCR for E. coryli was run in parallel with S. cerevisiae primers S14F/S17R, to exclude the presence of the latter. The efficiency and reliability of the qPCR protocol was evaluated on kernel hazelnut and on different body parts of the stink bug Halyomorpha halys. Results show the presence of E. coryli in the mouthparts of H. halys fed with infected hazelnuts, suggesting that the transmission of this fungus probably occurs through this way.
This work was supported by HCo Ferrero Company through the research project NOCEREHAL “Possibile trasmissione ed associazione del fungo Eremothecium coryli agente causale del dry rot della nocciola con la cimice asiatica (Halyomorpha halys)”.
Biochar potentialities for suppression of diseases caused by viruses, bacteria, oomycetes, fungi and parasitic plants
G. Iacomino 1 , M. Idbella 1 , S. Laudonia 1,4 , F. Vinale 1,2 , G. Bonanomi 1,3
1 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici (NA), Italy; 2 Department of Veterinary Medicine and Animal Productions, University of Naples “Federico II”, 80137 Naples, Italy; 3 Task Force on Microbiome Studies, University of Naples “Federico II”, Naples, Italy; 4 Center for Studies on Bioinspired Agro-environmental Technology (BAT Center), University of Naples “Federico II”, Naples, Italy. E-mail: [email protected]
Soilborne pathogens and pests in agroecosystems represent a serious problem limiting crop yield. With a view to developing more ecologically sustainable agriculture, the possibility of using biochar to control the damaging agents has been increasingly studied in recent years. This work provided a general overview of the suppression of diseases and infestations by biochar. We presented an update of literature from 2015 to 2022 based on 61 articles, including 117 experimental cases. We assessed how several biochar production feedstocks, pyrolysis temperatures, application rates, and the examined pathosystems affect diseases and infestations incidence. Fungal pathogens represented 55% of the case studies, followed by bacteria (15%), insects and nematodes (8%), oomycetes and viruses (6%), and only 2% parasitic plants. The most studied fungal species is Fusarium oxysporum f. sp. radicis lycopersici, Ralstonia solanacearum for bacteria, Meloidogyne incognita for nematodes, Epitrix fuscula for insects, Phytophthora capisci for oomycetes and Phelipanche aegyptiaca as the only parasitic plant. Biochar showed 85% suppression efficiency for fungi, 50% for oomycetes, 60% for viruses, 70% for bacteria and 50% for nematodes. Most studies used an application rate between 1 and 3%, a pyrolysis temperature between 500 °C and 600 °C, and a feedstock based on sawdust and wood debris. Different mechanisms were proposed to explain disease suppression by biochar, including induction of systemic resistance, improving rhizosphere competence of microbial community, and sorption of phytotoxic compounds of plant and/or microbial origin. Overall, it is important to standardize biochar feedstock and rate application and to improve beneficial effect on plant in terms of disease control.
Survey on the presence of fungal epiphytic, saprophytes, and leaf pathogens present on a collection of grapevine rootstocks in Italy
A. Infantino 1 , K. Kutya 1,3 , M.E.M. D’Arcangelo 2 , C. Biselli 2 , A. Zombardo 2 , L. Pirone 1 , A. Grottoli 1 , M.T. Valente 1
1 Council for Agricultural Research and Economics—Research Centre for Plant Protection and Certification (CREA—DC), Via C. G. Bertero 22, 00156 Roma, Italy; 2 Council for Agricultural Research and Economics—Research Centre for Viticulture and Enology, Viale Santa Margherita 80, 52100 Arezzo; 3 The Department of Innovation in Biology, Agri-food and Forest systems (DIBAF), Via San Camillo de Lellis snc., 01100 Viterbo, Italy. E-mail: [email protected]
Vitis rootstocks are cultivated in Italy on about 2,500 Ha, with a predominance of 5 varieties constituting 95% of the total (SO4, 110 Richter, 1103 Paulsen, 140 Ruggeri and K5BB). Besides conferring resistance to Daktulosphaira vitifoliae, recent breeding efforts have been devoted to new rootstocks tolerant to salinity, nematode, and water saving. No data are available on the characterization of fungal species potentially affecting the health of the propagation material. The aim of the present work is to provide a preliminary characterization of fungal species associated with different leaf symptomatologies, ranging from discolorations to necroses of variable extent. A first survey was conducted in 2020 at the CREA-VE experimental vineyard. Conventional diagnosis through isolation of Potato Dextrose Agar amended with antibiotics, and pathogenicity test on detached Vitis vinifera leaves, allowed to identify fungal species (Botrytis cinerea, Coniella diplodiopsis, Pilidium concavum, Diaporthe helianthii, Neopestalotiopsis spp.) already known as pathogenic on other crops. For a second surveys, done in 2021 in the same farm, a metabarcoding strategy using Illumina MiSeq amplifying the ITS2 region, allowed to classify a total of 110 OTUs; 27.7% of the total reads were identified as Pseudopezicula tracheiphila, agent of red fire disease of grapevine, followed by Cladosporium sp., Alternaria alternata, and Saccotheciaceae sp., a family including potentially phytopathogenic genera. The use of some rootstocks as donors of genetic resistance towards the most common grapevine diseases (powdery mildew, black rot blight), could imply the risk of transfer, through crossing, susceptibility to these neglected but potentially dangerous fungal species.
This work was founded by the Italian Ministry of Agriculture (MiPAAF) within the research project PATHORES, “Studio della resistenza a patogeni fungini e batterici per lo screening di varietà ottenute mediante genome editing”.
SUStaINable use of bioactive compounds from Brassicaceae and Solanaceae wastes for CEReal crop protection (SUSinCER)
C. Lanzanova 1 , D. Pacifico 2 , F. Nicoletti 2 , R. LoScalzo 3 , E. Pagnotta 2 , L. Bassolino 2 , L. Ugolini 2 A.M. Mastrangelo 4 , D. Marone 4 , D.B.M. Ficco 4 , M.A. Russo 4 , S. Locatelli 1 , F. Modica 1 , C. Balconi 1
1 CREA Centro di ricerca di Cerealicoltura e Colture industriali, Via Stezzano 24, Bergamo, Italy; 2 CREA Centro di ricerca di Cerealicoltura e Colture industriali, Via Corticella 133, Bologna, Italy; 3 CREA Centro di Ingegneria e trasformazioni Agroalimentari, Via Venezian 26, Milano, Italy; 4 CREA Centro di ricerca di Cerealicoltura e Colture industriali, S.S 673- Km 25,200, Foggia, Italy. E-mail: [email protected]
Defatted seed meals of oleaginous Brassicaceae, such as Eruca sativa, and potato peel are excellent plant matrices to recover potentially useful biomolecules from industrial processes in a circular economy perspective aiming at crop protection. These biomolecules, mainly glucosinolates for Brassicaceae and glycoalkaloids for potato, have been proven to be effective against pests like bacteria and fungi. Their role in plant protection is investigated, together with the molecular basis of their synthesis in plant, and their mechanisms of action. Possible genetic and biotechnological strategies are presented to increase their content in plants. The application of these bioprotectors will deal with new global challenges, such as reducing food waste and increasing sustainability and food safety for the consumer. SUSinCER project aims to investigate the biological activity of secondary metabolites from defatted seed meals of oleaginous Brassicaceae and potato peel, against development of fungal pathogens and mycotoxin production by fungi that contaminate cereals, mainly maize and wheat. In particular, during the first year of project activity a radial growth inhibition assay has been performed in vitro to evaluate the effect of different bioactive compounds on Fusarium verticillioides and Fusarium graminearum, important mycotoxigenic fungal pathogens responsible for maize and wheat rot. In field activities are also in course for wheat and maize to study the effect of bioactive compounds on resistance to fungal diseases and agronomic performance.
This work was supported by the project SUSINCER (2020-2023) “SUStaINable use of bioactive compounds from Brassicaceae and Solanaceae wastes for CEReal crop protection”. Fondazione Cariplo- Bando Economia Circolare: Ricerca per un Futuro Sostenibile - Area Scientifica Project code 2019-2538.
Multispecies Trichoderma consortium in combination with hydrolyzed lignin extract improve growth, yield, and nutritional quality of tomato
S. Lanzuise 1 , G. Manganiello, C. Vincenzo 1 , C. El-Nakhel 1 , Y. Rouphael 1 , P. Carillo 2 , V. Macchia 3 , S Pietarinen 4 , G.M. Fusco 2 , E. Dell’Aversana 2 , M. Lorito 1,4 , S.L. Woo 4,5,6
1 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 2 Department of Environmental, Biological and Pharmaceutical Sciences and Technologies, University of Campania “Luigi Vanvitelli”, Via Vivaldi 43, 81100 Caserta, Italy; 3 Green Innovation GMBH, Grabenweg 68, 6020 Innsbruch, Austria; 4 UPM-Kymmene Oyj, Alvar Aallon katu 1, FI-00101 Helsinki, Finland; 4 Task Force on Microbiome Studies, University of Naples “Federico II”, 80055 Portici, Italy; 5 Institute for Sustainable Plant Protection, National Research Council, 80055 Portici, Italy; 6 Department of Pharmacy, University of Naples “Federico II”, 80131 Naples, Italy. E-mail: [email protected]
The introduction of beneficial synthetic communities with functional and metabolic complementarity can modify the rhizosphere microbial composition and substantially improve plant yield and pathogenic microorganism containment. Many studies aimed to evaluate the potential association of microbial consortia and bioactive molecules belonging to "botanicals", plant extracts able of inhibiting pathogen development, inducing systemic resistance and promoting plant growth. The aim of this research was to develop new biological formulations combining Trichoderma spp. multispecies consortia with a lignin-derived polyphenolic mixture, for improving plants health status and productivity, both in quantitative and qualitative terms. To this scope, the compatibility between strains belonging to T. asperellum, T. virens, T. atroviride and commercial hydrolyzed lignin extract (Solargo UPM Solargo™) was tested in vitro and in vivo assays. The effect of lignin-Trichoderma formulations was assessed in greenhouse and field experiments, by evaluating the main plant biometric parameters, yield, and quality of tomato fruits. T. virens GV41 + T. asperellum + T. atroviride + Solargo formulation was the most effective in terms of growth promotion, increasing root and stem dry weight compared to control (45.4 and 43.9%, respectively). The same formulation determined a 63% increase in yield, compared to the control, resulting the better performing treatment also compared to single or coupled constituents. In addition, significant differences in terms of lycopene, GABA, ornithine, total, essential and branched-chain amino acid contents were revealed in fruits from treated tomato plants, confirming that beneficial microorganisms and polyphenolic mixture can positively interact in improving crop nutritional quality.
Is the combination of DNA metabarcoding and conventional leaf-baiting isolation techniques the best strategy to unravel the Phytophthora variability in natural, semi natural and horticultural ecosystems?
F. La Spada 1 , F. Aloi 1 , M. Riolo 1,2,3 , A. Pane 1 , P.J.A. Cock 4 , E. Randall 4 , D.E.L. Cooke 4 , S.O. Cacciola 1
1 Dipartimento di Agricoltura, Alimentazione e Ambiente, Via Santa Sofia 100, 95123 Catania; 2 Dipartimento di Agraria, Università degli Studi "Mediterranea" di Reggio Calabria, località Feo di Vito, 89122 Reggio Calabria; 3 Centro di Ricerca Olivicoltura, Frutticoltura e Agrumicoltura (CREA-OFA), Contrada Rocchi, 87036 Rende (Cosenza); 4 The James Hutton Institute, Invergowrie, Errol Road, Dundee DD2 5DA Dundee, Regno Unito. E-mail: [email protected]
Over the past two decades, molecular diagnostic tools for identification of cultured Phytophthora specimens supported by traditional isolation techniques provided efficient tools for the monitoring of Phytophthora communities within different kinds of ecosystems. More recently, metabarcoding approaches emerged as new paradigms for the detection and surveillance of target organisms within an environmental sample. In this study, the Illumina metabarcoding and a conventional baiting isolation technique complemented each other to describe the variability in Phytophthora communities from three small-scale ecosystems: (i) Nature Reserve, (ii) Botanical Garden, (iii) Citrus Orchard. Overall, 28 out of 39 collected rhizosphere soil samples, processed by both leaf baiting and Illumina metabarcoding, were classified as Phytophthora-positive. In total, the 1,406,613 ITS1 sequences obtained by metabarcoding together with the 155 baited isolates made it possible to record 21 Phytophthora taxa, five exclusively by baiting (P. bilorbang, P. cryptogea, P. gonapodyides, P. parvispora and P. pseudocryptogea), 12 exclusively by metabarcoding (P. asparagi, P. occultans, P. psycrophila, P. syringae, P. aleatoria/P. cactorum, P. castanetorum/P. quercina, P. iranica-like, P. unknown taxon 1, P. unknown taxon 2, P. unknown taxon 3, P. unknown taxon 4, P. unknown taxon 5) and four with both techniques (P. citrophthora, P. multivora, P. nicotianae and P. plurivora). Results suggested that the combined use of an efficient leaf baiting technique and a reliable metabarcoding detection method is the best approach to unravel the variability of Phytophthora communities from both natural and managed ecosystems.
High Throughput Sequencing testing of small RNAs empowers the simultaneous detection of citrus viruses and viroids
G. Licciardello 1 , G. Scuderi 2 , M. Russo 2 , M.C. Bazzano 2 , R. Ferraro 2 , C. Oliveri 3 , A.F. Catara 2
1CREA Research Centre for Olive, Citrus and Tree Fruit, Corso Savoia 190, 95024, Acireale, Italy; 2Agrobiotech soc. coop. ZI Blocco Palma I, Str.le V. Lancia 57, 95121 Catania; 3Agronomic consultant of Novarancia project. E-mail: [email protected]
Citrus are affected by a relevant number of viruses and viroids routinely tested by bioindexing and molecular assays. Anyway, the selection of virus free mother trees candidate for the foundation block requires a process of shoot tip grafting and biological indexing in greenhouse and molecular assays. A challenge which needs adequate professional and technical resources. Recent papers have shown the adoption of High Throughput Sequencing (HTS) technology is effective for the simultaneous detection and identification of viruses and viroids in Citrus and other crops, as pre-screening to conventional detection methods to solve ambiguous results. To explore the reliability of a potential HTS-based virus detection protocol in association of bioinformatic strategies, within the project Novarancia we developed a pilot testing of different field trees in parallel with molecular and biological methods. Three sweet orange trees and two alemow seedlings previously indexed by conventional methods were re-analyzed by HTS of small RNAs. High-quality reads, depleted of the Citrus sinensis genome, were aligned with reference genomes of 13 viruses and 6 viroids and analyzed by a bioinformatic pipeline. A genome coverage above 90% was assumed indicative of the presence of a specific genotype, and coverage > 50% of potential presence of variants not represented in the read mapping reference list. The results show the implementation of conventional certification scheme with HTS sequencing of small RNAs speeds up the time over all required and reduces the greenhouse footprint, labor, time, and cost needed for bioindexing.
This work was supported by the project “Novarancia” CUP G39J22000880009, funded by the Rural Program of Sicily 2014–2022.
Differences in the endophytic microbial communities of two olive cultivars infected by Pseudomonas savastanoi pv. savastanoi
G. Licciardello 1 , A. Mosca 2 , S. Di Silvestro 1 , M.P. Russo 1 , V. Catara 3 , P. Caruso 1
1 CREA Research Centre for Olive, Citrus and Tree Fruit, Corso Savoia 190, Acireale (CT), Italy; 2 University of Catania, Dipartimento di Fisica e Astronomia, Via Santa Sofia 68, 95123 Catania, Italy; 3 University of Catania, Dipartimento di Agricoltura, Alimentazione e Ambiente, Via Santa Sofia 100, 95123 Catania, Italy. E-mail: [email protected]
Olive knot disease (OKD) seriously affects olive production in the Mediterranean basin. The characteristic symptom is the formation of tumorous overgrowths deeply colonized by Pseudomonas savastanoi pv. savastanoi (Psv). Microorganisms in nature live as members of complex multispecies communities and an articulated pathobiome has been described for OKD. During a 3-years survey at the International Olive Germplasm Collection “Villa Zagaria” (Sicily, Italy), difference in resistance to Psv has been detected in trees harbouring a similar epiphytic bacterial population titre. In order to investigate on the possible interactions with endophytic microbial communities the cultivars Zaituna and Giarraffa, showing high and low resistance to OKD, respectively, were selected for an amplicon-based metagenomics analysis. The DNAs were extracted from surface sterilised twigs of field-grown olive trees. The endophytic bacterial and fungal communities were analyzed using Illumina-based sequencing of the V3–V4 hyper variable regions of bacterial 16S rRNA genes and the ITS2 regions of fungal rRNA genes, respectively. In the whole, Proteobacteria and Ascomycota were the most abundant bacterial and fungal phyla respectively. A significantly higher bacterial richness was detected in the shoots of the susceptible ‘Giarraffa’ although with a lower diversity. The opposite trend was observed for fungal communities. Differential abundance analysis of bacterial communities showed that Amnibacterium (Actinobacteria), unidentified Mollicutes (Tenericutes), Methylobacterium and Sphingomonas (Proteobacteria) were enriched in ‘Zaituna’ whereas Pseudomonas (Proteobacteria), unidentified Plankthothricoides and unidentified Microcoleaceae (Cyanobacteria) were depleted. Moreover, in ‘Zaituna’ the fungal community resulted enriched in Neofabraea (Ascomycota) and unidentified Tremellomycetes (Basidiomycota) and depleted in Alternaria, Neofusicoccum and Pseudocercospora (Ascomycota).
The work has been financially supported by SALVAOLIVI project (MIPAAF, DM 59 del 10/01/2018).
Biological control of Pseudomonas savastanoi pv. savastanoi by using an extract from pomegranate by-products
G. Licciardello, L. Signorello, P. Caruso, S. Di Silvestro, M.C. Strano, M.P. Russo, N. Timpanaro, G. Ballistreri, F.V. Romeo
Consiglio per la Ricerca in agricoltura e l’analisi dell’Economia Agraria-Centro di ricerca Olivicoltura, Frutticoltura e Agrumicoltura (CREA), Corso Savoia 190, Acireale (CT), Italy. E-mail: [email protected]
Olive knot disease (OKD) seriously affects olive production inducing the formation of tumorous overgrowths. Pseudomonas savastanoi pv. savastanoi (Psv) is the causal agent and lives epiphytically on the surface of leaves and twigs. The pruning of infected material and copper treatments are the unique currently available strategies to control further spread. Because sometimes copper treatments can have a reduced efficacy and can cause undesirable effects, alternative biological methods are under evaluation. A standardized dry powdered pomegranate extract (PPE) derived from the by-products (peels, exhausted arils and seeds) of pomegranate juice industrial processing, developed by the researchers of CREA (Acireale), has been tested to contain Psv epiphytic population in olive plants. Preliminary in vitro experiments revealed relevant inhibitory activities of PPE at different concentrations (10%, 5%, 2.5%, 1%, 0.5%) against Psv (108–106 log cfu ml−1). Furthermore, in vivo treatments were applied on the canopy of ‘Biancolilla’, ‘Moresca’ and ‘Calatina’ olive plants with different susceptibility to OKD. Three treatments of PPE at 5% concentration were applied during spring (after 30 days each), few days after pruning. The overall epiphytic bacterial population obtained by plate count on King’s B medium showed no statistical differences between PPE, copper treatment and control (water). A significant reduction of epipthyic Psv population, estimated by quantitative real-time PCR, has been detected in ‘Biancolilla’ and ‘Calatina’ treated with PPE and copper (2.2 log cfu mL−1) in comparison to the control (3.7 log cfu mL−1). No efficacy was detected in ‘Moresca’. The Psv population counts were statistically different among the different groups and were influenced by the fixed factors taken into account (i.e. cultivar, treatment, cultivar x treatment).
The work has been financially supported by SALVAOLIVI project (MIPAAF , DM 59 del 10/01/2018).
Gene expression of sour oranges seedlings inoculated with an aggressive seedling yellow CTV and a homologous asymptomatic isolate
G. Licciardello 1 , D. Puglisi 1 , G. Lopatriello 2,3 , V. Grosso 2,3 , M. Caruso 1 , M. Bazzano 4 , G. Scuderi 4 , A.F. Catara 4 , P. Caruso 1 , M. Rossato 2,3 , C. Licciardello 1
1 CREA Research Centre for Olive, Citrus and Tree Fruit, Corso Savoia 190, 95024 Acireale, Italy; 2 Genartis srl, via IV Novembre 24, 37126 Verona, Italy; 3 Department of Biotechnology, University of Verona, Strada Le Grazie 15, 37134 Verona, Italy; 4 Agrobiotech soc. coop. ZI Blocco Palma I, Str.le V Lancia 57, 95121 Catania (Italy). E-mail: [email protected]
Citrus tristeza virus (CTV) is a phloem-restricted aphid-transmitted virus, responsible for tremendous economic losses of sweet oranges and grapefruits grafted on sour orange (SO), leading many countries to the use of tolerant rootstocks. In earlier studies SO seedlings inoculated with a severe VT-SY isolate (SG29) showed severe seedling yellows, stunting and reduced root apparatus, while those inoculated with a homologous asymptomatic variant (M39) remained very similar to control non-inoculated plants. Moreover, the asymptomatic isolate inhibited the superinfection with the severe one. To understand the basis of this cross protection-like reaction we have investigated the transcriptomic response of SO seedlings inoculated with SG29 or the homologous asymptomatic and protective variant M39. Analysis of transcriptome obtained through Illumina RNA-seq, in comparison to uninoculated controls, showed 2,869 differentially expressed transcripts (DETs) after infection with SG29 and only 7 with M39. A total of 517 DETs were exclusively present in SG29 plants relative to M39, whereas 1,739 were shared also to the uninoculated control. Many DETs consist of immune receptors (i.e., nucleotide-binding-leucine-rich repeat proteins) able to initiate a resistance response through a cascade consisting of MAP kinases and chaperone protein complex (heat shock proteins). Transcripts associated to senescence, hormone metabolism (auxin, ethylene), cell wall and ribosomal composition are also included. These data reveal that SG29 is responsible of the activation of an effector-trigged immunity (ETI) response, giving rise to the marked symptomatic hypersensitive reaction observed in plants. Only few DETs linked with ETI are upregulated in M39 inoculated plants.
This research is part of the project S.I.R.P.A (n. 08CT7211000254) funded by PO FESR Sicilia 2014–2020, action 1.1.5.
Evaluation of copper-alternative products to control citrus anthracnose and Alternaria brown spot in Sicilian citrus orchards
M.F. Lombardo 1 , S. Panebianco 2 , S. Trovato 1 , A. Azzaro 1 , G. Cirvilleri 1
1 Dipartimento di Agricoltura Alimentazione e Ambiente, Università degli Studi di Catania, via Santa Sofia 100, 95123 Catania, Italy; 2 Dipartimento di Fisica ed Astronomia, Università degli Studi di Catania, via Santa Sofia 64, 95123 Catania, Italy. E-mail: [email protected]
Colletotrichum and Alternaria species, causal agents of pre- and postharvest anthracnose and Alternaria brown spot have been recently reported as emerging fungal pathogens on citrus in the Mediterranean area. The limitation on the use of copper-based fungicides in most European countries, in both integrated and organic farming, due to its demonstrated noxious effect on the environment, makes the control of these pathogens difficult. Thus, alternative products able to reduce/phase out copper in citrus farming are needed. In this work, copper-alternative products, previously identified and evaluated in vitro and in pre- and post-harvest trials, were tested for 2 consecutive years (2020–2021) in two integrated and one organic citrus orchards in different pedo-climatic conditions to control natural infections caused by Colletotrichum spp. and Alternaria spp. The alternatives, including basic products, active substances, biocontrol agents and their combinations, were applied on oranges cultivar “Tarocco Scirè” and “Tarocco Tapi” and on lemon cultivar Femminello 2KR. Even under different disease pressure levels, the basic products (chitosan and equisetum) and the active substance (sweet orange essential oil), alone and in mixture, significantly reduced disease incidence and severity compared with the untreated controls, often showing similar or better efficacy than copper compounds. The good efficacy of alternative products encourages their further evaluations in other in vivo trials and indicates the potential of their sustainable and large-scale use, useful for replacing or reducing the use of copper in integrated and organic citriculture in view of future limitation of its use.
This work was supported by H2020 project Organic plus, grant agreement No. 774340, 2018–2022 and by Programma Ricerca di Ateneo MEDITECO , UNICT 2020–2022 Linea 2-University of Catania (Italy).
Genome-wide analysis of G-type lectin family in Fragaria vesca and functional characterization of FaMBL1 gene in defense to fungal pathogens
L. Ma 1 , Z.M. Haile 1,2 , S. Sabbatini 3 , B. Mezzetti 3 , F. Negrini 1 , E. Baraldi 1
1 Department of Agricultural and Food Sciences (DISTAL), University of Bologna, Bologna, Italy, 2 Ethiopian Institute of Agricultural Research (EIAR), Addis Ababa, Ethiopia, 3 Department of Agricultural, Food and Environmental Sciences, Polytechnic University of Marche, Ancona, Italy. E-mail: [email protected], [email protected]
Strawberry (Fragaria × ananassa) is one of the most consumed and produced fleshy fruits in the world but due to its characteristic softening is highly susceptible to many fungal pathogens. Anthracnose and gray mold are two of the most destructive diseases of strawberry which lead to serious fruit rot. White unripe strawberry fruits are more resistant to Colletotrichum acutatum than red ripe fruits. Previous transcriptional analysis conducted during early stage infection has shown that a mannose-binding lectin gene FaMBL1 was the most upregulated in the white stage compared to the ripe susceptible one. FaMBL1 belongs to the G-type lectin family, having important roles in plant development and defense process. In this study genome-wide identification of G-type lectin gene family was carried out in F. vesca. Overall, 133 G-lectin genes were characterized for domain arrangement and expression patterns. Active transcription of G-lectin genes during development and under stresses suggested a potential role of G-lectin genes in strawberry defense. Hence, stable transgenic strawberry plants overexpressing the FaMBL1 gene were generated. Transformed strawberry plants were selected and molecularly characterized through droplet digital PCR and RT-PCR analysis. In total four overexpressing lines (OE) with different copy numbers were obtained and used for the subsequent studies, including the evaluation of disease-related phytohormones content and their reaction to biotic stresses. Accordingly, jasmonic acid (JA) was found decreased in OE-lines compared to wild type (WT). OE-plants petioles inoculated with C. fioriniae had lower disease incidence than WT, and leaves challenged by B. cinerea showed remarkably smaller lesion. Furthermore, expression of defence related genes as chitinase 2–1 (FaChi2-1) gene showed higher expression in OE-lines than in WT during B. cinerea infection development. Our results showed that G-type lectin genes, a big gene family in F. vesca, are involved in strawberry response to biotic and abiotic stresses, and in particular, FaMBL1 gene plays a role in inducing disease resistance, presumably in a JA-dependent pattern.
Fitness of Venturia inaequalis strains associated with resistance to different fungicides classes
G. Maddalena, B. Lecchi, D. Marcianò, S.L. Toffolatti
Studi di Milano, Dipartimento di Scienze Agrarie e Ambientali (DiSAA), via Celoria 2, 20133 Milano, Italy. E-mail : [email protected]; [email protected]
Venturia inaequalis, the causal agent of apple scab, represents one of the major threats to apple production. The repeated use of fungicides increases the selection pressure in favour of individuals resistant to different fungicide classes. To limit the spread of resistant individuals, it is mandatory to characterize their competitiveness inside the population. In this study, 35 V. inaequalis strains sampled in Lombardy and previously phenotyped for resistance towards five fungicides (dodine, cyprodinil, trifloxystrobin, boscalid, myclobutanil) of different classes, were characterized for their fitness components, to evaluate the spread and persistence potentials of resistant. For each isolate, the mycelial growth and conidial production, in terms of concentration and size, were evaluated and related to the phenotype and environment. Generally, the results highlighted the absence of significant differences between resistant and sensitive strains in terms of fitness. Only strains resistant to cyprodinil showed a significantly higher mycelial growth than sensitive ones, while strains resistant to dodine showed a significant reduced conidia size compared to sensitive ones. Moreover, the environmental conditions seemed to influence the fitness components, especially conidia differentiation. Indeed, strains sampled at higher elevation produced more conidia with reduced size, suggesting that genetic background and environmental conditions due to altitude could induce the pathogen to differentiate a higher number of conidia, providing a greater diffusion in the environment and increasing the probability to infect the host. Therefore, the resistant strains could persist and spread throughout the population, thus making essential to adopt adequate control strategies, taking into account anti-resistance practices.
The rules behind plant microbiome assembly: a big data approach
A. Malacrinò, N. Zakaria, M.G. Li Destri Nicosia, S. Mosca, G.E. Agosteo, L. Schena
Dipartimento di AGRARIA, Università degli Studi “Mediterranea” di Reggio Calabria, Loc Feo di Vito, 89131 Reggio Calabria, Italy. E-mail: [email protected]
The study of plant microbiomes is increasingly attracting a large interest worldwide, generating groundbreaking knowledge on the composition, function, and origin of the plant-associated microbial communities. However, we are only beginning to understand the rules behind the assembly of plant microbiomes. In this study, we leverage on the massive amount of publicly available data generated by over 14,000 studies (timespan 2017–2021), which we collected and re-analyzed under a common framework. Using a variety of metrics (microbiome diversity, structure, composition, co-occurrence networks) and approaches (Bayesian modelling, network comparison, machine learning), we tested several questions, including: (i) which are the major factors contributing to the assembly of plant microbiomes? (ii) do specific stressors (e.g., pathogens, drought) generate unique changes in the plant microbiomes? (iii) does plant phylogeny reconciliate with the diversity/structure of plant microbiomes? We also carefully searched each article, highlighting important methodological limitations that need to be considered in the future research endeavors. Our results contribute to a more general understanding of plant-microbiome interactions, which are thought to be key towards eco-sustainable plant protection, ecosystems conservation, and successful ecological restoration.
Selection of informing spectral bands and hyperspectral indices for the downy mildew ( Plasmopara viticola ) detection on Vitis vinifera
G. Manganiello 1 , C. Pane 2 , S. Lanzuise 1 , N. Nicastro 2 , R. Marra 1,3 , F. Carotenuto 4 , M. Lorito 1 , S.L. Woo 3,5
1Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 2Council for Agricultural Research and Economics (CREA), Research Centre for Vegetables and Ornamental Crops, Via Cavalleggeri 25, 84089 Pontecagnano Faiano, Italy; 3BAT Center—Interuniversity Center for Studies on Bioinspired Agro-Environmental Technology, University of Naples “Federico II”, Via Università 100, 80055 Portici, Italy; 4Department of Earth Science, Environment and Resources, University of Naples “Federico I”I, Via Vicinale Cupa Cintia, 21, 80126 Naples, Italy; Department of Pharmacy, University of Naples “Federico II”, 80131 Naples, Italy. E-mail: [email protected]
Downy mildew, caused by the obligate biotrophic oomycete Plasmopara viticola, is one of the most serious grapevine diseases worldwide, predominantly controlled by the application of copper-based fungicides. The large application of these products is responsible for copper (Cu) accumulation in the upper layers of soils and water sources and may cause phytotoxicity. With limitations in the use of Cu-based products imposed for organic agriculture by the European Union, research for alternative sustainable products is strongly encouraged. This project aims at reducing Cu usage by combining the application of microbial agents and natural compounds with the monitoring of disease appearing and progression by hyperspectral sensors. In this work informative spectral bands and vegetation indices describing differences between healthy and infected leaves were selected. Hyperspectral images of symptomless and diseased leaves were acquired by using the SPECIM IQ camera (Specim, Spectral Imaging Ltd., Oulu, Finland) working in the range of 400–1.000 nm. The presence of disease determined a flattening and a strongly reduction of reflectance between 570–670 and 710–1.000 nm, respectively. Vegetation indices discriminating healthy leaves were related to water content (WI, FWBI1, FWBI2) while diseased leaves were exhaustively described by chlorophyll and vegetation-associated indices (RGRcn, G, VOG3, PRI, PSRI). Spectral signature produced in this preliminary stage of research will be applied to estimate the effectiveness of bioformulation applications in ongoing field trials.
Hyperspectral imaging elucidates the ability of Trichoderma harzianum T22, Pseudomonas simiae WCS417r, and their combination to control wild rocket foliar bacterial blight through induction of systemic resistance
G. Manganiello 1,3 , A. Pascale 2 , N. Nicastro 3 , S.L. Woo 4,5,6 , C. Pane 3
1 University of Naples “Federico II”, Department of Agricultural Sciences, via Università 100, 80055, Portici, Italy; 2 Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Faculty of Science, Utrecht University Padualaan 8, 3584 CH Utrecht, The Netherlands; 3 Council for Agricultural Research and Economics (CREA), Research Centre for Vegetables and Ornamental Crops, Via Cavalleggeri 25, 84089 Pontecagnano Faiano, Italy; 4 Task Force on Microbiome Studies, University of Naples “Federico II”, 80055 Portici, Italy; 5 Department of Pharmacy, University of Naples “Federico II”, 80131 Naples, Italy; 6 Institute for Sustainable Plant Protection, National Research Council, 80055 Portici, Italy. E-mail: [email protected]
Selected members of plant microbiota can protect plants against pathogenic microorganisms through different mechanisms, including the induction of systemic resistance (ISR). Pseudomonas simiae WCS417r and Trichoderma harzianum T22 are two model beneficial microbes and their ability to induce ISR has been demonstrated on multiple plant species and against a wide range of phytopathogens. In this work we evaluated their ISR activity, in individual and combined applications on wild rocket against Xanthomonas campestris pv. campestris (Xcc) by assessing hyperspectral signature, ISR-associated gene expression and disease severity. Fifteen-dayold plantlets were transferred to peat mixed with T22, WCS417r and T22 + WCS417r, while untreated peat was used as control. After 2 weeks, 75 plants per treatment were infected with Xcc by shoot dipping. Starting from 24 h post inoculation (hpi) and for the following 4 days, each 24 h plants were subjected to hyperspectral image acquisition, disease monitoring and snap frozen for further gene expression analyses. Antagonists significantly reduced disease severity of 40 and 50% compared to the infected control, at 96 and 168 hpi, respectively. ISR activation was confirmed by analyzing marker genes of different hormone pathways which resulted differentially expressed compared to infected untreated control. Hyperspectral vegetation indices (22 out of 54 analyzed) proved to be able in discriminating different disease levels (none, low, intermediate, and high). The dynamic of vegetation indices resulted time-dependent and allowed the significant differentiation of treatments up to 72 h, before symptom appearing, thus favoring the early disease detection.
This research was funded by the MiPAAF, sub-project ‘AgroFiliere’ (AgriDigit programme) (DM 36,503.7305.2018 of 20/12/2018).
Isolation and partial characterization of fungi associated with cankers and necrosis on young holm oak ( Quercus ilex ) stands in the Salento Peninsula of Southern Italy
L. Marchese 1 , C. Del Grosso 1,2 , L. Melissano 3 , D. Palmieri 1 , G. Lima 1
1 Dipartimento Agricoltura, Ambiente e Alimenti, Università del Molise, Via F. De Sanctis, 86100 Campobasso, Italy; 2 Istituto per la Protezione Sostenibile delle Piante, Consiglio Nazionale delle Ricerche, via Amendola 122/D, 70126 Bari, Italy; 3 Regione Puglia, Dipartimento Agricoltura, Sviluppo Rurale e Ambientale, Servizio Territoriale di Lecce, via Libertà 70, 73100 Lecce, Italy. E-mail: [email protected]
Emerging plant diseases are an increasingly common problem for forest plants globally, and climate change is most likely a major predisposing factor. Such diseases may be caused by latent microorganisms cryptically associated with plants such as many fungal endophytes, originally described as generalist or widespread species and recently recognized as pathogens. The transition to a pathogenetic stage in fungi may depend on physiological alterations of the host, environmental changes and/or stress factors. The Botryosphaeriaceae family is one of the most representative examples of a very diverse group of fungi, including well-studied endophytes and latent pathogens of woody plants that typically cause stressor associated diseases. A sudden decline of young holm oak plants (Quercus ilex) has recently been observed during our inspections in the Salento peninsula (Apulia region, Italy). Samplings and isolations were carried out from symptomatic holm oak branches and trunks with cankers and necrosis. Representative fungal isolates were morphologically characterized and used in pathogenicity tests on artificially inoculated young holm oak plants. The result confirmed the aggressiveness of collected fungal isolates and their belonging to the Botryosphaeriaceae family, which include genera and fungal species generally known as weak pathogens. However, under favourable conditions, e.g., in plants affected by other biotic and/or abiotic factors, these organisms can cause serious infections, debilitating the host even to death. Further investigations are underway to identify the most virulent species of fungi involved and to better understand their role in the decline of holm oak stands.
H.J. Maree
Huanglongbing has been observed in South Africa for nearly 100 years. Unusual symptoms of yellow branches and green fruit were reported in the late 1920’s, and were linked to significant crop losses. The disease was named Greening with the aetiological agent unknown. They had no idea that these symptoms would later become part of an almost global pandemic of diseases collectively called HLB. The disease was shown to be graft transmissible and vectored by Trioza erytreae. It took more than 50 years of research to identify the causative agent as ‘Candidatus Liberibacter africanus’ (CLaf). Greening has been detected in more African countries and is spreading through infected planting material and the insect vector. Five CLaf subspecies have been identified in various Rutaceous species with one CLaf subsp. clausenae being the only subspecies for which a biovar was detected in citrus.
heat sensitive that prevents the establishment of Greening in unfavorable climatic regions. Greening has been successfully managed in regions deemed unsuitable for citriculture, due to established populations of CLaf and Trioza erytreae, through coordinated area-wide management strategies.
efforts are focused on the development of sensitive detection methods and monitoring strategies for HLB causing liberibacter species and their vectors.
Effects of mulch biofilms and biostimulants based on Trichoderma and Ascophyllum nodosum on industrial tomato plants
R. Marra 1,2 , R. Bottiglieri 1 , I. Di Mola 1 , A. Staropoli 1 , R. Loffredo 1 , E. Cozzolino 3 , A. Pironti 1 , N. Lombardi 1,2 , M. Mori 1 , S. Woo 2,4 , M. Lorito 1
1 University of Naples “Federico II”, Department of Agricultural Sciences, Via Università 100, 80055 Portici (NA), Italy; 2 BAT Center—Interuniversity Center for Studies on Bioinspired Agro-Environmental Technology, University of Naples “Federico II”, Via Università 100, 80055 Portici, Italy; 3 Council for Agricultural Research and Economics (CREA)-Research Center for Cereal and Industrial Crops, Viale Douhet, 8, 81100 Caserta, Italy; 4 University of Naples” Federico II”, Department of Pharmacy, Via Domenico Montesano 49, 80131 Naples, Italy, E-mail: [email protected]
The European Green Deal aims to reduce the use of synthetic chemical pesticides by 50% by 2030. Thus, novel strategies using low inputs of chemical products are highly desirable. Bioformulates consisting of living beneficial microbes or natural products used to improve crop production or control phytopathogens represent fundamental issues of an Integrated Pest Management (IPM) program. In this work two commercial products based on Trichoderma afroharzianum strain T22 (Trianum P®, Koppert), and a seaweed extract from Ascophyllum nodosum (Phylgreen®, Trade Corporation International) were tested on industrial tomato plants (Solanum lycopersicum var. Heinz 5108) in a field experiment. The effects of single and combined applications of microbial and plant biostimulants on the productive and qualitative traits were evaluated on plants grown on two biodegradable plastic mulch films. Results showed that mulch films increased up to 30% marketable yield compared to control, while the combined application of biostimulants increased yield up to 25% compared to untreated plants. Treatments affected also qualitative parameters, such as fruit texture and colours, and concentration of bioactive compounds (total polyphenols, carotenoid, lycopene). Finally, a LC–MS Q-TOF metabolomic analysis revealed that plant metabolic profiles varied according to the presence of biofilm as well as to the application of biostimulants. An increase in the concentration of some metabolites (alkaloids, flavonoids) in tomato berries and leaves was observed following the application of Trichoderma-based product.
Characterization of fungal pathogens associated with canker and dieback of apple trees in Northern Italy
I. Martino 1 , C. Agustí-Brisach 2 , M.L. Gullino 1 , V. Guarnaccia 1,3
1Centre for Innovation in the Agro-Environmental Sector, AGROINNOVA, University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy; 2Departamento de Agronomía, Unit of Excellence “María de Maeztu”2020–23, ETSIAM, Universidad de Córdoba, Campus de Rabanales, 14071 Córdoba, Spain; 3Department of Agricultural, Forest and Food Sciences (DISAFA), University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy. E-mail: [email protected]
Italy is one of the largest apple (Malus × domestica) producers in Europe, with 57.025 ha cultivated and more than 2 million tons of apple produced in 2021. Wood infections represent a serious threat to apple cultivation worldwide. Several pathogens can infect trunks, branches and shoots causing cankers, twig blight, wood rot and death of trees. Currently, the main fungal species found in association with canker and dieback of apple trees belong to Botryosphaeriaceae, Diaporthaceae and Diatrypaceae families. Considering the impact of wood pathogens and the scarcity of studies on wood diseases of apple in Italy, different surveys were conducted in Piedmont during 2020 − 2022. Six apple orchards and eight cultivars were surveyed. The species characterization was achieved through morphological features, optimum growth temperature assessment, and multilocus phylogenetic analyses. Seven fungal species were found: Botryosphaeria dothidea, Cadophora luteo-olivacea, Diaporthe rudis, Diplodia seriata, Eutypa lata, Kalmusia longispora and Paraconiothyrium brasiliense. A broad optimum temperature range was observed among the species (21 to 29 °C). Pathogenicity tests were conducted on plants of the representative cultivar ‘Gala’ and the virulence of all the fungal species was confirmed. During this study, the species Cadophora luteo-olivacea, Diplodia seriata, Eutypa lata, Kalmusia longispora and Paraconiothyrium brasiliense were reported for the first time as causal agents of apple canker and dieback in Italy. Moreover, the demonstrated insights on apple wood diseases lay the basis for further epidemiological and diagnostic investigations to assess the phytosanitary status of plant material and to implement effective preventive management strategies.
Development and validation of a SYBR Green qPCR assay for early and specific detection of Colletotrichum ocimi , causal agent of black spot on basil
I. Martino 1 , P. Crous 2 , A. Garibaldi 1 , M.L. Gullino 1 , V. Guarnaccia 1,3
1 Centre for Innovation in the Agro-Environmental Sector, AGROINNOVA, University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy; 2 Westerdijk Fungal Biodiversity Institute, Uppsalalaan 8, 3584 CT, Utrecht, The Netherlands; 3 Department of Agricultural, Forest and Food Sciences (DISAFA), University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy. E-mail: [email protected]
The genus Colletotrichum includes several important species pathogenic on several plant hosts. Colletotrichum ocimi is the causal agent of black spot of basil (Ocimum basilicum L.) and represents a serious threat to basil producers and seed-production companies as it can affect both leaves and seeds. Considering the importance of basil cultivation in the Mediterranean area and the absence of diagnostic methods to identify C. ocimi, a SYBR Green real-time PCR assays was developed to detect the pathogen on basil leaves and seeds. Two primer sets were designed on the partial β-tubulin (tub2) region. The selected primers pair, TubOc68fw − TubOc_197rev, produced amplicons of 130 bp. The developed SYBR Green real-time PCR assay was validated for specificity, sensitivity, selectivity, repeatability and reproducibility. The assay was specific for C. ocimi with respect to 10 Colletotrichum spp. and to another 12 species known as pathogens of basil plants. The sensitivity of the method was 1 pg µl−1 of genomic fungal DNA. The amplification analyses were not influenced by basil genomic DNA. The primer set was able to detect and quantify C. ocimi on artificially inoculated basil leaves and on artificially inoculated seeds. To our knowledge, this is the first specific primer set for the identification of C. ocimi and the developed assay can be useful for diagnostics on plants and seeds, with an implemented extraction method, thus contributing to improve the early detection of the pathogen on seeds or seedlings and the adoption of effective preventive management strategies.
Insights on grapevine varietal susceptibility to Flavescence dorée in relation with the vector Scaphoideus titanus
C. Marzachì 1 , M. Ripamonti 1,2,a , N. Bodino 1 , D. Bosco 1,2 , L. Galetto 1
1 Institute for Sustainable Plant Protection ‐ National Research Council of Italy, IPSP-CNR, Torino, Italy; 2 Department of Agriculture, Forest and Food Sciences (DiSAFA), University of Turin, Grugliasco (TO), Italy. a Present address: Environmental Research and Innovation Department (ERIN), Luxembourg Institute of Science and Technology (LIST), Belvaux, Luxembourg. E-mail: [email protected]
Flavescence dorée (FD) of grapevine is a disease associated with the homonymous phytoplasma (FDp), causing severe losses to European viticulture. The main insect vector is Scaphoideus titanus (Ball), a Nearctic leafhopper introduced into Europe. To investigate the role of the grapevine cultivar in contrasting the disease, 14 varieties were exposed to FDp, using infectious S. titanus for a one-week Inoculation Access Period (IAP). Five and eight weeks after the IAP, the presence of FDp was assessed by qPCR, drawing a complete susceptibility range for these 14 cultivars. To exclude the possibility that low susceptibility to FD may be due to poor vector fitness on the specific grapevine genotype, S. titanus fitness parameters and feeding behaviour were measure on three Vitis varieties selected to represent the extremes of the FD-susceptibility range. In particular, S. titanus fitness was studied on Barbera as FD-susceptible, Brachetto, and Moscato as FD-tolerant. Female prolificacy was chosen as fitness parameter, by means of number of beared mature eggs and vitellogenin expression at three different times after emergence. In parallel, also nymph and adult survival were measured on the three Vitis genotypes. Overall, S. titanus performed and fed better on the FD most susceptible cultivar. On the other hand, impaired vector feeding behaviour and performance may explain the poor susceptibility to the disease of the least susceptible cultivar, possibly through antibiosis and antixenosis defense mechanisms against the vector.
Molecular characterization of an Apulian isolate of tomato leaf curl New Delhi virus
M. Mastrochirico, R. Spanò, T. Mascia
Università degli Studi di Bari, Dipartimento di Scienze del Suolo, della Pianta e degli Alimenti, via Amendola 165/A , 70125 , Italy. E-mail: [email protected]
Tomato leaf curl New Delhi virus (ToLCNDV) is an emerging begomovirus (Geminiviridae family) whose economic importance in the Mediterranean basin correlates with recurrent outbreaks in tomato and in cucurbits with estimated losses between 30 and 100%. In Apulia (southern Italy), the virus was isolated in 2018 in commercial fields of zucchini squash and melon and in 2020 in protected crops of zucchini squash with heavy infestations of aleyrodids. ToLCNDV genome consists of two circular single-stranded DNA molecules of about 2.7–2.6 kb, referred to as DNA-A and DNA-B, encapsidated in geminate particles. Betasatellites could be associated with DNA affecting replication, movement within host, horizontal transmission between plants and inducing severe disease symptoms and suppression of transcriptional gene silencing. We sampled zucchini squash during 2020 outbreaks and obtained the whole sequence of the DNA-A of one of the isolates for which the name of ToLCNDV-Le is proposed. Phylogenetic analysis of the CP sequences of ToLCNDV-Le with those of 42 ToLCNDV isolates from different countries and plant species, retrieved from the public database placed ToLCNDV-Le among the Spain strains (ToLCNDV-ES) of the virus as it shares more than 94% nucleotide identity, according to ICTV strain demarcation criteria. Characteristics of the fully sequenced DNA-A of ToLCNDV-Le using the Illumina 2 × 150 bp reads platform and WGS application with a 100 × depth of coverage will be reported. Since recombination has been previously reported for ToLCNDV, we examined the ToLCNDV-Le DNA-A for any evidence of recombination as well as for its association with betasatellites.
Grafting to manage tomato leaf curl New Delhi virus in cucurbits
M. Mastrochirico, R. Spanò, T. Mascia
Università degli Studi di Bari, Dipartimento di Scienze del Suolo, della pianta e degli Alimenti, via Amendola 165/A , 70125 , Italy. E-mail: [email protected]
Tomato leaf curl New Delhi virus (ToLCNDV) is an emerging begomovirus (Geminiviridae family) listed in the EPPO Alert-list 2, present in the Mediterranean area and in Italy, where it was reported in 2015 in Sicilian courgette and since then its recurrent outbreaks generated justified concern among growers. The virus is particularly harmful in cucurbits, where it causes 100% production losses, thus a sustainable and environmentally friendly approach must be adopted. Genetic resistances have been identified in Cucurbita moschata and Luffa cylindrica, but the graft could provide a faster and more flexible solution inducing tolerance rather than resistance, as shown in tomato crops by grafting susceptible commercial tomato varieties onto the tomato wild ecotype Manduria (Ma). Here we report results of a screening among twenty-one local cucurbit ecotypes to evaluate tolerance levels against mechanical transmission of ToLCNDV. Results will lead to the identification of potential rootstocks to attain suitable levels of tolerance against the virus in commercial cucurbit varieties. Plants were challenged with ToLCNDV isolated in Apulia and observed for disease symptoms development and viral DNA accumulation by quantitative dot-blot assays at 14 and 28 days after inoculation. C. melo var. Retato standard (F1 commercial hybrid) and C. pepo var. Scuro di Milano proved the most susceptible, whereas C. melo var. Barattiere and C. pepo accession 5 the most tolerant. Tolerant plants did not show disease symptoms and very low level of virus accumulation, suggesting their use as rootstocks of grafted cucurbits against ToLCNDV outbreaks.
Raman spectroscopy detection of virus infection in asymptomatic tomato and grapevine plants
S. Matić 1 , C. D’Errico 1 , L. Mandrile 2 , G. Barzan 2 , A.M. Giovannozzi 2 , L. Miozzi 1 , A.M. Vaira 1 , F. Nuzzo 1 , G. Gambino 1 , A.M. Rossi 2 , E. Noris 1
1 Institute for Sustainable Plant Protection, National Research Council of Italy (IPSP-CNR), Strada delle Cacce 73, 10135 Turin, Italy; 2 Italian National Institute of Metrology Research, Strada delle Cacce 91, 10135 Turin, Italy. E-mail: [email protected]
Plants are exposed to a huge variety of biotic stresses caused by different pathogens and react to them by activating several metabolic pathways. Among pathogens, viruses are the most difficult to control and their reliable detection in the early stages of the disease may help to reduce their spread and alleviate the economic impact. Besides laborious, costly and destructive diagnostic serological and molecular techniques, Raman spectroscopy (RS) is an innovative alternative method for a quick, cheap and non-destructive pathogen detection by the creation of a sample chemical fingerprint. In this study, the efficiency of RS (in combination with chemometric analysis) in virus detection in asymptomatic samples and in the monitoring of the virus infection progress in two selected agricultural crops was investigated. Tomato plants infected by tomato yellow leaf curl Sardinia virus (TYLCSV) and tomato spotted wilt virus (TSWV), and grapevine plants infected by grapevine fanleaf virus (GFLV) and grapevine rupestris stem pitting-associated virus (GRSPaV) were analysed. RS successfully differentiated the RS profiles of healthy and virus-infected asymptomatic plants with 70 and 85% precision for TYLCSV and TSWV, respectively, and with 80 and 100% accuracy for GRSPaV and GFLV in grapevine, respectively. Metabolic changes in chlorophylls, carotenoids, and polyphenolic compounds occurring in asymptomatic infected leaves were identified as the principal biomarkers. The potential uses of this emerging and cutting-edge technique for real-time in-field virus detection in crops will be outlined.
Plant pest surveillance program and survey results in Italy
E. Maurizi 1 , L. Damiano 1 , M. Torrigiani 1 , F. Mosconi. 1 , L. Marianelli 2 , P. D’Amaro 1 , S. Rosati 1 , L. Campus 3 , F.P. Roversi 2 , L. Tomassoli 1
1 Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria – Centro di ricerca Difesa e Certificazione, Via Carlo Giuseppe Bertero, 22—00156 Roma, Italy; 2 Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria – Centro di ricerca Difesa e Certificazione, Via di Lanciola 12/a, 50125 Firenze, Italy; 3 Ministero delle politiche agricole alimentari e forestali – Dipartimento delle politiche europee ed internazionali e dello sviluppo rurale – Direzione generale dello sviluppo rurale, DISR V – Servizio fitosanitario centrale, Via XX Settembre 2, 00187 Roma, Italy. E-mail: [email protected]
Since the second half of the nineteenth century, the introduction of plant pests has been a critical issue for national agricultural production. First in Italy, Umberto I of Savoy created a network of consortia with the purpose of monitoring phylloxera infestations. Then, processes such as trade globalization and climate change have greatly facilitated the introduction of non-native pathogens. Nowadays, EU legislation obligates Member States for annual plant health surveys carried out by the Regional Phytosanitary Services. The aim of the surveys is to confirm the pest-free status in areas where their presence is not known, and that any infections in the territory can be promptly eradicated. In 2021, Italy surveyed 82 pests, 34 of which included fungi, bacteria, viruses, and phytoplasma listed either as priority pests (i.e., pests subject to Union emergency measures) or as pests listed in parts A and B of Annex II of Regulation (EU) 2019/2072, as well as other pests of EU and/or national interest. The methodology for pathogen surveys consisted primarily of visual examinations supported by sampling in case of suspicious symptoms. Trapping have been performed for pathogen vectors, such as Philaenus spumarius, Pityophthorus juglandis, and Scaphoideus titanus. Samples were analyzed in official laboratories, with specific tests such as morphological identification, serological and molecular tests. The major survey activity concerned Xylella fastidiosa, Phyllosticta citricarpa, Pantoea stewartii subsp. stewartii, potato bacteria, Flavescence dorée and tomato brown rugose virus. In case of new findings, official emergency measures have promptly applied.
This activity was supported by the National Phytosanitary Service and by European Union Grants.
Evaluation of Gnomoniopsis castaneae incidence in Campania region under different agricultural systems
F. Mele, L. Gualtieri, R. Bossa, S. Gargiulo, M. Castaldo, P.A. Pedata, M.M. Monti, M. Ruocco
Istituto per la Protezione Sostenibile delle Piante SS Portici (CNR-IPSP), Piazzale Enrico Fermi I , 80055 Portici (NA), Italy. E-mail: [email protected]
Gnomoniopsis castaneae (Gnomoniaceae, Diaporthales) is an ascomycete fungus capable of colonizing different plant species, belonging to the families Fagaceae, Onagraceae and Rosaceae. It is the major rot agent on chestnut fruits, also associated with cankers on both chestnut and hazelnut, as well as necrosis on leaves and chestnut galls caused by Dryocosmus kuriphilus. In the last 10 years, the presence of G. castaneae caused a severe plant health emergency in Campania region, so that some chestnut productions were totally lost. Aim of this work was to evaluate the best field treatment (organic or IPM) against G. castaneae diffusion in chestnut cultivation. Different field sites were treated following: organic protocol, Integrated Pest Management (IPM) and no treatments, sampling in two different seasons: autumn 2020 and 2021. The disease incidence was assessed both by visual inspections of chestnut fruits and by evaluating the occult presence of the pathogen (identified as potential rot) by axenic isolation of G. castaneae from asymptomatic chestnuts. Molecular analyses by PCR amplification of ITS region of ribosomal DNA, confirmed the identity of morphologically identified G. castaneae isolates. Results demonstrated that both treatments are effective in controlling G. castaneae development in chestnut fruits. In chestnut fields treated following organic system, visible infection reached 15% and the potential rot was around 50%. In IPM treated fields a very low percentage of chestnut showed visible rot (3–4%) while 30–35% were asymptomatic but infected samples. In control field, were no treatment were carried out, visible chestnut rot was higher than 50% and the potential rot reached 100% of samples.
This work was supported by the project MIGLIORCAST funded by PSR 2014-2020 CAMPANIA action 16.1.1.
Trichoderma afroharzianum causing seed rot on maize in Italy
M. Mezzalama 1,2 , M. Sanna 1 , M. Pugliese 1,2 , M.L. Gullino 1
1 Centre for Innovation in the Agro-Environmental Sector, AGROINNOVA, University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy; 2 Department of Agricultural, Forest and Food Sciences (DISAFA), University of Torino, Largo Braccini 2, 10095 Grugliasco (TO), Italy. E-mail: [email protected]
Maize (Zea mays L.) is a cereal crop of great economic importance in Italy. Trichoderma species are widespread filamentous fungi in soil, well known and studied as biological control. In September 2020 during a routine seed testing procedure to check the phytosanitary condition of seeds of a yellow grain hybrid (class FAO 700, 132 days) collected from an experimental field located in Carmagnola (TO, Italy: GPS: 44°53′11.0"N 7°40′60.0"E) the presence of green mycelium typical of the genus Trichoderma over the 400 seeds tested was greater than 50%. Due to the high and unexpected percentage of decaying kernels, representative strains were identified through morphological features assessment and by sequence comparison of ITS, rpb2, and tef-1α gene fragments, revealing 100% identity to Trichoderma afroharzianum. Pathogenicity tests were carried out by injecting into the silk channel 1 ml of the conidial suspension (106 conidia/ml) obtained from the strain of T. afroharzianum, seven days after the silk channel emergence (BBCH 65). Ears were removed four weeks after inoculation and disease severity, reaching up to 75% of the kernels of the cobs, was visually assessed. T. afroharzianum has been already reported on maize in Germany and France as causal agent of ear rot of maize. The potential production of mycotoxins and the losses that can be caused by the pathogen during post-harvest need to be explored. To our knowledge this is the first report of T. afroharzianum as a pathogen of maize in Italy.
Identification of apple rubbery wood virus 1 and apple rubbery wood virus 2 infecting pear and apple in Campania and development of a detection assay for rubodviruses
M. Minutolo 1 , M. Cinque 1 , F. Di Serio 2 , B. Navarro 2 , D. Alioto 1
1 Dipartimento di Agraria, Università degli Studi di Napoli “Federico II”, 80055 Portici, Italy; 2 Istituto per la Protezione Sostenibile delle Piante, Consiglio Nazionale delle Ricerche, 70126 Bari, Italy. E-mail: [email protected]
Apple rubbery wood virus 1 (ARWV-1) and apple rubbery wood virus 2 (ARWV-2) are two negative-stranded RNA viruses recently discovered in apple and pear. Both viruses have been classified as Apple rubodvirus 1 and Apple rubodvirus 2 in the genus Rubodvirus, which includes other two species Grapevine rubodvirus 1 and Grapevine rubodvirus 2, identified in grapevine. In 2021, during a survey to assess the sanitary status of pear and apple trees in Campania region (southern Italy), leaves were collected and assessed by RT-PCR using primer pairs designed for the specific detection of ARWV-1 and ARWV-2. Amplicons of 312 and 272-bp expected for ARWV-1 and ARWV-2, respectively, were obtained in pear and apple samples. Of the 50 tested pear trees, one resulted mixed infected by ARWV-1 and ARWV-2 and 19 by ARWV-2 only. Of the 70 analyzed apple trees, four were infected by ARWV-1, 11 by ARWV-2 and three by both ARWV-1 and ARWV-2. Virus identity was confirmed by sequencing the amplicons, which shared 95.1–97.7 and 95.6–99.6% nucleotide identity with ARWV-1 and ARWV-2 sequences from GenBank, respectively. To our knowledge, this is the first report of ARWV1 and ARWV-2 in Italy and the first report of ARWV1 in pear. A degenerate primer pair, targeting a conserved domain in the RNA-dependent RNA polymerase of known rubodviruses was designed to develop a RT-PCR assay able to diagnose these viruses at genus level. Preliminary results showed that the assay allows the detection of ARWV-1 and ARWV-2 in single and mixed infection. Further studies are necessary to evaluate the ability of the assay to detect the rubodviruses from grapevine and eventually new rubodviruses from other host plants.
This work was supported by Regione Campania (URCOFI project).
Results of three-year survey conducted on EU quarantine pests of Citrus in Campania territory
M. Minutolo 1 , L. Notaro 1,2 , M. Cinque 1 , A. Alioto 1 , E. Barra 1,3 , R. Griffo 2
1 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 2 Plant Protection Service of Campania Region, Centro Direzionale, 80124 Napoli, Italy; 3 Phytopathological Laboratory, Campania Region, 80141 Napoli, Italy. E-mail : [email protected]
In order to cope with the increasing introduction and spread of new harmful organisms, the European Union has issued a series of regulations that are the basis of the "new plant health regime". For the Member States, territorial controls are foreseen through the preparation of surveys both for EU quarantine pests and for the 20 priority pests (Regulation (EU) 2019/1702). For a better phytosanitary surveillance of the territory, the Campania region has activated a partnership with the Department of Agriculture of the University of Naples "Federico II", with the National Research Council – IPSP, Unit of Portici and the Council for Research in Agriculture and the Analysis of agricultural economics—CREA, going to establish the Regional Unit of Phytosanitary Coordination—URCoFi. The regional survey program includes specific activities (i.e., visual surveys, sampling, laboratory analysis, use of traps) to be carried out on about eighty quarantine pests. Given the agricultural, environmental and territorial importance of citrus and the high risks of introduction of new harmful organisms through the large quantities of citrus fruit imported from third countries, particular attention is paid on the protection of this culture. Here, we present the results of a large-scale survey carried out during 2019–2021 in Campania, southern Italy, on the occurrence of Phyllosticta citricarpa, the agent of citrus black spot disease, Candidatutus Liberibacter asiaticus, Ca. L. africanus and Ca. L. americanus, the agents of Huanglongbing, and their two vectors Diaphorina citri and Trioza erytreae, and non-EU genotypes of citrus tristeza virus (CTV) and its vector Toxoptera citricidus.
Nanoplate-based digital PCR for early detection of different quarantine pathogens of ornamental plants
N. Miotti 1 , A. Passera 1 , I. Ferraris 2 , F. Gaffuri 3 , A. Follador 1 , L. Giordano 3 , V. Grosso 2 , A. Leone 1 , M. Rossato 2 , P. Casati 1 , M. Delledonne 2 , P.A. Bianco. 1
1 Università degli Studi di Milano, Dipartimento di Scienze Agrarie e Ambientali, Via Celoria 2, 20133, Milano; 2 Università degli Studi di Verona, Dipartimento di Biotecnologie, Strada le Grazie 15, 37134, Verona; 3 Laboratorio del Servizio Fitosanitario Regione Lombardia, c/o Fondazione Minoprio, Viale Raimondi 54, 22070, Vertemate con Minoprio (CO). E-mail: [email protected]
The early detection of quarantine pathogens is pivotal to reduce the risk of its introduction into new ecosystems, where they may find susceptible hosts and/or environments favoring their pathogenic behavior, especially in the context of growing global mobility and trade. In this scenario, we tested the applicability of nanoplate-based digital PCR (dPCR) for early, accurate, and quantitative detection of different quarantine pathogens, including bacteria, oomycetes, and viruses, that can infect ornamental plants. The selected pathogens were Erwinia amylovora, Ralstonia solanacearum, Phytophthora ramorum, and tomato spotted wilt virus (TSWV). The assays were carried out using nucleic acids samples of healthy plants spiked with known quantities of bacterial DNA or, for obligate pathogens, nucleic acids extracted from experimentally infected plants. Results were compared with those obtained with validated quantitative PCR (qPCR) techniques. The nanoplate-based dPCR provided reliable detection and quantification for three of the four selected pathogens (E. amylovora, P. ramorum, TSWV), with a limit of detection of 10 target copies per microliter and 100% accuracy, sensitivity, and specificity. For the last pathogen, R. solanacearum, on the other hand, the results did not reach the expected performance criteria, therefore, other tests are needed in order to optimize the diagnostic assay. In conclusion, our results suggest that nanoplate-based dPCR could become an efficient upgrade of qPCR for the early detection of quarantine pathogens on ornamental plants.
This work was supported by Regione Lombardia, GARDING project.
Agrobacterium mediated production of CanCV particles in N. benthamiana
N. Miotti 1 , N. Zampolli 2 , D. Baldo 2 , R. Paris 3 , C. Ratti 2 , P. Casati 1 , M. Dall’Ara 2
1Department of Agricultural and Food Sciences—Production, Landscape, Agroenergy—University of Milan, Via Celoria 2, 20133, Milan, Italy; 2Department of Agricultural and Food Sciences—Plant Pathology—University of Bologna, Bologna, Viale Giuseppe Fanin 40, 40127 Bologna, Italy; 3Council for Research in Agriculture and Analysis of the Agricultural Economy (CREA), Center of Research for Industrial Crops, 40128 Bologna, Italy. E-mail: [email protected]
Cannabis cryptic virus (CanCV) belongs to the Partitiviridae family and to the Betapartitivirus genus that includes virus infecting both plants and fungi. CanCV is bipartite and has a dsRNA genome encapsidated in small, isometric, non-enveloped particles. Each single sense strand RNA (ssRNA( +)) is monocistronic: one codes for the RNA dependent RNA polymerase while the other for the coat protein. The lack of movement proteins makes these viruses incapable to move systemically in mechanical inoculated plants. CanCV life cycle is restricted to the cytoplasm and the virus propagates exclusively during cell divisions. Other betapartitiviruses infecting fungi are able to propagate in new cells during processes of anastomosis and sporogenesis. In this work, we demonstrate the possibility of obtaining CanCV particles using a Nicotiana benthamiana plants as bioreactors. For this purpose we cloned the cDNA copy of each CanCV ssRNA( +) into an binary expression vector to be used by Agrobacterium tumefaciens. Agroclone infectivity was tested by their co-infiltration in mature leaves with or without the transient expression of tomato bushy stunt virus p19 used to suppress the sense transgene-induced post-transcriptional gene silencing. The system turns out to be very economical and the particles yield, using purification techniques, is far higher than what could be obtained from the same amount of biological material infected by a natural infection. We believe this system could be applied to other partitivirids allowing to obtain a high number of viral particles to be used in future experiments of characterization/transfection in fungal or plant protoplasts.
Fungi associated with a trunk disease in young grapevine plants in Sicily (Italy)
G. Mirabile 1 , A. Agnello 2 , P. Bella 1 , N. Bova 1 , S. Bertacca, A. Caruso 1 , S. Panno 1 , S. Davino 1 , L. Torta 1
1 Dipartimento di Scienze Agrarie, Alimentari e Forestali, Università degli Studi di Palermo, Viale delle Scienze Ed. 4, 90128 Palermo, Italy; 2 Studio Agronomico, Via Botticelli n.11, 90144 Palermo, Italy. E-mail: [email protected]
In spring 2022, in the territory of Sommatino (CL), 10-year-old plants of grapevine cv. Sangiovese and Insolia grafted on 140 Ruggeri, showed late sprout, stunted and chlorotic vegetation and evident xylem browning, starting from the rootstock. In the field, this syndrome affected about 20% of the plants of both cultivars. At the Plant Pathology laboratories of the SAAF Department symptomatic samples were submitted to isolation tests aimed at identifying associated fungal microorganisms. In particular, 3 Sangiovese and 3 Insolia grapevines were dissected and subcortical tissue fragments of approximately 2–3 mm were placed in Petri dishes, on PDA. The fungal colonies were identified with traditional (macro- and microscopic observations) and molecular methods (ITS and partial β-tubulin gene sequences). The results of these observations ascertained the presence of Neofusicoccum vitifusiforme and Phoma sp. in the plants of cv. Sangiovese and Arthrinium arundinis, Libertasomyces platani and Seimatosporium vitis in those of cv. Insolia. Among these fungi, only a few Phoma species, N. vitifusiforme and S. vitis have been reported as causal agents of grapevine trunk diseases. Further investigations in the field, to study the evolution of the disease, and in the laboratory, both to evaluate the spread of isolated fungal microorganisms and to satisfy Koch's postulates, will clarify the role of these fungi in the manifestation of the described disease.
Aspergillus contaminating food and feed: biocontrol assays and new diagnostic method (LAMP)
G. Mirabile 1 , P. Bella 1 , M. Alberto Vazquez 2 , A.G. Caruso 1 , S. Panno 1 , S. Davino 1 , L. Torta 1
1 Università degli Studi di Palermo, Dipartimento Scienze Agrarie, Alimentari e Forestali, Viale delle Scienze Ed. 5, 90128 Palermo, Italy; 2 Instituto de Ciencia Animal, Carretera Central, Km 47 ½, 32 700 San José de las Lajas, La Habana. E-mail: [email protected]; [email protected]
Contamination by Aspergillus is considered the main cause of spoilage in food and feed and of toxicosis in consumers. A previous study aimed to assess fungal contamination in 48 feed samples allowed us to isolate 47 strains belonging to 5 sections and characterized by cellulolytic activity. In order to develop biocontrol strategies, dual culture assays were carried out by using four fungi (Penicillium italicum Pi, Trichoderma atroviride P1, T. harzianum T22, T. pleuroticola Tp) and a bacterium (Bacillus amyloliquefaciens AG1) against 7 Aspergillus isolates. AG1 showed the highest antagonistic activity against all the aspergilli, with percentage growth inhibition (PGI %) ranging from 46.53% to 60.92%. The ability of AG1 to reduce aspergilli growth and sporulation was also confirmed by turbidimetric assay. In the poison agar assay, T22 culture filtrate induced a percentage of growth inhibition from 0 to 27.41%. In in vivo test conducted on artificially contaminated maize kernels, Aspergillus growth was reduced when P1 was inoculated three days before fungal contamination. Moreover, in order to rapidly detect A. flavus, the most common aflatoxin-producing species worldwide, LAMP assay based on the amplification of its rmt-A gene, involved in the production of conidia and sclerotia as well as in the expression of regulators in the aflatoxin gene cluster, was developed. Assay specificity was tested using the genomic DNA of 10 Aspergillus species. Among them only genomic DNA of A. flavus was amplified and the detection limit of the assay was 1.03 pg of DNA/reaction and 1 conidium/reaction in artificially contaminated samples.
Pleurotus ostreatus associated with white rot decay in a monumental specimen of Ficus macrophylla subsp. columnaris in Palermo
G. Mirabile 1 , G. La Mantia 2 , P. Bella 1 , N. Bova 1 , S. Bertacca 1 , A. Caruso 1 , S. Panno 1 , S. Davino 1 , L. Torta 1
1Dipartimento di Scienze Agrarie, Alimentari e Forestali, Università degli Studi di Palermo, Viale delle Scienze Ed. 4, 90128 Palermo, Italy; 2Comune di Palermo, Area del Decoro Urbano e del Verde, Viale Diana, ex Scuderie Reali, 90142, Palermo, Italy. E-mail: [email protected]
Among the most famous monumental trees in Italy, the giant specimen of Ficus macrophylla subsp. columnaris (syn.: Ficus magnolioides var. magnolioides) which grows in Palermo, in the Giardino Garibaldi (Piazza Marina), is considered the largest European tree (30 m high, more than 21 m of circumference at the base of the trunk and 50 m of foliage diameter). Originating from Australia and planted in the 1863, the tree grew quickly and luxuriantly. Recently, however, some crashes of entire branches have raised serious concerns about the plant health. In the summer of 2021, pleurotus-like basidiomata were also observed, associated with white rot in the wood near the large wounds caused by the breaking of the branches. Samples of these basidiomata, subjected to microscopic observations, showed basidia, spores and cystidia typical for shape and dimension to Pleurotus ostreatus. Isolation tests permitted to obtain white fungal colonies with dense, fluffy mycelium and abundant dark orange exudates. Preliminary molecular analyzes, based on the amplification of the ITS region, confirmed the morphological identification. P. ostreatus, reported for the first time on this host, poses a serious risk to the stability and health of the monumental tree. Therefore, constant and accurate monitoring of the entire plant is considered necessary in order to define the most appropriate intervention strategies aimed to maintaining the best vegetative state of the tree.
Fungi associated with wood decay in olive trees in Trapani province (Sicily, Italy)
G. Mirabile 1 , J.G. Muratore 2 , M. Lamendola 1 , P. Bella 1 , N. Bova 1 , S. Bertacca 1 , A. Caruso 1 , S. Panno 1 , S. Davino 1 , L. Torta 1
1Dipartimento di Scienze Agrarie, Alimentari e Forestali, Università degli Studi di Palermo, Viale delle Scienze Ed. 4, 90128 Palermo, Italy; 2Ente Sviluppo Agricolo, Servizi allo Sviluppo – Regione Sicilia, Via Libertà n° 203, 90143 Palermo, Italy. E-mail: [email protected]
Recently, in several localities of the province of Trapani (Sicily, Italy), many olive growers have reported cases of decay of olive trees cv. Nocellara del Belice, characterized by defoliation, drying of the branches, browning of the internal xylem tissues and reduced production. Internal symptoms consisted also in white and brown wood rot, starting from the base of the trunk. These alterations were observed in plants irrigated with a pipe system at the trunk with spray sprinklers. A collaboration was therefore started with the Ente di Sviluppo Agricolo of the Sicilian Region to investigate on the presence of fungi associated with this syndrome. Isolation tests were started on sections of symptomatic plants. In particular seven decaying olive trees were selected in the territories of Castelvetrano and Campobello di Mazara. Three trunk disks (thickness 10 cm) for tree were taken from the collar, at about 100 cm and 200 cm from soil. For each trunk disk, five plates containing PDA were used, placing five wood fragments (3 × 3 mm) for plate. Among the grown colonies, the most recurrent were identified by morphological (macro- and microscopical observations) and molecular (ITS and β-tubulin amplification) analyses. The first results showed the presence of Coriolopsis gallica and Fomitiporia mediterranea, two basidiomycetes well-known as agent of white wood rot, and Pleurostomophora richardsiae, already reported as an aggressive fungal pathogen of decaying olive trees in Southern Brazil and in Puglia (Southern Italy). Regarding C. gallica and P. richardsiae, this is the first report in olive trees in Sicily.
Survey of oomycetes associated with the Kiwifruit Vine Decline Syndrome using a baiting approach
S. Mosca, M.G. Li Destri Nicosia, A. Malacrinò, N.Z.M. Mohamed, V. Cianci, V. Vadalà, S. Procopio, G.E. Agosteo, L. Schena
Dipartimento di Agraria, Università degli Studi “Mediterranea” di Reggio Calabria, Loc. Feo di Vito, 89124 , Reggio Calabria, Italy. E-mail: [email protected]
Kiwifruit Vine Decline Syndrome (KVDS), initially detected in northern Italy in 2012, is a severe destructive disease currently spread within the major kiwifruit growing areas in Italy. The syndrome causes the collapse of plants due to root rot. The etiology of KVDS is still far from being clearly defined, although symptoms and pattern of spread suggest that biological agents might be involved. Different microorganisms have been associated with the syndrome, although oomycetes are those more frequently isolated, in particular species of Phytopythium. Based on these observations, we used a leaf baiting approach to characterize the oomycete community of the rhizosphere soils from the Gioia Tauro plain (Reggio Calabria, Italy), one of the major national kiwifruits growing areas. The use of bating with fresh leaves of carob (Ceratonia siliqua) was strategic to aid the isolation of pathogens with motile spores, which might be hindered by other fast-growing microorganisms on culturing media. The survey was carried over spring 2021 and 2022, and confirmed the abundant presence of potential plant-pathogenic oomycetes, preferentially but not exclusively associated with plants showing severe KVDS symptoms. Interestingly, the identification of representative isolates through barcoding the ITS1-5.8S-ITS2 region of the rDNA highlighted the abundant presence of Phytophthora species, including P. citricola, P. pseudocryptogea, P. plurivora, P. crassamura, and P. megasperma. A high incidence of Phytopythium vexans was also detected, while Pythium species were less frequent. The prevalence of Phytophthora species, widely known as aggressive plant pathogens, might contribute to clarify the etiology of KVDS.
This work was supported by "Programma Operativo Nazionale (PON)—D.M. 10 agosto 2021, n. 1062, Azione IV.6, Contratti di ricerca su tematiche Green" and by "Orizzonte Soc. Coop. Agr., Melicucco (RC)" with a research contract entitled "Indagini sulle cause della Moria del Kiwi in Calabria e possibili strategie di contenimento".
Conservation Detection Dogs for the detection of Tilletia indica and Xylella fastidiosa : a training project
F. Mosconi 1 , V. Bergamaschi 1,2 , A. Infantino 1 , S. Loreti 1 , A. Matere 1 , L. Riccioni 1 , P.F. Roversi 3
1 Council for Agricultural Research and Economics—Research Centre for Plant Protection and Certification (CREA—DC), Via C. G. Bertero 22, 00156 Roma, Italy; 2 Department of Environmental Biology, Sapienza University of Rome, P.le Aldo Moro 5, 00185 Rome, Italy; 3 Council for Agricultural Research and Economics—Research Centre for Plant Protection and Certification (CREA—DC), Via Lanciola 12A, 5,125 Impruneta (FI), Italy. E-mail: [email protected]
In recent years, dogs are increasingly used in the search for biological targets of economic interest, otherwise difficult to detect. Actually, phytosanitary is one of the fields of application of dogs mainly developed, in which dogs are a useful tool both for the interception of infected plants across borders and to detect and characterize possible outbreaks of new diseases. Only a few cases of dogs trained to recognize plant pathogens are known, including Xanthomonas citri, Candidatus Liberibacter asiaticus, some species of the Phytophthora genus and Xylella fastidiosa (unpublished data). The low number of studies concerning quarantine plant pathogens is probably due to the difficulties in their handling, mainly related to the need to manipulate these organisms only in accredited quarantine laboratories with specific security protocols that could be incompatible with training. A workflow aimed at training two dogs that will target two quarantine pathogens, Tilletia indica and Xylella fastidiosa, is proposed. As a first step, the imprinting phase will be carried out transferring the target odor of both T. indica and X. fastidiosa to an adsorbent support or using naturally infected plants as target (only for X. fastidiosa). Subsequently, the discrimination phase will be carried out by teaching the dogs to locate X. fastidiosa in association with various plant species and to distinguish T. indica from other species of the genus Tilletia which can be present in grain loads. The correspondence between the VOCs of the target species and the targets used for training will be validated using an electronic nose.
This work is funded under the PROTEGGO 1.4 (Point 15) collaboration agreement between MIPAAF – CREA (ex art. 15 L. 241/90).
Bois noir disease incidence is reduced by grafting of shoots from recovered grapevines
A. Moussa 1 , N. Miotti 1 , M. Faccincani 2 , F. Serina 2 , S. Torcoli 2 , A. Passera 1 , P. Casati 1 , P.A. Bianco 1 , N. Mori 3 , F. Quaglino 1
1Department of Agricultural and Environmental Sciences – Production, Landscape, Agroenergy, University of Milan, via Celoria 2, 20133 Milan, Italy; 2The Consortium for the protection of Franciacorta, via G. Verdi 53, 25030 Erbusco, BS, Italy; 3Department of Biotechnology, University of Verona, Strada Le Grazie 15, 37134 Verona, Italy. E-mail: [email protected]
Bois noir (BN), the most widespread disease of the grapevine yellows complex, is associated with ‘Candidatus Phytoplasma solani’ (CaPsol). Due to its multifaceted ecology, BN control is extremely difficult. Several studies showed that BN recovery can be elicited by abiotic stresses and treatment with resistance inducers. In this study, field trials along with molecular analyses have been conducted to evaluate if grafting of shoots from recovered grapevine plants can increase the BN recovery rate in symptomatic grapevines and decrease the new CaPsol infection rate on asymptomatic grapevines. Field trials were performed in two BN-affected vineyards (cv. Chardonnay/Kober 5BB) in Franciacorta (Lombardy Region, northern Italy). Grafting effects were evaluated for three consecutive years by symptom observation and CaPsol detection by nested-PCR amplification of stamp gene and compared with non-grafted control vines. Obtained data showed that BN incidence was lower in grafted plants, mainly due to a statistically significant increase of recovery rate, four times higher than in non-grafted plants. These data indicated that grafting of recovered shoots can efficiently induce BN recovery, opening an interesting scenario for its utilization in sustainable strategies of vineyard management.
This work was supported by The Consortium for the protection of Franciacorta.
Establishment of a new concept agronomic protocol to push eco-sustainability in the hazelnut agroecosystem
A. Napolitano, A. Sacco, F. Palomba, L. Gualtieri, D.G. Crispo, M.M. Monti, M. Ruocco
Istituto per la Protezione Sostenibile delle Piante SS Portici (CNR-IPSP), Piazzale Enrico Fermi I, 80055 Portici (NA), Italy. E-mail: [email protected]
Hazelnut (Corylus avellana) cultivation has become predominant over other crops, so much so that in some areas of Campania region it is considered ubiquitous. As the most adopted approach for hazelnut cultivation is intensive agriculture, the aim of the present study is to create a synergy between anthropogenic action with the natural balance of the agroecosystem, optimizing the use of productive inputs and maximizing environmental and economic sustainability. To this end a new eco-sustainable agronomic protocol, carrying out environmental and agronomic monitoring aimed at assessing the establishment of new ecological balances was applied. The experimental protocol is taking place in one hectare hazelnut grove located in Presenzano (CE), compared to the organic and adopted one of the partner company. The experimental protocol involves different aspects, as well as soil management by using grassing and adding to the soil manure previously inoculated with biostimulants (Trichoderma spp., rhizosphere bacteria, mycorrhizal fungi). To control pests and pathogens biofungicides, bioinsecticides and natural bactericides conjugated to corroborants are used. Evaluation of the eco-sustainability of the experimental protocol, is carried out with meta-genomics analysis after DNA extraction from manure and soil, in order to characterize the microbiome, pre- and post-treatment and denote any increase in microbial biodiversity and its new, possible interactions. During the trial, agro-climatic conditions at the site are monitored with a micro-climatic station equipped with a decision support system by correlating soil climate data with all the microbiome recorded data in order to better define the new concept agronomic crop management.
This work was supported by the project TESTIMONE funded by PSR 2014–2020 CAMPANIA-GAL ALTO CASERTANO action 16.1.1.
Viroid-like RNAs with hammerhead ribozymes in both polarity strands identified in a metagenomic study on citrus
B. Navarro 1 , S. Li 1,2 , A. Gisel 3,4 , M. Chiumenti 1 , M. Minutolo 5 , D. Alioto 5 , F. Di Serio 1
1 Institute for Sustainable Plant Protection, Consiglio Nazionale delle Ricerche, Bari, Italy; 2 Institute of Plant Protection of Chinese Academy of Agricultural Sciences, Beijing, China; 3 Istituto di Tecnologie Biomediche, Consiglio Nazionale delle Ricerche, Bari, Italy; 4 International Institute of Tropical Agriculture, 200001 Ibadan, Nigeria; 5 Dipartimento di Agraria, Università degli Studi di Napoli Federico II, Portici (NA), Italy. E-mail: [email protected]
Applying a specific pipeline aimed to search for hammerhead ribozymes (HHRzs) in RNAseq libraries, a novel small (550 nt) non-protein coding viroid-like RNA (Vd-LRNA) containing a HHRz in each polarity strand and adopting a rod-like conformation was identified in a library generated from young leaves and stem tissues of a Citrus reticulata tree. The new Vd-LRNA was named hammerhead viroid-like RNA 1 (HVd-LR1) and its existence confirmed by Northern-blot hybridization assays. HVd-LRNA1 circular forms of both polarity strands were detected. Cloning and sequencing of full-length amplicons generated by RT-PCR using three pairs of specific primers showed a high sequence variability preserving the rod-like secondary structure of minimal free energy and the conserved motifs of HHRzs. Self-cleaving activity of HVd-LR1 mediated by the HHRz in each polarity strands was confirmed during transcription and in the absence of any protein, providing evidence of their major role during replication through a symmetric pathway of the rolling circle mechanism. The existence of a DNA counterpart of HVd-LR1 was excluded based on PCR amplification assays. HVd-LR1 was not graft-transmissible to other citrus indicator seedlings (sour orange and grapefruit) and attempts of identifying it in the original citrus source tree by RT-PCR and/or HTS failed so far. Altogether these data show that HVd-LR1 was only transiently associated with the field tree, thus excluding that it is a new plant viroid. The possibility that Hvd-LR1 may be an infectious agent of another organism (i.e., a fungus) will be discussed.
This project has been partially funded from the European Union’s Horizon 2020 Research and InnovationProgram under the Marie Skłodowska-Curie grant agreement no. 734736. This article reflects only the author’s view, and the agency is not responsible for any use that may be made of the information it contains.
Bio-based treatments with laminarin, chitosan and Trichoderma harzianum to prevent alternariosis and grey mold on tomato in Mediterranean greenhouse system
N. Nicastro 1 , A. Navarro 1 , A. Affinito 2 , N. Loret 2 , F. Scotto di Covella 1 , C. Pane 1
1 Consiglio per la Ricerca in Agricoltura e l’Analisi dell’Economia Agraria, Centro di ricerca Orticoltura e Florovivaismo, Via Cavalleggeri 25, 84089 Pontecagnano Faiano, Italy; 2 EVJA s.r.l., Via Emilia, 35, 80011 Acerra, Italy. E-mail: [email protected]
Tomato is one of the most widespread crops in the Mediterranean basin, where new protocols for managing plant diseases with the use of bioprotection means is required to increase sustainability. In this work, three spray treatments with the resistance inducers, laminarin (Vacciplant, 2‰), chitosan (Biorend, 4‰), and Trichoderma harzianum strain TL23 (106 spores mL−1), were compared to chemical (cyprodinil + fludioxonil (0.6‰) and untreated controls in preventing Botrytis cinerea and Alternaria alternata leaf infections of tomato cv. Crovarese. One-day after the treatment, pregerminated fungal spores (5 mL, 105 spores mL−1) were inoculated on 5 leaves per plant. The design included, for each pathogen, 5 treatments replicated 7 times (pots) each containing 3 plants accounting for a total of 35 pots, 105 tomato plantlets and 525 inoculated leaves; the trial was repeated for 5 cycles. Three days after inoculum the developed necrotic area was measured. The disease control effect of the fungicides was, on average, 31 and 10%, for B. cinerea and A. alternata spots, respectively. Although, with a variability among the cycles, the control efficacy of the bio-based ingredients, Trichoderma, chitosan and laminarin, proved to be 17, 18 and 10%, respectively, for B. cinerea and 4, 5 and 13% for A. alternata infections. Levels of air relative humidity and temperature recorded continuously during the assays, were used to develop a forecasting model able to alert when favorable conditions for pathogenesis of the two fungi occur. The model was validated against natural infection events in a dataset of tomato greenhouse.
This research is carried out in the framework of the PRIMA project iGUESS-MED.
Mobile DNA elements within the genome define Pseudomonas syringae pv. tomato race-specific genotypes
B. Orfei 1 , J.F. Pothier 2 , L. Fenske 3 , J. Blom 3 , C. Moretti 1 , R. Buonaurio 1 , T.H.M. Smits 2
1 Dipartimento di Scienze Agrarie, Alimentari e Ambientali, Università degli Studi di Perugia, Perugia, Italy; 2 Environmental Genomics and Systems Biology Research Group, Institute of Natural Resource Sciences (IUNR), Zurich University of Applied Sciences ZHAW, Wädenswil, Switzerland; 3 Bioinformatics and Systems Biology, Justus-Liebig University Giessen, Giessen, Germany. E-mail: [email protected]
Bacterial speck of tomato is an important disease that can result in severe crop losses in all the tomato growing areas. For Pseudomonas syringae pv. tomato, the causative agent of this disease, two different races have been described: race 0 and race 1. While race 0 strains have avirulence genes for the expression of type III system-associated effectors AvrPto and AvrPtoB that are recognized and targeted by the effector-triggered immunity by tomato cultivars which have the Pto race-specific resistance gene, race 1 strains instead lack the avrPto and avrPtoB genes, and are therefore capable of infect also the resistant cultivars with Pto. In this work we have performed the sequencing and the analysis of the whole genome of the strain DAPP-PG 215, which was isolated and described as race 0 strain in 1996. Its whole genome comprises a 6.2 Mb circular chromosome and two plasmids (107 kb and 81 kb). The results indicate also that the strain is phylogenetically closely related to the T1 strain. The chromosome encodes race 1-associated genes like avrA and hopW1 and lacks race 0-associated genes like hopN1, giving it a race 1 genetic background. However, the genome harbors a complete ortholog of avrPto1, that allows the strain to display a race 0 phenotype. DAPP-Pg 215 is not the first strain reported to show intermediate virulence characteristics between race 0 and race 1, but comparative genomics with several PG1A genomes revealed that mobile DNA elements are potentially involved in the evolution of the two different races.
Identification of the main hemp diseases in Italy within the UNIHEMP Project
L. Orzali, L. Pirone, L. Riccioni
Council for Agricultural Research and Economics (CREA), Research Centre for Plant Protection and Certification (CREA-DC). Via C.G. Bertero 22, Rome 00156 , Italy. E-mail: [email protected]
The recent and rapid expansion of hemp cultivation has seen the increase of various phytosanitary problems. Many diseases and/or pests are already known, while other unreported are emerging and constitute a challenge for their identification and management. Within the UNIHEMP project, one of the aims is the identification of the main diseases affecting the cultivation of industrial hemp in open fields in different areas of Italy. Inspections were carried out in the experimental fields of the various partners involved in this project (Caserta, CREA-CI; Battipaglia (SA), CREA-DC; Lecce and Rutigliano (BA), CREA-AA; Foggia, CREA-CI; Rovigo, CRE-CI) and laboratory analysis completed the diagnosis. A complex phytopathological picture was highlighted, in which the main adversities encountered were at the root system (southern blight, Athelia rolfsii; fusarium wilt, Fusarium spp.) and at the stem (corn borer, Ostrinia nubilalis), with extensive damages to cultivation. Less harmful symptoms were found at the foliar and inflorescence level (caused by Alternaria spp.), with the related problematic of possible seed-borne diseases outbreak. It was also found that the phytosanitary state of the crops is linked to the cultivation area, with the southern fields (Caserta, Battipaglia, Lecce and Bari) mainly affected by A. rolfsii that can even cause the total loss of plants and harvest. In areas with different climatic conditions, such as Rovigo, the root system is instead more subject to root rot caused mostly by fusarium wilt causing plants wilting and death. On the other hand, the presence of the corn borer O. nubilaris is widespread throughout the territories.
This work was supported by the project UNIHEMP, funded by PON “Research and Innovation” 2014 – 2020 – Azione II – OS 1.b.
Persistence of Aspergillus flavus MUCL54911, active ingredient of the aflatoxin biocontrol product AF-X1, in maize fields in Italy
M.A Ouadhene 1 , A. Ortega-Beltran 2 , P.J. Cotty 3 , P. Battilani 1
1 Department of Sustainable Crop Production, Università Cattolica del Sacro Cuore, Piacenza, Italy; 2 International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria; 3 College of Food Science and Engineering, Ocean University of China, Qingdao, China. E-mail: [email protected]
AF-X1 is an aflatoxin biocontrol product containing the atoxigenic strain of Aspergillus flavus MUCL54911, endemic to Italy, as an active ingredient. The strain belongs to the vegetative compatibility group (VCG) IT006. AF-X1 is commercially available in Italy since 2015 with a temporary authorisation for use to prevent aflatoxin contamination in maize. The aim of this study was to evaluate the persistence of VCG IT006 in the soil after treatment. Soil samples were collected in 2020 and 2021 from seven locations placed in North Italy. Fields were selected as follows: i) not treated, intended as field where AF-X1 was never applied; ii) treated with AF-X1 in the year n-2, iii) treated with AF-X1 in the years n-2/n-1, and iv) treated in the year n-1. A total of 399 isolates of A. flavus were recovered from soil samples following serial dilution technique. Densities of A. flavus ranged between 54 to 191 Colony Forming Unit (CFU)/g, significantly (p < 0.005) influenced by location and treatment. The VCG analysis showed IT006 occurrence in the studied fields treated n-2/n-1 and treated n-1 ranging between 20–83% and 40–93%, respectively, whilst in not-treated fields IT006 incidence ranged from 0 to 50%. The data suggests that weather conditions and cropping context impact on the strain recovery. The active ingredient persists over the years, its dispersal in untreated fields is possible, but in some treatments/locations there was a decrease of IT006 occurence. Therefore, it is still suggested a yearly application of AF-X1 in order to provide adequate protection to aflatoxin contamination.
PLANTHEAD - A Remote PLANT HEAlth Diagnostic network to foster sustainable agricultural intensification in Eastern and Western Africa
S. Oufensou 1 , L. Altea 1 , V. Balmas 1 , G. Enne 1 , S. Merafina 2 , S.A. Okoth 3 , M. Rizzu 1 , P.P. Saporito 2 , G. Seddaiu 1 , Q. Migheli 1
1 Dipartimento di Agraria / Nucleo di Ricerca sulla Desertificazione, Università degli Studi di Sassari, Via E. De Nicola 9, 07100 Sassari, Italy; 2 OCCAM, Observatory on Digital Communication, Via Privata Duccio di Boninsegna 21, 20145 Milano, Italy; 3 Department of Biology, University of Nairobi, P. O. Box 30197–00100 Nairobi, Kenya. E-mail: [email protected]
African countries suffer from high levels of food and nutrition insecurity exacerbated by an increasing impact of climate change on agricultural production. The H2020 EWA-BELT project (https://www.ewabelt.eu/) aims to design and implement research activities involving relevant stakeholders willing to adopt innovations based on sustainable intensification (SI) approaches in order to increase agricultural yields, while preserving natural resources. In collaboration with the University of Nairobi and OCCAM, a remote PLANT HEAlth Diagnostic (PLANTHEAD) network is being developed, based on a platform hosting photographic database and georeferentiation. An image repository is being built-up with vouchered pictures of symptoms of the most relevant diseases and pests affecting major crops as indicated by the African partners. The local farmer sends an alert by mobile phone directly to the central HUB, providing relevant information such as localization, crop management, pesticide treatments, along with pictures (macro- or microscopic) and a short description of the problem. If the HUB can solve the problem, the platform sends the diagnosis and suggested actions directly to the farmer. If not, the platform sends the request to the first node, which takes charge of it. The request goes to a higher node if the lower one is unable to provide a solution. The node can be local, national or international. Once the solution is found, the competent node formulates a diagnosis and the farmer receives a notification of successful response. This response is stored in the database for future use and the images feed an AI-based image recognition system.
This work was supported by the EU H2020 EWA-BELT project [862848] “Linking East and West African farming systems experience into a BELT of sustainable intensification” coordinated by the Desertification Research Centre of the University of Sassari.
Intracellular, intercellular and vascular trafficking of plant viral RNAs
V. Pallas, Jose A. Navarro
Instituto de Biología Molecular y Celular de Plantas, (CSIC-UPV), Valencia, Spain. E-mail: [email protected]
To overcome the rigidity of the cell wall, plant viruses normally take advantage of the way of life of different biological vectors to enter and replicate within plant cells. Once inside the cell, plant viruses hijack endogenous host transport machinery to aid their intracellular spread. In addition, taking advantage of the characteristic symplastic continuity of plant cells, viruses need to remodel and/or modify the restricted pore size of the plasmodesmata (channels that connect plant cells). In a successful interaction for the virus, it can reach the vascular tissue to systematically invade the plant. We use one of the simplest multicomponent transport systems identified to date, such as the Carmovirus genus, which is extremely reduced in the case of melon necrotic spot virus (MNSV), as a model to study intra- and inter-cellular movement of plant viruses. The MNSV genome codes for five functionally characterized proteins and two small proteins are directly involved in the virus movement, p7A and p7B. p7A shows RNA-binding capabilities whereas p7B has a single-TMD domain, which allows protein insertion into the ER-derived microsomal membranes obtained in vitro. By the other hand, MNSV CP localizes to both mitochondria and chloroplasts in ectopic expression and during MNSV infection. By using different experimental approaches, we will show results that collectively demonstrate that a functional Golgi-mediated secretory pathway is essential for the intra- and intercellular movement of a plant virus and that a specific region/domain of MNSV CP can act as an ambiguous transit peptide driving dual targeting of MNSV CP to mitochondria and chloroplasts.
Molecular investigations on the role of extracellular polysaccharides released by the antagonist yeast Papiliotrema terrestris PT22AV in the biological control of Penicillium expansum
D. Palmieri, G. Ianiri, R. Castoria, T. Conte, G. Lima, F. De Curtis
Università degli Studi del Molise, Dipartimento Agricoltura, Ambiente e Alimenti, Via F. De Sanctis snc, 86100 Campobasso, Italy. E-mail: [email protected]
Exopolysaccharides (EPS), secreted by microbes during their growth play a wide range of biological roles, including protection against environmental stresses such as salinity and drought, adherence to surfaces, cell-to-cell interactions, nutrient storage, etc. Some plant-associated microbes are capable to synthesize EPS during vegetable surface colonization extending the benefits also for their plant hosts. While enough is known about the role of microbial EPS in overcoming abiotic stress in plants, little is known about the involvement of these compounds in plant protection against pathogens. The yeast Papiliotrema terrestris strain PT22AV is a strong producer of mannose-based EPS characterized for its functional properties and as biomaterial. The present study aimed to elucidate key factors in the modulation of EPS biosynthesis in PT22AV, and the role of these complex compounds in the protection of apple wounds from Penicillium expansum infection. For this purpose, two PT22AV-derived random mutant strains unable to synthesize EPS were selected and characterized for their biocontrol activity. Moreover, the EPS production in the wild-type strain was characterized under different nutritional conditions identifying the nutrients that mainly affect this biosynthetic process. Our results show as the yeast mutants defective in EPS production have a reduced ability to protect apple wounds against Penicillium expansum infection compared to the parental strain (PT22AV), especially when these strains are applied at low cells concentration. Furthermore, we found that the EPS production was strongly stimulated in nitrogen limiting conditions, especially when there is a marked imbalance between sugars and nitrogenous substances.
Hyperspectral imaging coupled with microclimatic-based alert help targeted management of downy mildew ( Hyaloperonospora parasitica (Pers.:Fr) Fr.) of wild rocket ( Diplotaxis tenuifolia L. [D.C.])
C. Pane 1 , N. Nicastro 1 , A. Pentangelo 1 , G. Manganiello 1,2 , G. Ragosta 1 , B. Rapi 3 , M. Romani 3 , F. Sabatini 3 , M. Chiesi 3 , M. Pieri 3 , F. Maselli 3 , P. Battista 3
1Consiglio per la Ricerca in Agricoltura e l’Analisi dell’Economia Agraria, Centro di ricerca Orticoltura e Florovivaismo, Via Cavalleggeri 25, 84089 Pontecagnano Faiano, Italy; 2Università degli Studi di Napoli “Federico II”, Dipartimento di Agraria, via Università 100, 80055, Portici, Italy; 3Istituto per la BioEconomia (IBE), Consiglio Nazionale delle Ricerche, Via Madonna del Piano 10, 50019 Sesto Fiorentino, Italy. E-mail: [email protected]
Wild rocket, Diplotaxis tenuifolia L. (D.C.) (Brassicaceae), is a widely cultivated baby-leaf salad crop for the fresh high-convenience food chain. Downy mildew, caused by Hyaloperonospora parasitica (Pers.:Fr) Fr. here identified by Internal Transcribed Spacer region sequencing, is the key disease of this crop, favored by environmental and agronomic conditions occurring during cultivation. Digital monitoring by optoelectronic sensors coupled with the assessment of microclimatic parameters can help more targeted disease management. A plastic tunnel trial, carried out comparing plot treatments with Trichoderma atroviride TA35, laminarin and cell-wall extract of Saccharomyces cerevisiae LAS117, to fungicides and no treatment, was subjected to hyperspectral imaging and recording of microclimatic parameters, with the aim of defining symptom spectral features on the canopy and setting phyto-pathological alert, respectively. Correlation analysis between spectral signatures and disease incidence allowed for the selection of four hyperspectral vegetation indices, ZTM, VOG1, Red-Edge NDVI and HVI (working in the range 692–750 nm) able to distinguish early symptoms from intermediate and advanced ones. A Random Forest machine learning model including a few of predictive variables (400, 536, 696, 756, 948, 963 and 962 nm), allowed to classify healthy or diseased samples with 92% accuracy. The analysis of the micrometeorological data allowed the spatial–temporal characterization of the experimental area, with the definition of the points necessary for the correct monitoring of the parameters for the evaluation of the spatial trend in disease risk. In addition, a leaf wetness estimation model and a disease alert model were implemented in MATLAB.
This research was funded by the MiPAAF, sub-project ‘AgroFiliere’ (AgriDigit programme) (DM 36,503.7305.2018 of 20/12/2018).
Steps to improve molecular detection of barley loose smut ( Ustilago nuda ) in seedlings
C. Panzetti 1,2 , E. Jenny 3 , K.E. Sullam 1
1Agroscope, Research Group Molecular Ecology, Reckenholzstrasse 191, CH-8046 Zürich (ZH); 2Institute of Biology, University of Neuchâtel, Rue Emile-Argand 11, CH-2000 Neuchâtel (NE); 3Agroscope, Research Group Extension Arable Crops, Reckenholzstrasse 191, CH-8046 Zürich (ZH). E-mail: [email protected]
Agricultural policies encourage a reduction of synthetic chemical plant protection products (PPPs). Cereal seeds are often prophylactically treated to prevent seed-borne diseases, including the internally located barley loose smut (Ustilago nuda). Cultivar resistance or biological seed treatments can contribute to chemical PPP reduction, but evaluating the effectiveness of new varieties or treatments can be cumbersome. U. nuda infections are symptomless in seeds and plants until its teliospores develop at emergence. Previously described detection methods are laborious and/or yield a non-target product. We investigated steps to improve U. nuda detection by optimizing seedling growth and qPCR conditions, including using filter paper as a substrate, growing the seedling without light, shortening the growth time, and testing new primers. The protocol alterations were tested on seedlings with either a low or high U. nuda infection level. Untreated, infected seed was directly used for the high infection rate. To reduce the infection, the same seed was treated with warm water, an effective U. nuda treatment. The less laborious seedling growth conditions, including on filter paper and without light, seems to yield more U. nuda DNA than growth in soil. Preliminary results suggest that the seedling growth period can be reduced. The newly designed COX3 primers appeared to be more sensitive than the ITS primers. Using the optimized seedling growth procedures and COX3 as the target gene can potentially streamline the molecular detection of U. nuda in seedlings, which in turn, can aid in the evaluation of new seed treatments or varieties.
Chemical characterization and biological properties of essential oils from different bioformulations treatment sweet basil plants
M. Papaianni 1 , E. Comite 1 , D. Gallo 2 , D. Rigano 2 , S.L. Woo 2
1 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 2 Department of Pharmacy, University of Naples “Federico II”, 80131 Naples (NA), Italy. E-mail: [email protected]
The development of innovative formulations as a replacement to synthetic chemicals is an increasing interest to improve agricultural production and resource use efficiency. Alternatives can consist of natural products containing beneficial microorganisms and bioactive metabolites able to prevent or reduce plant pathogens infection. The efficacy of the bioformulations can be improved by polymers as adjuvants or carriers. The variation in the chemical composition of essential oils (EOs) from sweet basil (Ocimum basilicum L.) after the treatments with Trichoderma afroharzianum T22 (T22), Azotobacter chroococcum 76A (76A), and 6-pentyl-α-pyrone (6PP), singularly or in a consortium, with or without a carboxymethyl cellulose-based biopolymer (BP) were performed. Moreover, the results of the application under in vitro conditions of the EOs against plant pathogen bacteria, showed that the oil treated with T22 + 76A, which contains a higher concentration of linalool compared to the untreated essential oil, has the highest antimicrobial activity against Xanthomonas campestris pv. campestris and Citrobacter freundii (80% of inhibition growth). Furthermore, the EOs under investigation displayed the same activity as the linalool standard, one of the major constituents in basil essential oil and a wellknown antimicrobial compound. This antimicrobial activity was also confirmed by cellular material release assay. The results are in line with the antimicrobial test, samples treated with T22 + 76A displayed a release of nucleic acids similar to the control treated with ethanol. In addition, the anti-biofilm activity of the essential oils on mature biofilm was performed and demonstrated that T22 + 76A reduced the biofilm biomass by 90%, greater than pure linalool.
Effect of field plant coverage on virus infection of seed potato ( Solanum tuberosum L.) tubers in two Italian up-land environments
B. Parisi 1 , A. Taglienti 2 , A. Tiberini 2 , F. Nicoletti 1 , C. Lanzanova 3 , D. Pacifico 1
1 Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di ricerca Cerealicoltura e Colture Industriali, Via di Corticella 133, 40128 Bologna, Italy; 2 Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di ricerca Difesa e Certificazione, Via C.G. Bertero 22, 00156 Roma, Italy; 3 Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di ricerca Cerealicoltura e Colture Industriali, Via Stezzano 24, 24126 Bergamo, Italy. E-mail: [email protected]
Potato is an economically important crop, which is severely affected by virus infection. The transmission of viruses, through seed potato tubers used as propagation material and by aphid vectors, represents a serious threat to potato yield production worldwide. In the frame of “RESILIENT” project, two multiplication techniques: 1) open field; 2) plant coverage (BBCH 31 301, beginning of crop cover growth stage: 10% of plants meet between rows) with woven fabric or insect-proof net tunnels were compared to establish the potential of seed potato tuber production in two up-land environments. Certified virus-free potato seeds of cv. Desirée and cv. Kennebec were evaluated in two areas in northern Italy (Romagnese, PV, 800 m asl and Madesimo SO, 1550 m asl) under two growing conditions. Sprouted seed tubers were analyzed by Double Antibody Sandwich—Enzyme-Linked ImmunoSorbent Assay (DAS-ELISA) for the presence of potato virus Y (PVY), potato leafroll virus (PLRV) and potato virus S (PVS). Statistical analysis highlighted different infection rates depending on the three variables: PVY was found prevalent, with higher incidence in ‘Kennebec’ and in the area at lower altitude (where aphid vectors are more widespread). Generally, a low impact of the cultivation technique on the infection rate was observed. PLRV was not detected in any sample, while a low PVS infection rate was found only in ‘Kennebec’ cultivated in Romagnese in open field. Thus, this study concluded that a higher altitude and prompt insect-proof plant coverage (until crop emergence from the soil) play a key role in the production of seed tubers with low rates of virus infections.
This work was supported by the project “Resilient” (Buone pratiche per la salvaguardia e la coltivazione di varietà locali lombarde tradizionali di patata e mais in aree interne) funded by Operazione 1.2.01 “Progetti dimostrativi e azioni di informazione” Programma di Sviluppo Rurale 2014–2020 della Regione Lombardia.
Neofusicoccum parvum causal agent of bot gummosis on lemon trees in Italy
R. Parlascino 1,2 , M. Riolo 1,2,3 , F. Aloi 1 , S. Conti Taguali 1,2,3 , E. Santilli 2 , A. Pane 1 , S.O. Cacciola 1
1 Department of Agriculture, Food and Environment (Di3A), University of Catania, Via Santa Sofia 100, 95123 Catania, Italy; 2 Council for Agricultural Research and Agricultural Economy Analysis, Research Centre for Olive, Citrus and Tree Fruit- Rende CS (CREA- OFA), 87036 Rende, Italy; 3 Department of Agricultural Science, Mediterranean University of Reggio Calabria, Localitá Feo di Vito, 89122 Reggio Calabria, Italy. E-mail: [email protected]
Bot gummosis, traditionally considered a minor citrus disease, has been found to be common and widespread in lemon groves in southern Italy. It is a disease whose causative agent causes premature aging of the plant with a consequent reduction in productivity. A fungal pathogen belonging to the Botryosphaeriaceae family, was isolated from the sampling carried out on symptomatic lemon plants in Sicily and Calabria. The main symptoms associated with the disease were gummy cancers on the trunk with abundant exudate and on the branches of the scaffolding of the trees. The identification of Neofusicoccum parvum was performed on the basis of morphological characters and the phylogenetic analysis of three loci, i.e., the internal transcribed spacer of nuclear ribosomal DNA (ITS) as well as the translation elongation factor 1-alpha (TEF1) and β-tubulin (TUB2) genes. To confirm Koch’s postulates pathogenicity tests were performed on 'Femminello 2kr', 'Monachello’ and Citrange ‘Carrizo’ cultivars. Symptoms (14 d.a.i.) were more severe on lemon cultivars ‘Femminello 2kr’and ‘Monachello’ while less on Citrange ‘Carrizo’. This is the first report of N. parvum as a pathogen of citrus in Italy. The incidence and frequency of these outbreaks are expected to increase in the future as a result of both climate change and the introduction of more sensitive lemon cultivars. It is suspected that nursery plants are the main vehicle for the wide spread of this fungus whose lifestyle as a latent pathogen may have favoured its spread through asymptomatic plants.
Role of enniatins as emerging mycotoxins and their association with deoxynivalenol in plant, insect, animal and human systems
M. Pasquali 1 , M. Saracchi 1 , A. Agazzi 2 , L. Ederli 3 , I. Bellezza 4 , R. Romani 3 , G. Beccari 3 , L. Covarelli 3
1 Dipartimento di Scienze per gli Alimenti, la Nutrizione e l’Ambiente, Università degli Studi di Milano, Via Celoria, 2, 20133, Milano, Italy; 2 Dipartimento di Medicina Veterinaria e Scienze Animali, Università degli Studi di Milano, Via dell’Università, 6, 26900, Lodi, Italy; 3 Dipartimento di Scienze Agrarie, Alimentari e Ambientali, Università degli Studi di Perugia, Borgo XX Giugno, 74, 06121, Perugia, Italy; 4 Dipartimento di Medicina e Chirurgia, Università degli Studi di Perugia, Piazzale Severi, 1, 06132, Perugia, Italy. E-mail: [email protected]
Deoxynivalenol (DON), produced by F. graminearum (FG), a causal agent of Fusarium Head Blight (FHB) of wheat, is notoriously toxic to animals and humans. Enniatins (ENNs), produced by F. avenaceum (FA), have also received attention for chronic exposure to contaminated feeds and foods but little is known about their effects. Therefore, this project aims to investigate the role of ENNs [enniatin B (ENB) in particular] and their association with DON: 1) in fungal virulence towards wheat; 2) in fungal competition; 3) towards the wheat microbiome and biocontrol agents (BCAs); 4) on wheat pests and their natural enemies; 5) on dairy cows health; 5) on human intestinal barrier permeability. So far, our research has shown that ENB plays a marginal activity in FA virulence but its role in defense priming is under evaluation. ENB also shows involvement in FA competition against FG. Streptomyces strains modulate DON production by FG in wheat being effective as BCAs. The effect of DON + ENB on BCAs fitness and on the whole microbiome are under investigation. While DON shows a toxic contact effect towards wheat aphids, ENB does not play the same activity. DON + ENB show a toxic effect on the topical application on lacewing larvae. ENNs in livestock diets are primarily from consuming cereal grains. Little information is available on their effects on dairy cows concerning chronic or acute exposure and well-balanced or acidotic diets. ENB and DON exert time and concentration-dependent toxic effects on human intestinal Caco2 cells. However, when combined, no additive effects are observed.
This study is supported by the “PRIN – 2020” project “Role of enniatins as emerging mycotoxins and their association with deoxynivalenol in plant, insect, animal and human systems – MYCENDEA (MYCotoxins ENniatins DEoxynivalenol Association) (2020ZAYHKA) funded by the Italian Ministry of University and Research.
Use of tomatine extracts from tomato industry wastes against Botrytis cinerea highlights potential for production of biopesticides from circular economy workflow
A. Passera, P. Abbasi, A. Follador, D. Marcianò, B. Scaglia, P. Casati
Università degli Studi di Milano, Dipartimento di Scienze Agrarie e Ambientali, Via Celoria 2, 20133 Milano, Italy. E-mail: [email protected]
Tomato is one of the most developed horticultural commodities in Italy. The tomato industry produces several wastes: pomaces have been widely investigated as substrates for reutilization, while residues of the washing and sorting selection (WSR) have received little interest. Despite this, the WSR include unripe tomato fruits, stems and leaves, boasting a high concentration of tomatines. Tomatines are defense glycoalkaloids produced by tomato plants that can reduce the incidence of biotic stresses thanks to the toxicity that these compounds possess against a broad range of pathogens and pests. Most of the studies carried out in the past focused on using standard molecules, while information regarding tomatines extracted from plant residues is currently very limited. In this study we tested two different extracts from tomato WSR for their efficacy in reducing the growth of Botrytis cinerea, a major pathogen that affects tomato both in field and after harvest, in vitro. The results obtained using different concentrations of the two extracts (ranging from 0.06 mg/mL to 2 mg/mL) showed antifungal effect, obtaining inhibition of fungal growth that ranged from 50 to 100% on two different B. cinerea strains. These results were comparable to those obtained using a mix of tomatine and tomatidine analytical standards with similar concentrations and ratios as those found in the extracts. These results suggest that the use of these extracts, obtained in the context of a circular economy approach, could see application towards the reduction of the environmental impact of the defense of horticultural crops.
This work was supported by the “Piano di Sostegno alla Ricerca 2020 Seal of Excellence” (SEED) project titled “Produzione di biofitofarmaci a base di tomatine da scarti dell’industria di trasformazione del pomodoro” (BRIO) funded by University of Milan.
Short-term effects of biological and chemical treatments against Heterobasidion irregulare on bacterial and fungal communities of Pinus pinea stumps
M. Pellicciaro, G. Lione, P. Gonthier
University of Turin, Department of Agricultural, Forest, and Food Sciences (DISAFA), Largo Paolo Braccini 2, I-10095 Grugliasco, Italy. E-mail: [email protected]
Stumps are the targets of chemical or biological treatments against airborne infections of Heterobasidion spp. Here, we report on the short-term impact of biological (Proradix®, the cell-free filtrate of Pseudomonas protegens [strain DSMZ 13134], the conidial suspension of Phlebiopsis gigantea [strain MUT 6212], Rotstop®) and chemical (urea) treatments targeting the alien invasive H. irregulare on both bacterial and fungal communities inhabiting the surface of Pinus pinea stumps. Microbial communities of 15 stumps per each treatment and for controls were characterised four months after treatments using amplicon metagenomic sequencing of 16S and ITS gene. Our results demonstrated that treatments with Rotstop® and urea had a significant impact on fungi, decreasing significantly both diversity and richness of fungal communities compared to other treatments and control stumps. Fungal communities inhabiting stumps were less sensitive to the other tested treatments. The levels of diversity and richness in bacteria were never significantly different from those of control stumps, revealing a short-term stability of bacterial communities. However, correspondence analysis indicated that the composition of both bacterial and fungal communities was affected by treatments. Our findings provide additional insights on the risk associated with stump treatments against Heterobasidion spp. on natural ecosystems and improve our understanding on microbial communities inhabiting stumps of P. pinea.
This work was funded by SP Sourcon Padena GmbH.
Polyphasic characterization of Xanthomonas strains isolated from hazelnut symptomatic plants
S. Perez Fuentealba 1 , E. Biondi 2 , N. Cortéz 1 , A. Checcucci 2 , F. Quaiotto 2 , N. Biondo 2 , J. Guerrero 3 , P. Minardi 2
1Institute of Agri-Food, Animal and Environmental Sciences (ICA3), Universidad de O’Higgins, Ruta 90 km. 3, 3070000 San Fernando, Chile; 2Department of Agricultural and Food Sciences (DISTAL), Alma Mater Studiorum—University of Bologna, V. le G. Fanin 42, 40127 Bologna, Italy; 3Facultad de Ciencias Agropecuarias y Forestales, Universidad de La Frontera, 4811230 Temuco, Chile. E-mail: [email protected]
The Xanthomonas arboricola (Xa) species has a high genetic variability and, among the 9 pathovars, pv. corylina causes the bacterial blight of hazelnut which is one of the main bacterial diseases present in all areas where hazelnut is cultivated. In this study, 19 Xa and 21 X. arboricola pv. corylina (Xac) strains isolated in Chile from European hazelnut symptomatic plants, were characterised through Rep-PCR analysis, where 13 X. arboricola reference strains belonging to the pathovars celebensis, pruni, juglandis and populi were used as controls, while DISTAL 9081 strain of X. axonopodis pv. vitians as outgroup; then, Biolog analysis, bioassays (hypersensitive response) and pathogenicity tests were carried out. The UPGMA cluster analysis of Rep-PCR fingerprints discriminated the 51 X. arboricola strains into 5 statistically significant groups corresponding to the relative pathovars: pruni, corylina, juglandis, celebensis and populi. The X. arboricola pv. corylina strains showed a high genetic variability (approx. 45%); similarly, the Rep-PCR fingerprints of the X. arboricola strains isolated from symptomatic hazelnut shoots and leaves gave similar results with a similarity value of approx. 50% among strains which were included in the corylina pathovar. The hypersensitive response on French bean resulted positive for all Xa/Xac strains, and the selected Xa/Xac strains resulted pathogenic on hazelnut plants confirming the Koch’s postulates. At last, most of metabolic activity data obtained with Biolog GenIII™ analysis confirmed the molecular characterization of Xac strains, pooled in the same group. Furthermore, the Xa isolates were here combined in a sub-group of the Xac cluster.
This work was partially funded by the National Agency for Research and Development (ANID) / FONDECYT Posdoctorado / 3,180,629.
Monitoring Calabrian (Southern Italy) citrus orchards for Phyllosticta citricarpa
L. Pirone 1 , V. Bergamaschi 1,2 , M.T. Valente 1 , L. Orzali 1 , L. Riccioni 1
1 Council for Agricultural Research and Economics (CREA), Research Centre for Plant Protection and Certification (CREA-DC). Via C.G. Bertero 22, Rome 00156, Italy, 2 Department of Environmental Biology, Sapienza University of Rome, P. le Aldo Moro 5, 00185 Rome, Italy. E-mail: [email protected]
Citrus Black Spot (CBS) is a foliar and fruit disease caused by a quarantine fungus Phyllosticta citricarpa (McAlpine). It is one of the most important disease affecting many commercial Citrus species. CBS was first reported in Australia in 1895 and subsequentially observed in many citrus producing areas of Asia, Africa, Australia, North and South America. P. citricarpa produces lesion on fruits and leaves; it has an economic impact because the external blemishes it generates, make fruits unsuitable for the market. The pathogen has not been reported in Europe until 2017 when it was described in Italy, Malta and Portugal. A network funded by European Food Safety Authority (EFSA) was created for the surveillance of this pathogen in the reported areas, including the Southern Italian area of Trebisacce (Calabria). Since leaves are reported to be a source of primary inoculum in CBS disease, we focused our analysis on both green leaves and leaf litter, developing an efficient DNA extraction method. From 2019 to 2021 leaf litter and green leaves of citrus trees were collected in Torre Albidona (Trebisacce), the site corresponding to geographical coordinates indicated in the first report of P. citricarpa in Italy. The vegetal material was analyzed by conventional methods (humid chamber and fungal isolation) and by molecular methods (qPCR). Samples analysed were negative for the presence of P. citricarpa fruiting body after morphological observation, as well for the presence of Phyllosticta citricarpa/paracitricarpa DNA after molecular test.
This work was supported by the project GP/EFSA/ALPHA/2018/03 – ITALY “Pilot application of smart surveillance tools for citrus black spot pathogen in Italy”.
Application of Trichoderma strains and Benzothiazole derivatives to tomato plants ( Solanum lycopersicum ) as biostimulants
A. Pironti 1 , S. Marcone 1 , M. Spychalski 2 , R. Kukawka 2,3 , M. Smiglak 2,3 , S.L. Woo 4
1Department of Agricultural Sciences, University of Naples “Federico II”, Via Università 133, 80055 Portici (NA), Italy; 2Poznan Science and Technology Park, Adam Mickiewicz University Foundation, ul. Rubie˙z 46, 61–612 Pozna´ n, Poland; 3Innosil Sp. z o.o., ul. Rubie˙z 46, 61–612 Pozna´ n, Poland; 4Department of Pharmacy, University of Naples “Federico II”, Via D. Montesano 49, 80131 Naples, Italy. Email: [email protected]
Synthetic chemical compounds are commonly used for the resistance induction in plants to provide crop protection, even if many disadvantages are known such as toxic effects to human health, environment and soil biodiversity. In recent years, alternative choices to chemical use in agriculture includes the use of beneficial microorganisms from the rhizosphere. The aim of this work was to evaluate different application strains of Trichoderma spp. in combination with benzothiazole compounds to obtain healthy and disease-resistant plants. Experiments were conducted using three Trichoderma strains: T. afroharzianum strain T22, T. harzianum strain M10, T. atroviride strain P1 according to their biocontrol abilities against plant pathogens and plant growth-promoting effects. Three derivatives of benzothiazole, CH, BTH (BION, Syngenta), and ILAGRO, are known as inducers of plant resistance. The first part evaluated compound compatibility with Trichoderma strains by in vitro tests. Results showed that the growth of M10 and T22 were not inhibited by the presence of benzothiazole compounds, and even promoted in some cases. Subsequently, plants were treated singly and in combination with treatments of Trichoderma strains (soil irrigation) and the compounds (foliar spray); with water and/or only fungal strains were used as controls. Plant growth promotion was confirmed by in vivo tests on tomato plants cv. San Marzano nano in the greenhouse by measuring biometric parameters (fresh and dry weight of stem and root; stem and root lengths; number, fresh and dry weight of leaves). These results may contribute to develop improved beneficial microbes to be used as plant protection products or pant biostimulants.
Extracellular acidification is not necessary for Beauveria bassiana biocontrol activity
M. Ranesi 1 , S. Vitale 2 , A. Staropoli 1,2 , S.L. Woo 3 , D. Turrà 1
1 Department of Agricultural Sciences, University of Naples “Federico II”, Via Università 100, 80055, Portici (Na), Italy; 2 CNR – Istituto per la Protezione delle Piante (IPP-CNR), Piazzale Enrico Fermi 1, 80055, Portici (Na), Italy; 3 Department of Pharmacy, University of Naples “Federico II”, Via Domenico Montesano 49, 80131, Naples, Italy. E-mail : [email protected]
Entomopathogenic fungi (EPF) belonging to Beauveria bassiana (Bb) apart from being widely used in agriculture for the control of insect pests they also play different roles in natural agroecosystems, including endophytism, plant growth promotion and disease control. Rhizosphere pH alkalinization represents a renowned pathogenicity mechanism of several fungal pathogens including the soil-borne ascomycete F. oxysporum f. sp. lycopersici (Fol), the causal agent of vascular wilt disease on tomato plants. Here, to understand if Bb biocontrol activity against fungal pathogens might be related to its ability to modify rhizosphere pH, we characterized the pH modulating and biocontrol activity in vitro and in vivo of ten Bb isolates. Bb acidification in vitro was evaluated by using the pH indicator bromophenol blue and an acidification index was calculated for each of the tested isolates. Nine isolates out of ten were able to acidify the culture medium and six of them produced a more intense acidification halo around the colony. However, when biocontrol activity was tested in vitro only three highly-acidifying and one non-acidifying isolate greatly inhibited Fol growth. Unexpectedly, when Bb biocontrol activity was evaluated in vivo against Fol, all isolates similarly protected tomato plants from wilting, thus suggesting that rhizosphere acidification might be only one of the biocontrol mechanisms used by Bb in vivo. Further experiments are required to determine which additional mechanisms apart from acidification are involved in Bb biocontrol activity.
Root and collar rot caused by Phytophthora nicotianae and Phytophthora palmivora , a new disease of Paulownia in Europe
M. Riolo 1,2,3 , F. Aloi 2 , S. Conti Taguali 1,2,3 , F. La Spada 2 , R. Parlascino 1 , A. Pane 2 , E. Santilli 1 , S.O. Cacciola 2
1 Council for Agricultural Research and Economics, Research Centre for Olive, Fruit and Citrus crops (CREA- OFA), 87036 Rende, Cosenza, Italy; 2 Department of Agriculture, Food and Environment, University of Catania, 95123 Catania, Italy; 3 Department of Agricultural Science, Mediterranean University of Reggio Calabria, 89122 Reggio Calabria, Italy. E-mail: [email protected]
Paulownia species, characterized by fast growth, ability to re-sprout rapidly after cutting as well as a good tolerance to drought and high soil acidity, are trees native to China. They are grown in managed plantings in several European countries for the production of wood and biomasses. In 2018, 40% of trees of a 2-year-old commercial planting of P. elongata × P. fortunei in Calabria (Southern Italy) were affected by a severe disease characterized by wilting, stunting, leaf yellowing, and final collapse of the entire tree as a consequence of root and crown rot. Phytophthora species were consistently recovered from basal stem bark, roots and rhizosphere soil of symptomatic trees and identified as Phytophthora nicotianae and P. palmivora. The identification was performed on the basis of both morphological characteristics and phylogenetic analysis of rDNA ITS sequences. To confirm Koch’s postulates, potted paulownia saplings were transplanted into infested soil and stem-inoculated by wounding. Both Phytophthora species were pathogenic and caused root rot and stem cankers. Even though P. palmivora was the only species recovered from roots of naturally infected plants, in pathogenicity tests through infested soil P. nicotianae was more virulent. This is the first report of Phytophthora root and crown rot of Paulownia in Europe. The use of healthy nursery plants is crucial for preventing this disease in commercial plantations. Strategies to prevent this emerging disease include the use of healthy nursery plants, well-drained soils for new plantations and proper irrigation management.
Identification of the genes involved in early response to Flavescence dorée infection in the model plant Arabidopsis thaliana
M. Rossi, L. Galetto, N. Bodino, S. Abbà, S. Palmano, C. Marzachì
Istituto per la Protezione Sostenibile delle Piante, CNR , Torino, Italy. E-mail: [email protected]
Flavescence dorée (FD) is a grapevine disease associated with a quarantine phytoplasma, representing a severe threat to the European wine-growing areas. FD management, consisting mainly of compulsory insecticide treatments and roguing of infected plants, has relevant environmental and economic impacts. Even if grapevine cultivars show different degrees of susceptibility to the disease, no genetic resistance to FD has been described yet. Arabidopsis thaliana (plant)/FD phytoplasma (FDp, pathogen)/Euscelidius variegatus (FDp insect vector) was the model system adopted to study the plant early response to FDp exposure. Literature search and a bioinformatic pipeline to analyze RNAseq data of healthy vs FD-infected A. thaliana and Vitis vinifera plants allowed selection of a set of 15 genes putatively involved in plant response to FDp infection. The expression of the selected genes was analyzed by Real-Time PCR comparing phytoplasma-exposed and not-exposed plants. The involvement in FDp infection of a subset of 3 differentially expressed genes (jaz7, pad4 and gsl5) was further investigated using knockout A. thaliana mutants in a reverse genetics approach. Phytoplasma multiplication and infection rates were measured in the mutant lines and compared with the wild type. Moreover, E. variegatus feeding behavior on wild type and a double-mutant line Pad4(-)/Pmr4(-) was investigated and described. Developing sustainable strategies for FD control requires expanding the knowledge of plant-pathogen interactions. Identifying genes involved in susceptibility to FDp infection in a model system is a preliminary step for the characterization of grapevine genes potentially involved in disease resistance.
Lessons learned from plant viruses: SARS-CoV-2 nsp3, nsp4 and nsp6 expression in Saccharomyces cerevisiae
L. Rubino 1 , A. De Stradis 1 , C. Telegrafo 2 , A. Antonacci 1 , S. Giannattasio
1 Consiglio Nazionale delle Ricerche, Istituto per la Protezione Sostenibile delle Piante, Via Amendola 165/A, 70126 Bari, Italy; 2 Consiglio Nazionale delle Ricerche, Istituto di Biomembrane, Bioenergetica e Biotecnologie Molecolari, Via Amendola 122/D, 70126 Bari, Italy. E-mail: [email protected]
Positive-strand RNA [( +)RNA] viruses are agents of important diseases in humans, animals and plants, including COVID-19. Regardless of the host, the replication of all ( +)RNA viruses occurs in association with the host endomembrane system. Based on this common replication mechanism, yeast has been successfully employed as a surrogate host to study plant ( +)virus replication. In this work the yeast model was used to express the replication-associated proteins of human SARS-CoV-2, having a role in the formation of the double-membrane vesicles and in virus replication, to decipher virus-membrane interactions. ( +)RNA SARS-CoV-2 non-structural proteins nsp3, nsp4 and nsp6 were cloned under the control of the inducible GAL1 promoter and the effects of their expression were studied in Saccharomyces cerevisiae strains W303-1B and YPH499. Yeast cells expressing nsp3 showed a low but significant decrease of growth rate, whereas nsp4 expression dramatically reduced cell growth. Moreover, nsp4-expressing cells showed a loss of cell viability compared to control cells harboring empty plasmid. Nsp6 expression did not have significant effect on cell growth and viability. All proteins sedimented in a membrane-enriched protein fraction. Immunofluorescence analysis showed that nsp3, nsp4 and nsp6 localize on the endoplasmic reticulum. Elucidating ( +)RNA virus-host cell interaction complexity will allow the identification of novel druggable targets for the development of broad-spectrum antivirals.
Tomato- Beauveria bassiana interaction for bioinspired strategies of pathogen control
A. Russo 1,2 , S. Pollastri 2 , M.M. Monti 2 , A. Ghirardo 3 , J.B. Winkler 3 , M. Ruocco 2 , J.P. Schnitzler 3 , F. Loreto 2, 4
1 University of Naples “Federico II”, Department of Agricultural Sciences, 80055 Portici, Italy; 2 Institute for Sustainable Plant Protection, National Research Council of Italy (CNR-IPSP); 3 Helmholtz Zentrum München, Research Unit Environmental Simulation, Institute of Biochemical Plant Pathology, 85764 Neuherberg, Germany; 4 University of Naples Federico II, Department of Biology, 80126 Naples, Italy. E-mail: [email protected]. E-mail: [email protected]
Plants are central to the complex networks of multitrophic interactions. Increasing evidence suggests that beneficial microorganisms (BMs) may be used as an alternative to chemically synthesized pesticides to control pathogens and pests while improving plant fitness and stress tolerance. Among BMs the entomopathogenic fungus Beauveria bassiana can endophytically colonize plant tissues, but its effect on plant physiology and resistance to pathogens has been poorly investigated. In particular, it is not known whether the colonization of B. bassiana changes plant photosynthesis, stomatal opening and the emission of Volatile Organic Compounds (VOCs) released by plants constitutively or in response to abiotic or biotic stresses. In the present work we used tomato (Solanum lycopersicum L.) to investigate whether endophytic colonization by B. bassiana strain ATCC 74,040 (Naturalis®) affects plant physiology and induces resistance against the pathogen Botrytis cinerea. The efficacy and persistence of endophytic colonization of B. bassiana was confirmed by PCR and colony formation on Potato dextrose agar (PDA) medium. Tomato plants responded to B. bassiana treatment with a significant but transient (1–2 day-long) reduction of stomatal conductance and photosynthesis, indicating rapid activation of defensive (rejection) responses, followed by host recognition and onset of a symbiotic relationship. We are now examining whether B. bassiana can control in vitro the infection of B. cinerea. Finally, we aim to investigate whether B. bassiana changes VOCs blend emitted by tomato and whether such a change is also a factor improving plant resistance to B. cinerea.
This work was supported by the PRIN grant (PROSPECT 2017JLN833_005) by Ministero dell’Istruzione, dell’Università e della Ricerca.
Tomato yellow leaf curl Sardinia virus confers tolerance to drought in tomato
C. Sacco Botto 1, 2 , A. Moine 2 , W. Chitarra 2, 3 , L. Nerva 2,3 , C. Pagliarani 2 , C. D’Errico 2 , S. Matić 2 , E. Noris 2
1 Department of Agriculture, Forestry and Food Science DISAFA—Turin University, Largo Braccini 2, 10095 Grugliasco, Italy; 2 Institute for Sustainable Plant Protection, National Research Council, Strada delle Cacce 73, 10135 Turin, Italy; 3 Council for Agricultural Research and Economics Centre of Viticultural and Enology Research (CREA-VE), Viale XXVIII Aprile 26, 31015 Conegliano (TV), Italy. E-mail: [email protected]
Viruses can alter the response to abiotic stresses of plants and their ability to tolerate them. Geminiviruses, particularly begomoviruses, have been shown to increase tolerance to heat and drought in Arabidopsis thaliana. In this study, we investigated the drought tolerance in tomato (Solanum lycopersicum L.), which is considered one of the most important crops in agriculture. Tomato plants infected by the begomovirus tomato yellow leaf curl Sardinia virus (TYLCSV) and mock-inoculated plants were grown under well-watered, water-stressed, and recovered conditions. Morphological and physiological parameters, and hormone levels were investigated in both groups of plants. In addition, the transcriptional response of candidate genes known to be involved in the synthesis and degradation of water-stress related metabolites (abscisic acid, auxin, proline, and salicylic acid) was studied by qRT-PCR. Based on the results obtained, the TYLCSV infection positively influenced the ability of tomato plants to tolerate drought by postponing the onset of stress-related traits, altering transcriptional responses of the water-stress pathway, enhancing the plant water use efficiency, and permitting the plant quick recovery after the rehydration phase. These results open up new possibilities to design new strategies allowing to cope with climate change in agriculture, particularly in locations with limited access to water and high temperatures.
Phytophthora spp. associated with kiwifruit vine decline syndrome in north-eastern Italy
A. Sadallah 1 , C. Bregant 2 , A. Di Francesco 1 , G. Firrao 1 , M. Martini 1 , L. Sella 2 , P. Ermacora 1 , B.T. Linaldeddu 2
1 Department of Agricultural, Food, Environmental and Animal Sciences, University of Udine, Via delle Scienze 206, 33100 Udine, Italy; 2 Department of Land, Environment, Agriculture and Forestry (TESAF), Viale dell’Università 16, 35020 Legnaro (PD), Italy. E-mail: [email protected]
Since 2012, kiwifruit vine decline syndrome has been threatening several orchards in the most important growing areas of north-eastern Italy. A field survey, aimed to detect the presence of Phytophthora spp. in roots of symptomatic plants, was conducted during the 2021 growing season in 5 orchards located in the Veneto and Friuli Venezia Giulia (FVG) regions. Pathogen isolation was performed on selective PDA media (supplemented with: 100 ml/L of carrot juice, 0.05 g/L of hymexazol, and 0.013 g/L of pimaricin) from decaying kiwifruit roots obtained from 39 and 25 plants collected in Veneto and FVG orchards, respectively. Phytophthora spp. isolates were identified by morphological and DNA-based techniques. In both regions, the species most frequently isolated were P. cinnamomi and P. citrophthora, while the least was P. nicotianae. Besides, P. acerina, P. plurivora, and P. palmivora were rarely isolated only from samples collected in Veneto. To fulfill Koch’s postulates, the pathogenicity and aggressiveness of Phytophthora spp. were tested on one-year old plants of Actinidia chinensis var. deliciosa, following a previously set-up protocol consisting of inoculation and 3 waterlogging events. The species Phytopythium vexans and Phy. chamaehyphon, whose pathogenicity was already demonstrated, were used as positive control. After three months from the inoculation, the most aggressive species caused the same symptoms observed in field, showing a root rot disease severity ranging from 70 to 100%. The results obtained in this study could help to better understand the complex etiology of this disease and to develop a reliable integrated strategy for its management.
A mechanistic weather-driven model to predict Puccinia graminis f. sp. tritici infection and disease development in wheat
I. Salotti 1 , F. Bove 2 , V. Rossi 1
1 Università Cattolica del Sacro Cuore, Department of Sustainable Crop Production (DI.PRO.VES.), Piacenza, Italy, 2 Horta Srl, Piacenza, Italy. E-mail: [email protected]
Stem rust (or black rust) of wheat, caused by Puccinia graminis f. sp. tritici, is a re-emerging, major threat to wheat production worldwide. A mechanistic, weather-driven model for the stem rust pathosystem was developed in order to quantitatively synthesize the literature knowledge and to provide a tool to guide tactic decisions for disease management. The model considers infections caused by uredospores that are responsible for disease epidemics in temperate climates and assumes that sites of the crop move from being healthy, latent, infectious, and finally removed during the epidemic. The model was implemented and validated in STELLA®, a visual programming language for modeling system dynamics. The ability of the model to predict the first seasonal infections was evaluated using field data collected in three wheat-growing areas of Italy (Emilia-Romagna, Apulia, and Sardinia) from 2016 to 2021. The model showed good accuracy and specificity, with correct predictions occurring in the 90% of the cases. The model’s ability to predict disease progress during the growing season was evaluated by using data of six epidemics occurred in the USA between 1968 and 1986. Comparison of observed versus predicted data gave a concordance correlation coefficient of 0.96 and an average distance between real data and the fitted line of 0.09. The model could therefore be considered accurate and reliable for predicting epidemics of wheat stem rust and it has the potential for being used for scheduling fungicide applications to control the disease.
This project has received funding from the LIFE Programme of the European Union (grant agreement LIFE17 CCM/IT/000062).
Study of the oxylipinic profile of the model plant Arabidopsis thaliana artificially infected with Xylella fastidiosa
C. Santaera 1 , A. Cacciotti 2 , M. Beccaccioli 2 , N. Pucci 1 , L. Loreti 1 , M. Reverberi 2 , V. Scala 1
1 Council for Agricultural Research and Agricultural Economic-Research Centre for Plant Protection and Certification, Roma, Italy; 2 Department of Environmental Biology, University of Rome, Sapienza, Roma, Italy. E-mail: [email protected]
Recently we demonstrated that oxylipins are hallmarks of Xylella fastidiosa (Xf) subsp. pauca infection in olive; in particular the 13-LOX derived oxylipins accumulated in Xf-infected Olea europea cv. Ogliarola salentina, one of the most susceptible variety. This study aims at deciphering how and if plant 13-LOXs can modulate Xf virulence in the model plant Arabidopsis thaliana that has been proposed as potential host for Xf. The first issue was to evaluate the ability of Xf subsp. pauca De Donno and subsp. fastidiosa Temecula 1 to colonize A. thaliana. The infection of A. thaliana ecotypes Col-0 and the mutant lox2 deleted for a 13-lox with Xf was confirmed by real-time PCR evaluating the ability of Xf to move from the inoculation point and colonize systemically the tissue plant. Oxylipins profile and the modulation of oxylipin genes of infected plants were monitored at different time intervals after the pathogen infection under controlled conditions. Preliminary results report that the bacterium is able to colonize the plant tissue systemically. In the infected tissue the oxylipins generated by 9-lipoxygenase and 13-lipoxygenase is down modulated in the infected tissue in Col-0 and in mutant. In the mutant infected by Xf it is possible to observe a down modulation of different lipidic entities.
Protein hydrolysates of different origins to improve resistance to stresses
S.M. Sanzani, A. Ippolito
Department of Soil, Plant and Food Sciences, University of Bari “Aldo Moro”, Via Amendola 165/A , 70126 Bari, Italy. E-mail: [email protected]
In the last two decades, several products generically referred as biostimulants appeared on the market to promote plant growth and induce defence responses against biotic and abiotic stresses. Particularly, protein hydrolysates (PHs), processed from the protein fractions of both plant and animal matrices, are commonly used in the food and cosmetic industry as the hydrolysis process allows to obtain amino acids while destroying allergens. Furthermore, in recent times they gained popularity as plant biostimulants for their efficacy and ease of use. In the present overview, the activity of soybean and casein protein hydrolysates, both applied before and after harvest, against biotic and abiotic stresses is reported and discussed. These activities seem mainly related to the bioactive peptides released after enzymatic degradation process, displaying multifunctional properties rooted mainly to influence plant hormonal activity; in response, plants may initiate or increase physiological processes that end up as resistance to stresses. Specific hints on their mode of action are provided.
The studies were partially funded by “A. Costantino & C.” S.p.a. (Favria, Turin, Italy).
Streptomyces spp. on wheat reduce disease impact and deoxynivalenol accumulation caused by F. graminearum
M. Saracchi 1 , C. Gardana 1 , M. Cerea 2 , C. Pizzatti 1 , A. Kunova 1 , P. Cortesi 1 , M. Pasquali 1
1 Università degli Studi di Milano, Dipartimento di Scienze per gli Alimenti, la Nutrizione e l’Ambiente – DeFENS, via Celoria 2, 20133 Milano, Italy; 2 Università degli Studi di Milano, Dipartimento di Scienze Farmaceutiche – DISFARM, via Mangiagalli 25, 20133 Milano, Italy. E-mail: [email protected]
Fusarium Head Blight causes important losses to wheat production due to reduced productivity and contamination with trichothecenes, in particular deoxynivalenol (DON). The disease is caused by different Fusarium species, among which the most significant and widespread worldwide is F. graminearum. Effective biocontrol against FHB shall limit both the disease incidence as well as DON accumulation in grains. Streptomyces spp. can act as endophytes of the wheat plant. Moreover, they produce a wide range of secondary metabolites that can hinder specifically toxin production by the fungus and/or limit the growth of the fungus. Our previous work identified some strains effective against the disease and toxin accumulation in controlled conditions. The aim of the work was to assess the efficacy of three strains of Streptomyces spp. to limit disease and DON accumulation in the field (using two types of treatment: seed coating and spike inoculation). The field trial, carried out in 2021, exploited artificial inoculation of the pathogen to obtain a high level of infection and toxin accumulation. The disease and the DON content were compared to the untreated control on a durum wheat and a spring wheat cultivar. In both cultivars, the three streptomycetes showed the ability to reduce disease and DON accumulation. Higher efficacy of the strains was observed in spring wheat (DON reduction > 60%) compared to durum wheat (DON reduction > 40%). The three Streptomyces strains under scrutiny showed promising results under field conditions. Future research will focus on the mechanism of action of the strains.
This work was partially funded by the University of Milan, One Health Action HUB.
Pantoea stewartii subsp. stewartii and seed trade: “trick or treat” for Italian sweet corn?
V. Scala 1 , F. Costantini 1 , A. Alessio 2 , L. Faino 2 , N. Pucci 1 , S. Loreti 1
1 Council for Agricultural Research and Agricultural Economic-Research Centre for Plant Protection and Certification, Roma, Italy; 2 Department of Environmental Biology, University of Rome, Sapienza, Roma, Italy. E-mail: [email protected]
Pantoea stewartii subsp. stewartii (Pss) caused Stewart’s vascular wilt of sweet corn and maize and it is responsible for serious crop losses. Pss, indigenous of North America spreads with maize seeds and its presence was notified in Italy since 2015. The risk assessment of the entry of Pss in the EU from the USA through seed trade is in the order of magnitude of hundred introductions per year. Several methods, based on molecular or serological methodology, were developed for the detection of Pss for commercial seed certification and official analysis but some of these are unable to correctly distinguish between P. stewartii subsp. indologenes (Psi), with Pss. Psi is avirulent on corn and it is not a regulated corn pest even if occasionally present on corn seed. The galE gene can differentiate Pss from Psi based on SNPs. Pal and colleagues develop a real-time PCR assay for the detection of Pss without the cross reaction of Psi, from corn seeds without the need for bacterial isolation. In this study, we characterized the Italian isolates recovered in 2015 and 2018 by molecular, biochemical, pathogenicity tests and their genome was assembled through hybrid MinION and Illumina – sequencing procedures. Through this, we were able to individuate by in silico analysis a sequence that can detect Pss until 103 CFU/mL in spiked seeds (bypassing bacteria isolation). The assay was proven specific for the sole Pss. This method with good sensitivity and specificity can improve the detection in maize seed to go through a critical problem associate to the import of maize seed potentially affected by Pss coming from region where Stewart's disease is endemic.
Cellulose Nanocrystals as innovative and sustainable tool to control bacterial plant pathogens
D. Schiavi, G.M. Balestra
Department of Agriculture and Forest Sciences (DAFNE), University of Tuscia, Via S. Camillo de Lellis snc, 01100 Viterbo, Italy. Email: [email protected]; [email protected]
Among the huge variety of nanomaterials usable in crop protection, recent promising results were obtained studying the antibacterial properties of Cellulose Nanocrystals (CNC). CNC can be obtained through enzymatic and chemical protocols, even from agro-food lignocellulosic biomasses. CNC have an acicular shape with dimension ranging from 80–400 nm and 5–20 nm for length and width respectively. CNC have showed some interesting antimicrobial properties when tested against several plant pathogenic bacteria: 0.5% w/v CNC suspension is able to inhibit the in vitro growth of Pseudomonas syringae pv. tomato (Pst), Pseudomonas savastanoi pv. savastanoi (Psav), Xanthomonas arboricola pv. corylina (Xac), Xylella fastidiosa subsp. pauca (Xf). The biological mechanisms behind these properties are still largely unknown; however, it has been recorded the ability of CNC to inhibit Pst and Psav biofilm formation at 24 h when applied at concentration lower than 1% w/v; while no generation of intracellular ROS was observed in treated cells. When in vivo applied, CNC showed also the capability of suppressing the epiphytic survival of Psav and Pst on olive and tomato leaves respectively, while a severity reduction was observed in Xac-artificially inoculated hazelnut seedlings. These results open a new scenario in bacterial disease management since CNC can be also used as nanocarriers for other active molecules. Promising results were obtained in controlling tomato bacterial speck disease with nanostructured microparticles containing CNC, starch and chitosan, by reducing the symptoms in a comparable way to copper. Similar results were obtained against Psav and Xf using CNC and thyme extract nanocapsules.
This work was founded by “Ministero dell’Università e Ricerca. PON Ricerca e Innovazione 2014–2020. Nanotecnologie chimiche green per la protezione sostenibile delle piante (NEMESI) ARS01_01002 CUP: F36C1800018000X.
Hormonal and antioxidant orchestrated responses of canker-resistant and susceptible Cupressus sempervirens clones inoculated with Seiridium cardinale
G. Scimone 1 , C. Pisuttu 1 , L. Cotrozzi 1 , E. Pellegrini 1 , S. Secci 2 , A. Frascella 2 , R. Danti 2 , G. Della Rocca 2
1 Department of Agriculture, Food and Environment, University of Pisa, Via del Borghetto 80, 56124 Pisa, Italy; 2 Institute for Sustainable Plant Protection (IPSP-CNR), National Research Council, Via Madonna del Piano 10, 50019 Sesto Fiorentino, Firenze, Italy. E-mail: [email protected]
Cypress Canker Disease is a pandemic affecting several species in the family Cupressaceae, caused by the invasive necrotrophic fungus Seiridium cardinale. As a result of a genetic improvement program a series of canker-resistant clones of Cupressus sempervirens have been patented. The present study intends to increase the understanding of early biochemical changes in the response to S. cardinale infection of canker-resistant and susceptible cypress clones. In infected ramets of the resistant material, a production of ethylene was observed at 3 and 13 days post inoculation (dpi) in the bark (more than twofold higher than in uninoculated controls). Conversely, an accumulation of salicylic acid was reported at 1 dpi in the foliage, twinned with an increase of abscisic acid which reached a maximum at 9 dpi (more than tenfold). These phytohormones and signaling molecules were able to counteract oxidative burst and prevent lipid peroxidation by offering protection to bark tissues and foliage. In infected ramets of the susceptible material, a concomitant synthesis of ethylene and jasmonic acid was recorded at 3 and 4 dpi in the bark (more than fourfold higher), whereas a partial and transient redox signaling activation was observed in the foliage. However, the synergistic interaction between phytohormones and antioxidants was not able to modulate the defense responses and limit the progression of the oxidative pressure (as confirmed by the increase of hydrogen peroxide levels and malondialdehyde-by products at 13 dpi) so suggesting an explanation of the fungal detrimental effects in the susceptible clone.
Laboratory investigation of the bacterial microflora of the seed of Linum usitatissimum L.
G. Serratore, V. Spina, A. Mondillo, B. De Angelis, E. Frusciante
CREA Centro di ricerca Difesa e Certificazione (Research Centre for Plant Protection and Certification) – SS 18, n° 242, 84091 Battipaglia (SA), Italy. E-mail: [email protected]
In Italy, textile and oilseed flax seeds can be marketed only if they comply with the certification standards. The national legislation refers to the provisions of EU regulation 2016/2031 and EU 2017/625. For the phytosanitary aspect, the reference protocol adopted is the ISTA 7–007 protocol. Bacteriological analyses are not foreseen. In 2021, bacterial colonies were detected during the mycological screening of a sample. The presence of Pseudomonas lini was initially hypothesized. From the sequencing of the 16S rRNA gene, the strain is not associated with the genus Pseudomonas. It was decided to extract the bacterial colonies from the seed. Since there is no specific ISTA protocol, 7–019 (b) was used as a reference. The changes were the use of substrates such as Nutrient Agar and King's medium B. Standard methods were used to characterize colonies at the genus level. Based on some morphological and biochemical characters, the isolate is identified as belonging to the genus Pantoea. Pathogenicity was tested in vivo on flax plants. Control plants were asymptomatic. The bacterium was re-isolated to demonstrate Koch's postulates. To date, there are no studies or reports that demonstrate the spread of this pathogen in Italy. The environmental conditions of our areas seem not to be predisposing for the development of the disease. The high level of homoplasia in the sequences of the 16S rRNA gene complicates the phylogenetic support of the polyphyletic genus Pantoea. The future prospect is therefore an in-depth study of DNA gyrase sequences.
Transcriptome profiling of sanitized artichoke ecotypes and characterization of genes involved in the biosynthesis of secondary metabolites
R. Spanò 1 , S. Fortunato 2 , V. Linsalata 3 , A. Cardinali 3 , I. D’Antuono 3 , M.C. De Pinto 2 , T. Mascia 1
1 Dipartimento di Scienze del Suolo della Pianta e degli Alimenti, Università degli Studi di Bari “Aldo Moro”, Via Amendola 165/A, 70126 Bari, Italy; 2 Dipartimento di Biologia, Università degli Studi di Bari “Aldo Moro”, Via E. Orabona 4, 70124 Bari, Italy; 3 Istituto di Scienze delle Produzioni Alimentari (ISPA) – CNR Via Amendola 122/O, 70126 Bari, Italy. E-mail: [email protected]
Globe artichoke (Cynara cardunculus L. var. scolymus) is gaining commercial interest for the high content of health-promoting bioactive compounds (BACs) such as inulin, polyphenols and antioxidant molecules. Availability of ecotypes sanitized by viral and fungal infections, characterized by vegetative vigor, productivity and quality has relaunched the crop in the market. We report the effect of sanitation protocol, based on in vitro culture and thermotherapy, on BACs accumulation by transcriptomic analysis of two late-flowering artichoke ecotypes Locale di Mola tardivo and Troianella. The analysis highlighted remarkable differences in genetic, environmental information processing and primary cell metabolisms between sanitized and wild-type plants with about 2% of differentially expressed genes (DEGs) mainly involved in biosynthesis of phenylpropanoid, carotenoid and other secondary metabolites. Comparison between the two sanitized ecotypes showed only 75 DEGs, with respect to a higher percentage of DEGs between wild-type ecotypes. This difference could be related to plant response and oxidative burst against pathogen infection in wild-type plants. BACs analysis by HPLC–DAD showed a significant decrease of polyphenols accumulation in sanitized-plants compared to wild-type, speculating a different modulation of the biosynthetic pathway by sanitation protocol. A different response of the two ecotypes was also observed by the analysis of peroxidase activity. A similar approach is being conducted on the early-flowering ecotype Brindisino. This study represents the unique investigation of transcriptome profile and BACs accumulation in artichoke ecotypes exposed to sanitation protocol and aims to promote the soilless cultivation of vigorous and pathogen-free artichokes, even for BACs extraction.
This work was supported by the project ERiCa—6E389E5E funded by Regione Puglia—POR Puglia 2014/2020—Asse X—Azione 10.4.-Research for Innovation – REFIN.
Metabolomics for the selection of beneficial microorganisms and/or their metabolites for a new generation of plant protection products
A. Staropoli 1,2 , S.L. Woo 2,3 , F. Vinale 2,4
1 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 2 Institute for Sustainable Plant Protection, National Research Council, 80055 Portici, Italy; 3 Department of Pharmacy, University of Naples “Federico II”, 80131 Naples, Italy; 4 Department of Veterinary Medicine and Animal Production, University of Naples “Federico II”, 80137 Naples, Italy. E-mail: [email protected]
Due to an increasing awareness about negative impacts to the environment, there has been a shift from conventional to sustainable agriculture, that has led to a change in regulations and the promotion in the use of microbial-based products as alternatives to synthetic chemical products. Metabolomics is a useful tool to guide the selection of beneficial microbes (single-strain inoculants, synthetic consortia, co-application of two or more microorganisms) and their secondary metabolites as active substances of new bio-formulates, and to evaluate treatment performance. With the aim to discover new formulations for agricultural uses, a metabolomic approach was used to investigate the compatibility of different beneficial microbial strains (i.e., Trichoderma, Streptomyces, Azotobacter) and test selected combinations in applications to different crops (parsley, basil, olive drupes). Moreover, the chelating properties of harzianic acid (a Trichoderma siderophore) were studied in order to enhance soil nutrient quality. Results showed that the tested combinations had significant beneficial effects on crops, in terms of pathogen control and improved nutritional value. Statistical analysis of plant extracts revealed a modulation of metabolic profiles depending on the treatment based on beneficial microbe application. It was possible to identify several metabolites (e.g., petroselinic acid, quinic acid, caffeic acid, rosmarinic acid, coumaric acid, oleuropein, luteolin and other phenolic compounds), whose relative abundance was increased in treated samples compared to the water control. Moreover, the analyses of the formulations consisting of harzianic acid and bivalent metal cations (biologically relevant or toxic), highlighted the formation of neutral of charged complexes depending on pH of the solution.
A novel methoxychromone derivative from Trichoderma harzianum M10
A. Staropoli 1,2 , G. Iacomino 1 , A. Vassetti 2 , C. Gigliotti 2 , R. Donelli 1 , L. Di Costanzo 1 , F. Vinale 2,3
1 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 2 Institute for Sustainable Plant Protection, National Research Council, 80055 Portici, Italy; 3 Department of Veterinary Medicine and Animal Production, University of Naples “Federico II”, 80137 Naples, Italy. E-mail: [email protected]
Fungi of the genus Trichoderma are characterized by a high versatility and some selected species are plant symbionts, capable of producing multiple positive effects on plants, such as disease control, growth promotion, increased resistance etc. Several strains of Trichoderma are well-known producers of bioactive secondary metabolites (SMs). Recently, it has been demonstrated that Trichoderma produces natural compounds that can be involved in the beneficial interactions directly established with the plants. In the course of an ongoing search for bioactive secondary metabolites from T. harzianum M10, a methoxychromone derivative has been isolated. After growing M10 in an inductive liquid media for 21 days, the cultural filtrate was exhaustively extracted with ethyl acetate and the crude extract was fractionated by direct phase column chromatography (silica gel, atmospheric pressure). A compound, obtained as pure crystal from one of the fractions, was identified as 5-hydroxy-2,3-dimethyl-7-methoxychromone by X-ray analysis (from data collected at the XRD1, Elettra Synchrotron, Trieste) and spectroscopic studies. The compound, isolated for first time from a T. harzianum culture, was tested at different concentrations for antifungal activity against the phytopathogenic agent Rhizoctonia solani. The methoxychromone showed a 45% of growth inhibition after 24 h of incubation at a concentration of 100 ng plug −1 whereas the maximum inhibition percentage was found to be 61% at 200 µg plug−1. These results suggest a possible role of 5-hydroxy-2,3-dimethyl-7-methoxychromone in antibiosis mechanism of T. harzianum M10 against fungal pathogens.
This research was funded by MISE (Protection F/050421/01–03/X32 and VIABio F/200095/01–03/X45) and PSR Campania 2014–2020 (DIONISO B98H19005010009).
Visualization and analysis of grapevine downy mildew disease progress data on partially resistant varieties
O. Taibi 1 , I. Salotti 1 , F. Bove 2 , T. Caffi 2 , V. Rossi 2
1 Università Cattolica del Sacro Cuore—Dipartimento delle Produzioni Vegetali Sostenibili (DIPROVES)— Via E. Parmense 84, 29122 Piacenza, Italia; 2 Horta srl – Spin-off Università Cattolica del Sacro Cuore— Via E. Gorra, 55, 29122 Piacenza, Italia. E-mail : [email protected]
Grapevine downy mildew (DM), caused by Plasmopara viticola, is a threatening disease in viticulture, especially when weather conditions are conducive and the host is highly susceptible. DM-resistant cultivars express varying degrees of partial resistance and are important for integrated disease management. Still, they require plant protection interventions throughout the season to prevent yield losses and protect the durability of host resistance. Most studies on grapevine varieties' resistance durability and variability were conducted in controlled conditions with monocyclic experiments, artificial inoculations and primarily on leaves. Therefore, an understanding of the disease progress on the resistant varieties on both leaves and clusters under field conditions, where many infection cycles are concatenated, would add precision to the scheduling of fungicide interventions. At the experimental vineyard (University campus), we monitored weekly DM infections on leaves and clusters for four years, from the bud break to the harvest of 16 varieties registered in the Vitis International Variety Catalogue (VIVC). Using a Linear Mixed Model (LMM), we analyzed the Area Under the Disease Progress Curve (AUDPC) to assess the influence of different varieties on disease development. Resistant varieties such as Bronner, Cabernet Volos, Calardis Blanc, Johanniter, Merlot Kanthus, Merlot Khorus, Regent, or Solaris had a lower AUDPC compared to Calandro, Felicia, Fleurtai, Reberger, Rkatsitelli, and Villaris, with Palava and Merlot showing the highest AUDPCs. The resistant varieties carrying quantitative treat loci (QTLs) such as Rpv 3, Rpv 10 and/or Rpv 12 reduced the AUDPC on leaves and clusters by > 90% on average, compared to the susceptible Merlot.
The complete genome of Fusarium musae strain F31, pathogen of banana fruits
V. Tava 1 , L. Degradi 1 , M. Saracchi 1 , A. Kunova 1 , C. Pizzatti 1 , P. Cortesi 1 , M. Brock 2 , M. Pasquali 1
1 DeFENS, Università degli Studi di Milano, Milano, ITALY; 2 School of Life Sciences, University of Nottingham, University Park, NG72RD Nottingham, UK. E-mail: [email protected]
F. musae is a recently described species belonging to Fusarium fujikuroi Species Complex and sister to F. verticillioides. It causes crown rot on banana fruits and it can also be opportunistic human pathogen. By exploring its genome we aim to explore diversities that might facilitate the identification of F. musae species within the genus Fusarium. Identification of pathogenic genes and comparative genomic studies could help us to understand peculiarities of this pathogen. We obtained complete genome at chromosome level of the representative strain of F. musae F31, isolated in 2013 in Dominican Republic from a diseased banana. The genome is 44.07 Mb, divided into 12 chromosomes (11 have both telomers), the circular mitochondrial DNA (mtDNA) and one unplaced contig. Functional annotation led to a total of 13,963 annotated genes, of which 13,661 were proteins and 302 transfer RNA (tRNA) for nuclear DNA. EffectorP 3.0 estimated the presence of 214 effectors (probability 0.9 or higher) and antiSMASH (version 6.1.1) analysis predicted 44 secondary metabolite clusters. F31 potentially produces mycotoxins typical of Fusarium species such as fusarin and fusaric acid but also pigments such as fusarubin and bikaverin. The fumonisin biosynthesis gene cluster, distinctive of F. verticillioides strains, is absent in F. musae, whereas a gibberellin biosynthesis gene cluster is present (usually absent in its “sister species”). Genome data allowed the generation of fluorescent F. musae reporter strains expressing E2-Crimson under control of the constitutively active F. musae enolase promoter Peno1. Reporter strains will be used to explore the infection mechanism in bananas.
Response of seed germination to heat treatment required to reduce the risk of ‘ Candidatus Liberibacter solanacearum’ vertical transmission in Apiaceae crops
L. Tizzani, A. Gentili, A. Taglienti, G. Bertinelli, V. Ilardi
Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di Ricerca Difesa e Certificazione sede di Roma (CREA-DC), Via C. G. Bertero 22, 00156 Roma, Italia. E-mail: [email protected]
In Europe and the Mediterranean basin, ‘Candidatus Liberibacter solanacearum’ (Lso) is associated with vegetative diseases of Apiaceae crops, mainly carrot (Daucus carota L.), inducing yellow or red leaf discoloration, reduction in size of the main root with lateral root proliferation. In some countries (e.g. Japan and Australia), mandatory heat treatment (50 °C) or PCR negative testing for Lso is required as a phytosanitary measure for carrot seed importation. Carrot seed movement is economically important for the Mediterranean region, and Lso is extensively widespread in Apiaceae seeds. Hence, heat treatment plays a key role as a disinfection method for ensuring international seed trade; nonetheless, the treatment could reduce seeds germination. In this work, the impact of heat treatment on germination of 41 Apiaceae seed lots (24 carrot, 10 parsley, 5 fennel, and 2 celery) belonging to different varieties and hybrids was investigated: healthy and Lso-infected seed lots were heat-treated and then germinated according to International Seed Testing Association. Germinability data of treated and untreated seeds were compared in both healthy and infected lots; overall, the heat treatment significantly decreased the germination percentage in 17% of the analysed seed lots of carrot, 30% of parsley, 40% of fennel, and 50% of celery. Our data show that heat treatment negatively affects the germination rate of Apiaceae seeds, although to a different extent depending on the species and variety. These results may envisage an economic burden not only for the heat treatment procedure per se, but also for decreasing the seed germinability.
This work was supported by the project PROTEGGO, funded by MiPAAF (Ministero delle Politiche Agricole, Alimentari e Forestali) and performed in the frame of Euphresco PhyLib III project.
Detection and absolute quantification of the fungus Fusicladium oleaginum in olive tree by droplet digital-PCR (ddPCR)
L. Tizzani, V. Lumia, G. Bertinelli, C. Arcuri, M. Pilotti, V. Ilardi
Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di Ricerca Difesa e Certificazione sede di Roma (CREA-DC), Via C. G. Bertero 22, 00156 Roma, Italia. E-mail: [email protected]
Olive leaf spot, caused by the fungus Fusicladium oleaginum (= Venturia oleagina), is a leaf disease, widespread in all olive-growing regions of the world. This pathogen is often present also in asymptomatic leaves depending on environmental factors (e.g. temperature, humidity). In appropriate conditions the attack is severe and it can lead to massive defoliation, reduced productivity and poor oil quality. The aim of this study was to develop real time-qPCR (qPCR) and droplet digital PCR (ddPCR) for an accurate, early and robust detection of F. oleaginum. ddPCR indeed overcomes some limitations of qPCR (e.g. the need of a standard, the influence of inhibitors). Symptomatic leaves of 28 olive trees from 10 different Italian regions were collected. The ITS FusiR Fusicladium-specific and the ITS5 primers enabled to amplify the region ITS 1—5.8S rDNA—ITS2, without fungal isolation. The obtained sequences showed that this region is highly conserved among different fungal isolates. Primer and probes were designed to be suitable for both EvaGreen®, TaqMan®-qPCR and ddPCR techniques. All genomic extracts, obtained from symptomatic leaves, tested positive in all developed methods. The exclusivity was tested against twenty-eight epiphytic and endophytic olive leaf fungi, isolated from symptomatic and asymptomatic leaves. Tests are underway to fully develop and validate these methods. Monitoring climate parameters together with a sensitive molecular detection of F. oleaginum will be crucial for leaf spot control by establishing timely chemical interventions, which combines effectiveness and dose reduction.
This work was supported by the project SALVAOLIVI funded by MIPAAF (Ministero delle Politiche Agricole, Alimentari e Forestali)
Population genetics and fitness of Plasmopara viticola in the winegrowing region of South Tyrol
S. Tomada, A. Moffa, G. Tonini, H.S. Shevkar, S. Baric
Libera Università di Bolzano-Bozen, Facoltà di Scienze e Tecnologie, Piazza Università 5, 39051 Bolzano. E-mail: [email protected]
Grapevine downy mildew, caused by Plasmopara viticola, is considered one of the most devastating oomycete diseases globally, causing significant yield losses in vineyards. It is generally accepted that effective control of the pathogen relies on solid knowledge of P. viticola population genetics for each specific wine-growing region. Up to date, no data are available about the genetic composition of P. viticola populations in South Tyrol (Northern Italy). For this reason, the aim of the present study was to investigate the population genetics and fitness of P. viticola in dependence on distinct farming management systems differing in the chemical classes and dosages of applied fungicides. Nine vineyards located in an area of 30 km of linear distance were selected. Fifteen samples per vineyard were collected in 2020 at the end of the growing season and genotyped with 15 microsatellite markers. The fitness parameters were evaluated in vitro by the leaf disk assay. A total of 35 different alleles were found in the dataset, with an average of 2.33 alleles per locus. Among the 135 isolates, 125 distinct multilocus genotypes were identified. The elevated genetic variability of P. viticola within vineyards points to a high degree of sexual reproduction in the area. Instead, population differentiation between the vineyards was low and no differentiation was found among the different management systems, also concerning the fitness parameters, suggesting the presence of a single metapopulation. An in-depth analysis of molecular data will allow us to elucidate processes taking place in the P. viticola populations of South Tyrol.
Hybrid assembly and comparative genomics of three different isolates of Gnomoniopsis castaneae
S. Turco 1 , M.I. Drais 1 , G. Bastianelli 2 , C. Morales-Rodriguez 2 , P. Gonthier 3 , A. Vannini 2 , A. Mazzaglia 1
1 Dipartimento di Scienze Agrarie e Forestali, Università degli Studi della Tuscia, via S. Camillo de Lellis snc, 01100 Viterbo, Italy; 2 Dipartimento per l’innovazione nei sistemi biologici, agroalimentari e forestali, Università degli Studi della Tuscia, via S. Camillo de Lellis snc, 01100 Viterbo, Italy; 3 Dipartimento di Scienze Agrarie, Forestali e Alimentari, Università degli Studi di Torino, Largo Braccini 2, 10095 Grugliasco, Italy. E-mail: [email protected]
Gnomoniopsis castaneae (syn. G. smithogilvyi) is the causal agent of the brown rot of chestnut fruits and is responsible for leaf, shoot blight, and bark cankers as well. The pathogen spends part of its life cycle as an endophyte in the host tissues. Symptoms on fruits are mainly expressed post-harvest when peculiar conditions of temperature and humidity occur. Thus, healthy fruits may hide a latent infection that can rapidly switch to symptoms expression. Recently, a rapid and reliable detection method based on a qPCR assay was developed to reveal the presence of the pathogen in all the symptomatic and asymptomatic specimens. Nonetheless, to better understand the genetics behind the endophytic behavior and the pathogenic mechanisms of this fungus, here we provide the first complete genome sequence of three different isolates of Gnomoniopsis castaneae: the Italian ex-type MUT-401, a second Italian MW494885 isolate and the ICMP 14,040 isolate from New Zealand. The three genome sequences were obtained through a hybrid assembly using both short Illumina reads and long Nanopore reads, the coding sequences annotated and compared with other genomes of the Diaporthales. The identification of specific traits related to the three isolates with different geographic origins could give insights into the interaction between the pathogen, its host, and the environment. These premises would also define the foundation for future analysis, with particular focus on the relationship between the endophytes and the outer world and on developing customized control strategies.
The research was carried out in the frame of the Italian MIUR (Ministry for education, University and Research) initiative “Department of Excellence” (Law 232/2016).
Rapid molecular assay for the evaluation of seed treatments with clove essential oil against wheat common bunt
M.T. Valente 1 , G. Manetti 1,2 , L. Orzali 1 , F. Magnanimi 1,3 , A. Matere 1 , V. Bergamaschi 1,2 , L. Riccioni 1
1 Council for Agricultural Research and Economics (CREA), Research Centre for Plant Protection and Certification (CREA-DC), Via C.G. Bertero 22, Rome 00156, Italy; 2 Department of Environmental Biology, “Sapienza” University of Rome, P. le Aldo Moro 5, 00185 Rome, Italy; 2 Department of Biology and Biotechnology “Charles Darwin”, “Sapienza” University of Rome, P. le Aldo Moro 5, 00185 Rome, Italy. E-mail: [email protected]
Common bunt, caused by the fungus Tilletia laevis, is an important disease of wheat, causing considerable yield loss and reduction in seed quality. The spread of organic and low-input farming, led to the increase of the disease incidence and to a growing search for bio-compounds replacing chemical treatments for pest management. In this study, the antifungal activity of clove oil was evaluated as seed treatment against T. laevis. The trial was set up with 6 different treatments, performed either by seed immersion or spray application, consisting in clove essential oil and two clove oil-based experimental formulations, used in different concentrations, plus two controls (infected/untreated and infected/copper-treated). For each treatment 90 seeds of durum wheat (cultivar Grifoni) naturally infected were sown in 3 repetitions, for a total of 30 seeds/ pot. The germination rate was evaluated: the clove oil treatment concentrated at 0.5% v/v showed phytotoxic effects. Before the tillering stage, 25 seedlings were sampled for each pot and separated in 5 bulks of 5 plants each; the DNA was extracted from each bulk. A qPCR test has been developed for the specific quantification of T. laevis in the seedling bulks, obtaining optimal results both for amplification efficiency (95–110%) and for the detection sensitivity (10 fg). The results were statistically analysed (ANOVA) and showed that immersion and spray treatments with both the experimental formulations significantly reduced the percentage of plant infection. Given the promising results, further tests are ongoing to confirm the data obtained.
This work was supported by the project DIBIO, funded by MIPAAF.
Metabolomic characterization of a Clonostachys rosea strain active against nematodes
A. Vassetti 1 , E. Troiano 1 , A. Staropoli 1,2 , C. Gigliotti 1 , C. Vincenzo 2 , R. Donelli 2 , G. Parrella 1 , G. d’Errico 2 , F. Vinale 3
1 Institute for Sustainable Plant Protection, National Research Council, 80055 Portici, Italy; 2 Department of Agricultural Sciences, University of Naples “Federico II”, 80055 Portici, Italy; 3 Department of Veterinary Medicine and Animal Productions, University of Naples “Federico II”, 80137 Naples, Italy. E-mail: [email protected]
The rhizosphere is occupied by a variety of organisms that coexist by sharing space and nutrients. Beneficial microbial communities are well-known for their capacity to improve plant growth and resistance to biotic and abiotic stresses and studied to achieve a better understanding of microbe-microbe, microbe-plant and microbe-pathogen interactions. Since secondary metabolites, which are released in the rhizosphere, play an important role in these interactions, a detailed characterization of microbial metabolome is essential. For this study, soil from selected vineyards located in Avellino province (Campania region, southern Italy) were collected to characterize the microfauna associated to grapevines rhizosphere by extensive application of morphological and molecular techniques. Among the beneficial microbes, a Clonostachys rosea strain isolated in suppressive soil, showed a significant in vitro nematicidal activity. Therefore, to chemically characterize the exo metabolome of this isolate, the crude extracts of cultural filtrates were fractionated by direct and indirect chromatography to identify the bioactive secondary metabolites. Mass spectrometry, NMR and crystallography techniques were used to characterize the isolated natural compounds. The fraction containing a mixture of diketopiperazines exhibited a strong nematicidal activity demonstrating that these exometabolites have a significant role on the improvement of soil quality and to control pests and/or pathogens on vineyard production. More investigations are ongoing in order to obtain pure molecules for in vivo test against plant pathogens and nematodes.
This work has been supported by DIONISO Project (PSR Campania 2014–2020—Azione 2 “Sostegno ai POI”), 2019–2021.
First evidence of Pantoea ananatis infections on strawberry in Italy
B. Vicelli 1 , G. Puopolo 2,3 , G. Angeli 1 , V. Gualandri 1
1 Fondazione Edmund Mach, Centro di Trasferimento Tecnologico, Via Edmund Mach 1, 38010 San Michele all’Adige, Trento, Italy; 2 Università degli Studi di Trento, Centro Agricoltura Aliementi Ambiente, Via E. Mach 1, 38010, San Michele all'Adige, Italy; 3 Fondazione Edmund Mach, Centro di Ricerca e Innovazione, Via Edmund Mach 1, 38010 San Michele all’Adige, Trento, Italy. E-mail: [email protected]
Recently, symptoms resembling fire blight such as brownish lesions, necrosis, and production of exudates were observed on a new cold-resistant strawberry variety (Ania®). Thus, the aim of this study was to isolate, identify and characterize the causative agents. Bacterial isolation was carried out from symptomatic plant tissues (leaves and stems) collected from four strawberry plants. Tissues were surface-sterilized, homogenized in 0.85% (w/v) NaCl and serially diluted onto Nutrient Agar. One hundred twenty bacterial isolates were tested for the induction of the hypersensitive response (HR) in tobacco leaves. Bacterial isolates with yellow colonies able to induce HR were identified as Pantoea ananatis by 16S rDNA gene sequencing. PCRs using the primer pair PANA_1080 specific for P. ananatis species validated these results. Subsequently, pathogenicity tests were carried out by wounding detached ripe pseudo-fruits from strawberry plants (Ania®) and inoculating five µL of bacterial cell suspension (∼1 × 108 CFU/mL) of a selected P. ananatis strain. Distilled water and Escherichia coli DH5α were used as control treatments. Strawberries were incubated in the dark at 27 °C and 90% RH and appearance of symptoms was monitored daily. No symptoms were observed on the control treatments. On P. ananatis inoculated strawberries, symptoms appeared after five days and the bacterial strain was re-isolated from the sympotmatic pseudo-frutits. Furthermore, P. ananatis strains were evaluated for the production of plant cell wall degrading enzymes (i.e. cellulases). To the best of our knowledge, this is the first report of P. ananatis on strawberry in Italy.
Insights on KP4 killer toxin-like proteins of Fusarium species in interspecific interactions
I. Vicente 1 , G. Quaratiello 1 , R. Baroncelli 2 , G. Vannacci 1 , S. Sarrocco 1
1 Laboratory of Plant Pathology, Department of Agriculture, Food and Environment, University of Pisa, Via del Borghetto 80, Pisa, Italy; 2 Laboratory of Plant Pathology, Department of Agricultural and Food Sciences (DISTAL), University of Bologna, Viale Fanin 50, Bologna, Italy. E-mail: [email protected]
KP4 killer toxins are secreted proteins inhibiting cell growth and inducing cell death in target organisms. First identified in Ustilago maydis, homologues have been also found in yeasts, in one species of moss, and in ascomycetes. In Fusarium graminearum, KP4-like (KP4L) proteins contribute to fungal virulence in wheat seedling rot and are expressed under stress conditions and during Fusarium head blight development. However, fungal KP4L proteins are also hypothesized to support fungal antagonism by permeabilizing cell walls of competing fungi, thus enabling penetration of toxic compounds. Here, we report differential expression patterns of four KP4L genes of F. graminearum (Fgkp4l-1, -2, -3 and -4) in a competitive interaction, using Trichoderma gamsii as antagonist. Results from time-course experiments performed in dual cultures revealed that Fgkp4l-3 and Fgkp4l-4 could participate in the recognition at distance of the antagonist, while all the four Fgkp4l genes were highly activated in the pathogen during the physical interaction of both fungi. When on wheat spikes, only Fgkp4l-4 was differentially expressed during the interaction with T. gamsii. This suggests a possible role of KP4L proteins in supporting F. graminearum interspecific interactions, even in living plant tissues. Database searches enabled the identification of KP4L orthologous proteins in 21 Fusarium genomes. Distribution of KP4L proteins across five phylogenetic groups within the genus Fusarium revealed they are more represented in species with broad host-plant range than in host-specific species. Phylogeny inferred provides evidence that KP4L genes evolved through gene duplications, gene loss and sequence diversification in the genus Fusarium.
The diversity of Xanthomonas spp. from Brassicaceae, diagnostics and disease control
J.G. Vicente1,2, S. Greer2, J. Harrison3, E. Haynes1, B. Carter1, A. Aspin1, A. Bryning1, J. Cole1, S. Carroll1, M. Kennedy1, S. Gosh2, D. Bown2, V. Ntoukakis2, D. Studholme3, M. Grant2
1 Fera Science Ltd, York, UK 2 The School of Life Sciences, University of Warwick, Coventry, UK 3 College of Life and Environmental Sciences, University of Exeter, Exeter, UK. E-mail: [email protected]
Plants from the Brassicaceae family can be attacked by different Xanthomonas pathogens. Xanthomonas campestris pv. campestris (Xcc), the cause of black rot of brassicas, is one of the most important diseases of crops like cabbage and cauliflower worldwide. Ornamental Brassicaceae plants like wallflowers and garden stocks and some common weeds can also be infected by X. campestris pathogens including X. campestris pv. incanae. In contrast to pvs campestris and incanae, that cause vascular diseases, pv. raphani causes leaf spots in a larger range of Brassicaceae and some other hosts including tomato. At least two of these pathovars, campestris and raphani, have race structures defined by the ability of causing disease in different brassica differential accessions. A disease that causes symptoms in watercress that resemble black rot, is caused by a different species, X. nasturtii. We are characterising the pathogenicity and sequencing a collection of Xanthomonas spp. isolates from different origins and spanning over 60 years, including the different pathovars of X. campestris and also the watercress pathogen, X. nasturtii. The results of our work will allow us to clarify the taxonomy of X. campestris and related species and to design diagnostic testing for some important species, pathovars and possibly some groups of races of Xanthomonas. We are screening brassica collections to identify useful sources of resistance and we are developing chlorophyll imaging and whole plant imaging techniques to allow us to study Xcc infections. We aim to characterise and map resistances and identify resistance-linked markers that can assist future breeding.
This work is part of the project ‘Xanthomonas plant diseases: mitigating existing, emerging and future threats to UK agriculture’ funded by UK Research and Innovation (UKRI) through the Strategic Priorities Fund (SPF), by grants (BB/T01924/1, BB/T010908/1 and BB/T010916/1) from the Biotechnology and Biological Sciences Research Council (BBSRC) of the UK with support from the Department for Environment, Food and Rural Affairs (Defra) and the Scottish Government.
Evaluation of tolerance of a walnut progeny ( Juglans regia x J. major x J nigra and J. microcarpa ) to crown and root rot disease caused by Phytophthora cinnamomi
S. Vitale 1 , L. Luongo 1 , I. Garaguso 1 , E. Degl’Innocenti 1 , L. Angeloni 2 , F. Menta 2 , G. Pignatti 2 , G. Pontuale 2 , A. Matere 1 , M. Petrucci 1 , A. Haegi 1
1 CREA DC Research Centre for Plant Protection and Certification, Via C. G. Bertero 22, 00156 Roma, Italy. 2 CREA FL Research Centre for Forestry and Wood, Via Valle della Quistione 27, 00166 Roma. E-mail: [email protected]
The cultivation of common walnut (Juglans regia) is very sensitive to Phytophthora cinnamomi. Several commercial walnut orchards were decimated by P. cinnamomi in northeastern Italy. In response to Phytophthora attacks, several Juglans species in addition to J. regia were considered as rootstocks. The Italian Ministry of Agriculture financially supported a project named PORTNOC addressed to respond to this issue. After an intense selection, 4 Juglans microcarpa genotypes and 5 hybrids showed good level of tolerance towards P. cinnamomi. A total of 392 progenies derived from these mother plants were subjected to artificial inoculation with P. cinnamomi (isolate CREADC-Om274). Based on field observations and necrosis length caused by the pathogen, 23 progenies of hybrid 10/43 (J. regia x J. major x J. nigra), 11 progenies of hybrid 6/22 (J. regia x J. major x J. nigra) and 3 progenies of J. microcarpa M7 showed significantly higher tolerance levels towards P. cinnamomi. In addition, evaluation of seedling progenies of tolerant genotypes grafted with commercial walnut varieties is currently underway in naturally infected fields and will provide further information for successful usage.
This work was founded by the Italian Ministry of Agriculture (MiPAAF) within the research project PORTNOC “Valutazione di portinnesti per la tolleranza/resistenza a Phytophthora e blackline e valorizzazione di varietà di Juglans regia compatibili”.
Influence of carbon and nitrogen sources on Trichoderma pH modulation and biocontrol activity
S. Vitale 1 , F. Salzano 2 , R. Marra 2 , S.L. Woo 3 , D. Turrà 2 , F. Vinale 4
1 National Research Council, Institute for Sustainable Plant Protection (IPP-CNR), Piazzale Enrico Fermi 1, 80055, Portici (Na), Italy; 2 Department of Agricultural Sciences, University of Naples “Federico II”, Via Università 100, 80055, Portici (Na), Italy; 3 Department of Pharmacy, University of Naples “Federico II”, Via Domenico Montesano 49, 80131, Naples, Italy; 4 Department of Veterinary Medicine and Animal Productions, University of Naples “Federico II”, 80137 Naples, Italy. E-mail: [email protected]
Trichoderma spp. are the most widely used biocontrol agents in agriculture protecting plants against pathogens through different mechanisms: mycoparasitism, secretion of lytic enzymes and bioactive metabolites, nutrient competition or antibiosis. The Trichoderma harzianum isolate M10 is known to produce harzianic acid (HA), a metabolite with both antimicrobial and plant growth promoting activities. To test the environmental conditions influencing the production of HA and thus its activity against plant pathogens, M10 was grown in different liquid media (rich or minimal media). Here, we found that M10 initially induces a slight extracellular acidification followed by a strong alkalinization at later time points in rich media. On the contrary, in poor media the pH is maintained constant over time. Since rhizosphere pH alkalinization is an important mechanism used by plant pathogens to infect the host plant, we decided to test if different concentrations of carbon sources (glucose or sucrose) or different nitrogen sources (nitrate or ammonia) influence M10 extracellular alkalinization and bicontrol activity in vitro. Our results suggest that M10 mediated alkalinization is independent of the carbon source but not of the nitrogen source supplemented to the medium. Indeed, ammonia supplements reverted extracellular alkalinization to acidification. Importantly, ammonia-mediated acidification also results in higher biocontrol activity against the soil born fungal pathogen Fusarium oxysporum. Further experiments are required to provide a putative correlation between extracellular pH modulation and HA production in M10.
This work was supported by PROTECTION Project (MISE CRESO grant number Protection no. F/050421/01–03/X32).
Volatile-mediated inhibitory activity of the biocontrol agent Lysobacter capsici AZ78: a result of multiple factors interaction
A. Vlassi 1 , A. Nesler 2 , A. Parich 1 , R. Schuhmacher 1 , G. Puopolo 3,4
1 Department of Agrobiotechnology (IFA-Tulln), Institute of Bioanalytics and Agro-Metabolomics, University of Natural Resources and Life Sciences Vienna (BOKU), Tulln, Austria; 2 Bi-PA nv (Biological Products for Agriculture), Londerzeel, Belgium; 3 Center of Agriculture, Food, Environment, University of Trento, San Michele all’Adige, Italy; 4 Research and Innovation Centre, Department of Sustainable Agro-Ecosystems and Bioresources, Fondazione Edmund Mach, San Michele all’Adige, Italy. E-mail: [email protected]
Plant beneficial rhizobacteria are able to inhibit the growth of soilborne phytopathogenic fungi and oomycetes through the release of a relevant number of volatile compounds. Based on this, we investigated the ability of the biocontrol agent Lysobacter capsici AZ78 (AZ78) to produce volatile organic compounds (VOCs) that may contribute to its efficacy in controlling soilborne phytopathogenic fungi and oomycetes. AZ78 significantly reduced the growth of Pythium ultimum, Rhizoctonia solani and Sclerotinia minor through the release of VOCs in split Petri dish assays. The GC–MS analysis revealed that AZ78 produced 22 VOCs and most of them were putatively identified as mono- and dialkylated methoxypyrazines. Exposure to 2,5-dimethylpyrazine, 2-ethyl-3-methoxypyrazine and 2-isopropyl-3-methoxypyrazine determined a drastic reduction of P. ultimum, R. solani and S. minor mycelium growth in split Petri dish assays. However, a remarkable difference between the toxicity of the pyrazines and the AZ78 total volatile blend was observed. This difference led us to further investigate the volatile-mediated inhibitory activity of the biocontrol bacterium. Further experiments revealed the ability of AZ78 cells to produce ammonia that caused the alkalinization of the physically separated growth medium in split Petri dishes assays. As a consequence, the mycelium growth of the tested phytopathogenic fungi and oomycetes was negatively affected on the alkalinized growth medium. Results achieved in this work clearly showed that volatile-mediated inhibitory activity of AZ78 mainly relies on the interaction between the toxicity of VOCs, ammonia and the alkalinization of growth medium.
This work was supported by European Union’s Horizon 2020 Research and Innovation Program under the Marie Skłodowska-Curie grant agreement no. 722642 (project INTERFUTURE).
Genetic characterisation of Fusarium culmorum isolates from Çanakkale, Turkey
E. Yörük 1 , S. Oufensou 2 , Ö. Sefer 1 , Q. Migheli 2 , V. Balmas 2
1Department of Molecular Biology and Genetics, Faculty of Arts and Sciences, Istanbul Yeni Yuzyil University, postal code:34010, no: 26, Zeytinburnu, Istanbul, Turkey; 2Dipartimento di Agraria, Desertification Research Centre, Università degli Studi di Sassari, Viale Italia 39, 07100 Sassari, Italy. E-mail: [email protected]
Fusarium head blight (FHB) is one of the most important fungal diseases of small grain cereals worldwide. Fusarium culmorum is the predominant causal agent of FHB in many Mediterranean regions, leading to reduction in crop quality and quantity. Mycotoxin contamination of grain, primarily with deoxynivalenol and its derivatives, may occur in fields hit by FHB epidemics. In this study, a detailed genetic characterisation of F. culmorum isolates obtained from scabby wheat samples collected from the Çanakkale province (Turkey) in the year 2021, was carried out. A total of 45 monosporic cultures were obtained and were identified at species level by morphological characteristics. Tef1-α sequencing was also carried out to confirm the identification at species level. For this purpose, genomic DNA was extracted from 7-day-old potato dextrose agar (PDA) cultures. A portion of the Tef1-α gene was amplified from all isolates and then sequenced. The sequences were subjected to BlastN analysis. All 45 monosporic isolates, obtained from Sutlucekoy and Bayirkoy regions (Çanakkale Province), were identified as F. culmorum. Class B chemotype assays were also carried out in Turkish F. culmorum isolates by PCR assay. The TRI12 gene was amplified to distinguish isolates as 3-acetyldeoxinivalenol (3-ADON) or nivalenol (NIV) producers. All isolates belonged to 3-ADON chemotype and to MAT1-2 idiomorph. Further studies will be extended to a larger number of isolates from different regions in order to provide a comprehensive set of data on the Fusarium populations causing FHB in this area.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
| 0 | PMC9707090 | NO-CC CODE | 2022-12-09 23:15:01 | no | J Plant Pathol. 2022 Nov 29; 104(4):1207-1280 | utf-8 | null | null | null | oa_other |
==== Front
J Gastrointest Surg
J Gastrointest Surg
Journal of Gastrointestinal Surgery
1091-255X
1873-4626
Springer US New York
36443554
5537
10.1007/s11605-022-05537-0
Original Article
Barriers and Facilitators in Implementation of an Esophagectomy Care Pathway: a Qualitative Analysis
http://orcid.org/0000-0002-0954-1052
Madsen Helen J. [email protected]
12
Lambert-Kerzner Anne 13
Mucharsky Ellison 1
Gergen Anna K. 2
Dyas Adam R. 2
McCarter Martin 4
Stewart Camille 4
Pratap Akshay 2
Mitchell John 5
Randhawa Simran 5
Meguid Robert A. 135
1 grid.430503.1 0000 0001 0703 675X Surgical Outcomes and Applied Research Program, Department of Surgery, University of Colorado School of Medicine, Aurora, CO USA
2 grid.430503.1 0000 0001 0703 675X Department of Surgery, University of Colorado School of Medicine, Aurora, CO USA
3 grid.430503.1 0000 0001 0703 675X Adult and Child Center for Health Outcomes Research and Delivery Science, University of Colorado School of Medicine, Aurora, CO USA
4 grid.430503.1 0000 0001 0703 675X Department of Surgery, Division of Surgical Oncology, University of Colorado School of Medicine, Aurora, CO USA
5 grid.430503.1 0000 0001 0703 675X Department of Surgery, Division of Thoracic Surgery, University of Colorado, Aurora, CO USA
28 11 2022
19
9 7 2022
1 11 2022
© The Society for Surgery of the Alimentary Tract 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Introduction
A new postoperative esophagectomy care pathway was recently implemented at our institution. Practice pattern change among provider teams can prove challenging; therefore, we sought to study the barriers and facilitators toward pathway implementation at the provider level.
Methods
This qualitative study was guided by the Theoretical Domains Framework (TDF) to study the adoption and implementation of a post-esophagectomy care pathway. Sixteen in-depth interviews were conducted with providers involved with the pathway. Matrix analysis was used to analyze the data.
Results
Providers included attending surgeons (n = 6), advanced practice providers (n = 8), registered dietitian (n = 1), and clinic staff (n = 1). TDF domains that were salient across our findings included knowledge, beliefs about consequences, social influences, and environmental context and resources. Identified facilitators included were electronic health record tools, such as note templates including pathway components and a pathway-specific order set, patient satisfaction, and preliminary data indicating clinical benefits such as a reduced anastomotic leak rate. The major barrier reported was a hesitance to abandon previous practice patterns, most prevalent at the attending surgeon level.
Conclusion
The TDF enabled us to identify and understand the individuals’ perceived barriers and facilitators toward adoption and implementation of a postoperative esophagectomy pathway. This analysis can help guide and improve adoption of surgical patient care pathways among providers.
Keywords
Esophagectomy
Postoperative pathway
Qualitative study
Implementation
==== Body
pmcIntroduction
Postoperative clinical pathways are frequently created to characterize and standardize the critical steps in caring for patients who have undergone a specific operation. Clinical pathways have been shown to have several benefits, including reducing hospital length of stay (LOS), improving patient quality of life and satisfaction, reducing postoperative complications, organizing and standardizing care, and minimizing delays in care and resource waste.1–3 However, the implementation of actual practice change among provider teams can prove challenging. In order to accomplish practice change, medical providers have to both de-implement, e.g., “unlearn,” the old practice as well as to adapt the new practice.4 Practice change in surgery can be particularly difficult as randomized controlled trials of surgical procedures are infrequent, and there is often a volume-based learning curve that can impact outcomes.5,6 While many studies have described patient outcomes after implementation of new clinical care pathways, there is a paucity of studies that detail the actual implementation and practice change process.2,5
Esophagectomy is the surgical cornerstone of multi-modal therapy for esophageal carcinoma, yet this operation is associated with high morbidity.7,8 Leak from the esophagogastric anastomosis is one of the most dreaded complications.9,10 After careful evaluation of clinical outcomes at our institution, we saw the opportunity to improve patient outcomes by implementing a revised evidence-based clinical care pathway for patients undergoing esophagectomy. This pathway was primarily aimed at reduction of the anastomotic leak rate. This pathway entails the steps in caring for the patient perioperatively, with time- and criteria-based progression through the components of the pathway. The pathway also involves a major surgical change from the conventional one stage esophagectomy to a staged procedure where gastric preconditioning is followed by esophagectomy. A diagnostic laparoscopy with gastric ischemic pre-conditioning involves partial devascularization of the proximal stomach by division of the short gastric vessels and the left gastric artery and vein to and from the fundus, carefully preserving the right gastroepiploic arcade, intended to improve perfusion of the gastric conduit and esophagogastric anastomosis and consequently reduce anastomotic leakage.11,12
In order to assess the success of the implementation process, we interviewed providers to understand their perspectives on the roll-out of a new clinical pathway. Interviews were analyzed using the theoretical domains framework (TDF), a widely used and validated theoretical framework for understanding facilitators and barriers associated with the implementation of evidence-based practices.13 The purpose of this study was to describe the process of implementing a surgical practice change at our institution centered on a clinical care pathway for esophagectomy. The results from this study illustrate the utility of TDF as a powerful tool to enable and improve the implementation of future practice change and interventions.
Methods
This study was approved by the Colorado Multiple Institutional Review Board (#21–3164).
Study Setting and Participants
The study was conducted at the University of Colorado Hospital, an academic medical center that serves as a tertiary care referral center for local hospitals as well as surrounding states. The esophagectomy pathway was developed to replace an existing esophagectomy pathway. It was created by a panel of surgeon experts at our institution who performed a collaborative review of the literature to develop an evidence-based clinical care pathway for patients undergoing esophagectomy. This revised pathway was discussed and approved by the thoracic, general, and surgical oncology attending surgeons. The major changes were as follows: addition of a diagnostic laparoscopy with gastric ischemic preconditioning procedure (4 weeks prior to esophagectomy) in all patients, routine placement of feeding jejunostomy tube (J-tube) during the esophagectomy in all patients, initiation of tube feeds on all patients postoperative day (POD) zero, allowance of comfort ice chips by mouth on POD 1 (250 cc), removal of nasogastric tube on POD 2 if output < 200 ml (previously recommended but left to attending preference), goal to remove all chest tubes by POD 4 to 7 if output < 200 ml total, and standardization of the oral diet regimen with staged advancement by postoperative day. This pathway was implemented on January 1, 2021. Quarterly review meetings were held to review patient outcomes data and to review compliance with the protocol components and to facilitate discussion about provider concerns with the pathway.
The study included providers and personnel involved in creating or implementing the new postoperative esophagectomy pathway. Providers came from both inpatient and outpatient settings, and included physicians, advanced practice providers, registered nurse, and dietitian (Table 1).Table 1 Provider roles and demographics
Provider role Sex Years in practice
Attending surgeon (thoracic) M 26
Attending surgeon (thoracic) M 10
Attending surgeon (thoracic) F 1
Attending surgeon (surgical oncology) F 2
Attending surgeon (surgical oncology) M 21
Attending surgeon (surgical oncology) M 5
Advanced practice provider (thoracic inpatient) F 8
Advanced practice provider (thoracic inpatient) F 5
Advanced practice provider (thoracic inpatient) F 10
Advanced practice provider (ICU) F 12
Advanced practice provider (ICU) F 12
Advanced practice provider (surgical oncology inpatient) F 2
Advanced practice provider (surgical oncology inpatient) F 9
Advanced practice provider (oncology outpatient) F 10
Registered nurse (outpatient) F 24
Registered dietitian F 13
ICU intensive care unit
Pathway Compliance
At the time of the interviews, the new pathway had been implemented for 1 year. At this time, 37 patients had been evaluated for the pathway. Of these, 29 patients had gastric preconditioning with the pathway. The patients who did not have gastric preconditioning were for the following reasons: one scheduled but had not yet had preconditioning surgery, four patients decided they wanted no surgical treatment whatsoever, two had progressive disease and were no longer surgical candidates, and one had surgery at an outside hospital. Of the 29 patients, 22 had completed their esophagectomy at the 1-year mark. Compliance with pathway components in January 2022 (toward the end of the interviews) was the following (Table 2): 9% started J-tube feeds on POD zero, 68% had J-tube feeds advanced as intended (by 10 ml every 4 h), 54.5% had removal of the nasogastric tube by POD 2, the average day the last chest tube was removed was day 8.5 (goal day 4–7), 59% did not have a routine esophagram, and 68% were discharged on the appropriate diet for their postoperative day.Table 2 Compliance with pathway components
Pathway component Patients with pathway component compliance
J-Tube feeds (POD zero) 9.0% (n = 2/22)
J-Tube feed advancement as planned 68.2% (n = 15/22)
Nasogastric tube removal (POD 2) 54.5% (n = 12/22)
Chest tube removal (POD 3–7) Mean day 8.5
No routine esophagram ordered 59.1% (n = 13/22)
Appropriate discharge diet 68.2% (n = 15/22)
J-tube feeding jejunostomy tube
Recruitment
A purposive sample of providers were recruited via an email that was sent in September 2021 to all Department of Surgery faculty providers involved with the esophagectomy pathway asking for voluntary participation in a qualitative study analyzing the implementation of the new esophagectomy care pathway. One additional reminder email was sent in October 2021. Those who chose to partake in the study were sent a postcard consent form to review.
Study Design
The study team developed the interview guide to evaluate participant opinions and beliefs surrounding the implementation of a new esophagectomy clinical care pathway. Each volunteer participated in a single semi-structured interview lasting between 10 and 30 min. All interviews were conducted individually via Zoom by team members experienced in qualitative methodology (ALK, HJM). The interviews were audio recorded with participant permission for transcription. One member of the study team (EOM) transcribed verbatim the audio for the interviews. Transcriptions were not returned to participants for comment. The Consolidated Criteria for Reporting Qualitative Research (COREQ) guided the rigor of this study.14
Study Team
The study team was made up of the four of the authors: RAM, MD MPH is a thoracic surgeon and was the principal investigator. ALK, PhD MSPH is a health services researcher and was the principal qualitative methodologist. HJM, MD is a general surgery resident. EM, GED is a transcriptionist. All team members were trained in qualitative interviewing and analysis by ALK. All interviews were conducted by the first author (HJM). The first author utilized field notes throughout the interviews.
Data Analysis
The 10–30-min individual interviews were audio recorded and transcribed verbatim by the transcriptionist. The analyses were conducted by a qualitative methodologist, ALK, and the first author trained in such analysis, HJM. The study team performed an inductive and deductive team-based analytical approach, drawing primarily on matrix analysis8 , and the data was organized with Microsoft Excel version 16.59 (Microsoft Inc, Redmond, WA). A matrix analysis is a tabular format that arranges data for easy viewing in one place, permits detailed analysis, and sets the stage for later cross-case analysis.8 Matrix analyses provide a visual format for showing and telling simultaneously. Two members of the research team (HJD, ALK) first independently coded the initial interviews and then reviewed the documents together to reach a consensus, thus defining the initial codebook. During the coding process, the data were examined and organized into categories, domains, and themes. Emergent codes (those that are newly found in the subsequent interviews or further data collection from medical records) were added to the codebook until thematic saturation occurred, e.g., no new concepts were identified.15,16 All analyses and findings were integrated and documented with an audit trail.14
After matrix analysis completion, TDF was used to interpret the data. TDF is a framework that was developed by a multi-disciplinary group of psychological theorists, psychologists, and health service researchers. TDF includes 14 different domains and 84 constructs within these domains. The domains are knowledge, skills, social/professional role and identity, beliefs about capabilities, optimism, beliefs about consequences, reinforcement, intentions, goals, memory and attention and decision processes, environmental context and resources, social influences, emotion, and behavioral regulation.13 The purpose of these domains and constructs is to provide a framework for analyzing and explaining behavior (e.g., practice) change in a standardized fashion. Creation of the TDF from the matrix analysis was performed by ALK.
Results
A total of 16 out of 17 invited providers (94%) were recruited and completed qualitative interviews regarding their thoughts and opinions on the implementation of this pathway from September 2021 to February 2022 (Table 1). Interview participants included attending surgeons who perform esophagectomies (thoracic surgeons, general surgeons, and surgical oncologists) (n = 6), and inpatient advanced practice providers (APPs) from both thoracic surgery and surgical oncology (n = 5). We also interviewed intensive care unit (ICU) APPs from the cardiothoracic surgery ICU (n = 2), a registered dietitian (n = 1), and outpatient clinical staff from thoracic and surgical oncology clinics (n = 2).
TDF Domains
Interviews were used to identify facilitators and barriers to adoption and implementation. The main TDF domains that were relevant to implementation of the clinical pathway in this study were knowledge, beliefs about consequences, environmental context and resources, and social influences (Table 3). These domains will be discussed within the context of facilitators and barriers to pathway implementation (Fig. 1).Table 3 Examples of major TDF domains and associated quotes from participants (barriers to implementation are in bolded text, and facilitators of implementation are in italicized text)
TDF domain [Constructs] Sample quote
Knowledge
[– Knowledge of esophagectomy care/scientific rationale
– Knowledge of pathway components]
“I’m not really sure why that’s [discharge on incorrect diet] been happening, possibly just confusion about what is supposed to happen.” (#4)
“I think that probably what would be the most effective at enabling the protocol to be successful, would be to… review that data… if we get everyone in a room and here’s the data… To say… you’re either doing this or not doing this without a clear reason. Other than that you feel like it” (#13)
“I was initially anxious about it [no routine esophagram] but the more data I’ve looked… it has shown that esophagrams can be erroneously normal.” (#2)
“I do think the APPs are the ones who really understand the pathway and are consistent and don’t change all of the time.” (#5)
Beliefs about capabilities/self-confidence
[– Perceived competence
– Professional confidence]
“I think that it was a challenge for the person who views themselves as in charge and then ask them to produce data to support a decision that they’re making because we make tons of decisions without any data all the time” (#13)
“I think that we did a good job on the front end, creating patient education materials. I think those have been super helpful to patients” (#12)
Beliefs about consequences
[– Beliefs
– Characteristics of outcome expectancies]
“I think that’s [not starting tube feeds on POD0] just attending preference and historically what the attendings have done” (#1)
“I know two of our attendings are kind of like I don’t care what the data shows I need the esophagram” (#1)
“I just don’t think it’s a good idea, and the attendings don’t like it. No one seems comfortable with it” (#5)
“I don’t always look at something that someone publishes and automatically say, ‘Oh, you can do this. Oh, great. And now we'll just dive in.’ If it doesn’t make sense to me, I’m less accepting of doing it. I don’t believe everybody” (#6)
“I really think that it’s really a good thing. I think the patients are more prepared when they go to the MIE” (#9)”
“And actually, patients are way more accepting than I thought they would be of having another procedure” (#12)
“Yeah, so I would say, I’m very encouraged. At least just subjectively looking at the quality of the stomach and the conduits when we go to do the definitive operation, looking much healthier, just by the eyeball test” (#3)
“My impression, anecdotally, is that gastric preconditioning has made a big difference” (#6)
Environmental context and resources
[– Resources
– Organization culture/climate]
“Especially with the ICU, I know that they’ve had such turnover with their staffing, that getting on the same page with this new pathway has taken a little bit longer” (#1)
“I think that’s the main barrier and one of the challenges I see is the high rotation of trainees through the teams” (#2)
“But there’s been PA turnover as well” (#3)
“Nurses know there’s a pathway and want to stick to it, but it ultimately still must be verified by the surgical team. ICU nurses are for all for using it” (#14, #15)
“I think that communication has to be improved. I just need to know when the patient is going to be discharged” (#8)
“We have a printout in our office (of the pathway). The note is super helpful too, the template helps us say okay this patient’s here this is what we should be doing as a little reminder” (#1)
“The PAs are the ones that really keep track of all this stuff [pathway components]… even if a resident knows about the pathway, they may not remember it and it’s just hard for them to implement something that they don’t know much about” (#13)
“I mean, all of them end up on the sixth floor, which I think is good. They’re [nurses] pretty good at keeping track… on other floors, they don’t really keep track” (#16)
Social influences
[– Social norms
– Group norms
– Intergroup conflict]
“It feels like it’s hard to follow the pathway for us, because you still have to ask the surgical team permission to do things on the pathway so even if we want to stick to the pathway it’s not our choice” (#14, #15)
“In non-procedural fields it usually will go when you try something as a part of a study where you look for a difference better than the standard of care, but within surgical technique, it’s almost never that way… You just start trying it out, you see if you like it, if you kind of like it, then you do it a little more, than you publish a retrospective paper about how you did it, and then if somebody feels like it then they’ll compare, or it’ll just become your new standard practice” (#13)
“We’re just trying to kind of toe that line of sticking with the pathway but also making sure the attendings are happy” (#1)
“Trying to get all the attendings on board with a consistent pathway. That’s been a problem” (#3)
“It should be patient by patient and surgeon to surgeon… I think because I have been trained that way, I will be doing that for a while. It makes it even more complicated that we have two teams operating with opinions” (#6)
“I think we [surgical oncology and thoracic surgery] both give insight into things and talk about things… we do a good job of being collaborative about it (#16)
“Discussing this with other providers and other institutions and finding out what they do and what they found, it has shown that esophagrams can be erroneously normal” (#2)
Fig. 1 Barriers to pathway implementation, categorized by the four major TDF domains
Facilitators
Overall, participants felt cautiously optimistic about the efficacy of the pathway. Five participants reported an overall mixed opinion and 10 shared a positive opinion of the pathway. A major reason for a positive opinion was an impression of favorable outcomes for patients with the pathway: “I actually like it. I think overall we’ve seen less complications and patients seem to go home quicker and I think just tend to be doing a little bit better” (#11). TDF domains: beliefs about consequences.
Supportive Data
Providers acknowledged the critical role of supporting data for the pathway, mainly evidence behind pathway interventions: “I think that probably what would be the most effective at enabling the protocol to be successful, would be to… review that data… if we get everyone in a room and here’s the data… To say… you’re either doing this or not doing this without a clear reason” (#13).
For providers who were reluctant with the initial roll-out of the pathway, data was a main contributor for altering their initial opinion: “I was initially anxious about it [no routine esophagram] but the more data I've looked… it has shown that esophagrams can be erroneously normal” (#2). Provider opinions were also changed in a positive direction by favorable preliminary patient outcomes data. One provider who reported an initial negative opinion of the pathway changed their opinion based on patient outcomes: “My impression, anecdotally, is that gastric preconditioning has made a big difference… it seems to me that we’re dealing with fewer leaks than we did a year ago” (#6). Another provider shared that patient outcomes supported their belief in the pathway: “At least just subjectively looking at the quality of the stomach and the conduits when we go to do the definitive operation, looking much healthier, just by the eyeball test” (#3). TDF domains: knowledge, social influences, emotion, beliefs about consequences.
Perception of Patient Opinion
Overall, providers perceived patient opinion of the new pathway to be positive, which served as a critical support for implementation of the pathway. The addition of routine feeding J-tube placement allowing early initiation of tube feeds was felt to be beneficial for patients: “Patients seem overall way more prepared to be without food and to feel better with the tube feeds” (#4), and: “I think the feeding tube helps them. It’s reassuring, you know, if you’re not eating by mouth, you have a feeding tube” (#8). Improved preoperative patient education was also felt to have a positive impact with patients being perceived as more prepared for the procedure: “They also show the J-tube and what the J-tube looks like, and then the flush and all that other stuff. And also with the education part, they feel better about it because we give them the whole packet [explains post-operative care] and it’s posted out day one to seven… and week by week” (#9). Patients seemed to be more accepting of a long hospital stay postoperatively, accepting of the plan for tube feeds with delayed oral nutrition, and to not have significant resistance to the additional gastric preconditioning procedure. TDF domains: knowledge, beliefs about consequences, social influences. “So far, I have not had any patient resist or reject the idea of doing gastric preconditioning… there’s the well, here’s what we’re doing to try to prevent these complications of leaks, so that has added significant time to the discussion” (#3).
Resources
Consistency of informed staff provided immense support for the pathway. APPs were noted to be essential advocates, especially given the high turnover of housestaff, who rotate through teams on a monthly to semimonthly basis: “I do think the APPs are the ones who really understand the pathway and are consistent and don’t change all of the time” (#5). Another important resource was the development of electronic health record (EHR) tools to assist with implementation. These include an EHR order set for the pathway and a templated note to be used each day detailing the steps of the pathway. TDF domains: knowledge, environmental context and resources.
Barriers
Practice Patterns and Emotional Influence
A commonly cited barrier was friction between the new pathway and previous provider practice patterns. Discordance with practice patterns, at the attending surgeon level specifically, was cited as a barrier to several specific pathway components, including J-tube feeds starting on POD zero (n = 6), early chest tube removal prior to beginning oral intake (n = 4), and not obtaining a routine esophagram postoperatively (n = 5). While practice patterns were a major barrier, the ability for practice to slowly shift was also acknowledged: “I’m becoming more and more comfortable with that. Again, it’s my slow acceptance of things that make sense on paper, but I haven’t done clinically, and so these barriers are more emotional than intellectual” (#2).
Prior clinical experiences leading to disagreement with certain aspects of the pathway was also a barrier: “Attendings again have their practice patterns they’re used to and have been doing for a long time. However, I do think the attendings are starting to change and getting more used to it” (#4).
Providers reported hesitancy starting J-tube feeds on POD zero due to the risk of conduit distention (n = 4). In addition, not obtaining a routine esophagram precludes assessment of conduit emptying function (n = 2). The combination of early nasogastric tube removal and allowance of ice chips with no gastric emptying procedure was felt to be an issue among two attendings. TDF domains: knowledge, emotion, social influences, beliefs about consequences.
Pathway Education
Some elements of the pathway were not adhered to simply due to poor awareness of the pathway. This issue is compounded by new providers (e.g., housestaff): “The main barrier and one of the challenges I see is the high rotation of trainees through the teams, it seems like the deviation that I see early on is not necessarily mindful but it’s people not being aware of the pathway and adhering to it” (#2). Through interviews, it was also revealed that the order for a routine esophagram study was still pre-checked in the order set, causing it to be unintentionally performed in some patients. Also, it was found that outpatient and inpatient clinical staff were often not communicating clearly about a patient’s discharge diet, causing some difficulty with clinic follow-up. TDF domains: knowledge, environmental context and resources.
Patient Disagreement
Overall, while most providers indicated that patients seemed accepting of the pathway and gastric preconditioning, one provider did mention patients pushing back against a second operation: “Let’s say somebody comes in sees us and he’s already looked into Google… nobody talks about preconditioning operation… this is hard for them to understand why we’re doing this two-part operation” (#10). TDF domains: beliefs about consequences, social influences.
Discussion
This study presents a qualitative assessment using matrix and TDF analysis of the facilitators and barriers to the implementation of a new esophagectomy clinical care pathway. The main TDF domains that were salient to implementation were knowledge, beliefs about consequences, environmental context and resources, and social influences. Facilitators of pathway implementation included positive preliminary patient outcomes data, supporting evidence for pathway interventions, perception of patient satisfaction with the pathway, and resources such as consistent provider staffing and EHR tools. The main reported barriers were discordance with prior practice patterns, particularly at the attending level, disagreement with pathway components, and staffing turnover combined with lack of knowledge and education about the pathway components.
The salient TDF domains in this study are in agreement with a recent scoping review of surgical practice change, in which the authors similarly identified environmental context and resources, social influence, knowledge, and beliefs about consequences to be critical for practice change.5 The authors postulated that in a setting of supportive infrastructure, “It is individual knowledge, technical skill, and personal beliefs that encourage initiation and sustainment of practice change”.5 Emotion was an uncommonly cited barrier and facilitator due to difficulty in measuring this factor, but nevertheless was believed to play a major role in practice change. Provider aversion has been shown to be the strongest barrier in the implementation of enhanced recovery after surgery (ERAS) protocols as well.17 Our study provides evidence that emotion and personal beliefs have a role in surgical practice change (e.g., in providers not feeling “comfortable” with the changes) and demonstrates that over time these may slowly dissipate as barriers.
Prior studies have postulated models for change, which essentially involve several phases of readiness for change from acknowledging the possibility of change, preparation for change, and finally acting on change.4,18,19 It was clear in our study that although providers discussed and agreed upon the clinical care pathway, providers fell on a spectrum of readiness for change. Our study provides a unique qualitative assessment of the opinions of attending interviewees who, despite being involved with development of the pathway, experienced hesitancy to change their practice as a barrier to implementation. This is not unexpected, as even published evidence-based guidelines are not always followed after dissemination, indicating that implementation requires a more defined approach to be successful.20–22 Movement along the spectrum of change was seen in this study as a result of positive preliminary patient outcomes data shared with the provider group at quarterly meetings. Other studies have shown that evidence behind pathway interventions is critical for successful implementation at the provider level23 , and that providing information on pathway compliance throughout the rollout of a clinical pathway improves implementation.24 However, the significance of providing early patient outcomes data on the provider implementation process has not been previously shown, and it was found to be a facilitator of implementation in our study. Also, we found that improved comfort with interventions appeared to occur as providers gained experience with the pathway over time. On pathway rollout, we planned the quarterly provider meetings to mitigate hesitancy regarding pathway rollout and to create a mechanism for critical discussion of any necessary revisions to the pathway.
While ongoing discussion and staff education can facilitate compliance with clinical pathways25 , we found that staffing turnover was a barrier to ongoing pathway education. Interviewees mentioned that both expected staffing changes such as rotating housestaff and unexpected changes such as higher than typical registered nurse and advanced practice provider (APP) turnover, likely related to COVID-19 changes, made pathway implementation challenging. The presence of consistent staff to educate new providers and ensure pathway compliance was essential to addressing staffing turnover as a barrier. However, in contrast to other studies where education consistency is achieved at the surgeon level,26 our study found this role was more often fulfilled by advanced practice providers. It is likely that the pathway will act as a standardized guide and eventually help to mitigate staffing turnover as a barrier to care of esophagectomy patients, but in the early implementation phase, turnover was a challenge.
The potential limitations of this study include (1) the small sample size; (2) the study which was performed at a single institution; and (3) possible attribution bias, which is interpretation based on judgments and assumptions about behaviors (although as the data analyst was not a surgeon, objectivity was addressed). (4) We acknowledge that the interview results may have been affected by respondent bias given that we interviewed providers who were invested in the success of the postoperative pathway. Finally, interview questions were limited by our selection of determinants and therefore may not have captured all relevant factors influencing implementation. We also did not include housestaff in interviews, but it would be interesting to include this group in future implementation studies. Although the sample size was small, we believe the diversity of provider role, immersion of those who directly participated in the process, and transferability to academic surgical practice are strengths.
Conclusion
Our study illustrates the facilitators and barriers to a surgical practice change of implementing a new clinical care pathway for esophagectomy patients. We found factors such as education on pathway components, consistent provider staffing, EHR order set and note templates, and presence of supporting evidence and data to be important facilitators. However, even in the presence of these facilitators, there were barriers that stymied implementation, namely, surgical attending prior practice patterns and comfort with new interventions. There was evidence that with additional time for adjustment combined with demonstration of positive patient outcomes, providers became more open to change. Implementation of evidence-based interventions requires time and persistence to be successful.
Data Availability
The data generated/analyzed during the current study are available in anonymized format from the corresponding author on reasonable request.
Declarations
Conflict of Interest
The authors declare no competing interests.
Meeting Information: Presented as a quickshot oral presentation at American College of Surgeons Annual Meeting October 2022, San Diego, CA, USA
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Mhawej R, Harmych BM, Houlton JJ, Tabangin ME, Meinzen-Derr J, Patil YJ. The impact of a post-operative clinical care pathway on head and neck microvascular free tissue transfer outcomes. J Laryngol Otol. 2020:1–9.
2. Rotter T, Kinsman L, James EL, Machotta A, Gothe H, Willis J, et al. Clinical pathways: effects on professional practice, patient outcomes, length of stay and hospital costs. Cochrane Database Syst Rev. 2010.
3. Coffey RJ Richards JS Remmert CS LeRoy SS Schoville RR Baldwin PJ An introduction to critical paths Qual Manag Health Care. 2005 14 1 46 55 10.1097/00019514-200501000-00006 15739581
4. Gupta DM, Boland RJ, Aron DC. The physician’s experience of changing clinical practice: a struggle to unlearn. Implementation Science. 2017;12(1).
5. Arroyo NA Gessert T Hitchcock M Tao M Smith CD Greenberg C What promotes surgeon practice change? A scoping review of innovation adoption in surgical practice Ann Surg. 2021 273 3 474 482 10.1097/SLA.0000000000004355 33055590
6. Broekman ML Carrière ME Bredenoord AL Surgical innovation: the ethical agenda: a systematic review Medicine (Baltimore). 2016 95 25 e3790 10.1097/MD.0000000000003790 27336866
7. Coupland VH Lagergren J Lüchtenborg M Jack RH Allum W Holmberg L Hospital volume, proportion resected and mortality from oesophageal and gastric cancer: a population-based study in England, 2004–2008 Gut. 2013 62 7 961 966 10.1136/gutjnl-2012-303008 23086798
8. Hanna GB Boshier PR Knaggs A Goldin R Sasako M Improving outcomes after gastroesophageal cancer resection: can Japanese results be reproduced in Western centers? Arch Surg. 2012 147 8 738 745 10.1001/archsurg.2012.983 22911070
9. Schuchert MJ, Abbas G, Nason KS, Pennathur A, Awais O, Santana M, et al. Impact of anastomotic leak on outcomes after transhiatal esophagectomy. Surgery. 2010;148(4):831–8; discussion 8–40.
10. Kassis ES Kosinski AS Ross P Jr Koppes KE Donahue JM Daniel VC Predictors of anastomotic leak after esophagectomy: an analysis of the Society of Thoracic Surgeons General Thoracic Database Ann Thorac Surg. 2013 96 6 1919 1926 10.1016/j.athoracsur.2013.07.119 24075499
11. Heger P Blank S Diener MK Ulrich A Schmidt T Büchler MW Gastric preconditioning in advance of esophageal resection–systematic review and meta-analysis J Gastrointest Surg. 2017 21 9 1523 1532 10.1007/s11605-017-3416-z 28439770
12. Michalinos A, Antoniou SA, Ntourakis D, Schizas D, Ekmektzoglou K, Angouridis A, et al. Gastric ischemic preconditioning may reduce the incidence and severity of anastomotic leakage after οesophagectomy: a systematic review and meta-analysis. Dis Esophagus. 2020;33(10).
13. Cane J O’Connor D Michie S Validation of the theoretical domains framework for use in behaviour change and implementation research Implementation Science. 2012 7 1 37 10.1186/1748-5908-7-37 22530986
14. Tong A Sainsbury P Craig J Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups Int J Qual Health Care. 2007 19 6 349 357 10.1093/intqhc/mzm042 17872937
15. Patton M. Qualitative research and evaluation methods, 3rd edition. Sage Publications Ltd. 2002.
16. Creswell J. Qualitative inquiry and research design, 2nd edition. Sage Publications Ltd. 2007;Thousand Oaks, Ca.
17. Page AJ Gani F Crowley KT Lee KHK Grant MC Zavadsky TL Patient outcomes and provider perceptions following implementation of a standardized perioperative care pathway for open liver resection Br J Surg. 2016 103 5 564 571 10.1002/bjs.10087 26859713
18. Manchester J Gray-Miceli DL Metcalf JA Paolini CA Napier AH Coogle CL Facilitating Lewin's change model with collaborative evaluation in promoting evidence based practices of health professionals Eval Program Plann. 2014 47 82 90 10.1016/j.evalprogplan.2014.08.007 25192609
19. Cabana MD, Rand CS, Powe NR, Wu AW, Wilson MH, Abboud P-AC, et al. Why don't physicians follow clinical practice guidelines? JAMA. 1999;282(15):1458.
20. Grol R. Successes and failures in the implementation of evidence-based guidelines for clinical practice. Med Care. 2001;39(8 Suppl 2):Ii46–54.
21. Grol R Grimshaw J From best evidence to best practice: effective implementation of change in patients' care Lancet. 2003 362 9391 1225 1230 10.1016/S0140-6736(03)14546-1 14568747
22. Greenberg CC Lipsitz SR Neville B In H Hevelone N Porter SA Receipt of appropriate surgical care for Medicare beneficiaries with cancer Arch Surg. 2011 146 10 1128 1134 10.1001/archsurg.2011.141 21690439
23. Pearsall EA Meghji Z Pitzul KB Aarts MA McKenzie M McLeod RS A qualitative study to understand the barriers and enablers in implementing an enhanced recovery after surgery program Ann Surg. 2015 261 1 92 96 10.1097/SLA.0000000000000604 24646564
24. McLeod RS Aarts MA Chung F Eskicioglu C Forbes SS Conn LG Development of an enhanced recovery after surgery guideline and implementation strategy based on the knowledge-to-action cycle Ann Surg. 2015 262 6 1016 1025 10.1097/SLA.0000000000001067 25692358
25. Lyon A Solomon MJ Harrison JD A qualitative study assessing the barriers to implementation of enhanced recovery after surgery World J Surg. 2014 38 6 1374 1380 10.1007/s00268-013-2441-7 24385194
26. Kahokehr A Sammour T Zargar-Shoshtari K Thompson L Hill AG Implementation of ERAS and how to overcome the barriers Int J Surg. 2009 7 1 16 19 10.1016/j.ijsu.2008.11.004 19110478
| 36443554 | PMC9707093 | NO-CC CODE | 2022-12-01 23:20:20 | no | J Gastrointest Surg. 2022 Nov 28;:1-9 | utf-8 | J Gastrointest Surg | 2,022 | 10.1007/s11605-022-05537-0 | oa_other |
==== Front
Fish Sci
Fish Sci
Fisheries Science
0919-9268
1444-2906
Springer Japan Tokyo
1639
10.1007/s12562-022-01639-5
Original Article
Antioxidant and protective effects of a peptide (VTAL) derived from simulated gastrointestinal digestion of protein hydrolysates of Magallana gigas against acetaminophen-induced HepG2 cells
Ulagesan Selvakumari [email protected]
1
Park Su-Jin [email protected]
2
Nam Taek-Jeong [email protected]
3
http://orcid.org/0000-0002-0091-7893
Choi Youn-Hee [email protected]
123
1 grid.412576.3 0000 0001 0719 8994 Division of Fisheries Life Sciences, Pukyong National University, Nam-gu, Busan, 48513 Republic of Korea
2 grid.412576.3 0000 0001 0719 8994 Department of Fisheries Biology, Pukyong National University, Nam-gu, Busan, 48513 Republic of Korea
3 grid.412576.3 0000 0001 0719 8994 Institute of Fisheries Sciences, Pukyong National University, Gijang-gun, Busan, 46041 Republic of Korea
29 11 2022
111
15 6 2022
26 8 2022
© The Author(s), under exclusive licence to Japanese Society of Fisheries Science 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Oxidative stress is an automatic mechanism responsible for the commencement and continuance of liver injury. In this study, an antioxidative peptide Val-Thr-Ala-Leu (VTAL) was purified from simulated gastrointestinal digestion of protein hydrolysates of the triploid oyster Magallana gigas. Significant antioxidant activity was identified, as well as a protective effect against acetaminophen (APAP)-induced human liver cancer (HepG2) cells. The results suggested that the antioxidant activity improved in a dose-dependent manner. The highest cell viability (88.105 ± 3.62%) was observed in 15 mM APAP-induced cells when treated with 25 μg/mL M. gigas peptide [M.g (pep)]. The peptide sequences include hydrophobic amino acids, which could be responsible for its chemoprotective and antioxidant activities. Treatment with M.g (pep) significantly promoted the proliferation of HepG2 cells, thus protecting them against APAP and imbuing them with significant antioxidant capacity. M.g (pep) could be beneficial for treating drug-induced oxidative stress and liver damage. Additionally, M.g (pep) could serve as an alternative to synthetic antioxidant drugs.
Keywords
Magallana gigas
Enzymatic digestion
Bioactive peptide
Antioxidant activity
Chemoprotective effect
Acetaminophen
Simulated gastrointestinal digestion
http://dx.doi.org/10.13039/501100002701 Ministry of Education NRF- 2020R1F1A1074614 Choi Youn-Hee
==== Body
pmcIntroduction
Oysters are highly nutritious marine organisms with medicinal value. They have a high protein, active polysaccharide, taurine, vitamin, and mineral content, and are also low in fat (Guo et al. 2020). Bioactive materials have been extracted from various oyster species: oyster protein can be defragmented as multiple peptides with high bioactivity. Oyster protein hydrolysates (OPHs) and peptides are valuable owing to their stability and diverse biological activities (Guo et al. 2020; Xie et al. 2018). Many studies have described the bioactive properties of oyster peptides (OPs), including their antioxidant, antitumor, antimelanogenic, immunomodulatory, antifatigue, antimicrobial, antithrombotic, antiviral, antiwrinkle, antihypertensive, antiinflammatory, antifungal, anticancer, anticoagulant, and osteogenic effects (Ulagesan et al. 2022; Asha et al. 2016; Wang et al. 2014; Wang et al. 2010; Gueguen et al. 2006; Liu et al. 2008; Seo et al. 2013; Zeng et al. 2008; Hwang et al. 2012; Qian et al. 2020; Liu et al. 2007; Cheong et al. 2013; Han et al. 2019; Kim et al. 2015; Hao et al. 2013; Miao et al. 2018; Cheng et al. 2018; Cheng et al. 2021; Chen et al. 2019a, b A; Chen et al. 2019a, b B; Ngo et al. 2012; Bharathiraja et al. 2017). Oyster peptides may also enhance spatial learning and memory, acetylcholinesterase activity, and sexual function, and they also serve as angiotensin-converting enzyme (ACE) inhibitors (Wang et al. 2020; Zhang et al. 2021; Hao et al. 2022).
Acetaminophen (APAP) is a widely available antipyretic and painkiller medicine that has been used extensively since 1955. It is considered safe at therapeutic levels, but overdose or therapeutic misuse may cause hepatotoxicity and liver damage. In the last few decades, there have been concerns regarding the role of APAP in acute liver failure (ALF) in adults, with APAP-related ALF accounting for about 30–50% of cases worldwide (Bunchorntavakul and Reddy, 2013). In the USA, APAP hepatotoxicity was reported as the cause of 50% of all cases of overdose-related ALF and 20% of liver transplants. Toxic ingestion of APAP induces hepatic failure at concentrations above 150 mg/kg, but even minor doses may result in acute liver injury or failure (Yoon et al. 2016). During the COVID-19 epidemic, infected patients were widely treated with antipyretic agents that contained APAP (Feng et al. 2020). Misuse of these medicines could lead to liver injury (Zhang et al. 2022). Additionally, an APAP overdose will increase oxidative stress and levels of reactive oxygen species (ROS).
ROS are produced in the body by several endogenous systems (Lobo et al. 2010). The formation of free radicals, such as superoxide anions (O2−) and hydroxyl radicals (·OH), are inevitable in aerobic organisms during respiration, and play a major role in many diseases. These radicals are highly unstable and so they react with other groups and elements in the body and cause cell or tissue injury (Kim et al. 2007; Bharathiraja et al. 2017). Excessive amounts of these free radicals induce oxidative stress, which may damage cells and lead to chronic diseases such as hepatitis, as well as alcoholic and nonalcoholic fatty liver diseases. Toxins, drugs, and xenobiotics are metabolized in the liver, while liver cells balance ROS production. Generally, the metabolic functions of the liver, and its relationship to the gastrointestinal tract, make it susceptible to these toxins (Cichoz-Lach and Michalak 2014; Bharathiraja et al. 2018; Santha Moorthy et al. 2017). Thus, antioxidant therapy with bioactive peptides could be useful to maintain the balance between oxidants and antioxidants during liver disease, as well as for protecting hepatocytes from oxidative stress.
Oysters are an abundant marine resource with a high protein content. Oyster proteins are digested by various proteases, producing low-molecular-weight protein hydrolysates and amino acids. Various enzymes are used to digest oyster proteins, and bioactive peptides have been purified from OPHs. Oyster peptides possess a wide range of bioactive properties that vary according to the receptors involved. Many studies have shown that OPs are derived from gastrointestinal enzymatic hydrolysates with high antioxidant activity (Qian et al. 2008, 2020; Nam et al. 2015). However, limited studies have been carried out to assess the chemoprotective effects of OPs. This study aims to investigate the ability of OPs derived from Magallana gigas to repair the damage caused by APAP-induced hepatotoxicity using HepG2 cells.
Materials and methods
Chemicals
The following enzymes and chemicals were used in this study: trypsin, pepsin, α-chymotrypsin, fluorescein (FL), 2,2′-azobis(2-methylpropionamidine) dihydrochloride (AAPH), ( ±)-6-hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid (Trolox), 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid (ABTS), and APAP. All of these analytical grade reagents and enzymes were purchased from Sigma-Aldrich (St. Louis, MO, USA). Reactions were performed in deionized water.
Sample collection and preparation
The broodstock of cultivated Pacific oysters M. gigas was procured from the Myeongdeung Fisheries in Geoje, Gyeongsangnam-do, Korea. The shell length (SL), shell height (SH), and shell width (SW) were measured using a Vernier caliper. The total wet weight, soft tissue weight (STW), and total shell weight were calculated using an electronic balance. The oysters were dissected and the adductor muscle, digestive gland, gills, labial palps, mantle edge, and gonads were isolated. Each specimen was frozen in liquid nitrogen and stored at −80 °C.
Preparation of M. gigas protein hydrolysate
The protein was extracted from the gill tissue of M. gigas. Gill tissue (1 g) was homogenized with deionized water at a ratio of 1:3 v/w. The homogenate was modified to the optimum temperature and pH for the protein hydrolysate preparation.
Protein hydrolysates were prepared with three different enzymes as single hydrolysates and in combination, with an enzyme/substratum ratio of 1/100 (w/w) using buffers under optimum conditions.
Four different protein hydrolysates were prepared as follows:Pepsin (10 mM glycine buffer; pH 2 at 37 °C for 2 h at 100 rpm)
Trypsin (50 mM sodium phosphate buffer; pH 8 at 37 °C for 4 h at 100 rpm)
α-Chymotrypsin (50 mM sodium phosphate; pH 8 at 37 °C for 4 h at 100 rpm)
Pepsin + trypsin + α-chymotrypsin (digested under optimum pepsin conditions for 2 h; the pH was changed to 8, and trypsin and α-chymotrypsin were added and incubated for 4 h at 37 °C and 100 rpm)
After digestion, the enzymes were deactivated by heating for 10 min at 100 °C. The hydrolysates were centrifuged at 10,000 × g (4 °C) for 10 min, and the supernatants were stored for further study.
Antioxidant activity
Oxygen radical absorbance capacity (ORAC) assay
The ORAC assay of M. gigas protein hydrolysates was conducted using FL as the probe. The programmed ORAC assay was performed on a Synergy HTX multi-mode microplate reader (BioTek, Winooski, VT, USA) with fluorescence filters, at excitation and emission wavelengths of 485 nm and 535 nm, respectively; 96-well black-bottomed plates were used. The reaction was carried out at 37 °C, and was initiated by thermal decomposition of AAPH (Zulueta et al. 2009).
FL stock solution was prepared by dissolving of 44 mg of FL in 100 mL of phosphate-buffered saline (PBS;75 mM, pH 7.0), and was stored in a dark refrigerated condition. Additionally, the working solution (78 nM) was freshly prepared by diluting 0.167 mL of the stock solution with 25 mL phosphate buffer. The AAPH radical (221 mM) was freshly made by making up 600 mg of AAPH to 10 mL with PBS. The reference standard was a 20 μM Trolox solution prepared daily with PBS from a 1 mM stock standard solution, and kept in the freezer at –20 °C. Since the ORAC assay is very sensitive, samples should be diluted properly prior to the analysis to avoid interference. Each well was filled with 150 μL of FL (78 nM) and 25 μL of the sample, blank (PBS), or standard (Trolox, 20 μM). Then 25 μL of AAPH (221 mM) was added. Owing to the low conductivity of polypropylene plates, measurement deviations between wells can occur. To avoid this, the plates were warmed to 37 °C for 15 min before adding AAPH. Fluorescence was measured instantly after the addition of AAPH and measurements were then taken every 2 min up to 60 min. The measurements were performed in triplicate. The ORAC value, expressed as mM Trolox equivalent (mM TE/mg) was calculated by applying the following equation:ORACμM TE=CTrolox.AUCSample-AUCBlank.k/(AUCTrolox-AUCBlank)
where C Trolox is the concentration (μM) of Trolox (20 μM), k is the sample dilution factor, and AUC is the area under the fluorescence decay curve of the sample, blank, and Trolox.
Trolox equivalent antioxidant capacity (TEAC) assay
This study utilized the method described by Re et al. 1999 with minor modifications, to measure an individual sample’s ability to inhibit the ABTS radical (ABTS +) when compared with a reference antioxidant standard (Trolox). ABTS + is produced by a chemical reaction with potassium persulfate (K2S2O8). The radical was prepared by adding 25 mL of ABTS (7 mM) to 440 μL of K2S2O8 (140 mM) in the dark and allowing it to stand at room temperature for 12–16 h (the amount of time required for radical formation). The working solution was prepared by diluting a volume of the previous solution in PBS (pH 7.4) until the absorbance at k = 734 nm was 0.7 ± 0.02. Further dark incubation was performed for 5 min, followed by diluted ABTS working solution (200 μL + sample or PBS, or 10 μL of Trolox) added, and the absorbance at 734 nm (Zulueta et al. 2009) was determined using the Synergy HTX multi-mode microplate reader. In the range of 0–400 μM, a Trolox calibration curve was generated, and inhibition percentages for the samples were interpolated to determine the concentration in Trolox equivalents (mM TE/mg). Each measurement was performed in triplicate.
Molecular weight cutoff for separation
The M. gigas protein hydrolysate was separated by an ultra-filtration membrane (Vivaflow 200 Laboratory Cross Flow Cassette; Sartorius, Gottingen, Germany).
The protein hydrolysate was separated into four different molecular weight fractions: > 10 kDa
10–5 kDa
5–3 kDa
< 3 kDa
Analyses of antioxidant activity were performed on the four different fractions. Compared with the other three molecular weight fractions, < 3 kDa exhibited greater antioxidant activity. To further purify the activity-rich fraction < 3 kDa, anion exchange chromatography was performed.
Fast protein liquid chromatography (FPLC)
After further purification, the > 3 kDa hydrolysate was run on a HiPrep 16/10 diethylaminoethyl ion-exchange column (DEAE-FF) with the FPLC system (AKTA Prime Plus; GE Healthcare, Piscataway, NJ, USA). The freeze-dried 3 kDa protein hydrolysates (20 mg/mL) was dispersed in 20 mM Tris–HCl buffer (pH 9.5). Further, 5 mL of samples were loaded onto a HiPrep 16/10 DEAE FF ion-exchange column equilibrated in 20 Mm Tris–HCl buffer (pH 9.5), followed by an elution into a linear gradient of NaCl (0–1 M) at a flow rate of 2 mL/min. For each fraction (5 mL), the peaks were monitored at 214 nm, and fractions corresponding to those peaks were collected and pooled. The freeze-dried pooled fractions were subsequently analyzed for antioxidants and hydroxyl radical-scavenging activity. The fraction with the pooled fractions that exhibited highest antioxidant and hydroxyl radical-scavenging activities was additionally purified with high-performance liquid chromatography (HPLC) to find a pure peptide.
Preparative reverse phase (RP)-HPLC
Fractions with higher antioxidant were further purified by RP-HPLC using a Luna 5 µm C18(2) 100 Å, 250 × 10 mm LC column (Phenomenex., Torrance, CA, USA), attached to the Dionex Chromeleon 7.2 chromatography system (Thermo Fisher Scientific, Waltham, MA, USA.). Separation was performed with solution A (0.1% formic acid in deionized water) and solution B (0.1% formic acid in acetonitrile) eluted with a linear gradient of acetonitrile (0–5 min, 90% of solution A and 10% of solution B; 6–55 min, 20% of solution A and 80% of solution B; 56–60 min, 20% of solution A and 80% of solution B; 60–66 min, 90% of solution A and 10% of solution B) at a flow rate of 1 mL/min. Absorbance was monitored at 214 nm.
Determination of peptide sequences by liquid chromatography–tandem mass spectrometry (LC-MS/MS)
The amino acid sequence was confirmed by de novo sequencing of peptides derived from proteolytic substances or extracted through LC–MS/MS. A micro quadrupole time of flight (Q-TOF) III mass spectrometer (255,748; Bruker Daltonics, Hamburg, Germany) was used to perform the MS analysis. Filtered (Minisart syringe filter; Sartorius; pore size, 0.45 μm) samples (10 μL) were injected through a Poroshell 120 EC-C18 (2.1 × 100 mm, 2.7 μm; Agilent) LC column and separated using the UltiMate 3000 system (Thermo Fisher Scientific). The separation was attained at a flow rate of 0.2 mL/min by using a gradient elution program, starting with 98% solvent AA [H2O/FA = 100/0.1 (v/v)) and 2% solvent B (acetonitrile/FA = 100/0.1 (v/v)] for up to 2 min. This was gradually changed to 70% solvent A and 30% solvent B at 20 min. The gradient then changed rapidly to 5% solvent A and 95% solvent B from 20 to 21 min, which was then maintained up to 26 min. The program then rapidly changed to 98% solvent A and 2% solvent B at 27 min, which was maintained up to 35 min.
Further samples were subjected to the mass spectrometer via electrospray ionization (ESI +) via the subsequent MS parameters: capillary voltage, 3500 V; nebulizer flow, 0.8 L/min; dry gas flow rate, 5.5 L/min; dry temperature, 180 °C; transfer, funnel 1RF at 400 Vpp and funnel 2RF at 400 Vpp; ISCID energy, 0 eV; hexapole radiofrequency setting, 250; quadruple, ion energy of 5.0 eV; low mass of 300 m/z; collision cell collision energy, 7 eV; collision radiofrequency, 600; transfer time, 80 μs; and pre-pulse storage, 10 μs. The collected spectra were scanned from 50 to 2000 m/z. The normalized collision energy of 100 Vpp was used to produce the MS spectra by the collision-induced dissociation of the metabolite ions. Peptide de novo sequencing was performed to analyze the amino acid sequence based on the MS/MS spectra. A single peptide amino acid sequence was identified.
Cell culture
HepG2 human liver cancer cells (cat. no. HB-8065) were procured from the American Type Culture Collection (Manassas, VA, USA) with the culture conditions of 37 °C with 5% CO2 in minimum essential medium (MEM; Sigma-Aldrich) enhanced with 10% fetal bovine serum (FBS) (GenDEPOT, Katy, TX, USA) containing 100 U/mL penicillin and 100 mg/mL streptomycin. The medium was changed on alternative days.
Cell viability assay
Cell viability was assessed with the EZ-Cytox cell viability assay kit (cat. no. EZ-1000). Cells were seeded in 96‑well plates (2 × 104 cells/well in 100 μL medium) and allowed to attach for 24 h at 37 °C. The attached cells were then treated with M. gigas (pep) (1.25, 2.5, 5, or 25 µg/mL) and 15 mM APAP (A7085; Sigma‑Aldrich) in serum‑free MEM (SFM) for 18 h at 37 °C. After adding the Cytox solution to the cells, they were incubated at 37 °C for 1 h, and the absorbance was measured at a wavelength of 450 nm using a FilterMAX F5 microplate reader (Molecular Devices LLC, San Jose, CA, USA) (Kim et al. 2019).
Apoptosis assay
Apoptosis was evaluated using Muse annexin V and a dead cell assay kit (cat. no. MCH100105; BD Biosciences, Franklin Lakes, NJ, USA). The harvested cells were washed twice with PBS, and stained with FITC annexin V and propidium iodide for 15 min at room temperature. The apoptotic cell percentage was determined using annexin V and the Muse cell analyzer system (Merck Millipore, Burlington, MA, USA).
Statistical analysis
All the results are presented as the mean with the standard deviation of three independent experiments. The significant difference among the means was calculated by one‑ or two‑way ANOVA followed by Dunnett’s multiple comparison test using GraphPad Prism software (version 9/0; GraphPad Software Inc., San Diego, CA, USA). A p-value < 0.05 was considered to indicate a statistically significant difference.
Results
Protein extraction and purification
M. gigas (triploid oyster) was dissected and crude protein was extracted from the gill tissue.
Antioxidant activity of M. gigas protein hydrolysate
The oyster protein was digested with multiple proteases, either singly or in combination. The digested protein hydrolysates were analyzed to determine their antioxidant activities (ORAC and TEAC). The three enzymes (pepsin + trypsin + α-chymotrypsin) combined in the simulated gastrointestinal digestion (SGID) had higher antioxidant activity than the other protein hydrolysates and unhydrolyzed M. gigas protein (Fig. 1a and b). The ORAC and TEAC values in the SGID were 3437.22 ± 44.7 and 392.96 ± 9.3 mM TE/mg, respectively, and the corresponding values for unhydrolyzed M. gigas were 311.371 ± 27.3 and 117.28 ± 9.3 mM TE/mg.Fig. 1 Antioxidant activity of M. gigas protein and the different protein hydrolysates of M. gigas. a ORAC activity, b TEAC activity. Values are expressed as mean ± SD (n = 3). *denote significant differences (p < 0.05), while ns indicate no significant difference (p > 0.05)
Antioxidant activities of different molecular weight fractions
The gastrointestinal digested protein hydrolysate was further purified using different molecular weight cutoffs (> 10, 10–5, 5–3, and < 3 kDa). The lowest molecular weight fraction (< 3 kDa) had the highest antioxidant activity. The ORAC and TEAC values were 4,298.51 ± 82.69 and 5,44.91 ± 7.73 mM TE/mg, respectively (Fig. 2a and b).Fig. 2 The antioxidant activity of different molecular weight fractions of M. gigas protein hydrolysates. a ORAC activity, b TEAC activity. Values are expressed as mean ± SD (n = 3). * denote significant differences (p < 0.05), while ns indicate no significant difference (p > 0.05)
Purification and antioxidant activity of < 3 kDa molecular weight fraction
Anion exchange chromatography (FPLC) was used to purify the antioxidant-rich (3 kDa) molecular weight fraction (Fig. 3a). Gradient elution was performed using 1 M NaCl. Six peaks were observed and pooled, and then analyzed for antioxidant activity. Fraction 3 exhibited the maximum antioxidant activity compared with other fractions (Fig. 3b and c).Fig. 3 a Fast protein liquid chromatography profile of < 3 kDa peptide. A 5 mL sample was loaded onto a HiPrep 16/10 DEAE FF anion-exchange column. Separation was performed at 2 mL/min. Elution was monitored at 214 nm. Antioxidant activity of the fractions collected from FPLC. b The ORAC value of the fractions, c the ABTS activity of the fractions. Values are expressed as mean ± SD (n = 3). *denote significant differences (p < 0.05), while ns indicate no significant difference (p > 0.05)
Purification of the antioxidant-rich fraction 3
Antioxidant-rich fraction 3 was further purified by RP-HPLC, with two peaks analyzed separately for antioxidant activity (Fig. 4a) and compared with glutathione (GSH) (naturally occurring antioxidant). Peak two had the highest antioxidant activity (Fig. 4b and c). Peak two (p2) had the maximum antioxidant activity and was subsequently analyzed by LC-MS/MS; a single peak was observed (Fig. 5) and the peptide sequence was identified by de nova sequencing (Table 1).Fig. 4 a Results of RP-HPLC analysis using a Luna 5 µm C18 (2) 100 Å, 250 × 10 mm LC column attached to a Dionex system. Elution was performed with solution A (0.1% formic acid in deionized water) and solution B (0.1% formic acid in acetonitrile), eluted with a linear gradient of acetonitrile (0–80% in 0–66 min) at a flow rate of 1 mL/min. Absorbance was monitored at 214 nm. Antioxidant activity of peaks identified by RP-HPLC. b The ORAC value of the fractions, c the ABTS activity of the fractions are compared with the glutathione (GSH)
Fig. 5 The LC-MS/MS analysis of Peak 2. a MS spectrum of the peptide, b MS/MS sequencing
Table 1 The isolated peptide sequence
Charge m/z Sequence
1 403.25 VTAL
Analysis of M. gigas peptide [M.g (pep)] in APAP-induced HepG2 cell
Cell viability
The cell viability of M.g (pep) was analyzed relative to the APAP-induced toxicity in HepG2 liver cells. The APAP-treated groups had a 60% survival rate after an 18 h treatment. The cell viability of the APAP-treated group was 61.67 ± 2.32% that of the control (Fig. 6). Subsequent treatment with 15 Mm APAP and 1.25, 2.5, 5, and 25 μg/mL M.g (pep) was associated with significantly higher cell viability compared with the control group: 69.20 ± 3.66%, 79.78 ± 7.3%, 85.97 ± 2.5%, 88.10 ± 3.6%, respectively. These results suggested that the increased cell viability was evident in a dose-dependent manner.Fig. 6 Cell viability was measured by Cytox assays, and the results are presented as the percentage of surviving cells compared with the control (non‑treated group). Protective effects of M.g (pep) against APAP-induced liver damage involving HepG2 cells. a Cell viability of M.g (pep); b Protective effect of M.g (pep) against APAP c HepG2 cells were incubated with 15 mM APAP with or without various concentrations of PYP1– 4 for 18 h. FITC Annexin V flow cytometry was employed to determine the percentages of apoptotic and necrotic cells. Data are presented as the mean ± SD of three independent experiments and were analyzed using two‑way ANOVA. *Indicate the significance value p < 0.05 versus control group, #denotes the significance value p < 0.05 versus the 15 mM APAP group
Discussion
Oyster proteins generally contain various bioactive sequences, which can be released by enzyme hydrolyses to act as bioactive peptides. Thus, enzyme hydrolyses increase the nutritional properties of the oyster protein (Wang et al. 2010). The present study also elucidated the protective effects of a bioactive peptide derived from protein hydrolysates of the oyster Magallana gigas against acetaminophen-induced HepG2 cells. Oyster meat accounts for about half of the dry weight of an oyster (Linehan et al. 1999). The oyster gill is a complex ciliated organ with major roles in feeding, respiration, and excretion. Cilia of the gill are involved in the generation of strong water currents that pass through numerous branchial chambers to ensure gas exchange between organs and the surrounding medium, as well as the transport of food particles. Crude protein was isolated from the gill tissue of M. gigas. Protein hydrolysates digested with the combined SGID showed higher antioxidant activity. The antioxidant activity of gastrointestinally digested protein hydrolysates were significantly higher than other enzyme-digested protein hydrolysates and unhydrolyzed M. gigas protein. The antioxidant activities of protein hydrolysates mainly depend on the amino acid composition and sequence, as well as the peptide’s configuration and size (Chen et al. 1996). Generally, antioxidant activity is characterized by the enzymes when using the same protein substrates. Protein hydrolysates derived from different proteases had different antioxidant activities, suggesting that the enzymes were a major factor influencing the antioxidant properties of the protein hydrolysates. The results indicated that M. gigas undergoing SGID released more amino acids than other enzyme-digested protein hydrolysates. A previous study on SGID of mulberry leaf protein and its neutrase hydrolysates found similar results (Sun et al. 2021). Further, the antioxidant activity of different molecular weight fractions indicated the lower fraction (< 3 kDa) exhibited the maximum activity, demonstrated with ORAC and TEAC assays. The molecular size of a peptide is also a major factor influencing the absorption of food protein hydrolysates. Peptides with lower molecular weights were the main contributors to the antioxidant activity of protein hydrolysates (Garcia-Mora et al. 2015). These results are in accordance with those of previous studies reporting that low-molecular-weight peptides have greater antioxidant activity (Ren et al. 2008; Zhu et al. 2008; Cao et al. 2009). The FPLC purified > 3 kDa fraction with higher antioxidant activity (fraction 3) was subjected to RP-HPLC analysis and further separated into two peaks. The peak with higher activity (P2) was further purified by LC-MS/MS and the peptide sequence was identified as Val-Thr-Ala-Leu (VTAL). A strong correlation exists between the amino acid composition and the bioactivities of peptides (Mendis et al. 2005a, b A; Mendis et al. 2005a, b; Shen et al. 2010). Generally, the higher the content of hydrophobic amino acids (Pro, Tyr, Val, Ala, Leu, Ile, Phe, and Met) in the peptide sequence, the stronger the antioxidant activity; this may be due to the interaction with lipid-soluble free radicals and a delay in lipid peroxidation (Harnedy and FitzGerald, 2012; Bunchorntavakul and Reddy, 2013; Zou et al. 2016). The peptide sequence identified in the present study contained Val-Thr-Ala-Leu, which indicated that the peptide isolated from M. gigas was rich in amino acids implicated in antioxidant activity, and could therefore be a potential antioxidative peptide. Further, the protective effect of protein hydrolysate was tested with APAP-induced HepG2 cells and increased viability was observed in a dose-dependent manner. Park et al. 2014 also revealed the protective effects of enzymatic oyster hydrolysate against APAP-induced HepG-2 cell damage, and the results show that the oyster hydrolysate has potential as a health food or liver-protecting drug.
This study showed that M. gigas protein hydrolysate subjected to SGID had significantly more antioxidant activity than other enzyme-digested protein hydrolysates and unhydrolyzed proteins. Furthermore, the low-molecular-weight peptide fractions derived using molecular weight cutoffs exhibited significant antioxidant activity. However, the purified peptide (VTAL) also had remarkable antioxidant activity, among other positive effects (i.e., promotion of cell viability and reduction of oxidative stress) against APAP-induced liver damage involving HepG2 liver cells. The identified peptide sequence, which encompasses hydrophobic and aromatic amino acids, could promote antioxidant and chemoprotective activity. Further synthesis and in vivo studies of the peptide will be valuable for the food and pharmaceutical industries.
Acknowledgements
This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF- 2020R1F1A1074614). The English in this document has been checked by at least two professional editors, both native speakers of English. For a certificate, please see:http://www.textcheck.com/certificate/3vC5LX
Author contributions
Conceptualization, SU, TJN, and YHC; formal analysis, SU and SJP; investigation, SU and TJN; data curation, SU and SJP; writing—original draft preparation, SU; writing—review and editing, YHC and TJN; supervision, YHC and TJN; project administration, YHC; funding acquisition, YHC. All authors have read and agreed to the published version of the manuscript.
Funding
Ministry of Education, NRF- 2020R1F1A1074614, Youn Hee Choi.
Declarations
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Asha KK Remya Kumari KR Ashok Kumar K Sequence determination of an antioxidant peptide obtained by enzymatic hydrolysis of oyster Crassostrea madrasensis (Preston) Int J Pept Res Ther 2016 22 421 433 10.1007/s10989-016-9521-0
Bharathiraja S Manivasagan P Santha Moorthy M Photo-based PDT/PTT dual model killing and imaging of cancer cells using phycocyanin-polypyrrole nanoparticles Eur J Pharm Biopharm 2018 123 20 30 10.1016/j.ejpb.2017.11.007 29154833
Bharathiraja S Moorthy MS Manivasagan P Chlorin e6 conjugated silica nanoparticles for targeted and effective photodynamic therapy Photodiagnosis Photodyn Ther 2017 19 212 220 10.1016/j.pdpdt.2017.06.001 28583295
Bunchorntavakul C Reddy KR Acetaminophen-related hepatotoxicity Clin Liver Dis 2013 17 587 607 10.1016/j.cld.2013.07.005 24099020
Cao W Zhang C Hong P Ji H Optimising the free radical scavenging activity of shrimp protein hydrolysate produced with alcalase using response surface methodology Int J Food Sci Technol 2009 44 1602 1608 10.1111/j.1365-2621.2008.01901.x
Chen H Cheng S Fan F Identification and molecular mechanism of antithrombotic peptides from oyster proteins released in simulated gastro-intestinal digestion Food Funct 2019 10 5426 5435 10.1039/c9fo01433k 31402368
Chen H Muramoto K Yamauchi F Nokihara K Antioxidant activity of designed peptides based on the antioxidative peptide isolated from digests of a soybean protein J Agric Food Chem 1996 8561 2619 2623 10.1021/jf950833m
Chen H Xu Z Fan F Identification and mechanism evaluation of a novel osteogenesis promoting peptide from Tubulin Alpha-1C chain in Crassostrea gigas Food Chem 2019 272 751 757 10.1016/j.foodchem.2018.07.063 30309606
Cheng S Tu M Chen H Identification and inhibitory activity against α-thrombin of a novel anticoagulant peptide derived from oyster (Crassostrea gigas) protein Food Funct 2018 9 6391 6400 10.1039/c8fo01635f 30457135
Cheng S Tu M Liu H A novel heptapeptide derived from Crassostrea gigas shows anticoagulant activity by targeting for thrombin active domain Food Chem 2021 334 127507 10.1016/j.foodchem.2020.127507 32688180
Cheong SH Kim EK Hwang JW Kim YS Lee JS Moon SH Jeon BT Park PJ Purification of a novel peptide derived from a shellfish, Crassostrea gigas, and evaluation of its anticancer property J Agric Food Chem 2013 61 11442 11446 10.1021/jf4032553 24199654
Cichoż-lach H Michalak A Oxidative stress as a crucial factor in liver diseases World J Gastroenterol 2014 20 8082 8091 10.3748/wjg.v20.i25.8082 25009380
Feng G Zheng KI Yan Q COVID-19 and liver dysfunction: current insights and emergent therapeutic strategies J Clin Transl Hepatol 2020 8 1 18 24 10.14218/JCTH.2020.00018 32274342
Garcia-Mora P Peñas E Frias J High-pressure improves enzymatic proteolysis and the release of peptides with angiotensin I converting enzyme inhibitory and antioxidant activities from lentil proteins Food Chem 2015 171 224 232 10.1016/j.foodchem.2014.08.116 25308663
Gueguen Y Herpin A Aumelas A Characterization of a defensin from the oyster Crassostrea gigas: Recombinant production, folding, solution structure, antimicrobial activities, and gene expression J Biol Chem 2006 281 313 323 10.1074/jbc.M510850200 16246846
Guo Z Zhao F Chen H Heat treatments of peptides from oyster (Crassostrea gigas) and the impact on their digestibility and angiotensin I converting enzyme inhibitory activity Food Sci Biotechnol 2020 29 961 967 10.1007/s10068-020-00736-4 32582458
Han JH Bang JS Choi YJ Choung SY Anti-melanogenic effects of oyster hydrolysate in UVB-irradiated C57BL/6J mice and B16F10 melanoma cells via downregulation of cAMP signaling pathway J Ethnopharmacol 2019 229 137 144 10.1016/j.jep.2018.09.036 30273735
Hao G Cao W Hao J Zhang C In vitro antioxidant activity and in vivo anti-fatigue effects of oyster (Ostrea plicatula gmelin) peptides prepared using neutral proteinase Food Sci Technol Res 2013 19 4 623 631 10.3136/fstr.19.623
Hao L Wang X Cao Y A comprehensive review of oyster peptides: Preparation, characterisation and bioactivities Rev Aquac 2022 14 120 138 10.1111/raq.12588
Harnedy PA Fitzgerald RJ Bioactive peptides from marine processing waste and shellfish: A review J Funct Foods 2012 4 6 24 10.1016/j.jff.2011.09.001
Hwang JW Lee SJ Kim YS Purification and characterization of a novel peptide with inhibitory effects on colitis induced mice by dextran sulfate sodium from enzymatic hydrolysates of Crassostrea gigas Fish Shellfish Immunol 2012 33 993 999 10.1016/j.fsi.2012.08.017 22960100
Kim HA Park SH Lee SS Choi YJ Anti-wrinkle effects of enzymatic oyster hydrolysate and its fractions on human fibroblasts J Korean Soc Food Sci Nutr 2015 44 11 1645 1652 10.3746/jkfn.2015.44.11.1645
Kim IH Choi JW Nam TJ PYP1 - 4 peptide from Pyropia yezoensis protects against acetaminophen-induced hepatotoxicity in HepG2 cells Exp Ther Med 2019 19 2 849 860 10.3892/etm.2019.8304 32010245
Kim SY Je JY Kim SK Purification and characterization of antioxidant peptide from hoki (Johnius belengerii) frame protein by gastrointestinal digestion J Nutr Biochem 2007 18 31 38 10.1016/j.jnutbio.2006.02.006 16563720
Linehan LG O’Connor TP Burnell G Seasonal variation in the chemical composition and fatty acid profile of Pacific oysters (Crassostrea gigas) Food Chem 1999 64 211 214 10.1016/S0308-8146(98)00144-7
Liu Z Dong S Xu J Production of cysteine-rich antimicrobial peptide by digestion of oyster (Crassostrea gigas) with alcalase and bromelin Food Control 2008 19 231 235 10.1016/j.foodcont.2007.03.004
Liu Z Zeng M Dong S Effect of an antifungal peptide from oyster enzymatic hydrolysates for control of gray mold (Botrytis cinerea) on harvested strawberries Postharvest Biol Technol 2007 46 95 98 10.1016/j.postharvbio.2007.03.013
Lobo V Patil A Phatak A Chandra N Free radicals, antioxidants and functional foods: Impact on human health Pharmacogn Rev 2010 4 118 126 10.4103/0973-7847.70902 22228951
Mendis E Rajapakse N Byun H Kim S Investigation of jumbo squid (Dosidicus gigas) skin gelatin peptides for their in vitro antioxidant effects Life Sci 2005 77 2166 2178 10.1016/j.lfs.2005.03.016 15916780
Mendis E Rajapakse N Kim SK Antioxidant properties of a radical-scavenging peptide purified from enzymatically prepared fish skin gelatin hydrolysate J Agric Food Chem 2005 53 581 587 10.1021/jf048877v 15686405
Miao J Liao W Kang M Jia Y Wang Q Duan S Xiao S Cao Y Ji H Anti-fatigue and anti-oxidant activities of oyster (Ostrea rivularis) hydrolysate prepared by compound protease Food Funct 2018 9 12 6577 6585 10.1039/c8fo01879k 30488053
Nam BH Seo JK Lee MJ Functional analysis of Pacific oyster (Crassostrea gigas) β-thymosin: Focus on antimicrobial activity Fish Shellfish Immunol 2015 45 167 174 10.1016/j.fsi.2015.03.035 25842181
Ngo DH Vo TS Ngo DN Wijesekara I Kim SK Biological activities and potential health benefits of bioactive peptides derived from marine organisms Int J Biol Macromol 2012 51 378 383 10.1016/j.ijbiomac.2012.06.001 22683669
Park SH Moon SS Xie CL Choung SY Choi YJ Protective effects of enzymatic oyster hydrolysate on acetaminophen-induced hepg-2 cell damage J Korean Soc Food Sci Nutr 2014 43 1166 1173 10.3746/jkfn.2014.43.8.1166
Qian B Zhao X Yang Y Tian C Antioxidant and anti-inflammatory peptide fraction from oyster soft tissue by enzymatic hydrolysis Food Sci Nutr 2020 8 3947 3956 10.1002/fsn3.1710 32724655
Qian ZJ Jung WK Byun HG Kim SK Protective effect of an antioxidative peptide purified from gastrointestinal digests of oyster, Crassostrea gigas against free radical induced DNA damage Bioresour Technol 2008 99 3365 3371 10.1016/j.biortech.2007.08.018 17904358
Re R Pellegrini N Proteggente A Pannala A Yang M Rice-Evans C Antioxidant activity applying an improved abts radical cation decolorization assay Free Radic Biol Med 1999 26 1231 1237 10.1016/s0891-5849(98)00315-3 10381194
Ren J Zhao M Shi J Optimization of antioxidant peptide production from grass carp sarcoplasmic protein using response surface methodology Food Sci Technol 2008 41 1624 1632 10.1016/j.lwt.2007.11.005
Santha Moorthy M Subramanian B Panchanathan M Fucoidan-coated core-shell magnetic mesoporous silica nanoparticles for chemotherapy and magnetic hyperthermia-based thermal therapy applications New J Chem 2017 41 15334 15346 10.1039/c7nj03211k
Seo JK Lee MJ Nam BH Park NG CgMolluscidin, a novel dibasic residue repeat rich antimicrobial peptide, purified from the gill of the Pacific oyster, Crassostrea gigas Fish Shellfish Immunol 2013 35 480 488 10.1016/j.fsi.2013.05.010 23711469
Shen S Chahal B Majumder K You SJ Wu J Identification of novel antioxidative peptides derived from a thermolytic hydrolysate of ovotransferrin by LC-MS / MS J Agric Food Chem 2010 10.1021/jf101323y 21155571
Sun C Shan Y Tang X Effects of enzymatic hydrolysis on physicochemical property and antioxidant activity of mulberry (Morus atropurpurea Roxb) leaf protein Food Sci Nutr 2021 10.1002/fsn3.2474 35035930
Ulagesan S Krishnan S Nam T-J Choi Y-H A review of bioactive compounds in oyster shell and tissues Front Bioeng Biotechnol 2022 10.3389/fbioe.2022.913839 35733526
Wang Q Li W He Y Novel antioxidative peptides from the protein hydrolysate of oysters (Crassostrea talienwhanensis) Food Chem 2014 145 991 996 10.1016/j.foodchem.2013.08.099 24128574
Wang X Yu H Xing R Optimization of Oyster (Crassostrea talienwhanensis) protein hydrolysates using response surface methodology Molecules 2020 10.3390/molecules25122844 33396754
Wang YK He HL Wang GF Oyster (Crassostrea gigas) hydrolysates produced on a plant scale have antitumor activity and immunostimulating effects in BALB/c mice Mar Drugs 2010 8 255 268 10.3390/md8020255 20390104
Xie CL Kang SS Lu C Choi YJ Quantification of multifunctional dipeptide YA from oyster hydrolysate for quality control and efficacy evaluation Biomed Res Int 2018 10.1155/2018/8437379 30671456
Yoon E Babar A Choudhary M Review article acetaminophen-induced hepatotoxicity: a comprehensive update J Clin Transl Hepatol 2016 4 131 142 27350943
Zeng M Cui W Zhao Y Liu Z Dong S Guo Y Antiviral active peptide from oyster Chinese J Oceanol Limnol 2008 26 307 312 10.1007/s00343-008-0307-x
Zhang R Wang Q Yang J Impact of liver functions by repurposed drugs for COVID-19 treatment J Clin Transl Hepatol 2022 10 4 748 756 10.14218/JCTH.2021.00368 36062269
Zhang W Wei Y Cao X Enzymatic preparation of Crassostrea oyster peptides and their promoting effect on male hormone production J Ethnopharmacol 2021 264 113382 10.1016/j.jep.2020.113382 32918991
Zhu L Jie C Tang X Xiong YL Reducing, radical scavenging, and chelation properties of in vitro digests of alcalase-treated zein hydrolysate J Agric Food Chem 2008 10.1021/jf703697e 19007129
Zou T He T Li H The structure-activity relationship of the antioxidant peptides from natural proteins Molecules 2016 10.3390/molecules21010072 28035970
Zulueta A Esteve MJ Frígola A ORAC and TEAC assays comparison to measure the antioxidant capacity of food products Food Chem 2009 114 310 316 10.1016/j.foodchem.2008.09.033
| 36465482 | PMC9707094 | NO-CC CODE | 2022-12-01 23:20:20 | no | Fish Sci. 2022 Nov 29;:1-11 | utf-8 | Fish Sci | 2,022 | 10.1007/s12562-022-01639-5 | oa_other |
==== Front
J Mol Model
J Mol Model
Journal of Molecular Modeling
1610-2940
0948-5023
Springer Berlin Heidelberg Berlin/Heidelberg
5391
10.1007/s00894-022-05391-6
Original Paper
In silico evaluation of flavonoids as potential inhibitors of SARS-CoV-2 main nonstructural proteins (Nsps)—amentoflavone as a multitarget candidate
Portilla-Martínez Andrés 1
Ortiz-Flores Miguel 1
Hidalgo Isabel 2
Gonzalez-Ruiz Cristian 3
Meaney Eduardo 1
Ceballos Guillermo 1
Nájera Nayelli [email protected]
1
1 grid.418275.d 0000 0001 2165 8782 Sección de Posgrado, Escuela Superior de Medicina, Instituto Politécnico Nacional, Plan de San Luis Y Diaz Mirón S/N, Col Santo Tomás, 11340 Mexico City, Mexico
2 grid.9486.3 0000 0001 2159 0001 Laboratorio de Investigación en Inmunología Y Salud Pública, Facultad de Estudios Superiores Cuautitlán, Unidad de Investigación Multidisciplinaria Universidad Nacional Autónoma de México, Estado de México, Mexico City, Mexico
3 grid.9486.3 0000 0001 2159 0001 Facultad de Estudios Superiores Iztacala, Universidad Nacional Autónoma de México, Mexico City, Mexico
29 11 2022
2022
28 12 40417 2 2022
11 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Despite the development of vaccines against COVID-19 disease and the multiple efforts to find efficient drugs as treatment for this virus, there are too many social, political, economic, and health inconveniences to incorporate a fully accessible plan of prevention and therapy against SARS-CoV-2. In this sense, it is necessary to find nutraceutical/pharmaceutical drugs as possible COVID-19 preventives/treatments. Based on their beneficial effects, flavonoids are one of the most promising compounds. Therefore, using virtual screening, 478 flavonoids obtained from the KEGG database were evaluated against non-structural proteins Nsp1, Nsp3, Nsp5, Nsp12, and Nsp15, which are essential for the virus-host cell infection, searching for possible multitarget flavonoids. Amentoflavone, a biflavonoid found mainly in Ginkgo biloba, Lobelia chinensis, and Byrsonima intermedia, can interact and bind with the five proteins, suggesting its potential as a multitarget inhibitor. Molecular docking calculations and structural analysis (RMSD, number of H bonds, and clustering) performed from molecular dynamics simulations of the amentoflavone-protein complex support this potential. The results shown here are theoretical evidence of the probable multitarget inhibition of non-structural proteins of SARS-CoV-2 by amentoflavone, which has wide availability, low cost, no side effects, and long history of use. These results are solid evidence for future in vitro and in vivo experiments aiming to validate amentoflavone as an inhibitor of the Nsp1, 3, 5, 12, and 15 of SARS-CoV-2.
Graphical Abstract
Supplementary Information
The online version contains supplementary material available at 10.1007/s00894-022-05391-6.
Keywords
SARS-CoV-2
Non-structural protein
Docking
Molecular dynamics
Multitarget
Amentoflavone
issue-copyright-statement© Springer-Verlag GmbH Germany, part of Springer Nature 2022
==== Body
pmcIntroduction
The recent severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic that began in Wuhan (China) has expanded worldwide with an estimated 539,893,858 confirmed cases and 6,324,112 deaths on June 23, 2022, worldwide [1]. This virus causes coronavirus disease 2019 (COVID-19), whose clinical manifestations are not specific but somewhat like many viral illnesses [2]. The most common symptoms after viral incubation between 4 and 14 days are cough, fever, fatigue, anosmia, dysgeusia, headache, and sometimes nausea and diarrhea [3]. Clinical manifestations can range from mild to very severe and even fulminant disease [3]. Pandemic has promoted several strategies in all science branches to understand and solve COVID-19 conditions. Vaccine development is one of the main strategies adopted by different research institutes and pharmaceutical companies. At this moment, at least six vaccines have concluded the clinical trial phase III and are approved worldwide for emergency use [4].
Nevertheless, vaccines have less than 95% efficacy, depending on age and comorbid-mortality factors of the population. In addition, due to the short development time and technological novelty adopted, these vaccines have been deployed with several unresolved issues such as technical problems associated with the production of billions of doses and ethical issues related to the availability of these vaccines in the poorest countries [5]. In addition, the SARS-CoV-2 virus is evolving fast, and there is no complete guarantee that vaccines will protect the population against new virus variants [6]. Furthermore, political, economic, and social (vaccine mistrust, asymptomatic individuals (30%), and not following security methods) constraints may limit vaccine access [5].
Drug development for COVID-19 disease treatment continues to be necessary. Several investigational drugs, aiming to inhibit viral entry into the host mechanisms and thus block subsequent viral replication, are currently explored in clinical trials. An example is remdesivir [7], which the Food and Drug Administration (FDA) approved for COVID-19 treatment requiring hospitalization in patients 12 years of age and older. Several drugs must complete clinical trials phase II and III to be approved [8]. So, it is still a long way to have fully approved drugs to treat and prevent this disease.
SARS-CoV-2 possesses potential targets for developing new drugs; some are the family of non-structural proteins (Nsps), which are essential macromolecules involved in the viral genome replication and transcription. Sixteen Nsps have been identified, participating in different host cell infection processes [9].
The viral proteases Nsp3 (PLPro) and Nsp5 (MPro) cleave proteolytically viral polyproteins to form the rest of the Nsps. Most of them integrate the machinery of the replication/transcription complex, highlighting the role of Nsp12 (RdRp) as an RNA-dependent polymerase that in complex with Nsp7 and Nsp8 greatly stimulates polymerase activity; and Nsp15 (EndoU) an endoribonuclease which helps the virus evades the immune system, preventing the detection of viral dsRNA by the host. Furthermore, Nsp1 interferes with the host cell protein synthesis, binding to the 40S ribosomal subunit and endonucleolytic cleavage of host mRNA [9].
In this sense, it is mandatory to continue the research to find and develop nutraceutical/pharmaceutical (easy to insert in our diets or of feasible acquisition) directed to inhibit these important proteins for COVID-19 treatments, such as flavonoids. Flavonoids are molecules with variable phenolic structures widely distributed in the plant kingdom; they are found in leaves, seeds, roots, stems, fruits, and barks [10]. These molecules are considered promising biologically active substances to prevent SARS-CoV-2 due to their possible beneficial action in different mechanisms of this virus infection: inhibiting essential enzymes for virus replication and regulation of immune system functions [11].
Amentoflavone (C30H18O10) belongs to the flavonoid family (a C3′-C8″ apigenin dimer). Ginkgo biloba, Lobelia chinensis, and Byrsonima intermedia are some plants where amentoflavone can be found; all have been used in traditional Chinese or American medicine [12]. Among the biological effects associated with the use of amentoflavone are anti-inflammatory [13, 14], antifungal activity [15], effects on the nervous system [16] and the cardiovascular system [17], and antiviral effects [11]. These facts made amentoflavone an excellent candidate for interacting and possibly modulating different target proteins of SARS-CoV-2 virus.
In the present work, we use one of the important parameters that are currently part of the rational drugs design process, such as the ability to predict ΔG (free energy) binding values, since the affinity of a drug for any receptor is closely related to its biological activity. We use molecular dynamics to predict amentoflavone’s structural dynamics, stability, and binding affinity and predict the feasibility of using it as a SARS-CoV-2 non-structural protein (Nsp1, 3, 5, 12, and 15) inhibitor.
Methods
Protein preparation
Crystal structures of non-structural protein 1 from SARS-CoV-2 structure were obtained from Zhang lab server https://zhanggroup.org/COVID-19/, ID: QHD43415_1 [18]; the structures of SARS-CoV-2 papain-like protease (PLPro) with inhibitor GRL0617 (PDB ID: 7JRN) [19], SARS-CoV-2 3CL protease (3CL pro) in complex with a novel inhibitor (PDB ID: 6M2N) [20], SARS-Cov-2 RNA-dependent RNA polymerase in complex with cofactors (PDB ID: 6M71) [21], and crystal structure of Nsp15 endoribonuclease form SARS-CoV-2 in complex with potential repurposing drug tipiracil (PDB ID: 6WXC) [22] were downloaded from the Protein Data Bank database (https://www.rcsb.org/) [23] and were prepared in USCF Chimera [24], removing water atoms, co-crystallized ligands, cofactors, and/or extra chains from each structure; furthermore, each designed protein was minimized energetically with conjugate gradient and steepest descent algorithms of this software, using AMBER ff14SB force field. The structures were subjected to 100 ns of molecular dynamics, according to method further described, and 200 ns in the case of Nsp1, to understand its conformational profile; after dynamics completion RMSD and cluster analysis were performed to obtain the main structure or/and the pdb files for the most representative clusters; after that, proteins structure were saved in pdbqt format in AutoDock Tools [25] after adding polar hydrogens and Kollman charges to perform virtual screening assay.
Ligand preparation
A specific search was performed on the advanced module of the PubChem database, using “flavonoid” as keyword in the phytochemical compound subclassification of the KEGG database. KEGG is a comprehensive drug information resource for approved drugs in Japan, the USA, and Europe, unified based on active ingredients’ chemical structure and/or chemical components [26].
On the other hand, 3D conformer of co-crystallized ligands was separated from its protein: GRL0617 for Nsp3, baicalein from Nsp5, for Nsp12 the novel AT9 inhibitor was obtained from PDB ID: 7ED5 [27], tipiracil for Nsp15, and no ligand for Nsp1, because no structures with co-crystallized ligand(s) are available for this protein. In the same sense, the 3D conformers of the 478 flavonoids found were downloaded in sdf (structure data file) format and converted to pdb (protein data bank) using the module of OpenBabel software [28], applying the conjugate gradient and steepest descent algorithms for energy minimization using Universal Force Field (UFF). The energy minimized structures were converted to pdbqt format by the AutoDock Tools script prepligand4.py included in the software package [25].
Virtual screening
For the virtual screening, a directed methodology was employed, with the grid box centered on each protein catalytic site (based on co-crystallized ligand), covering a 20 × 20 × 20 Armstrong volume on each. AutoDock Vina virtual screening script written in Linux Shell was used to perform this process [29], generating a single docking pose, which were analyzed based on protein–ligand interactions and binding affinity values.
Molecular docking
Results of virtual screening analysis, amentoflavone was only the molecule showing interactions with nearly all the Nsps; for this reason, this flavonoid was selected as a possible multitarget drug. The amentoflavone molecule was processed with AutoDock Tools to add the polar hydrogens and Gasteiger charges. The grid box dimension used in this section was the same as described above. The amentoflavone vs. Nsps docking was performed with AutoDock Vina with 1000 independent replicates using a script written in Shell to evaluate the best pockets and how the compounds are arranged in them. Afterward, based on AutoDock Vina energy scores and the root-mean-square deviation (RMSD) calculation, the most probable conformation and interaction site was chosen. With these coordinates, molecular dynamic (MD) simulations were carried out to explore the structural and conformational stability of protein–ligand complexes. Docking methodology was validated by performing independent assays for the co-crystallized ligands. 2D interactions images were built with Discovery Studio [30]. Ligand efficiencies (LE) were estimated using the previous reported equation [31] where the Kd (dissociation constant of a ligand–protein complex) calculated values by the formulas:Kd=e-ΔGRT
LE=-RTNHAln(Kd)
where ΔG corresponds to binding energy (kcal mol−1) obtained from docking, R is the gas constant (0.001987207 kcal mol−1 K−1), T is the temperature at standard conditions of aqueous solution (298.15 K), and NHA denotes the number of non-hydrogen atoms in a ligand.
Molecular dynamics
Protein and protein–ligand complexes were subjected to molecular dynamics using the CHARMM36-jul2020 force field in GROMACS 5 [32, 33], also for amentoflavone and co-crystalized ligands the topologies were generated using CGenFF online server (https://cgenff.umaryland.edu/). Before system solvation using simple point charge (SPC) and TIP3P water model in a cubic box with a minimum distance of 1 nm from the edge of the protein with periodic conditions, the solvent molecules were replaced with NaCl 0.15 M and then neutralized with the necessary counter ions. The solvated systems were minimized using the steepest descent function for 50,000 steps with a maximum force of 10 kJ mol−1. Then, the systems were balanced in the NVT ensemble for 100 ps followed by equilibration in the NPT ensemble for an additional 100 ps with protein and ligand position restraint. Finally, production dynamics were carried out in the NPT ensemble for 100 ns at 300 K and 1 atm pressure using the V-rescale temperature coupling method and Parrinello-Rahman coupling method, respectively. The particle mesh Ewald method was used for computing long-range electrostatic interactions with a non-bonded cutoff of 10 Å and the LINCS algorithm. The non-bonded interactions, Coulomb (electrostatic potential), and Lennard Jones (Pauli repulsion and hydrophobic/van der Waals attractions) interactions were truncated at 10 Å using the Verlet cut-off scheme. The leap-frog algorithm was used to compute the equation of motion with a time step of 2 fs. All the parameters used to perform the molecular dynamics were based on the widely used methodology of Justin Lemkul, which is constantly updated according to the technological and algorithmic advances of these methods [34].
For the analysis, from all the trajectories, the root mean square deviation (RMSD) of the complex, the protein, and ligand was calculated using the initial ligand and Nsp structures as reference protein and the Cα from the backbone. The number of hydrogen bonds was also calculated between protein and ligand, the most representative (repeated) conformational structures (clusters) with a distance limit of 0.25 nm were obtained, electrostatic and van der Waals interaction energy were computed, and solvent accessible surface area (SASA) also were obtained. GROMACS package was used for all cases. Images were built using PyMOL [35].
Ligand–protein interactions
Interaction fingerprints of protein–ligand complexes were generated by Prolife software [36], using whole molecular dynamics trajectories and graphing those with more than 0.1 ns of occurrence.
Results and discussion
Protein molecular dynamics
RMSD and cluster analysis
After molecular dynamics completions, the stability of Nsp1 (Fig. 1A) and for the rest of the non-structural proteins (Fig. 1B–E), relative to its conformation was determined by the deviations produced during its simulation (RMSD). The smaller the deviations, the more stable the protein structure. In the same sense, cluster analysis was performed to find the main structure conformation, based on their time frequency along the whole trajectory (Table 1).Fig. 1 The RMSD plot of the five non-structural proteins A Nsp1 after 200 ns of molecular dynamics, and 100 ns of molecular dynamics for B Nsp3, C Nsp5, D Nsp12, and E Nsp15
Table 1 Results of cluster analysis of the molecular dynamics of the five non-structural proteins
Protein Cluster (0.25 cutoff) Cluster frequency (ns)
Nsp1 1 19.2
2 15.72
3 11.70
Nsp3 1 97.29
Nsp5 1 87.26
2 11.38
Nsp12 1 99.76
Nsp15 1 96.92
Regarding Nsp1, its trajectory has a sudden movement at the beginning (0 to ~ 0.75 nm), keeping ~ 0.75 nm as the mean value; however, fluctuations were visible throughout the simulation, like the variation of ~ 0.25 nm in ~ 150 ns of the DM simulation. We can appreciate that the protein keeps a main conformation between ~ 80 and ~ 104 ns, which corresponds to the structure of cluster 1. In a similar way, correlation of RMSD with the main structures of cluster analysis, it is possible to distinguish cluster 2 from the ~ 170 to ~ 190 ns and cluster 3 from the ~ 105 to ~ 130 ns. Based on the protein structure and the RMSD analysis, we can conclude that relatively large motions over time are due to the inherent flexibility of the structure, mainly in the last fifty residues of the protein, which do not possess a stable secondary structure (i.e., α-helix or β-strand).
On the other hand, RMSD analysis of Nsp3 (Fig. 1B), Nsp12 (Fig. 1D), and Nsp15 (Fig. 1E) shows a more stable protein throughout the simulation, except for the first ns RMSD jump of Nsp15 (~ 0.3 nm). We can notice that Nsp12 (Fig. 1D) reaches a stable conformation approximately at ~ 20 ns; at this point the three proteins fluctuate between the same range all along and without movements that exceed ~ 0.05 nm from each other, except for a few moments of their trajectory, where peaks that stand out (from ~ 0.2 to ~ 0.3 nm) are observed and then return to the movement trend until the end. Based on this, and according to the cluster analysis, the three proteins keep a main structure along the whole trajectory (Table 1).
Finally, according to RMSD analysis of Nsp5 (Fig. 1C), protein stabilizes after ~ 25 ns of the molecular dynamics, at that point and before ~ 80 ns, the protein has no important fluctuations greater than ~ 0.05 nm, the main structure of this period of time corresponds to cluster 1; after ~ 80 ns and until the end of the simulation, protein flexibility have important changes > ~ 0.1 nm, generating a different conformation which is represented by the cluster 2.
In brief, Nsp3, Nsp12, and Nsp15 have a cluster which occupancy is on more than 95% of the dynamics, so this structure was selected for the subsequent analysis; on the other hand, Nsp1 and Nsp5 showed three and two clusters, respectively, the structures of these conformations were considered for the later analysis, taking into account that they represent different conformational states of the protein, and thus can be important to characterize ligand–protein interaction at each conformation.
Virtual screening
Virtual screening was performed with the 478 flavonoids obtained from the search in the KEGG database of PubChem, to determine a possible mechanism by which these flavonoids act on the SARS-CoV-2 Nsp1, 3, 5, 12, and 15 main structures; as mentioned above, the coordinates for virtual screening were centered on each protein catalytic site, based on that from the co-crystallized ligands (Nsp3, 5, 12, and 15) and in case of Nsp1, centered on the alpha carbon of His165, known as an important residue for its activity [37–39].
The distribution of all tested compounds energies for each cluster is represented in Fig. 2, after analysis of the results and seeking for a ligand that interacted with the five proteins to consider it as a probable multitarget flavonoid onto the 10 compounds with the highest binding score (Fig. 3), amentoflavone (Fig. 4) was the only flavonoid that binds with an acceptable binding score (between − 7.3 and − 9.6 kcal/mol) to the five non-structural proteins.Fig. 2 Energy distribution after virtual screening of the 478 flavonoids with the five non-structural proteins at their respective cluster A Nsp1, B Nsp3, C Nsp5, D Nsp12, and E Nsp15
Fig. 3 Binding energy of the best ten molecules and co-crystallized ligands of each protein
Fig. 4 Amentoflavone molecular structure
Molecular docking
Based on docking results, Fig. 5A shows the main residues (including those well-described residues important for the enzymatic activity of each protein) involved in the interaction between amentoflavone and Nsp1, 3, 5, 12, and 15, showing the potential of this molecule as a probable inhibitor; our analysis also shows the two most probable complex conformations based on frequency and RMSD calculations (Fig. 5B). The most frequent cluster, the one with the highest binding energy, was used to perform molecular dynamics studies and the second more frequent cluster represents an alternative binding pose with similar binding score values; data used for this analysis could be found in Table S2.Fig. 5 A Main and reported residues important for activity of each protein involved in the interaction between amentoflavone and the five Nsps. B Frequency and binding score of the two most probable complex conformations
Analyzing the interactions of each Nsp1 cluster and amentoflavone (Fig. 6A–C), we can see that binding pockets are different for the three clusters, and so the residues that interact with the ligand. In all cases, the binding pocket of amentoflavone includes residues of the flexible C-terminal region of the protein. For the cluster 1 (Fig. 6A), important interactions include the H-bonds with Arg124 and Gln158, π interactions with Asp75 and Gly168 (Fig. 4A1); it has been demonstrated that Arg124 strongly interacts with the phosphate backbone of SARS-CoV-2 RNA 5′-untranslated region and also Asp75 sometimes formed hydrogen bonds with the bases of it [40], furthermore protein uS5 of the 40S ribosome subunit interacts within a hydrophobic surface which involves this residue and other adjacent ones, including Gln158 and Gly168 [41]. In the case of cluster 2 (Fig. 6B), we can appreciate in Fig. 4A2 that more H-bonds were formed compared to the previous cluster, standing out the interactions with Phe157 and Gly168 which are included in the hydrophobic surface above mentioned [41]. Finally for cluster 3 (Fig. 6C), hydrogen bonds are still being present, now with Asn162 and Gly168, which are part of this important site for the interaction of the 40S ribosome subunit, as well as Trp161; moreover, a π-alkyl interaction is formed with Lys164 an important residue which is crucial for RNA cleavage and translation inhibition functions of SARS-CoV-2 [37–39]. In this sense, amentoflavone could abolish 40S binding and relieve translational inhibition, by the interactions described above, mainly formed with the residues of the region where the 40S subunit binds.Fig. 6 Molecular docking 2D representation. Each quadrant shows the binding site and the interactions of the results performed with amentoflavone and the co-crystallized ligands vs. all Nsps after the virtual screening. A Nsp1_1-amentoflavone, B Nsp1_2-amentoflavone, C Nsp1_3-amentoflavone, D Nsp3-amentoflavone, E Nsp3-GRL0617, F Nsp5_1-amentoflavone, G Nsp5_1-baicalein, H Nsp5_2-amentoflavone, I Nsp5_2-baicalein, J Nsp12-amentoflavone, K Nsp12-AT9, L Nsp15-amentoflavone, and M Nsp15-tipiracil
On the other hand, although Nsp3 is well known for its protease activity cleaving the viral polypeptide, it has also been observed that has an additional function stripping ubiquitin and interferon-stimulated gene 15 (ISG15) from host-cell proteins to aid SARS-CoV-2 in its host evasion innate immune responses (ISG15) [42, 43]. Two active sites blocking this protein activity have been reported: (1) the catalytic site which contains a canonical cysteine protease catalytic triad (Cys111, His272, and Asp286) and (2) the flexible loop BL2 or β-turn, an important mobile loop (Gly266–Gly271) adjacent to the active site that closes upon substrate and/or inhibitor binding [44]. Nevertheless, the cysteine protease drug-like inhibitors present several challenges, notably, for their toxicity and lack of specificity, due to untargeted cysteine residue covalent modification. For that, BL2 loop it is a better option to find probable non-covalent inhibitors, such as the already known compound GRL0617 [45]; the binding of this compound induces BL2 loop closure clamping the inhibitor to the body of the protein, limiting the catalytic triad movement and restricting access to the cysteine by reducing agents, thereby generating an inactive enzyme; also other inhibitor-independent motions are observed at the tip of the fingers domain, which may account for PLpro’s ability to interact with topologically different poly-Ub chains and the UBL modifier ISG15 [45, 46]. Reported interactions between GRL0617 and Nsp3 include the residues: Asp164, Gln269 (hydrogen bonds), Tyr264, Tyr268, Pro247, Pro248, Thr301, Arg166 (aromatic interactions) [47]; important interactions coincide with those in Fig. 6E and are comparable with those established with amentoflavone.
As shown in Fig. 6D, amentoflavone forms a hydrogen bond with Asp164, an interaction presumed to be important for ligand stabilization, which is complemented with the Tyr268 interaction, presumed to make a fitting induced movement to adopt a closer conformation, strengthen the interaction with the inhibitor and anchoring it to the binding site, and with Arg166 and Glu167, important for water and carbonyl coordination [46, 47]. Amentoflavone also interacts with the hydrophobic pocket residues formed by Pro247, Pro248, and Tyr268, which is a specific site for leucine side chain of LXGG consensus sequence recognition [44], for PLPro proteolytic cleavage activity.
In this sense, amentoflavone is a good candidate inhibitor because it binds to this site, making hydrophilic interactions with Nsp3 surface residues and interfering with the peptide recognition motif LXGG. Furthermore, our results show a binding score of amentoflavone against Nsp3 of − 8.69 ± 0.02 kcal/mol, comparable with other studies where this interaction is also addressed: − 7.5 kcal/mol for Nsp3 (PDB ID: 6W02) [48] and − 8.8 kcal/mol for PLPro (PDB ID: 7JN2) [49]; for this reason, amentoflavone has the advantage that not only inhibiting viral replication but also inhibiting the signaling cascades dysregulation in infected cells.
For the case of Nsp5 or 3CLpro, its catalytic site is composed by a dyad of Cys145 and His41 residues which are buried in an active site cavity located on the protein surface. This cavity consists of S1 to S4 subpockets crucial for substrate recognition; the active site is well characterized, and most crystallized ligands bound to Nsp5 (185 ligands known, including baicalein co-crystallized ligand of the protein used here) coincide at this site, which is comprised by the residues: Thr25, His41, Met49, Asn142, Ser144, Cys145, His163, His164, Gln189, Glu166, Pro168, His172, and Ala191 [50, 51]; our baicalein docking result with Nsp5 cluster1 (Fig. 6I) shows H-bond interaction with the catalytic dyad (Cys145 and His41), Met165 and Gly166, and π-sigma interaction with Met49 and Leu50, which seems to stabilize the molecule into the substrate-binding pocket core (through S1 and S2 subsites). In the case of cluster 2 (Fig. 6H), no interactions with the catalytic dyad were formed, in seems that ligand was pushed out of catalytic site (found in S1 subsite) and displaced to the long loop (residues 185–200), forming interaction with at least five residues of this region, and the S2 subsite, which includes the interactions with Met165 and Pro168; this could be interesting due to the orientation of this loop and II and III domains is important for maintaining a catalytically competent activity [52].
Similarly, our results show that amentoflavone can bind within the active site cavity of cluster 1 (Fig. 6F), carbonyl groups make hydrogen bonds with Thr169 and Gly170 and π electrons mediated interactions with the residues of catalytic dyad (Cys145 and His41) and surrounding amino acids, stabilizing molecule into the substrate-binding pocket core (through S1 and S2 subsites); moreover, the Leu50 side chain also established π-alkyl interaction from the residue nitrogen to the phenyl ring centroid, and Met165 contacted the middle amentoflavone ring via π-sulfur interaction. Talking about cluster 2 (Fig. 6G), a similar displacement, maybe provoked by a closer status of S1 subsite, as occurs with baicalein happened: amentoflavone was pushed to the limits of S2 subsite and the long loop, forming hydrogen bonds and π-sigma interactions, like those formed by baicalein. The interactions at this site could define a possible new target for a possible mechanism of the enzyme inhibition, due to the orientation of this region is important for a competent catalytic activity [52].
Considering the described results, amentoflavone is a great inhibitor candidate for Nsp5 because it interacts with two catalytic residues, Glu166, and the S1/S2 subsites, which are the key elements for the recognition of substrates [51, 52]. Our results shows a binding score of − 8.90 ± 0.01 kcal/mol for amentoflavone-Nsp5 complex; this value is comparable with other studies which report similar interacting residues and binding scores of − 7.589 kcal/mol (PDB ID: 6WNP) [53], − 9.4 kcal/mol (PDB ID: 6M2N) [49], − 9.2 kcal/mol (PDB ID: 6LU7) [54], − 27.0441 kcal/mol (PDB ID: 6LU7) [55], − 9.6 kcal/mol (PDB ID: 6LU7) [56], − 8.9 kcal/mol (protein modeled in Swiss Modeler) [57], − 10.2 kcal/mol (PDB ID: 6LU7) [58], and − 10.0 kcal/mol (PDB ID: 6LU7) [59].
Furthermore, inhibition of Nsp12 or RdRp could target its conserved polymerase motifs, highlighting motif A composed of residues from 611 to 626, including Asp618 important for the binding process. And motif C, which comprises residues from 753 to 767, includes the catalytic residues Ser759, Asp760, and Asp761 [21]. As Fig. 6J show, amentoflavone forms hydrogen bonds with Asp618, Asp623, Arg624, and Thr556, necessary for binding and stabilizing ligand within motif A and C; π-ion interactions with Lys621 and Asp760 create an electrostatic potential into the cavity, disfavoring electronic-based processes carried out by Nsp12. Besides the aforementioned interactions, it is also important to highlight those forms with Asp618 and Asp760, essential residues for catalytic site activity. Similar interactions are presented for AT9 binding (Fig. 6K), highlighting those with Lys621 and Arg624, necessary for binding and stabilizing; in addition, the guanine-like group occupies the site which recognizes the DNA strand [21]. Therefore, amentoflavone may be an excellent candidate to inhibit the Nsp12. Similar results have been published, reporting binding scores of − 9.4 kcal/mol (protein modeled in Swiss Modeler) [57] and − 9.3 kcal/mol (PDB ID: 6M71) [48].
Finally, Nsp15 or NendoU is associated with multiple functions and could be involved in allowing the virus to evade the immune system innate response by hiding the viral RNA from the macrophage dsRNA sensors [60]; the catalytic site of this protein is composed of the residue triad His235, His250, and Lys290. These amino acids have been demonstrated to be conserved in all viruses of the coronavirus family [61]. They are fundamental for the activity of the Nsp15 enzyme and the formation of its hexameric quaternary structure [62]. Our results, presented in Figs. 5A and 6L, show that amentoflavone interacts with His235 with a hydrogen bond, which is fundamental in substrate hydrolysis and complex stabilization, also supported by hydrogen bonds with Gln245, Val292, Ser294, and Thr341; moreover, this flavonoid forms hydrophobic interactions with Lys290 (part of the catalytic triad), Cys291, Tyr343, and Leu346, amino acids that serve as an anchor, promoting the binding and residence of the substrate onto the catalytic site, favored by the aromatic rings of amentoflavone. From these interactions, we must distinguish those with Ser294 and Tyr343 because these residues together are believed to govern U specificity, in RNase A base recognition [63]. Similar interactions were observed for tipiracil (Fig. 6M), highlighting the recognition of carbonyl group by Ser294, which is specific for uracil recognition. With the information analyzed above, Nsp15 is another protein that could be inhibited by amentoflavone.
Ligand efficiency
Ligand efficiency (LE) allows us to compare molecules according to their average binding energy [59]; very low values indicate that the compound binds tightly to the protein. LE represents the average binding energy per non-hydrogen atom, where fair values of LE for inhibitors candidates are LE > 0.3 kcal [31]. Considering this information, for all the evaluated proteins, amentoflavone has ligand efficiency values less than 0.3 kcal (0.23, 0.20, and 0.18 for Nsp1 clusters, 0.22 for Nsp3, 0.21 and 0.23 for Nsp5 clusters, 0.23 for Nsp12, and 0.23 for Nsp15), information that supports the proposal of using this flavonoid as a potential inhibitor of non-structural proteins of the SARS-CoV-2 virus. These values are smaller than those of the co-crystalized ligands, i.e., GRL0617 0.33, baicalein 0.34 and 0.35, and tipiracil 0.35, except for AT9 (0.22). In the same sense, amentoflavone exhibits low theoretical dissociation constants (Kd) for each of the evaluated proteins, reaffirming the stability of the complex and the favorable binding of the ligand, even promising a most efficient inhibitor of these non-structural proteins (fewer doses are required for reaching the inhibitory effect). The top 10 molecules interacting with the best docking scores for the selected clusters of each protein were selected for this comparison (Table S1).
Protein–ligand molecular dynamics
RMSD analysis
After evaluation of the three clusters of Nsp1-amentoflavone complex (Figure S1A, S2A and S3A), we can appreciate that for cluster 1 RMSD has the same behavior as the protein alone, just with a little high nm, due to the ligand binding; the ligand reaches its stability at 25 ns. More variations in ligand RMSD are observed for cluster 2, those sudden movements reflect on the global RMSD, mainly in the first 15 ns. Finally, about the third cluster, at the first ns of the complex RMSD has higher values than just protein caused by the fluctuations on ligand’s RMSD. In this sense, the first cluster of Nsp1-amentoflavone complex is the most stable, and the three represent three different stages of the protein behavior.
With respect to Nsp3 (Figure S4A and S5A), complex and protein have no significant differences, in both amentoflavone and GLR-0617, due to the stability of the ligand which does not suddenly elevate the RMSD values. Also, protein, ligand, and the complex reach their best conformation at 18 ns, keeping its flexibility near the RMSD mean value (0.35 nm); the little variations in RMSD of the ligand let us to find different complex clusters.
On the other hand, the RMSD of cluster 1 of Nsp5-amentoflavone (Figure S6A) has no significative variations between complex and single protein, regarding ligand, two possible states could be identified with approximately 0.05 nm of difference; cluster 2 (Figure S7A) shows higher variation in complex against protein, this is due to the rise of RMSD values of the ligand. Similarly, the second cluster of Nsp5-baicalein complex (Figure S9A) shows a significant variation of protein vs. complex, which correlates with the ligand fluctuation. First cluster (Figure S8A) shows no variation in RMSD of complex vs. protein, the ligands show a lot of fluctuations along the whole trajectory, but two probable states can be defined based on this.
Talking about nsp12 (Figure S10A), amentoflavone reaches a single conformation at 17 ns, fluctuations of the ligands have significant influence on the RMSD of the complex; the behavior of the Nsp12-inhibitor is contrary to that shown for AT9 (Figure S11A), because variations of complex, ligand and protein maintain on the 0.18 nm average RMSD.
Finally, for the Nsp15, respecting to the complex with amentoflavone (Figure S12A), the RMSD of the complex have several variations, these variations could be a result of the high flexibility of the protein or for a high or they could be a result of several conformation changes because of ligand binding, although ligand has a very stable RMSD, representing its permanence into the binding site. In contrast, Nsp15 in complex with tipiracil (Figure S13A) has sudden variations at 65 to 80 ns, representing an interesting cluster for this complex.
Hydrogen bond analysis
The hydrogen bonds that form during the drug-receptor interaction are characteristic of a high-affinity binding; this is due to the force exerted by this bond repelling the force exerted by water in the cavities, conferring greater stability to the complex [63]. In other words, this approach allows us to have a closer look at the complete phenomenon, because in MD simulations during the formation of hydrogen bonds, it necessarily implies an energetic process of desolvation of the hydrogen bonds previously formed between the protein-water system [64].
Regarding the formation of hydrogen bonds, during all MD simulations, the amentoflavone establishes a variable number of them with various residues of the SARS-CoV-2 Nsps. Throughout the simulation in the trajectories of the three trajectories of Nsp1 (Figure S1B, S2B and S3B), amentoflavone formed 2 hydrogen bonds on average and up to 5 for Nsp3 (Figure S4B) and Nsp15 (Figure S12B). The amentoflavone in complex with Nsp12 (Figure S10B) was able to form 5 hydrogen bonds in most of their conformations along the trajectory, in contrast with the co-crystallized ligand that formed up to 10 hydrogen bonds in the most of nanoseconds along the whole trajectory (Figure S11B). Finally, talking about Nsp15 for both amentoflavone and tipiracil, they reach a maximum of five hydrogen bonds, although there are ns with no hydrogen bonding, the permanence of ligands into binding pockets depend on hydrophobic interactions (Figure S12B and S13B).
Cluster analysis
The clustering process allows us to analyze the conformational behavior of the protein under conditions established on the MD simulations to find the most repeated conformation and thus the most representative protein–ligand binding process. In the following analysis, we have selected the most frequent, also representative, cluster from the best three of all five ligand–protein MD simulations (Fig. 7).Fig. 7 Results of cluster analysis of the molecular dynamics of the five Nsp-ligand complexes
In Figs. 8, 9, 10, 11, and 12, we can see the 2D and 3D best cluster for each complex representation. For Nsp1 cluster 1 (Fig. 8A), we appreciate two interactions that had also been formed in the previous docking assay analysis with residues Trp161, forming two π-π interactions with two different rings of amentoflavone; a more stable anchor is formed by the π interactions with Phe157 and Ser167. We can also see another hydrophobic carbon-hydrogen interaction with the residue Gly168 and another hydrogen bond with Gln158. In the case of cluster 2 (Fig. 8B), the main interactions are hydrophobic, i.e., carbon-hydrogen and π interactions, with Leu4, Val5, Leu145, Gly146, and Thr151, highlighting the only hydrogen bond with Leu149. Referring to cluster 3 (Fig. 8C), it is the poorest talking in interactions terms as it just has two interactions: a hydrogen bond with Ser40 and a π-alkyl with Leu140. By these, the most representative way of binding on the well described pocket is cluster 1.Fig. 8 3D and 2D models of the amentoflavone onto the most favorable cluster in complex with the three clusters of Nsp1 and their interaction fingerprint along whole dynamics
Fig. 9 3D and 2D models of the amentoflavone and GRL0617 onto the most favorable cluster in complex with the three clusters of Nsp3 and their interaction fingerprint along whole dynamics
Fig. 10 3D and 2D models of the amentoflavone and baicalein onto the most favorable two clusters in complex with the three clusters of Nsp5 and their interaction fingerprint along whole dynamics
Fig. 11 3D and 2D models of the amentoflavone and AT9 onto the most favorable cluster in complex with the three clusters of Nsp12 and their interaction fingerprint along whole dynamics
Fig. 12 3D and 2D models of the amentoflavone and tipiracil onto the most favorable cluster in complex with the three clusters of Nsp15 and their interaction fingerprint along whole dynamics
In the Nsp3 cluster binding site, amentoflavone (Fig. 9A) remains into the binding pocket, keeping all interactions of previous performed docking, highlighting those with Asp164, important for ligand stabilization; Tyr268 presumed to make a fitting induced movement to adopt a closer conformation and with Arg166 and Glu167, important for water and carbonyl coordination [46, 47]. Regarding to the inhibitor GRL0617 (Fig. 9B), this compound preserves π-π interactions with Tyr268 and hydrophobic π interactions with Pro247, Pro248, and Tyr264, residues within the site called the flexible loop BL2 or β-turn [44, 47], preserving that with Pro248 as in the docking assay.
For Nsp5, it can be observed that amentoflavone at cluster 1 (Fig. 10A) that the previous docking reported interactions are not preserved, instead hydrogen bonds are formed with Gly138 and Glu288, also π interactions with Tyr126 and Lys137 are included. In comparison with cluster 2 (Fig. 10B), just one interaction is preserved: Tyr126, and new ones are formed with residues near to Tyr126 (which seems to be crucial in the binding process of amentoflavone): Val114, Gln128, Cys128, Ser139, and Leu141, residues considered important for ligand recognition [65], and important for binding of other molecules with inhibitory activity, such as paritaprevir, lopinavir, and epigallocatechin-3-gallate [66]. In contrast, baicalein in both clusters (Fig. 10C and D), binds with less amino acids, totally different to those reported in docking assay: Gln19, Met49, and Gly120 for cluster 1 and for cluster 2, baicalein interacts with amino acids of the catalytic site: His41 and Cys145, reaffirming the mechanisms of inhibition of this flavone; complemented with the hydrophobic interactions of Met49, Leu50, Met165, and Ala191; also highlighting the hydrogen bond with Glu166.
Regarding the Nsp12 cluster, amentoflavone (Fig. 11A) remained at the binding site keeping the interactions reported in the docking assay such as Ala547, Arg555, and Asp760, the last one, essential for catalytic site activity (as mentioned above); additionally, interactions with Asp845 and Lys545 were formed, which have already been reported as capable of binding to at least three other drugs with inhibitory potential [67, 68]. On the other hand, AT9 (Fig. 11B) maintains in the binding site, very favorable by the phosphate groups of its structure (as mentioned above), most of the interactions of this cluster are preserved from the docking assay, highlighting electrostatic interactions with Lys551, Arg555, Lys621, and Arg624, reaffirming its mechanism as mimic of uracil.
Finally, in the cluster obtained from the amentoflavone-Nsp15 complex trajectory (Fig. 12A), we observed that the same hydrogen bonds formed in the docking performed prior MD simulation with residues His235, Ser294, and Tyr343, as well as π-π interactions with residues Trp333 and Tyr343. In addition to these interactions, here, the flavonoid forms a new hydrogen bond with the Glu340 residue. All these interactions remain within the binding site of known inhibitors, such as benzopurpurin 4B, which, in addition to sharing the same interactions shown here, has also been shown to inhibit Nsp15 in vitro [69]. In the case of tipiracil, it keeps into the binding pocket by interacting strongly with Glu267 by a salt bridge and π-π interactions of guanidine ring with Phe280 (Fig. 12B).
Protein–ligand fingerprint
The time that protein–ligand interactions maintain along the 100 ns of molecular dynamics simulation are represented in Fig. 4E, 5, 6, 7, 8, 9, 10, 11, 12. Analyzing the interactions of amentoflavone with the three clusters of Nsp1, we can appreciate that most of the time the main amino acids that keep the ligand at binding site are Trp161, Phe157, Gln158, His165, and Gly168, mainly through hydrophobic interactions (Fig. 8A); at the cluster 2 (Fig. 8B), which conformation displace some armstrongs the ligand, the main residues interacting with amentoflavone are Leu149, Asn128, Thr151, Asp144, and Gly150, by hydrophobic and hydrogen bonding, interestingly, the interactions with Phe157 and His165 diminished; at the end, amentoflavone interacts mostly via hydrophobic with Leu140, Gly146, Leu145, Glu148, and Leu149 in cluster 3 (Fig. 8C). By this way, amentoflavone binds into the hydrophobic cavity formed by α1 and α2 helices of Nsp1, which is confirmed by the residues between Asp148 and Gly180, important for the 40S ribosomal subunit binding [41].
Regarding Nsp3, the main reported interactions into the binding site involve Asp164, Gln269 with hydrogen bonds, and Tyr264, Tyr268, Pro247, Pro248, Thr301, Arg166 with aromatic interactions, mainly in the flexible loop BL2 or β-turn, an important mobile loop (Gly266–Gly271) [44, 47]. Amentoflavone remains in the binding site due to the interactions formed by Ser293, Tyr296, Lys297, Lys292, and Tyr264 mainly by hydrophobic and hydrogen bonding (Fig. 9A), fitting perfect with the flexible loop BL2 and the specific site for leucine side chain of LXGG consensus sequence recognition [44], for PLPro proteolytic cleavage activity. Similarly, GRL0617 interacts mainly with Tyr268, Tyr264, Pro247, Met208, and Pro248 (Fig. 9B).
In the same sense, Nsp5 hydrophobic interactions are the main type formed along the whole molecular dynamics, for both amentoflavone and baicalein. In case of amentoflavone, the five main residues for cluster 1 (Fig. 10A) includes Tyr126, Gly138, Ser139, Lys137, and Gln127; two of these amino acids are conserved for the cluster 2: Tyr126, Ser139 (Fig. 10B), complemented by the interactions with Tyr118, Phe140, and Cys128. Based on this information, we can mention that amentoflavone accommodates into the binding cavity, nevertheless, it does not reach the catalytic dyad residues, but the occupancy of the binding pocket mediated by the hydrophobic and π-stacking interactions with Tyr126 and Phe140, suggesting that the inhibition mechanism is through binding site competition and occupation instead electronic disruption of catalytic dyad. In contrast, baicalein forms mostly hydrophobic and π-stacking interactions with the residues His41 and Cys145, in both clusters (Fig. 10C and D), complemented by the interactions with the surrounding subsites, indicating that the main mechanism of this molecule is by disruption of the catalytic reaction.
On the other hand, the evaluated known antagonist (AT9) of Nsp12 (Fig. 11B) binds favorably into the binding pocket by electrostatic interactions such as anionic and hydrogen bonds with Arg566, Arg497, Lys563, and Lys493, provoked by the phosphate groups contain; in contrast mostly of the interactions between amentoflavone and Nsp12 (Fig. 11A) are hydrophobic and in some cases are hydrogen bonds. The main residues which favor the binding are Arg497, Asp787, Ile789, Val409, and Asp702. In both cases, the compounds occupy the main entrance of DNA coupling; the advantage of AT9 is that it possesses a couple of phosphate groups which mimics those that nucleic acid possess.
Finally, in the case of Nsp15 the interactions that favor the amentoflavone binding (Fig. 12A) and permanence into the binding pocket are with Gly152, Lys151, Val188, Gln246, and Val293. Although the amino acids of the catalytic triad interact with the ligand, but not for all the time of molecular dynamics, the interaction with surrounding residues lets us to propose that the inhibition method of amentoflavone is by the rearrangement of the binding pocket, leading to no catalytic reaction, supported by hydrophobic and hydrogen bonding. In the same way, tipiracil (Fig. 12B) keeps on the binding pocket, with no direct interaction with the catalytic residues, but interacting with surrounding amino acids like Tyr344, Val293, Lys291, His251, and Ser295, mainly with hydrophobic and electrostatic interactions.
Electrostatics and van der Waals energies
Contributions of electrostatic and van der Waals interactions to global binding energy play a major role in determining protein–ligand binding specificity and the rate of protein–ligand association [70]. Regarding to the calculated energies for the three clusters of Nsp1 (Figure S1C, Figure S2C and Figure S3C) in complex with amentoflavone, most of the contribution to the global binding energy is mediated by van der Waals, correlating to hydrophobic interactions reported in the above section, moreover at some times of the trajectory, electrostatic energy contributes in a greater manner (~ 25 and 50 ns), correlating with electronic π interactions.
Regarding Nsp3, a curious behavior because there are some moments in the trajectory that electrostatic energy does not contribute to the binding process (Figure S4C and Figure S5C), and the van der Waals contributions seems to be zero; this could be provoked by no residue-ligand interaction or for a remoteness of the ligand from its position. Apart from this curious fact, the co-crystallized ligand (Figure S5C) of Nsp3 binding is mediated by the contributions of van der Waals energy, in difference with amentoflavone (Figure S4C) which contributions are equative between electrostatic and hydrophobic, as correlated with the analysis of interaction fingerprint.
In a similar way, most of the contribution to global binding energy of the Nsp5-ligand complexes is derived from van der Waals; nevertheless, there are several nanoseconds where electrostatic energies contribute in a higher way, principally at the middle and the first quarter of simulation. These contributions are reflected into the interaction type that establishes each compound with residues of the binding pocket, which are mainly hydrophobic, and those peaks of electrostatic contributions correlate well with π-stacking interactions and hydrogen bonding. These results are similar between both clusters (Figure S6C-S7C and Figure S8C-S9C).
Regarding Nsp12 and amentoflavone (Figure S10C), at least half of the simulation time the electrostatic energy contributes more than van der Waals to the binding process, energies that correlates with the hydrogen bonding of Asp787, Asp702, and Arg497 and the hydrophobic interactions. In the case of AT9 (Figure S11C), electrostatic interactions contribute abysmally to the global binding constant (− 1100 kJ/mol); this correlates perfectly with the fully anionic and hydrogen bonding observed in the interaction fingerprint.
Finally, respecting to Nsp15, there are several moments, for both ligands, that energy contribution is zero (Figure S12C and S13C); it seems to be that only residual van der Waals energy remains. This phenomenon could be due to a displacement of ligand form the initial position, nevertheless, the RMSD values from both remain stable, so no correlation could be established between with the two variables, but yes with hydrogen bonding; nevertheless, the binding process is favored by other class of hydrophobic interactions.
SASA
Solvent-accessible surface area (SASA) is an important descriptor in ligand binding. The extent of ligand SASA value decreases upon binding indicates whether the ligand is deeply buried or not upon binding to the pocket. For all complexes (Figure S1D-S13D), SASA dynamic behavior suggests preferential confinement of the different ligands within the protein pocket, although Nsp15 is the one with the most variability in the area (~ 15 nm2), the rest of them just vary in ~ 15 nm2 from the mean SASA value.
Conclusion
The main idea of this work was to find a natural compound that could bind five different non-structural proteins of SARS-CoV-2 in order to stop virus infection by inhibition of key process targets, although finding a single target is a good option for the inhibition viral disease, a multitarget approach, as presented in this work, has some advantages such as inhibition mechanisms are multiple, combination with other drugs could lead to synergic effect, and multidrug assessment could be avoided.
In this way, amentoflavone was selected, among 478 flavonoids, as a multitarget inhibitor for being the only flavonoid capable of binding to the five non-structural proteins of SARS-CoV-2 (Nsp1, Nsp3, Nsp5, Nsp12, and Nsp15). We demonstrate with MD simulations high stability of the complex formed between the amentoflavone and the five Nsps throughout the 100-ns simulation. Furthermore, the information provided by the energy analysis showed that in order of affinity and efficacy, amentoflavone would have a greater capacity to inhibit Nsp5, followed by Nsp12, Nsp15, Nsp3, and finally Nsp1. Allowing us to emphasize that amentoflavone has the potential to function as a multitarget molecule. Nevertheless, due to the in silico approach, some of the variables influencing the binding process in vitro and in vivo could be avoided, so biological experiments are needed to corroborate the phenomena described here.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (PDF 2227 KB)
Acknowledgements
This work would not be achieved without the magnificent work done by all the people behind free software, who are also in charge of constantly developing updates and new applications that they share with the world in favor of free knowledge. Like the team behind PyMol, Discovery, Chimera, OpenBabel, GROMACS, AutoDock Tools, and Vina. Isabel Hidalgo would like to thank the Post-Doctoral Fellowship Program of CONACyT. Cristian Gonzalez-Ruiz would like to thank the Post-Doctoral Fellowship Program of UNAM.
Author contribution
All authors contributed to the study conception and design. Methodology development, data collection, and analysis were performed by Andrés Portilla-Martínez and Miguel Ortiz-Flores. Investigation, formal analysis, and results validation were performed by Cristian Gonzalez-Ruiz and Isabel Hidalgo. Supervision, project administration, and provision of resources were performed by Nayelli Nájera, Eduardo Meaney, and Guillermo Ceballos. The first draft of the manuscript was written by Miguel Ortiz-Flores and Andrés Portilla-Martínez and all authors commented on previous versions of the manuscript. Review and editing of the final version were performed by Nayelli Nájera and Guillermo Ceballos. All authors read and approved the final manuscript.
Data availability
The data sets used and/or analyzed during the current study are available from the corresponding author on request.
Code availability
All the software used and cited in the study is free and available online for use.
Declarations
Conflict of interest
Dr. Ceballos is a stockholder of Epirium, Inc.
Miguel Ortiz-Flores and Nayelli Nájera are co-senior authors of this work.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. World Health Organization (2022) WHO COVID-19 dashboard. In: https://covid19.who.int/
2. He F Deng Y Li W Coronavirus disease 2019: what we know? J Med Virol 2020 92 719 725 10.1002/jmv.25766 32170865
3. Bhavana V Thakor P Singh SB Mehra NK COVID-19: pathophysiology, treatment options, nanotechnology approaches, and research agenda to combating the SARS-CoV2 pandemic Life Sci 2020 261 118336 10.1016/j.lfs.2020.118336 32846164
4. Izda V Jeffries MA Sawalha AH COVID-19: a review of therapeutic strategies and vaccine candidates Clin Immunol 2021 222 108634 10.1016/j.clim.2020.108634 33217545
5. Brüssow H COVID-19: vaccination problems Environ Microbiol 2021 23 2878 2890 10.1111/1462-2920.15549 33928745
6. Giovanetti M Benedetti F Campisi G Evolution patterns of SARS-CoV-2: snapshot on its genome variants Biochem Biophys Res Commun 2021 538 88 91 10.1016/j.bbrc.2020.10.102 33199021
7. Rezagholizadeh A Khiali S Sarbakhsh P Entezari-Maleki T Remdesivir for treatment of COVID-19; an updated systematic review and meta-analysis Eur J Pharmacol 2021 897 173926 10.1016/j.ejphar.2021.173926 33549577
8. Sahebnasagh A Avan R Saghafi F Pharmacological treatments of COVID-19 Pharmacol Rep 2020 72 1446 1478 10.1007/s43440-020-00152-9 32816200
9. Rohaim MA el Naggar RF Clayton E Munir M Structural and functional insights into non-structural proteins of coronaviruses Microb Pathog 2021 150 104641 10.1016/j.micpath.2020.104641 33242646
10. Ross JA Kasum CM Dietary flavonoids: bioavailability, metabolic effects, and safety Annu Rev Nutr 2002 22 19 34 10.1146/annurev.nutr.22.111401.144957 12055336
11. Russo M, Moccia S, Spagnuolo C, et al (2020) Roles of flavonoids against coronavirus infection. Chem Biol Interact 328:. 10.1016/J.CBI.2020.109211
12. Yu S, Yan H, Zhang L, et al (2017) A review on the phytochemistry, pharmacology, and pharmacokinetics of amentoflavone, a naturally-occurring biflavonoid. Molecules 22:. 10.3390/MOLECULES22020299
13. Cai J, Zhao C, Du Y, et al (2019) Amentoflavone ameliorates cold stress-induced inflammation in lung by suppression of C3/BCR/NF-κB pathways. BMC Immunol 20:. 10.1186/S12865-019-0331-Y
14. Zhang Z Sun T Niu JG Amentoflavone protects hippocampal neurons: anti-inflammatory, antioxidative, and antiapoptotic effects Neural Regen Res 2015 10 1125 1133 10.4103/1673-5374.160109 26330838
15. Hwang IS Lee J Jin HG Amentoflavone stimulates mitochondrial dysfunction and induces apoptotic cell death in Candida albicans Mycopathologia 2012 173 207 218 10.1007/S11046-011-9503-X 22210020
16. Rong S, Wan D, Fan Y, et al (2019) Amentoflavone affects epileptogenesis and exerts neuroprotective effects by inhibiting NLRP3 inflammasome. Front Pharmacol 10:. 10.3389/FPHAR.2019.00856
17. Qin L Zhao Y Zhang B Li Y Amentoflavone improves cardiovascular dysfunction and metabolic abnormalities in high fructose and fat diet-fed rats Food Funct 2018 9 243 252 10.1039/C7FO01095H 29168869
18. Zheng W Zhang C Li Y Folding non-homologous proteins by coupling deep-learning contact maps with I-TASSER assembly simulations Cell Reports Methods 2021 1 100014 10.1016/j.crmeth.2021.100014 34355210
19. Ma C Sacco MD Xia Z Discovery of SARS-CoV-2 papain-like protease inhibitors through a combination of high-throughput screening and a FlipGFP-based reporter assay ACS Cent Sci 2021 7 1245 1260 10.1021/acscentsci.1c00519 34341772
20. Su H Yao S Zhao W Anti-SARS-CoV-2 activities in vitro of Shuanghuanglian preparations and bioactive ingredients Acta Pharmacol Sin 2020 41 1167 1177 10.1038/s41401-020-0483-6 32737471
21. Gao Y Yan L Huang Y (2020) Structure of the RNA-dependent RNA polymerase from COVID-19 virus Science 1979 368 779 782 10.1126/science.abb7498
22. Kim Y Wower J Maltseva N Tipiracil binds to uridine site and inhibits Nsp15 endoribonuclease NendoU from SARS-CoV-2 Communications Biology 2021 4 193 10.1038/s42003-021-01735-9 33564093
23. Burley SK Bhikadiya C Bi C RCSB Protein Data Bank: powerful new tools for exploring 3D structures of biological macromolecules for basic and applied research and education in fundamental biology, biomedicine, biotechnology, bioengineering and energy sciences Nucleic Acids Res 2021 49 D437 D451 10.1093/nar/gkaa1038 33211854
24. Pettersen EF Goddard TD Huang CC UCSF Chimera? A visualization system for exploratory research and analysis J Comput Chem 2004 25 1605 1612 10.1002/jcc.20084 15264254
25. Morris GM Huey R Lindstrom W AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility J Comput Chem 2009 30 2785 2791 10.1002/jcc.21256 19399780
26. Kanehisa M KEGG: Kyoto Encyclopedia of Genes and Genomes Nucleic Acids Res 2000 28 27 30 10.1093/nar/28.1.27 10592173
27. Shannon A Fattorini V Sama B A dual mechanism of action of AT-527 against SARS-CoV-2 polymerase Nat Commun 2022 13 621 10.1038/s41467-022-28113-1 35110538
28. O’Boyle NM Banck M James CA Open Babel: an open chemical toolbox Journal of Cheminformatics 2011 3 33 10.1186/1758-2946-3-33 21982300
29. Trott O Olson AJ AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading J Comp Chem NA-NA 2009 10.1002/jcc.21334
30. BIOVIA DS (2022) Discovery Studio, 2017R2
31. Abad-Zapatero C Ligand efficiency indices for effective drug discovery Expert Opin Drug Discov 2007 2 469 488 10.1517/17460441.2.4.469 23484756
32. Abraham MJ Murtola T Schulz R GROMACS: high performance molecular simulations through multi-level parallelism from laptops to supercomputers SoftwareX 2015 1–2 19 25 10.1016/j.softx.2015.06.001
33. Berendsen HJC van der Spoel D van Drunen R GROMACS: a message-passing parallel molecular dynamics implementation Comput Phys Commun 1995 91 43 56 10.1016/0010-4655(95)00042-E
34. Lemkul JA From proteins to perturbed Hamiltonians: a suite of tutorials for the GROMACS-2018 molecular simulation package, v1.0 Living J Comp Mol Sci 2018 1 1 5068
35. Schrodinger LLC (2015) The PyMOL Molecular Graphics System, version 1.8
36. Bouysset C Fiorucci S ProLIF: a library to encode molecular interactions as fingerprints Journal of Cheminformatics 2021 13 72 10.1186/s13321-021-00548-6 34563256
37. Mendez AS Ly M González-Sánchez AM The N-terminal domain of SARS-CoV-2 nsp1 plays key roles in suppression of cellular gene expression and preservation of viral gene expression Cell Rep 2021 37 109841 10.1016/J.CELREP.2021.109841 34624207
38. Schubert K Karousis ED Jomaa A SARS-CoV-2 Nsp1 binds the ribosomal mRNA channel to inhibit translation Nat Struct Mol Biol 2020 27 959 966 10.1038/S41594-020-0511-8 32908316
39. Shen Z, Zhang G, Yang Y, et al (2021) Lysine 164 is critical for SARS-CoV-2 Nsp1 inhibition of host gene expression. Journal of General Virology 102:. 10.1099/jgv.0.001513
40. Sakuraba S Xie Q Kasahara K Extended ensemble simulations of a SARS-CoV-2 nsp1–5’-UTR complex PLoS Comput Biol 2022 18 e1009804 10.1371/journal.pcbi.1009804 35045069
41. Thoms M Buschauer R Ameismeier M (2020) Structural basis for translational shutdown and immune evasion by the Nsp1 protein of SARS-CoV-2 Science 1979 369 1249 1255 10.1126/science.abc8665
42. Mielech AM Deng X Chen Y Murine coronavirus ubiquitin-like domain is important for papain-like protease stability and viral pathogenesis J Virol 2015 89 4907 4917 10.1128/JVI.00338-15 25694594
43. Ratia K Saikatendu KS Santarsiero BD Severe acute respiratory syndrome coronavirus papain-like protease: structure of a viral deubiquitinating enzyme Proc Natl Acad Sci U S A 2006 103 5717 5722 10.1073/PNAS.0510851103 16581910
44. Osipiuk J Azizi S-A Dvorkin S Structure of papain-like protease from SARS-CoV-2 and its complexes with non-covalent inhibitors Nat Commun 2021 12 743 10.1038/s41467-021-21060-3 33531496
45. Bhati S (2020) Structure-based drug designing of naphthalene based SARS-CoV PLpro inhibitors for the treatment of COVID-19. Heliyon 6:. 10.1016/J.HELIYON.2020.E05558
46. Báez-Santos YM st. John SE, Mesecar AD, The SARS-coronavirus papain-like protease: structure, function and inhibition by designed antiviral compounds Antiviral Res 2015 115 21 38 10.1016/j.antiviral.2014.12.015 25554382
47. Ratia K Pegan S Takayama J A noncovalent class of papain-like protease/deubiquitinase inhibitors blocks SARS virus replication Proc Natl Acad Sci U S A 2008 105 16119 16124 10.1073/PNAS.0805240105 18852458
48. Nallusamy S, Mannu J, Ravikumar C, et al (2021) Exploring phytochemicals of traditional medicinal plants exhibiting inhibitory activity against main protease, spike glycoprotein, RNA-dependent RNA polymerase and non-structural proteins of SARS-CoV-2 through virtual screening. Front Pharmacol 12:. 10.3389/FPHAR.2021.667704
49. Swargiary A Mahmud S Saleh MA Screening of phytochemicals as potent inhibitor of 3-chymotrypsin and papain-like proteases of SARS-CoV2: an in silico approach to combat COVID-19 J Biomol Struct Dyn 2020 10.1080/07391102.2020.1835729 33089730
50. Cho E Rosa M Anjum R Dynamic profiling of β-coronavirus 3CL M pro protease ligand-binding sites J Chem Inf Model 2021 61 3058 3073 10.1021/acs.jcim.1c00449 34124899
51. Su H, Yao S, Zhao W, et al (2020) Discovery of baicalin and baicalein as novel, natural product inhibitors of SARS-CoV-2 3CL protease in vitro; bioRxiv 2020.04.13.038687. 10.1101/2020.04.13.038687
52. Tan J Verschueren KHG Anand K pH-dependent conformational flexibility of the SARS-CoV main proteinase (Mpro) dimer: molecular dynamics simulations and multiple X-ray structure analyses J Mol Biol 2005 354 25 40 10.1016/j.jmb.2005.09.012 16242152
53. Patil R, Chikhale R, Khanal P, et al (2021) Computational and network pharmacology analysis of bioflavonoids as possible natural antiviral compounds in COVID-19. Inform Med Unlocked 22:. 10.1016/J.IMU.2020.100504
54. Ghosh R Chakraborty A Biswas A Chowdhuri S Computer aided identification of potential SARS CoV-2 main protease inhibitors from diterpenoids and biflavonoids of Torreya nucifera leaves J Biomol Struct Dyn 2020 10.1080/07391102.2020.1841680 33382027
55. Lokhande K Nawani NK Venkateswara S Pawar S Biflavonoids from Rhus succedanea as probable natural inhibitors against SARS-CoV-2: a molecular docking and molecular dynamics approach J Biomol Struct Dyn 2020 10.1080/07391102.2020.1858165 33300454
56. Varughese JK Joseph Libin KL Sindhu KS Investigation of the inhibitory activity of some dietary bioactive flavonoids against SARS-CoV-2 using molecular dynamics simulations and MM-PBSA calculations J Biomol Struct Dyn 2021 10.1080/07391102.2021.1891139 33618628
57. Mahmoudi S Balmeh N Mohammadi N Sadeghian-Rizi T The novel drug discovery to combat COVID-19 by repressing important virus proteins involved in pathogenesis using medicinal herbal compounds Avicenna J Med Biotechnol 2021 13 107 115 10.18502/AJMB.V13I3.6370 34484639
58. Sawant S, Patil R, Khawate M, et al (2021) Computational assessment of select antiviral phytochemicals as potential SARS-Cov-2 main protease inhibitors: molecular dynamics guided ensemble docking and extended molecular dynamics. In Silico Pharmacol 9:. 10.1007/S40203-021-00107-9
59. Yañez O, Osorio MI, Areche C, et al (2021) Theobroma cacao L. compounds: theoretical study and molecular modeling as inhibitors of main SARS-CoV-2 protease. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie 140:. 10.1016/J.BIOPHA.2021.111764
60. Deng X Hackbart M Mettelman RC Coronavirus nonstructural protein 15 mediates evasion of dsRNA sensors and limits apoptosis in macrophages Proc Natl Acad Sci 2017 114 E4251 E4260 10.1073/pnas.1618310114 28484023
61. Krishnan DA Sangeetha G Vajravijayan S Structure-based drug designing towards the identification of potential anti-viral for COVID-19 by targeting endoribonuclease NSP15 Informatics in Medicine Unlocked 2020 20 100392 10.1016/j.imu.2020.100392 32835078
62. Zhang L, Li L, Yan L, et al (2018) Structural and biochemical characterization of endoribonuclease Nsp15 encoded by Middle East respiratory syndrome coronavirus. Journal of Virology 92:. 10.1128/JVI.00893-18
63. Kim Y Jedrzejczak R Maltseva NI Crystal structure of Nsp15 endoribonuclease NendoU from SARS-CoV-2 Protein Sci 2020 29 1596 1605 10.1002/pro.3873 32304108
64. Ferreira De Freitas R Schapira M A systematic analysis of atomic protein-ligand interactions in the PDB Medchemcomm 2017 8 1970 1981 10.1039/c7md00381a 29308120
65. Hakami AR Bakheit AH Almehizia AA Ghazwani MY Selection of SARS-CoV-2 main protease inhibitor using structure-based virtual screening Future Med Chem 2022 14 61 79 10.4155/FMC-2020-0380 34814706
66. Farabi S Saha N Hasanuzzaman M Prediction of SARS-CoV-2 main protease inhibitors in medicinal plant-derived compounds by molecular docking approach J Adv Biotechnol Exp Ther 2020 3 79 10.5455/jabet.2020.d159
67. Ur Rehman MF, Akhter S, Batool AI, et al (2021) Effectiveness of natural antioxidants against SARS-CoV-2? Insights from the in-silico world. Antibiotics (Basel) 10:. 10.3390/ANTIBIOTICS10081011
68. Adegbola PI, Fadahunsi OS, Adegbola AE, Semire B (2021) In silico studies of potency and safety assessment of selected trial drugs for the treatment of COVID-19. In Silico Pharmacol 9:. 10.1007/S40203-021-00105-X
69. Elhady SS, Abdelhameed RFA, Malatani RT, et al (2021) Molecular docking and dynamics simulation study of Hyrtios erectus isolated scalarane sesterterpenes as potential SARS-CoV-2 dual target inhibitors. Biology (Basel) 10:. 10.3390/BIOLOGY10050389
70. Kukić P Nielsen JE Electrostatics in proteins and protein–ligand complexes Future Med Chem 2010 2 647 666 10.4155/fmc.10.6 21426012
71. Chen D, Oezguen N, Urvil P, et al (2016) Regulation of protein-ligand binding affinity by hydrogen bond pairing. Science Advances 2:. 10.1126/sciadv.1501240
| 36445575 | PMC9707096 | NO-CC CODE | 2022-12-08 23:16:06 | no | J Mol Model. 2022 Nov 29; 28(12):404 | utf-8 | J Mol Model | 2,022 | 10.1007/s00894-022-05391-6 | oa_other |
==== Front
Childs Nerv Syst
Childs Nerv Syst
Child's Nervous System
0256-7040
1433-0350
Springer Berlin Heidelberg Berlin/Heidelberg
36443475
5770
10.1007/s00381-022-05770-7
Review
Hydatid disease of the brain and spine
Padayachy L.C [email protected]
1
Ozek M.M 2
1 grid.49697.35 0000 0001 2107 2298 Department of Neurosurgery, Pediatric Neurosurgery Unit, Steve Biko Academic Hospital, University of Pretoria, Pretoria, South Africa
2 grid.411117.3 0000 0004 0369 7552 Department of Neurosurgery, Division of Pediatric Neurosurgery, Acibadem University, Istanbul, Turkey
29 11 2022
18
3 10 2022
16 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Hydatid disease of the central nervous system is relatively rare and comprises about 2–3% of all the hydatid cyst cases reported in the world. Spinal hydatid disease is an even rarer entity. It is endemic in sheep and cattle-raising regions, seen mainly in Mediterranean countries including Turkey and Syria. Pediatric neurosurgeons in non-endemic countries face a challenge when they encounter children with hydatid cysts of the central nervous system, mostly due to lack of awareness and the ensuing diagnostic dilemmas. It is also a significant socioeconomic problem in developing countries, due to improper hygiene and lack of dedicated veterinary practice. The clinical features are largely nonspecific and very according to location and severity of disease. However, with the advent of advances in MR imaging, the diagnostic accuracy of hydatic disease involving the brain and spine has increased. Intact removal of the cyst/s, without causing any spillage, and appropriate antihelminthic therapy is the goal and key to cure and prevention of recurrence. In this manuscript, the current literature on hydatid cyst of the brain and spine is reviewed to better understand the epidemiology, pathophysiology, diagnostic accuracy, and advances in therapeutic options. A heightened clinical suspicion, awareness of MR imaging features, improved surgical strategies, and options for prevention are discussed.
Keywords
Hydatid disease
Intracranial infestation
Spinal infection
Echinococcus
==== Body
pmcIntroduction
Echinococcosis is a disease of antiquity, with reports of human affliction by this condition dating back to Hippocrates and Galen, where it was thought to mainly affect the liver of slaughter animals [1, 2]. This disease, however, remains a relevant public health concern in the modern era, with considerable socioeconomic impact on human beings in several parts of the world [3]. The World Health Organization (WHO) has identified echinococcosis as one of the seventeen neglected diseases targeted for control or eradication by 2050 [4].
It is caused by the adult and larval stages of the tapeworm (cestode) belonging to the genus Echinococcus (family Taeniidae). The term Echinococcus was first described by Rudolphi [5]. Cestodes have a predilection for the central nervous system, with hydatid disease of the central nervous system being caused by infestation during the larval stage of the cestode Echinococcus granulosus. Other species, including Echinococcus multilocularis, Echinococcus vogeli, and Echinococcus oligarthrus, have been recognized, but have less relevance in the context of CNS infestation [6, 7]. Parasitic infestation of the central nervous system affects millions of people with a worldwide geographical distribution. A change in the distribution of these parasitic diseases has also become more evident in the era of globalization. Hydatid disease remains endemic in parts of Latin America, Australia, Mediterranean countries, the Middle East, and India, with prevalence rates ranging from 1.2 to 23 per 100,000 inhabitants, especially high where raising sheep, cattle, and other livestock are common practice [8]. The definitive hosts for Echinococcus are canines such as dogs, wolves, and foxes [9].
Hydatid disease of the brain
Pathology and pathophysiology
Echinococcus granulosus is the most common genus in humans, who inadvertently serve as intermediate hosts through the accidental ingestion of contaminated food. Viable parasite eggs form oncospheres in the human intestine. These viable Taenia ova later pass through the mucosa of the human intestinal tract into the portal system and spread hematogenously to other organs, mostly the liver, where 75% of the embryos are retained. A further 15% reach the lung and are retained by the pulmonary capillaries, with only 2% of these ova lodging in the brain and 1% in the spine [5, 10, 11]. The hydatid cyst itself has a wall composed of two layers, an inner layer of germinal epithelium (endocyst) and an outer layer laminated hyaline membrane (ectocyst). Hydatid cysts in the brain do not usually have the adventitial membrane (pericyst) surrounding the parasitic organism, which occurs in other parts of the body, unless there has been infection, injury, or rupture of the cyst after which a thick, even calcified adventitial membrane may develop (Fig. 1) [12–14]. Daughter cysts and scolices at various stages of development occur in the germinal layer and may detach from the cyst wall and settle at the base of the cyst, forming a sediment. The cyst fluid is generally colorless and contains albumin, glucose, chloride, lymphocytes, scolices, and hooks [5, 6, 8, 15].Fig. 1 Calcified adventitial membrane
Hydatid cysts of the brain are almost always spherical, with a wall that has a whitish to transparent, smooth, thin, slightly elastic consistency, and is made by the parasite and not the host [5]. Growth of the cyst in the brain usually occurs at about 1 cm/year but an increase of 5–10 cm within a year has been reported [16, 17]. While quite rare, hydatid cysts may degenerate and die, typically becoming turbid, with linear calcification, undergoing degeneration and finally shrinkage of the cyst [5, 16].
Cerebral involvement is quite rare, with the majority of cystic hydatid disease of the brain occurring in children, usually transmitted from dog saliva or uncooked contaminated vegetables [18–22]. Cysts are usually single, unilocular, and located supratentorially in the distribution of the middle cerebral artery, but can occur infratentorially [9, 23–26]. Cyst rupture is uncommon, but can occur spontaneously or secondary to trauma, usually resulting in the development of multiple secondary cysts [23, 25–28].
Clinical features
Most cerebral hydatid cysts occur in childhood and manifest clinically in early adulthood [29, 30]. The clinical presentation depends on the age of the child, size, number, and location of the cysts as well as the host immune response and temporal evolution of the cyst. Symptoms and signs related to raised intracranial pressure (ICP), i.e., headache, nausea, vomiting, papilledema, impaired level of consciousness, and focal neurological deficit, remain the most common presentation [17, 20, 22, 31, 32]. Seizures remain a less common presentation in this condition, with reports ranging from around 8 to 22% [33–37].
Diagnosis
Serological testing in hydatid disease has limited diagnostic accuracy, with reported sensitivities and specificities of 60–90% [4, 38]. Blood tests should include a C-reactive protein (CRP), estimated sedimentation rate (ESR), and complete blood count which are likely to only show an eosinophilia. The use of antigen tests, ELISA, indirect hemagglutination, and complement fixation tests demonstrate poor diagnostic accuracy, with sensitivities of 25–56% in extra-hepatic disease [4, 38].
Use of enriched antigen and recombinant antigen tests may increase the diagnostic accuracy, demonstrating sensitivities of over 90% on selected serum samples [4, 9, 10, 39]. Protein biomarkers and DNA-based detection methods, such as quantitative or nested PCR assays, provide better diagnostic accuracy. The Casoni skin test has been mostly replaced by serological tests, which are safer and have a better diagnostic accuracy.
Imaging forms the mainstay of diagnosis in patients with a suggestive history and clinical findings, and characteristic imaging findings may still support the diagnosis if serology is negative [9, 10, 39]. CT and MRI form the core imaging modalities for defining the features of cerebral hydatid cysts. Plain CT scan demonstrates a hypodense, cystic lesion with mass effect and midline shift, entrapment hydrocephalus is usually present. There is often not much surrounding edema. Cysts may be a single or multiple in appearance and calcification is thought to suggest previous cyst rupture or infection. Cerebral hydatid cysts appear as large, unilocular, thin-walled cysts, usually without calcification or surrounding edema. They contain fluid with a density similar to CSF on CT scan and MRI. Calcification on CT scan may signify death of the parasite, and irregularity of the cyst wall suggestive of previous rupture [14] (Fig. 2). T1W MRI usually demonstrates hypointense lesion/s, often with minimal surrounding perilesional edema, unless the cyst is infected and eliciting a host inflammatory response. The absence of contrast enhancement and edema are typical of hydatid cyst [40].Fig. 2 Axial CT scan demonstrating a large hydatid cyst in the left frontal hemisphere with irregular cyst wall
T2W MRI usually demonstrates a hyperintense cystic lesion. Intraoperative guidance using ultrasound to identify the cysts and guide the surgical approach and dissection can be a useful adjunct in this context [32, 41].
Management
Medical management
While the mainstay of hydatid disease of the brain involves surgical extirpation, the role of antihelminthic therapy in this context deserves unpacking. Mebendazole was initially used to treat hydatid cysts, but was later replaced by albendazole, which demonstrated improved gut absorption when compared to the former [42–44]. The recommended treatment regimens include albendazole at a dose of 12–15 mg/kg/day (given in 2 daily doses, ideally with fatty meals) for 3 months, or at 10–12 mg/kg/day for recurrent 1 month courses, followed by a “rest period” of 15 days after each month [34, 42–44]. The optimal duration of treatment is still unclear. Side effects include nausea, vomiting, diarrhea, headache, dizziness, reversible hepatotoxicity, and other gastro-intestinal disturbances, and laboratory monitoring, including liver function tests, should be checked at 2-week intervals for the first 3 months, then monthly is advised [45–47].
Praziquantel has demonstrated effectivity as an antihelminthic drug, but does not appear to have a definitive role as a primary therapeutic agent. There is some evidence to suggest that combination therapy with praziquantel and albendazole is more beneficial than albendazole alone [6, 7].
The use of antihelminthic therapy has been recommended in cases of recurrent disease, hydatid dissemination, lesions considered inoperable, or with intraoperative cyst rupture [32]. Use of systemic albendazole before and after surgery has been shown to reduce the recurrence rate [43, 48]. While seizures are rare in cerebral hydatid disease, treatment with appropriate anti-epileptic drugs may be required [32, 48].
Surgical management
Surgery remains the cornerstone of treatment for cerebral hydatid disease, with the goal being intact cyst removal without spillage of the contents. Dowling’s technique remains a popular method of “in toto” cyst extirpation using hydrostatic dissection to define the plane between the cyst wall and brain [49] (Fig. 3). The principles of surgery for intact extirpation of cerebral hydatid cysts mandate an adequate, large craniotomy, meticulous dural opening (especially for superficial cysts), and careful microsurgical dissection of the overlying and surrounding cortex to expose the cyst wall. The finest bore catheters are gently positioned between the cyst wall and the surgical patty-lined underlying cortex (Fig. 4). Normal saline (0.9%) can be used for hydro-dissection of the cyst wall in order to separate it from the brain parenchyma (Fig. 5). Subtle head position adjustment and elevation together with minimal Valsalva by the anesthetist, done timeously, may encourage cyst extirpation (Fig. 6). Hypertonic saline (5%) soaked swabs lining the cyst border are helpful to minimize spillage of the cyst content in cases where there is inadvertent cyst rupture. Always be vigilant to avoid excessive systemic absorption of the hypertonic saline solution. Intraoperative ultrasound may be a useful adjunct in defining the borders of the cyst, and identifying deeply located cysts. Cyst rupture during surgery is almost inevitably associated with recurrence and is an indication for commencing antihelminthic therapy [48]. The consequences of cyst rupture include dissemination, recurrence, and possibly an anaphylactic reaction. Cyst aspiration has been described, but should be considered only when intact removal of the cyst is not possible. Hemostasis, watertight dural closure, and layered tissue and skin closure are essential to limit surgical complications. Usually the brain re-expands to a remarkable extent within months but occasionally postoperative complications such as pneumocephalus, subdural collections or porencephalic cysts, seizures, and transient neurological deficits may occur. Mass effect from the subdural collection or porencephalic cyst may require shunt placement [48–50].Fig. 3 Dowliing's technique used to remove an intact hydatid cyst
Fig. 4 Intra-operative microsurgical dissection to remove a large hydatid cyst
Fig. 5 Intact hydatid cyst demonstrating exocyst membrane
Fig. 6 Large hydatid cyst post-evacuation
Hydatid disease of the spine
Spinal hydatid cysts in children account for 1% of all hydatid disease, and remains a very rare occurrence [51–63]. There are subsequently very few cases of spinal hydatid disease reported in children and adolescents [61, 62, 64–66], which may be partially due to underreporting of this condition, especially in developing countries. Hydatid disease affecting the spine occurs mostly extradurally, often as multiple cysts [55, 62]. Intradural extramedullary cysts are exceedingly rare, with vertebral body involvement occurring in around 0.5–2% of cases [52–55, 60, 62]. Intramedullary hydatid cysts are the rarest form of this disease, and have been reported in only a few case reports [67–69]. Spinal involvement usually results from direct extension from the abdomen, chest cavity, or pelvis, mostly affecting the dorsal area of the spine. The most commonly affected regions of the spine are thoracic (52%), lumbar (37%), cervical (5.5%), and sacral spine (5.5%) [38, 66, 70–72]. Spinal hydatid disease can be anatomically and radiologically classified into paraspinal, spinal, and intraspinal, with intraspinal lesions further divided into extradural, intradural extramedullary, or intramedullary (Table 1) [70], or classified according to the route of infection, i.e., primary or secondary disease [38].Table 1 Classification of spinal hydatid disease by Braithwaite and Lees [70]
Type 1 Hydatid cyst is intramedullary
Type 2 Hydatid cyst is intradural and extramedullary
Type 3 Hydatid cyst is extradural and intraspinal
Type 4 Hydatid cyst is in the vertebral body
Type 5 Hydatid cyst is paravertebral
Approximately 50% of cases involving the vertebral body also have involvement of the spinal canal, but as the disease spreads under the periosteum and ligaments, the intervertebral discs are spared. The cyst grows slowly, at the rate of 7 mm per month, and results in bone destruction by expansion and mechanical compression, causing ischemia of the nutrient vessels leading to bony necrosis, sequestra formation, and osteoclast proliferation [62]. Intradural extramedullary hydatid disease can present as a giant cystic lesion mimicking an arachnoid cyst [71, 72]. Extradural spread of hydatid cysts through widened neural foramina into the muscle planes may result in a grape-like appearance [54, 58, 66]. Hydatid cyst in the sacral region may mimic an anterior sacral meningocele [73], and purely ventral, extradural hydatid cyst of the spine with no extension into the dura has also been reported [53].
Clinical features
These features can be nonspecific and vary with the location of the cyst. The duration of clinical symptoms can range from acute to prolonged onset. Most commonly, these include back pain, paraparesis, radiculopathy, sensory disturbance, and sphincter involvement. Paraplegia is seen in 26% [55]. Spinal hydatid disease in children, while rarer than adults, has been reported [38, 74–77].
Imaging
There are no clearly distinguishing radiological findings. Plain X-ray findings are nonspecific, but may show bony destruction, typically of a moth-eaten, osteolytic nature and a soft tissue mass, at single or multiple levels.
CT scan of the spine more efficiently demonstrates the bony erosion and the extent of the lesion. Spinal deformity, if present, is also best demonstrated on CT scan. Calcification demonstrated as a “double layer of arcuate calcification” may be helpful in diagnosing echinococcus [38]. MRI remains the investigation of choice and shows well-circumscribed, cystic lesions, with CSF-like signal intensities, hypointense on T1-weighted imaging, and hyperintense on T2-weighted imaging. T2-weighted images show a low-intensity rim surrounding the homogeneous hyperintense cyst contents (Fig. 7a, b). The cyst wall may be thin and regular, isointense, or demonstrate a slightly lower signal than its contents. A markedly hypointense cyst wall on T1- and T2-weighted MR images is characteristic of hydatid disease. There may be mild enhancement after gadolinium injection, reflecting the vascularity of the pericyst. The differential diagnosis includes arachnoid cyst, arachnoiditis, cystic tumor, tuberculosis, and cysticercosis [62, 72].Fig. 7 a Sagittal T2-MRI demonstrating a cystic (neuroepithelial cyst) lesion ventral to the brainstem and spinal cord. b Post-operative sagittal T2-MRI demonstrating evacuation of the cyst decompression of the cord
Management
Treatment is fundamentally surgical [55, 61, 62, 66]. A laminectomy with total and intact cyst extirpation remains the goal. Precaution must always be taken to avoid spillage of the cyst contents into the intradural or intraspinal space. Hypertonic saline soaked swabs should line the cavity as it destroy the parasites from the cyst fluid. Intact cyst removal remains challenging with vertebral body involvement. When the invasive nature of the infestation in the spine precludes total removal, atraumatic cyst aspiration, especially in multiple extradural cysts, has been described, but remains a secondary option [51, 55, 60, 62]. Correction of spinal deformity in children, where this occurs, usually following multilevel laminectomies, in addition to intact removal of cysts, appropriate anterior, and posterior instrumented stabilization techniques may be indicated [55, 77]. It can often be done in a single setting, preferably under the guidance of intraoperative neuromonitoring. A posterior approach with costotransversectomy is preferred in a kyphotic deformity especially for the resection of the posterior wall of the bony apex [55].
Drug therapy in spinal hydatid disease
Albendazole remains the drug of choice, and the treatment regimens are the same as those for intracranial hydatid disease, as described earlier, where use of albendazole after surgery has been suggested to delay recurrence and reduce complications [51, 59, 61, 62]. Although the reported recurrence rate of spinal hydatid disease ranges between 40 and 90%, patients have an acceptable quality of life following surgery [38, 55, 61, 62, 65]. The risk factors for recurrence include infiltrative and extensive disease, intraoperative cyst rupture, and anatomical location of the disease. Peri-operative use of scolicidal and antihelminthic agents have been advocated to minimize the risk of recurrence in hydatid disease of the CNS [74–76].
Advances in management
While most are still in their early stages, there have been several developments in the application of diagnostic technologies which have already improved our understanding of hydatid disease and may be useful for informing new public health strategies aimed at prevention and improved intervention. These developments mostly involve gene transcriptome and proteomic analysis, identification of hormone and cytokine-activated pathways, and improved vaccine development [4].
Conclusion
Hydatid disease should be considered part of the differential diagnosis when dealing with cystic lesions of the brain and spine, especially in endemic regions. Early diagnosis with a high index of suspicion, definitive surgical treatment with intact cyst removal, avoiding intraoperative rupture, followed by a postoperative regimen of appropriate antihelminthic treatment, yields the best outcome.
Author contribution
LP — writing, editing, and images. MO — editing and images.
Availability of data and material
All data and material available as required.
Declarations
Ethics approval and consent to participate
Informed consent was not required as this is a review article.
Consent for publication
No consent required as this is a review article.
Conflict of interest
The authors have no conflict of interest or funding to declare for this manuscript.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Fuchs R (1895) Hippokrates. Sämtliche Werke, München: Lüneburg
2. Hosemann G (1928) Die Echinokokkenkrankheit. Enke
3. Eckert J Thompson RC Historical aspects of echinococcosis Adv Parasitol 2017 1 95 1 64
4. Wen H Vuitton L Tuxun T Li J Vuitton DA Zhang W McManus DP Echinococcosis: advances in the 21st century Clin Microbiol Rev 2019 32 2 e00075 e118 10.1128/CMR.00075-18 30760475
5. Abbasioun K Amirjam SA Diagnosis and management of hydatid cyst of the central nervous system Neurosurgery 2001 11 1 9 10.1097/00013414-200103000-00001
6. McManus DP Zhang W Li J Bartley PB Echinococcosis The lancet 2003 362 9392 1295 1304 10.1016/S0140-6736(03)14573-4
7. Moro P Schantz PM Echinococcosis: a review Int J Infect Dis 2009 13 2 125 133 10.1016/j.ijid.2008.03.037 18938096
8. Taratuto AL Venturiello SM Echinococcosis Brain Pathol 1997 7 1 673 679 10.1111/j.1750-3639.1997.tb01082.x 9034573
9. Eckert J Deplazes P Biological, epidemiological and clinical aspects of echinococcosis, a zoonosis of increasing concern Clin Microbiol Rev 2004 17 107 135 10.1128/CMR.17.1.107-135.2004 14726458
10. Beggs I The radiology of hydatid disease Am J Roentgenol 1985 145 639 648 10.2214/ajr.145.3.639 3895873
11. Tuzun M Hekimoglu B Hydatid disease of the CNS: imaging features Am J Roentgenol 1998 171 1497 1500 10.2214/ajr.171.6.9843277 9843277
12. Alvarez F Blazquez MG Oliver B Manrique M Calcified cerebral hydatid cyst Surg Neurol 1982 17 3 163 164 10.1016/0090-3019(82)90267-1 7079935
13. Obrador S Urquiza P Two cases of cerebral abscess of unusual nature: tuberculous abscess and suppurated hydatid cyst J Neurosurg 1948 5 6 572 576 10.3171/jns.1948.5.6.0572 18103788
14. Peter JC Domingo Z Sinclair-Smith C de Villiers JC Hydatid infestation of the brain difficulties with computed tomography and surgical treatment Pediatr Neurosurg 1994 20 78 83 10.1159/000120769 8142287
15. Dew HR Primary cerebral hydatid disease Aust N Z J Surg 1955 24 3 161 171 10.1111/j.1445-2197.1955.tb05089.x 14363064
16. Kemaloglu S Ozkan U Bukte Y Acar M Ceviz A Growth rate of cerebral hydatid cyst, with a review of the literature Child’s Nerv Syst 2001 17 743 745 10.1007/s003810100498 11862441
17. Sierra J Oviedo J Berthier M Leiguarda R Growth rate of secondary hydatid cysts of the brain: case report J Neurosurg 1985 62 5 781 782 10.3171/jns.1985.62.5.0781 3989600
18. Al Zain TJ Al-Witry SH Khalil HM Multiple intracranial hydatosis Acta Neurochir (Wien) 2002 144 1179 1185 10.1007/s00701-002-0987-5 12434174
19. Iplikcioglu A, Ozek M, Ozer A, Ozgen T (1992) Periventricular hydatid cyst presenting with hemichorea. Child's Nerv Syst 8:292–293
20. Fieggen G, Padayachy LC (2015) Chapter 80 - Tuberculosis, parasitic infestation and fungal infections. In: Principals and practice of pediatric neurosurgery, Thieme
21. Khaldi M Mohamed S Kallell J Brain hydatosis: report on 117 cases Childs Nerv Syst 2000 16 765 769 10.1007/s003810000348 11151730
22. Padayachy LC Fieggen AG Infections in the immunocompromised child Child’s Nervous System 2018 34 10 1989 1996 10.1007/s00381-018-3893-8 29995268
23. Arana-Iñiguez R San JJ Hydatid cysts of the brain J Neurosurg 1955 12 4 323 335 10.3171/jns.1955.12.4.0323 14392485
24. Alok R, Mahmoud J (2020) Successsful surgical treatment of a brain stem hydatid cyst in a child. Case Rep in Surg 2020
25. Banzo J Diaz FJ Pina JI Abós MD Rios G Garcia D Marin F Multiple cerebral hydatid cysts Eur J Nucl Med 1984 9 12 561 563 10.1007/BF00256857 6526056
26. Rahimizadeh A Hadadian K Hydatid cyst of the fourth ventricle Neurosurgery 1984 14 6 787 788 10.1097/00006123-198406000-00028 6462420
27. Abdulla K Tapoo AK AGHA HA Ruptured cerebral hydatid cyst: a case report J Trop Med Hyg 1988 91 6 302 305 3204645
28. Begg NC Begg AC Eobinson EG Primary hydatid disease of the brain-its diagnosis, radiological investigation, treatment and prevention N Z Med J 1957 56 312 84 98 13430962
29. Andronikou S Welman CJ Kader E Classic and unusual appearances of hydatid disease in children Pediatr Radiol 2002 32 11 817 828 10.1007/s00247-002-0785-5 12389111
30. Nemati A Kamgarpour A Rashid M Sohrabi NS Giant cerebral hydatid cyst in a child—a case report and review of literature BJMP 2010 3 3 a338
31. Ersahin Y Mutluer S Güzelbag E Intracranial hydatid cysts in children Neurosurgery 1993 33 2 219 225 10.1097/00006123-199308000-00006 8367043
32. Padayachy LC Dattatraya M Hydatid disease (Echinococcus) of the central nervous system Child’s Nervous System 2018 34 10 1967 1971 10.1007/s00381-018-3883-x 29961084
33. Altinörs N Bavbek M Caner HH Erdoğan B Central nervous system hydatidosis in Turkey: a cooperative study and literature survey analysis of 458 cases J Neurosurg 2000 93 1 1 8 10.3171/jns.2000.93.1.0001 10883898
34. Braunsdorf EW Schmidt D Rautenberg M Cerebral manifestation of hydatid disease in a child Child’s Nervous System 1988 4 4 249 251 10.1007/BF00270924 3167880
35. Ciurea AV Vasilescu G Nuteanu L Carp N Cerebral hydatid cyst in children Childs Nerv Syst 1995 11 12 679 685 10.1007/BF00262230 8750949
36. Kocaman S Ersabin Y Mutluer S Cerebral hydatid cysts in children J Neurosci Nurs 1999 31 5 270 10.1097/01376517-199910000-00003 10633303
37. Krajewski R Stelmasiak Z Cerebral hydatid cysts in children Child’s Nervous System 1991 7 3 154 155 10.1007/BF00776712 1878870
38. Sioutis S Reppas L Bekos A Soulioti E Saranteas T Koulalis D Sapkas G Mavrogenis AF Echinococcosis of the spine EFORT Open Reviews 2021 6 4 288 296 10.1302/2058-5241.6.200130 34040806
39. Turgut M Intracranial hydatosis in Turkey: its clinical presentation, diagnostic studies, surgical management and outcome. A review of 276 cases Neurosurg Rev 2001 24 4 200 208 10.1007/s101430100168 11778827
40. Kovoor JM Thomas RD Chandrashekhar HS Jayakumar PN Pillai S Shankar SK Neurohydatidosis Australas Radiol 2007 51 5 406 411 10.1111/j.1440-1673.2007.01860.x 17803790
41. Padayachy LC Fieggen G Intraoperative ultrasound-guidance in neurosurgery World Neurosurgery 2014 82 3 e409 e411 10.1016/j.wneu.2013.09.052 24099676
42. De Rosa F Teggi A Treatment of Echinococcus granulosus hydatid disease with albendazole Ann Trop Med Parasitol 1990 84 5 467 472 10.1080/00034983.1990.11812496 2256768
43. Saimot AG Cremieux AC Hay JM Meulemans A Giovanangeli MD Delaitre B Coulaud JP Albendazole as a potential treatment for human hydatidosis The Lancet 1983 322 8351 652 656 10.1016/S0140-6736(83)92533-3
44. Teggi A Capozzi A Rosa FD Treatment of Echinococcus granulosus hydatid disease with mebendazole J Chemother 1989 1 5 310 317 10.1080/1120009X.1989.11738914 2585032
45. Brunetti E Kern P Vuitton DA Expert consensus for the diagnosis and treatment of cystic and alveolar echinococcosis in humans Acta Trop 2010 114 1 1 6 10.1016/j.actatropica.2009.11.001 19931502
46. Nabarro LE Amin Z Chiodini PL Current management of cystic echinococcosis: a survey of specialist practice Clin Infect Dis 2015 60 5 721 728 10.1093/cid/ciu931 25422388
47. Vuitton DA Benzimidazoles for the treatment of cystic and alveolar echinococcosis: what is the consensus? Expert Rev Anti Infect Ther 2009 7 2 145 149 10.1586/14787210.7.2.145 19254162
48. Duishanbai S Dangmuren J Guo H Intracranial hydatid cysts in children: report of 30 cases Childs Nerv Syst 2010 26 821 827 10.1007/s00381-009-1008-2 20013283
49. Carrea R Dowling E Jr Guevera A Surgical treatment of hydatid cysts of the central nervous system in the pediatric age (Dowling’s technique) Childs Brain 1975 11 4 21
50. Izci Y Tüzün Y Secer HI Cerebral hydatid cysts: technique and pitfalls of surgical management Neurosurg Focus 2008 24 6 E15 10.3171/FOC/2008/24/6/E15 18518745
51. Bhake A Agrawal A Hydatid disease of the spine J Neurosci Rural Prac 2010 1 2 61 62 10.4103/0976-3147.71715
52. Dagtekin A Koseoglu A Kara E Karabag H Avci E Torun F Bagdatoglu C Unusual location of hydatid cysts in pediatric patients Pediatr Neurosurg 2009 45 5 379 383 10.1159/000260908 19940536
53. Dogan I Kahilogullari G Guner E Unlu A A rare and unexpected clinical progress and location on a primary extradural spinal hydatid cyst in a pediatric patient: a case report Childs Nerv Syst 2015 8 1407 1411 10.1007/s00381-015-2728-0
54. Eloqayli H Matalka I Daoud S Primary spinal extradural hydatid cyst in a 4-year-old child Br J Neurosurg 2010 24 5 602 603 10.3109/02688697.2010.487134 20632879
55. Güneçs M Akdemir H Tuğcu B Günaldi O Gümüçs E Akpinar A Multiple intradural spinal hydatid disease: a case report and review of literature Spine (Phila Pa 1976) 2009 34 9 E346 E350 10.1097/BRS.0b013e3181a01b0f 19531992
56. Kalkan E Cengiz SL Ciçek O Erdi F Baysefer A Primary spinal intradural extramedullary hydatid cyst in a child J Spinal Cord Med 2007 30 3 297 300 10.1080/10790268.2007.11753941 17684899
57. Karadereler S Orakdögen M Kiliç K Ozdogan C Primary spinal extradural hydatid cyst in a child: case report and review of the literature Eur Spine J 2002 11 5 500 503 10.1007/s00586-002-0411-0 12384760
58. Lakhdar F Arkha Y Rifi L Derraz S El Ouahabi A El Khamlichi A Spinal intradural extramedullary hydatidosis: report of three cases Neurosurgery 2009 65 2 372 376 10.1227/01.NEU.0000349751.39742.E3 19625917
59. Limaiem F Bellil S Bellil K Chelly I Mekni A Khaldi M Haouet S Zitouna M Kchir N Primary hydatidosis of the central nervous system: a retrospective study of 39 Tunisian cases Clin Neurol Neurosurg 2010 112 1 23 28 10.1016/j.clineuro.2009.09.001 19781849
60. Midyat L Gökçe S Onder A Ozdemir Y Mursalov G Mir S A very rare cause of childhood paraparesis: primary intradural extramedullary spinal hydatid cyst Pediatr Infect Dis J 2009 28 8 754 755 10.1097/INF.0b013e31819c8004 19512955
61. Rumana M Mahadevan A Nayil Khurshid M Kovoor JM Yasha TC Santosh V Indira B Shankar SK Cestode parasitic infestation: intracranial and spinal hydatid disease—a clinicopathological study of 29 cases from South India Clin Neuropathol 2006 25 2 98 104 16550744
62. Singh S Sardhara J Singh AK Srivastava AK Bhaisora KS Das KK Mehrotra A Sahu RN Jaiswal AK Behari S Spinal intradural hydatid cyst causing arachnoiditis: a rare etiology of cauda equina syndrome J Craniovertebr Junction Spine 2016 7 4 282 284 27891041
63. Turgut AT Turgut M Intradural extramedullary primary hydatid cyst of the spine in a child: a very rare presentation Eur Spine J 2009 18 8 1234 1235 10.1007/s00586-009-1006-9 19396476
64. Bhojraj SY Shetty NR Primary hydatid disease of the spine: an unusual cause of progressive paraplegia. Case report and review of the literature J Neurosurg 1999 91 2 Suppl 216 218 10505508
65. Jaiswal S Jaiswal AK Jain M Behari S Pandey R Primary spinal extradural hydatid cyst causing paraplegia Indian J Pathol Microbiol 2009 52 3 432 433 10.4103/0377-4929.55018 19679985
66. Sharma A Kashyap V Abraham J Kurian S Intradural hydatid cysts of the spinal cord Surg Neurol 1981 16 235 237 10.1016/0090-3019(81)90014-8
67. Ley A Marti A Intramedullary hydatid cyst: case report J Neurosurg 1970 33 4 457 459 10.3171/jns.1970.33.4.0457 5471936
68. Şenol MG Tekeli HK Mustafa Tansel K Serdar TV Güner S Saracoglu M Intramedullary hydatid cyst of the cervical spine Indian J Med Microbiol 2012 30 4 480 481 10.4103/0255-0857.103780 23183479
69. Zhang Z Fan J Dang Y Xu R Shen C Primary intramedullary hydatid cyst: a case report and literature review Eur Spine J 2017 26 1 107 110 10.1007/s00586-016-4896-3 27896531
70. Braithwaite PA Lees RF Vertebral hydatid disease: radiological assessment Radiology 1981 140 3 763 766 10.1148/radiology.140.3.7280247 7280247
71. Rashid M Kirmani S Rashid M Giant intradural extramedullary spinal hydatid cyst—a rare presentation Clin Imaging 2012 36 6 881 883 10.1016/j.clinimag.2011.12.011 23154030
72. Secer HI Anik I Celik E Daneyemez MK Gonul E Spinal hydatid cyst mimicking arachnoid cyst on magnetic resonance imaging J Spinal Cord Med 2008 31 1 106 108 10.1080/10790268.2008.11753990 18533421
73. Hemama M Lasseini A Rifi L Boutarbouch M Derraz S Ouahabi AE Khamlichi AE A sacral hydatid cyst mimicking an anterior sacral meningocele J Neurosurg Pediatr 2011 8 5 526 529 10.3171/2011.8.PEDS10555 22044380
74. Ndondo AP Fieggen G Wilmshurst J Hydatid disease of the spine in South African children J Child Neurol 2003 18 5 343 346 10.1177/08830738030180050401 12822819
75. Ozek MM Complications of central nervous system hydatid disease Pediatr Neurosurg 1994 20 84 91 10.1159/000120770 8142288
76. Pamir MN Ozduman K Elmaci I Spinal hydatid disease Spinal Cord 2002 40 153 160 10.1038/sj.sc.3101214 11965552
77. Thaler M Gabl M Lechner R Gstöttner M Bach CM Severe kyphoscoliosis after primary Echinococcus granulosus infection of the spine Eur Spine J 2010 19 9 1415 1422 10.1007/s00586-010-1398-6 20514501
| 36443475 | PMC9707099 | NO-CC CODE | 2022-12-01 23:20:20 | no | Childs Nerv Syst. 2022 Nov 29;:1-8 | utf-8 | Childs Nerv Syst | 2,022 | 10.1007/s00381-022-05770-7 | oa_other |
==== Front
J Intell Inf Syst
J Intell Inf Syst
Journal of Intelligent Information Systems
0925-9902
1573-7675
Springer US New York
768
10.1007/s10844-022-00768-8
Article
Multi-class classification of COVID-19 documents using machine learning algorithms
https://orcid.org/0000-0002-1212-0101
Rabby Gollam [email protected]
https://orcid.org/0000-0003-0464-2257
Berka Petr [email protected]
grid.266283.b 0000 0001 1956 7785 Department of Information and Knowledge Engineering, Prague University of Economics and Business, Prague, Czech Republic
29 11 2022
121
10 6 2022
16 11 2022
17 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
In most biomedical research paper corpus, document classification is a crucial task. Even due to the global epidemic, it is a crucial task for researchers across a variety of fields to figure out the relevant scientific research papers accurately and quickly from a flood of biomedical research papers. It can also assist learners or researchers in assigning a research paper to an appropriate category and also help to find the relevant research paper within a very short time. A biomedical document classifier needs to be designed differently to go beyond a “general” text classifier because it’s not dependent only on the text itself (i.e. on titles and abstracts) but can also utilize other information like entities extracted using some medical taxonomies or bibliometric data. The main objective of this research was to find out the type of information or features and representation method creates influence the biomedical document classification task. For this reason, we run several experiments on conventional text classification methods with different kinds of features extracted from the titles, abstracts, and bibliometric data. These procedures include data cleaning, feature engineering, and multi-class classification. Eleven different variants of input data tables were created and analyzed using ten machine learning algorithms. We also evaluate the data efficiency and interpretability of these models as essential features of any biomedical research paper classification system for handling specifically the COVID-19 related health crisis. Our major findings are that TF-IDF representations outperform the entity extraction methods and the abstract itself provides sufficient information for correct classification. Out of the used machine learning algorithms, the best performance over various forms of document representation was achieved by Random Forest and Neural Network (BERT). Our results lead to a concrete guideline for practitioners on biomedical document classification.
Keywords
Multi-class classification
Machine learning algorithms
Text mining
COVID-19
University of Economics Prague, Czech RepublicIGA 40/2021 Rabby Gollam
==== Body
pmcIntroduction
Text mining or text analytics can be understood as data mining on textual documents. So, text mining aims at discovering novel, interesting and useful patterns and new insights in large collections of texts. Typical text mining tasks are text categorization (i.e., classification of documents into different classes), document clustering (i.e., grouping documents according to their similarity), or document filtering (e.g., classification of documents into two classes like interesting vs. uninteresting documents, or spam vs. ham). The main problem in text mining is the question of the suitable representation of unstructured texts to be analyzed using machine learning (ML) algorithms. Most of the algorithms can process only structured data organized into a single data table containing a fixed number of columns – features representing the instances (documents in the case of text mining) that are stored as rows in the table. A typical pipeline of text processing that looks for suitable features consists of lexical analysis (to identify individual words), lemmatization or stemming (to transform inflected words to their base form), and stop-words removal (to remove words not related to the content of the document). This process results in obtaining tokens (terms, words) that are used as features in the data table. Each document is then represented as a vector of a fixed length that contains information about the tokens that occur in the document for the bag-of-word (BOW) representation. The tokens can be encoded in the vector as binary values (yes/no occurrence in the document), numbers of occurrences in the document, or TF-IDF (term frequency-inverse document frequency) values. So, each document is represented using a large (many components) sparse (most values are zero as a particular token does not appear in the document) numeric vector. As this yields the “Curse of dimensionality” phenomenon, some dimensionality reduction methods are usually applied to the document vectors, either feature selection or feature transformation.
There are several drawbacks to the BOW representation. BOW cannot handle multi-word phrases and, in its basic form, cannot reflect the different importance of various parts of the text. The first problem can be dealt with by n-grams; instead of using single tokens to create features, we can create tokens for N subsequent words. The second problem can be handled by weighting tokens according to their appearance in different parts of the document or by considering only important parts of the documents. So e.g., we can expect that title, abstract, and keywords, which are obligatory parts of scientific papers, are closely related to the content of the paper; so only these parts can be considered when defining the tokens. There are also some techniques that can enhance the basic BOW representation: entity detection or semantic expansion (Li et al., 2021). It is also possible to use word embeddings. This relatively new concept from computational linguistics aims at describing semantic similarities between linguistic items (words) using their co-occurrence in large textual databases.
Text mining can be applied in a wide range of application domains. As our work is oriented on the multi-class classification of medical papers, we review some work in this area. Gani et al. (2016) used Naïve Bayes, Decision Tree, Support Vector Machine, and Stochastic Gradient Descent algorithm to classify documents related to 23 cardiovascular diseases taken from the MEDLINE database. They used BOW representation, where tokens are represented using TF-IDF values, and achieved the best classification accuracy of 76% for SVM and 3000 selected features (Jindal and Taneja, 2015b). Yan et al. describes a Convolutional Neural Network framework (B-CNN) for biomedicine semantic indexing (Yan et al., 2018). The proposed CNN architecture can adaptively deal with features of documents and can capture context information. They extend the features created using the BOW model by word sequence embedding. As they reported, this extension can improve the classification performance of CNN models but not of simpler models like Naive Bayesian classifier or Logistic Regression (Balaji et al., 2020). Another example application of the CNN model to text classification is presented in Balaji et al. (2020). The authors of this paper used the word embedding method Word2vec for texts retrieved from the PubMed databases word embeddings were used to train a sentence-level classifier for texts that belong to one of 26 medical categories. Mujtaba et al. used a small subset of MEDLINE documents belonging to top-10 disease categories to Compare Bayesian Network, Decision Tree, and Random Forest models on a multi-class classification problem. They report, that in this particular task, Bayesian Network, when used on BOW representation without stemming, achieved the best performance (Mujtaba et al., 2019). Jindal and Taneja propose a lexical approach to text categorization in the biomedical domain. They represent the documents using tokens that are derived from words in the abstract by matching them with keywords taken from MeSH. They then used a modified K-Nearest Neighbor algorithm for the multi-class classification of the documents (Jindal and Taneja, 2015a). Elberichi, Amel, and Malika used MeSH ontology to enhance document representation of papers taken from biomedical benchmark text corpus Ohsumed (Elberrichi et al., 2012). They show that using hypernyms derived from ontology as additional features in document representation can improve classification performance.
Objectives
The work reported in the paper deals with the multi-class classification of biology and medical research papers published on issues related to the COVID-19 pandemic. As this pandemic threatens people in countries all over the world, a huge number of papers that refer to COVID-19 have been published. It is impossible to read all these research papers. So in our work, we try to classify these research papers depends not only on the bibliometric information but also on utilizing the internal context of these research papers. We then performed several ML experiments for different methods of document representation with the aim to identify the suitable combination of representation scheme and algorithm to classify COVID-19 related research papers into these classes.
Methods
The document processing pipeline is shown in Fig. 1. After the data enhancement, we extracted different types of entities from the different parts of the research papers: title, and abstract. We also consider various methods of encoding the features to represent the documents. These text pre-processing steps result in different data tables used in the machine learning experiments. We assess not only the quality of the created models but also aim at the interpretation of the results. The next subsections show details for each of these steps. Fig. 1 Overview of methodological pipeline
Corpora details
The LitCovid corpus is a collection of published PubMed research papers that are directly related to the novel Coronavirus that was discovered in 2019. The collection comprises over 23,000 articles, with about almost 2,000 new articles uploaded every week, making it a comprehensive resource for scholars keeping up to date on the current COVID-19 situation. The whole article or at least the abstract can be downloaded straight from LitCovid’s website for a major number of research papers. We chose 23,038 articles with full texts or abstracts from more than 35,000 articles for our document classification corpus. These articles have an average of 74 sentences and 1,399 tokens, indicating a fairly equal split between abstracts and full articles based on visual assessment. Prevention, Treatment, Diagnosis, Mechanism, Case Report, Transmission, Forecasting, and Generalcitep are the eight topic descriptors allocated to each research paper in LitCovid (Chen et al., 2021a, 2021b). Despite the fact that every research paper in the corpus can be tagged with numerous tags, the majority of research papers (about 76 %) only have one. We only utilized the research papers with one label in our experiment.
Data enhancement
We obtained and verify the authors, published year, type of paper, citations, references, type of journal, and journal name based on data from Google scholar (https://scholar.google.com/) and CORD-19 (CORD-19: COVID-19 Open Research Dataset)(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7251955/) in order to enhance the bibliometric data (Fig. 2). Fig. 2 Process of data enhancement
Target classes
We used the LitCovid topic descriptors to define classes for our multi-class classification task. Out of the eight topic descriptors, we worked only with the Prevention, Treatment, Diagnosis, and Case Report classes. As we described before, our focus is on these four classes, and the majority of research papers (about 76 %) only have these classes. Table 1 shows the distribution of classes in the data that we used for the experiment. The classes are highly imbalanced with the most commonly occurring label appearing about five times as frequently as the least frequent one. We used a state-of-the-art algorithm SMOTE (Chawla et al., 2002) for the over-sampling of minority classes, to handle this problem. In the oversampling technique with SMOTE, the synthetic samples are generated for the minority class. Instead of simply duplicating minority class examples, the algorithm generates new examples similar to those, existing in the data. To do this the algorithm first finds a close neighbor of a minority class example and then creates a new example that lies in between. So new minority class examples are obtained as interpolations of existing ones (Chawla et al., 2002). Table 1 Label count (used for this experiment) of the LitCovid corpus
Label Count
Prevention 2599
Treatment 1454
Diagnosis 658
Case_report 542
Data pre-processing and feature engineering
At the beginning of the feature engineering, we remove the unwanted characters from the title and abstract. This helps to get more accurate features from the corpus. Also, we convert all the uppercase words to lowercase. We also drop the duplicate data (abstract and title) and remove the “NaN” values from the entire corpus. Also, removing the stop words did not create any huge impact on the accuracy but it had a huge impact on improving the model interpretability. After completing the basic data cleaning, the first set of features was derived from the bibliometric indicators and placed into the “Bibliometric data” table. This included the following information: Authors’ number,
Age of the publication,
Type of paper,
Number of citation,
Number of references,
Type of publication,
Tokenized journal name
As we described before, for our experiment we use the LitCovid corpus where we utilize the abstracts, title, and bibliometric information but are not able to process full text because of the low computational power and the full text is always not easy to find for reproducibility, where abstracts are almost always available.
Entity extraction
We used ScispaCy (Neumann et al., 2019), to extract entities from the research papers. ScispaCy is one of the most robust model pipelines for a variety of natural language processing tasks focused on biomedical text. It contains modules for part of speech tagging, dependency parsing, named entity recognition, and sentence segmentation (Neumann et al., 2019). We used four pre-trained ScispaCy models for named entity recognition in the research papers. We employed a transition-based method based on the chunking model for entity extraction (Lample et al., 2016) as implemented in ScispaCy. All four pre-trained ScispaCy models are depicted in Table 2. A partial sample of entities for a research paper “Gene expression in epithelial cells in response to pneumovirus infection” could be Respiratory syncytial virus, RSV, pneumonia virus, mice, PVM, viruses, family Paramyxoviridae, subfamily pneumovirus, respiratory infections, etc. All the entities were employed in forming the model (see Section 3.2.2). Table 2 ScispaCy entity recognition systems used corpus
Training corpus Entity types
CRAFT GGP, SO, TAXON, CHEBI, GO, CL
JNLPBA DNA, CELL_TYPE, CELL_LINE, RNA, PROTEIN
BC5CDR DNA, CELL_TYPE, CELL_LINE, RNA, PROTEIN
BIONLP13CG AMINO_ACID, ANATOMICAL_SYSTEM,
CANCER, CELL, CELLULAR_COMPONENT,
DEVELOPING_ANATOMICAL_STRUCTURE,
GENE_OR_GENE_PRODUCT, IMMATERIAL_ANATOMICAL_ENTITY,
MULTI-TISSUE_STRUCTURE, ORGAN, ORGANISM,
ORGANISM_SUBDIVISION, ORGANISM_SUBSTANCE,
PATHOLOGICAL_FORMATION, SIMPLE_CHEMICAL, TISSUE
List adapted from https://allenai.github.io/scispacy/
Document representation
We used three alternative ways to represent the text-related features: binary word incidence approach (Zhang et al., 2010), TF-IDF approach (Aizawa, 2003; Muralikumar et al., 2017), and the embeddings-based approach (Tenney et al., 2019). The first two ways are related to the Bag-of-words (BOW) model. BOW is one of the simplest methods for document representation. In this model, a document is represented as a multi-set of words appearing in the document. In binary representation, only the presence or absence of a word in the document is encoded. TF-IDF reflects both the occurrence of a term in a particular document (as term frequency, TF) and the occurrence of this term in the whole collection of documents (as inverse document frequency, IDF). For both representations, we used the N-gram approach for the article titles and abstracts. According to multiple research projects, N-gram is one of the most used methods in the field of computational linguistics (Brown et al., 1992). For our experiment, we applied uni-gram, bi-gram, and tri-gram to find out the best possible entities from the title and abstract of a research paper.
Bidirectional Encoder Representations from Transformers (BERT) were employed for the embeddings-based method (Devlin et al., 2018). The BERT model is made by stacking up multiple encoders of the transformer architecture on top of one another. The BERT architecture is designed to pre-train deep bidirectional representations from the unlabeled text by jointly conditioning on both left and right contexts in all layers. We applied BERT Tokenizer based on WordPiece (Muller et al., 2019) for the title, and abstracts from the LitCovid corpus. We used a pre-trained model bert-base-uncased (Geetha & Renuka, 2021) and the pre-training was performed on a large corpus of English data (BookCorpus and English Wikipedia) in a self-supervised fashion (Geetha & Renuka, 2021; Turc et al., 2019).
Dimensionality of input corpus
To reduce the dimensionality, for the document representation input data table we used tree derived importance feature selection method, which is a very straight forward, fast, and general approach for selecting the good features for machine learning methods. In Table 3, we show the results of reduction applied to different input data tables. In their experiment, Beranova et al. demonstrated that 1500 features outperformed other numbers of features for the ScispaCy-related input data table (Beranová et al., 2022). For that reason, we utilize the 1500 features for all of our ScispaCy entity-based experiments. But for others (TF-IDF and BOW), we utilized all the features for our experiment. In Table 3, we also include the total feature count before the feature selection, which was considered as an input to reduce the size of the dataset to address the training time and scalability issues encountered with the different machine learning-based methods. Also, for the Neural Network-based model, the input data tables have been reduced because of training in a reasonable time. For the BOW Input data table and the TF-IDF Input data table, we utilize all features. As we described before, for the ScispaCy-related input data table we utilize the 1500 features and utilize the threshold value like Beranová et al. (2022). We also investigated the relationship between the dimension of the input data table and the accuracy of the Random Forest model trained on it. Table 7 shows the best accuracy with the different document representations with different machine learning-based methods, where the highest accuracy of 92% is most stably attained for a vector length of about 21264 for the TF-IDF document representation with the Random Forest method. Also, the ScispaCy related document representation (binary) with 1500 features got 79% accuracy. We used MDI feature importance scores to select the most 1500 important features. Other input data tables (such as Title and Abstract BibliometricFeatures) are constructed by merging the related input data tables. Table 3 Overview of input corpus in the machine learning methods
Corpus (Input data table) Used text Columns Original columns
ScispaCy Abstract 1500 195434
TF-IDF Abstract 20939 20939
BOW Abstract 20939 20939
ScispaCy Title 1500 3430
TF-IDF Title 326 326
BOW Title 326 326
ScispaCy Title_Abstract 1500 195760
TF-IDF Title_Abstract 21264 21264
BOW Title_Abstract 21264 21264
BOW Abstract_BibliometricFeatures 20945 20945
TF-IDF Abstract_BibliometricFeatures 20945 20945
BERT Title 3072 30522
BERT Title_and_Abstract 3072 30522
BERT Title_and_Abstract_and_ BibliometricFeatures 3072 30522
Machine learning experiments
For the machine learning experiments, we used the Random Forest (Liaw et al., 2002; Breiman, 2001), Linear Support Vector classifier (Linear SVC) (Suthaharan, 2016), Multinomial Naive Bayes (Kibriya et al., 2004), Logistic Regression (Sperandei, 2014), K-Nearest-Neighbors (Fukunaga and Narendra, 1975), Support Vector Classifier (SVC) (Suthaharan, 2016), Decision Tree (Safavian & Landgrebe, 1991), Multi-layer Perceptron (MLP) (Taud & Mas, 2018) and Adaptive Boosting (AdaBoost) (Margineantu & Dietterich, 1997) classifier as implemented using the scikit-learn (https://scikit-learn.org/), a free software machine learning library for python. For training and testing purposes, we used the title, abstract and bibliometric information from each research paper. Because of the limited computational power, we were not able to use the full-body text for our experiments. Table 3, corpus(Input data table) column shows different variants of article representations we used as input data tables for machine learning. Here TF-IDF, BOW, and ScispaCy refer to methods used to represent information taken from the abstract or title of the articles (BOW stands for binary representation), and BibliometricFeatures refers to bibliometric information about the articles.
Convolutional Neural Networks (CNNs) outperform alternative neural network architectures such as LSTMs and Recurrent Neural Networks for classification tasks (Gu et al., 2018; Prusa & Khoshgoftaar, 2017). To encode our data, we created a BERT tokenizer and a pre-trained BERT model configuration. We used a function called “batch_encode_plus” to encode all of the titles and abstracts from the research paper, and we trained and validated the data individually. Table 4 shows how the model was learned for various combinations of hyperparameters. Table 4 Overview of input Parameter grid
Machine learning algorithm Parameter grid
Random Forest ∙ ‘max_depth’: 10,150,500,1000
∙ ‘max_features’: 30,500,3000
∙ ‘min_samples_leaf’: 1,10,100
∙ ‘min_samples_split’: 2,10,100
∙ ‘n_estimators’: 10, 100
Logistic Regression ∙ ‘random_state’: 0
K-Nearest-Neighbors ∙ ‘n_neighbors’: 3
SVC ∙ ‘gamma’: 2
∙ ‘C’: 0.025, 1
∙ ‘kernel’: linear
Decision Tree ∙ ‘max_depth’: 5, 10, 15
Multi-layer Perceptron ∙ ‘alpha’: 1
∙ ‘max_iter’: 1000
Neural Network (BERT) ∙ ‘max_length’: 256
∙ ‘epochs’: 5
∙ ‘lr’: 1e-5
∙ ‘eps’: 1e-8
We created eleven variants of the input data table altogether as input for machine learning methods. In Section 4.2, we will present a comparison regarding the variants of data tables and find which data table provides the best accuracy in our full experiment.
We also tuned different types of hyperparameters to find the best result using different machine learning methods. We utilize the K-fold cross-validation from the scikit-learn.1 It provides us the cross-validation with random search and grid search hyperparameter optimization via the RandomizedSearchCV2 and GridSearchCV3 classes respectively. We used the inner loop of nested cross-validation where the training dataset was defined by the outer loop. We also configure the hyperparameter search to refit a final model with the entire training dataset using the best hyperparameters. As we describe before, we utilize nested cross-validation for fine-tuning the hyperparameters. Nested cross-validation is an approach for model hyperparameter optimization that attempts to overcome the problem of overfitting the training dataset. The procedure involves treating model hyperparameter optimization as part of the model itself and evaluating it within the broader K-fold cross-validation procedure for evaluating models for comparison and selection. A set of different hyperparameters for the different machine learning methods were optimized according to the grid that we present in Table 4.
Explanation algorithms
The question of interpretability or explainability of the created models becomes a hot topic in the area of machine learning. The end-users are interested not only in the quality of the models but also in the insight into the classification process. Some models are easy to understand by their nature (typical examples are Decision Trees), but some models, typically Neural Networks, work as a black-box model.
The Random Forest method was designed for the calculation of the feature importance scores. However, due to the number of trees with their complexity, and also because the multiple trees can take part in a decision, direct interpretation of the Random Forest models is not possible (Breiman, 2001). In our work, we adopt the original method for computing the feature importance scores of Random Forest, which is based on the Mean Decrease of Impurity (MDI). For this method, it has been shown that the MDI importance of a relevant feature is invariant with respect to the removal or addition of irrelevant features and that the importance of a feature is zero if and only if the feature is irrelevant (Louppe et al., 2013). LIME (Ribeiro et al., 2016) and SHAP (Lundberg et al., 2020) methods were also used to interpret the created models.
LIME
Local Interpretable Model-Agnostic Explanations (LIME) (Ribeiro et al., 2016) is an interpretability surrogate model which can be used on any black-box model to provide local interpretability for the result of prediction or classification of a single instance. The idea is to explain this prediction using a simpler (usually linear) model that has been created for a sample of the original data. Higher weights are assigned to examples that are like the instance in question. Based on the linear model, we can assess the contribution of each individual feature to the result of classification.
SHAP
SHapley Additive exPlanations (SHAP) (Lundberg et al., 2020) is a game theory-based method for interpreting any machine learning model’s output. It uses the traditional Shapley values from game theory and their related extensions to correlate optimal credit allocation with local explanations.
Results
Evaluation metrics
In order to evaluate the performance of our proposed model, we used overall Accuracy (A), Precision (P), Recall (R), and F1-score (F1). Overall accuracy is simply the proportion of correct classifications. Precision and Recall have been proposed in the area of information retrieval to evaluate the quality of search in a corpus of documents. In this original setting, Precision represents the proportion of relevant documents in the set of retrieved documents, and Recall represents the proportion of retrieved relevant documents in the set of all relevant documents. When adapted to evaluate results of classification, Precision is defined as the proportion of examples correctly classified to a given class in the set of examples classified to this class (formula (2)), and Recall is defined as the proportion of examples correctly classified to a given class in the set of all examples of this class (formula (3)). Here, TP stands for the number of examples correctly classified to a given class, FP stands for the number of examples incorrectly classified to a given class (so TP + FP is the number of examples classified to this class) and FN stands for the number of examples that were not classified to a given class (so TP + FN is the number of examples that belong to this class). Finally, F1-score is the harmonic mean of Precision and Recall (formula (4)). 1 Accuracy=TP+TNTP+TN+FP+FN
2 Precision=TPTP+FP
3 Recall=TPTP+FN
4 F1=2∗TP2∗TP+FP+FN
Precision, Recall, and F1-score are defined for each class separately. To evaluate a multi-class classifier, these characteristics can be extended to macro average or weighted average values. The macro average is just an arithmetic mean. A weighted average is obtained as a weighted sum where weights are relative frequencies of classes.
Predictive performance
We used 70% training data and 30% test data to evaluate the effectiveness of different machine learning based approaches. The overall accuracy was used to evaluate the results, but we also computed the per-class accuracy. As previously stated, this dataset was unbalanced, and we used the oversampling approach to solve this. Tables 5 and 6 shows the accuracy (per-class accuracy) of the Neural Network (BERT) and Random Forest approach, whereas Table 7 shows the overall accuracy of the other methods. Table 5 Accuracy with Neural Network (BERT)
Input data table Per class accuracy Accuracy
Title prevention 0.92
treatment 0.77 0.76
diagnosis 0.69
case_report 0.67
Title_and_Abstract prevention 0.95
treatment 0.80 0.87
diagnosis 0.86
case_report 0.85
Title_and_Abstract_and_biblo prevention 0.92
treatment 0.67 0.74
diagnosis 0.59
case_report 0.79
Table 6 Accuracy with Random Forest
Input data table Used text Class Per class accuracy Accuracy
ScispaCy Abstract prevention 0.76
treatment 0.69 0.74
diagnosis 0.75
case_report 0.74
TF-IDF Abstract prevention 0.96
treatment 0.92 0.92
diagnosis 0.90
case_report 0.89
BOW Abstract prevention 0.93
treatment 0.90 0.90
diagnosis 0.92
case_report 0.87
ScispaCy Title prevention 0.64
treatment 0.50 0.57
diagnosis 0.62
case_report 0.47
TF-IDF Title prevention 0.83
treatment 0.81 0.80
diagnosis 0.81
case_report 0.77
BOW Title prevention 0.71
treatment 0.66 0.70
diagnosis 0.78
case_report 0.64
ScispaCy Title_and_Abstract prevention 0.81
treatment 0.76 0.79
diagnosis 0.80
case_report 0.79
TF-IDF Title_and_Abstract prevention 0.96
treatment 0.92 0.92
diagnosis 0.91
case_report 0.9
BOW Title_and_Abstract prevention 0.93
treatment 0.90 0.91
diagnosis 0.92
case_report 0.87
BOW Title_and_Abstract_and_ prevention 0.82
Bibliometric Features treatment 0.73 0.73
diagnosis 0.67
case_report 0.72
Input data table Used text Class Per class accuracy Accuracy
TF-IDF Title_and_Abstract_and_ prevention 0.96
Bibliometric Features treatment 0.93 0.92
diagnosis 0.90
case_report 0.89
Table 7 Best accuracy with different machine learning algorithm for different text representation method
Input data table Used text Machine learning algorithms Accuracy Macro average (F1 score) Weighted average (F1 score)
ScispaCy Abstract Random Forest Classifier 0.74 0.74 0.74
TF-IDF Abstract Random Forest Classifier 0.92 0.92 0.92
BOW Abstract Random Forest Classifier 0.90 0.90 0.90
ScispaCy Title Multinomial NB 0.68 0.68 0.68
TF-IDF Title Random Forest Classifier 0.80 0.80 0.80
BOW Title Random Forest Classifier 0.70 0.70 0.70
ScispaCy Title and Abstract Random Forest Classifier 0.79 0.79 0.79
TF-IDF Title and Abstract Random Forest Classifier 0.92 0.92 0.92
BOW Title and Abstract Random Forest Classifier 0.91 0.91 0.91
BOW Abstract with Bibliometric Features Random Forest Classifier 0.73 0.73 0.73
TF-IDF Abstract with Bibliometric Features Random Forest Classifier 0.92 0.92 0.92
Bidirectional Encoder Representations Title_Abstract Neural Network (BERT) 0.87 0.87 0.87
As we see in Table 7, out of the traditional multi-purpose machine learning algorithms, the Random Forest method outperforms others. The BERT over Neural Network, a modern method used for NLP, achieved similar performance. But not only the classification accuracy should be considered when comparing the results of different methods. Another issue that should be taken into account is the complexity of the used data and the complexity of created models. Concerning the complexity of data, Table 3 indicates, that BERT was trained on less complex data representation than was the TF-IDF representation used for Random Forest. The complexity of the models can be assessed according to the number of nodes (neurons in the Neural Network or branching nodes in the trees in the forest). The last important issue when evaluating the models is their understandability and interpretability. More on this is presented in the next subsection.
Model interpretation
As stated earlier, we used SHAP and LIME to get more insight into the created models. As an illustration, we present an interpretation of the Random Forest model. Feature importance scores were used to analyze Random Forest models. We utilized the MDI method to find the relevant characteristics, as explained previously. Table 8 shows three example matrices from a total of eleven matrices. Table 8 Top most important features (MDI method) by input data table
ScispaCy Imp. TF-IDF Imp. TF-IDF and Bibliometric Imp.
Pandemic 0.004 year old 0.005 case 0.005
Fever 0.003 report case 0.004 year 0.005
Patient 0.003 fever 0.004 pcr 0.004
Respiratory distress syndrome 0.003 report 0.003 old 0.003
Drug 0.002 cough 0.003 present case 0.003
Covid-19 0.002 trials 0.003 acute 0.003
Hospital 0.002 measures 0.003 report 0.003
Patient 0.002 inhibitors 0.003 sars cov 0.003
Patient covid-19 0.002 cov 0.003 personal 0.003
Cell 0.002 polymerase chain reaction 0.003 sars 0.003
People 0.002 year 0.003 ct 0.003
Ace2 0.001 rt 0.002 inhibitors 0.003
Region 0.001 antiviral 0.002 covid 19 pandemic 0.002
Pneumonia 0.001 cov infection 0.002 clinical trials 0.002
Coronavirus 0.001 covid 19 pandemic 0.002 ace2 0.002
Protein 0.001 sars cov 0.002 cells 0.002
Recipient 0.001 receptor 0.002 specificity 0.002
Cytokine 0.001 angiotensin converting enzyme 0.002 distancing 0.002
Infection 0.001 therapeutic 0.002 report case 0.002
Protease 0.001 ground 0.002 healthcare 0.002
Coronavirus disease 2019 0.001 anti 0.002 therapeutic 0.002
SHAP can detect the direction of feature significance in the same way as the MDI technique can. A SHAP plot in Figs. 3 and 4 illustrates the significance of specific TF-IDF (title) and TF-IDF (abstracts). Additionally, it is the first evidence of the connection between a feature’s importance and its influence on a prediction. This SHAP plot combines the significance of the features with their impacts. A Shapley value for a feature and an instance may be found at each point on the summary plot. The feature determines the location on the y-axis, while the Shapley value determines the position on the x-axis. From low to high, the color indicates the value of the feature. We can observe how the distribution of the Shapley values for each feature is distributed since overlapping points are jittered in the y-axis direction. The features are listed in ascending order of significance. Fig. 3 SHAP plot for titles (using TF-IDF)
Fig. 4 SHAP plot for abstracts (using TF-IDF)
To get an overview of which features are most important for a model we plot the SHAP values of important features for titles and abstracts using TF-IDF. Figures 3 and 4 sort features by the sum of SHAP value magnitudes over all samples and uses SHAP values to show the distribution of the impacts the features have on the model’s output. The color represents the feature value whereas the red color shows greater values than values in blue. Case, patient, covid 19, infection, etc are one of the most important features for the titles using TF-IDF, and year old, case, stars cov, etc are one of the most important features for the abstracts using TF-IDF for this research paper’s classification task. We also include the LIME figure in Figs. 5 and 6 to explain the prediction of the Random Forest model for a sample research article based on its title and abstract. Fig. 5 LIME plot for title
Fig. 6 LIME plot for abstract
Discussion
The average overall accuracy was very different depending on the form of document representation; values ranged from 0.41 to 0.92. The document was represented using TF-IDF and bibliometric features in several input data tables with the abstract, title_abstract, and title_abstract_bibliometric features. The high-dimensional input data table is used for a wide range of purposes. The major goal is to see how the entity extraction method, document representation method, and bibliometric features affect the multi-class classification of the research paper. The interesting finding is that when we only use the retrieved entities (ScispaCy entities) from the abstract and title, we always obtain low accuracy. As a result, only utilized entities have little bearing on the multi-class classification task. As we discussed before, we used multiple input data tables for all of the available bibliometric features, and the best result was 0.92 (TF-IDF with title_abstract_bibliometric features). The abstract only and title-and-abstract combined input data tables yield comparable results. When we simply use the title data table, we receive less accuracy than when we only use the abstract data table. We may conclude from this experiment that the title has a smaller influence on classification accuracy than the abstract alone.
In order to generalize our findings by evaluating several document representation types, we discovered that The context (Such as title, abstract, etc) of a research paper creates more influence than the bibliometric information on the research paper’s classification.
The information derived only from abstracts is more relevant than information derived only from the title,
Extracted entities from either abstract or title with binary document representation are less effective than the use of abstract or title with TF-IDF document representation.
On the other side, the Neural Network (BERT) outperforms (with the exception of the Random Forest method) the other machine learning methods when using TF-IDF document representation, where we utilize three distinct input data tables (title, title_abstract, and title_abstract_some_bibliometric info). In this case, the title_abstract outperforms the others, and also it is closely similar to the Random Forest approach with TF-IDF document representation. We were also able to identify algorithms that outperformed the others for a specific document representation scheme. However, Random Forest and Neural Networks (BERT) display high performance across all document representation types.
Error analysis
To further understand the performance of the top-scoring Random Forest models, we additionally try to analyze the errors they made using TF-IDF document representation. First, we observe that these models frequently correlate several categories namely, diagnosis and treatment much more closely than is required. Despite the fact that there is some overlap and a semantic relationship between these groups. Future work should attempt to explicitly model correlation between categories to help the model recognize the particular cases in which labels should occur together. Finally, we observe that models have trouble identifying discriminative sections of the research paper due to how much introductory content on the pandemic can be found in most articles. Future work also should explicitly model the gap in relevance between introductory sections and crucial sentences such as research paper abstracts and titles. In Table 9, we provide three different example error samples. In the first research paper, the Random Forest method classified it as a treatment instead of prevention. Also for the second and third research papers, Random Forest classified it as a prevention and diagnosis instead of treatment. In the future, We also try to find the errors with the impact of the outside knowledge base in this classification task. In some cases, full text from the research papers is needed to make better predictions with the title, abstract and bibliometric information. We also include a LIME figure in Fig. 7 is an example of an error to explain the prediction of the Random Forest model for a sample research article based on its title. Table 9 LitCovid error samples
DOI Title Label Prediction
10.1136/emermed-2020-209797 strategic planning response covid-19 london em... Prevention Treatment
10.1007/s11606-020-06042-3 planning post-covid syndrome: how payers can m... Treatment Diagnosis
10.1007/s10877-020-00550-7 covid-19: pulse oximeters spotlight. Treatment Diagnosis
Fig. 7 LIME plot for an example error
Conclusion and future work
Our work deals with classifying the whole documents into classes that reflect important COVID-19 related issues like diagnosis, treatment, case reports, or prevention. The proposed pipeline can thus help medical practitioners to filter out research papers dealing with these issues from a massive amount of COVID-19 related research papers. In the future, we also plan to apply different rule mining methods to find the different patterns from the issues like diagnosis, treatment, case report, or prevention. Also, the found papers can present different opinions on the same issue bringing controversial information, the next step in using text mining techniques to support work with the documents can be sentiment analysis (Liu, 2012). Sentiment analysis can be turned into the question of whether a piece of text is expressing positive, negative, or neutral sentiment towards the discussed topic and can be thus understood as a knowledge-based classification problem. Sentiment analysis can be performed at the document level, at the sentence level, or at the aspect level. To apply sentiment analysis techniques, not only the texts themselves but also some sentiment lexicons that contain representative words for different opinion polarities must be used. Also, the MeSH thesaurus can be useful for sentiment analysis of COVID-19 related documents when considering the aspect level.
Acknowledgements
The authors also would like to thank Vojtěch Svátek, Tomáš Kliegr, and Lucie Beranová for providing helpful feedback on the first version of the manuscript and again Vojtěch Svátek for the second version of the manuscript.
Author Contributions
Gollam Rabby works with conception, writing code, experiments, manuscript text, and editing. Petr Berka works with the conception, manuscript text, managed the research, and editing of the article. All authors read and approved the final manuscript.
Funding
Gollam Rabby was partly supported by grant IGA 40/2021 “Action rules for text”. Petr Berka was supported by the Faculty of Informatics and Statistics, Prague University of Economics and Business, through long-term support for research activities.
Data Availability
The research has been conducted only with openly available software packages and all the scripts, results, dataset after pre-processing and feature engineering will be publicly available in (https://github.com/corei5/Multi-Class-Classification-of-COVID-19-Documents) from the authors after accepting the manuscript.
Declarations
Consent for Publication
We give our consent for the publication of identifiable details, which can include all the text (“Material”) to be published in the above Journal and Article.
Competing interests
All authors declare that they have no conflicts of interest.
1 https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.KFold.html
2 https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html
3 https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Aizawa A An information-theoretic perspective of TF–IDF measures Information Processing & Management 2003 39 1 45 65 10.1016/S0306-4573(02)00021-3
Balaji V Suganthi S Rajadevi R Skin disease detection and segmentation using dynamic graph cut algorithm and classification through naive bayes classifier Measurement 2020 163 107 122 10.1016/j.measurement.2020.107922
Beranová, L., Joachimiak, M.P., Kliegr, T., & et al. (2022). Why was this cited? explainable machine learning applied to COVID-19 research literature. Scientometrics, 1–37. 10.1007/s11192-022-04314-910.1007/s11192-022-04314-9.
Breiman L Random forests Machine Learning 2001 45 1 5 32 10.1023/A:1010933404324
Brown PF Della Pietra VJ Desouza PV Class-based n-gram models of natural language Computational Linguistics 1992 18 4 467 480
Chawla NV Bowyer KW Hall LO Smote: synthetic minority over-sampling technique Journal of Artificial Intelligence Research 2002 16 321 357 10.1613/jair.953
Chen, Q., Allot, A., Leaman, R., & et al. (2021a). Overview of the BioCreative VII LitCovid track: multi-label topic classification for COVID-19 literature annotation. In Proceedings of the 7th BioCreative challenge evaluation workshop. 10.1093/database/baac069.
Chen Q Allot A Lu Z LitCovid: an open database of COVID-19 literature Nucleic Acids Research 2021 49 D1 D1534 D1540 10.1093/nar/gkaa952 33166392
Devlin, J., Chang, M.W., Lee, K., & et al. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv:181004805, 10.48550/arXiv.1810.04805.
Elberrichi, Z., Amel, B., & Malika, T. (2012). Medical documents classification based on the domain ontology mesh. arXiv:12070446, 10.48550/arXiv.1207.0446.
Fukunaga K Narendra PM A branch and bound algorithm for computing k-nearest neighbors IEEE Transactions on Computers 1975 100 7 750 753 10.1109/T-C.1975.224297
Gani A Siddiqa A Shamshirband S A survey on indexing techniques for big data: taxonomy and performance evaluation Knowledge and Information Systems 2016 46 2 241 284 10.1007/s10115-015-0830-y
Geetha M Renuka DK Improving the performance of aspect based sentiment analysis using fine-tuned bert base uncased model International Journal of Intelligent Networks 2021 2 64 69 10.1016/j.ijin.2021.06.005
Gu J Wang Z Kuen J Recent advances in convolutional neural networks Pattern Recognition 2018 77 354 377 10.1016/j.patcog.2017.10.013
Jindal R Taneja S A lexical approach for text categorization of medical documents Procedia Computer Science 2015 46 314 320 10.1016/j.procs.2015.02.026
Jindal, R., & Taneja, S. (2015b). Ranking in multi label classification of text documents using quantifiers. In 2015 IEEE international conference on control system, computing and engineering (ICCSCE) (pp. 162–166). IEEE, DOI 10.1109/ICCSCE.2015.7482177, (to appear in print).
Kibriya, A.M., Frank, E., Pfahringer, B., & et al. (2004). Multinomial naive bayes for text categorization revisited. In Australasian joint conference on artificial intelligence. 10.1007/978-3-540-30549-1_43 (pp. 488–499). Springer.
Lample, G., Ballesteros, M., Subramanian, S., & et al. (2016). Neural architectures for named entity recognition. arXiv:160301360, 10.18653/v1/N16-1030.
Li W Saigo H Tong B Topic modeling for sequential documents based on hybrid inter-document topic dependency Journal of Intelligent Information Systems 2021 56 3 435 458 10.1007/s10844-020-00635-4
Liaw A Wiener M Classification and regression by randomforest R News 2002 2 3 18 22
Liu B Sentiment analysis and opinion mining Synthesis Lectures on Human Language Technologies 2012 5 1 1 167 10.2200/S00416ED1V01Y201204HLT016
Louppe, G., Wehenkel, L., Sutera, A., & et al. (2013). Understanding variable importances in forests of randomized trees. Advances in Neural Information Processing Systems, 26. 10.5555/2999611.299966010.5555/2999611.2999660.
Lundberg SM Erion G Chen H From local explanations to global understanding with explainable AI for trees Nature Machine Intelligence 2020 2 1 2522 5839 10.1038/s42256-019-0138-9
Margineantu, D.D., & Dietterich, T.G. (1997). Pruning adaptive boosting. In ICML. 10.5555/645526.757762 (pp. 211–218). Citeseer.
Mujtaba G Shuib L Idris N Clinical text classification research trends: Systematic literature review and open issues Expert Systems with Applications 2019 116 494 520 10.1016/j.eswa.2018.09.034
Muller, B., Sagot, B., & Seddah, D. (2019). Enhancing BERT for lexical normalization. In The 5th workshop on noisy user-generated text (W-NUT). 10.18653/v1/D19-5539.
Muralikumar J Seelan SA Vijayakumar N A statistical approach for modeling inter-document semantic relationships in digital libraries Journal of Intelligent Information Systems 2017 48 3 477 498 10.1007/s10844-016-0423-6
Neumann, M., King, D., Beltagy, I., & et al. (2019). Scispacy: fast and robust models for biomedical natural language processing. arXiv:190207669, 10.48550/arXiv.1902.07669.
Prusa JD Khoshgoftaar TM Improving deep neural network design with new text data representations Journal of Big Data 2017 4 1 1 16 10.1186/s40537-017-0065-8
Ribeiro, M.T., Singh, S., & Guestrin, C. (2016). “why should i trust you?” explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 10.1145/2939672.2939778 (pp. 1135–1144).
Safavian SR Landgrebe D A survey of decision tree classifier methodology IEEE Transactions on Systems Man, and Cybernetics 1991 21 3 660 674 10.1109/21.97458
Sperandei S Understanding logistic regression analysis Biochemia Medica 2014 24 1 12 18 10.11613/BM.2014.003 24627710
Suthaharan S Machine learning models and algorithms for big data classification Integr Ser Inf Syst 2016 36 1 12
Taud, H., & Mas, J. (2018). Multilayer perceptron (mlp). pp 451–455. 10.1007/978-1-4842-4470-8_31 .
Tenney, I., Das, D., & Pavlick, E. (2019). BERT rediscovers the classical NLP pipeline. arXiv:190505950, 10.18653/v1/P19-1452.
Turc, I., Chang, M.W., Lee, K., & et al. (2019). Well-read students learn better: On the importance of pre-training compact models. arXiv:190808962, https://paperswithcode.com/paper/?openreview=BJg7x1HFvB.
Yan Y Yin XC Yang C Biomedical literature classification with a CNNS-based hybrid learning network PloS ONE 2018 13 7 93 97 10.1371/journal.pone.0197933
Zhang Y Jin R Zhou ZH Understanding bag-of-words model: a statistical framework International Journal of Machine Learning and Cybernetics 2010 1 1 43 52 10.1007/s13042-010-0001-0
| 36465147 | PMC9707112 | NO-CC CODE | 2022-12-01 23:20:20 | no | J Intell Inf Syst. 2022 Nov 29;:1-21 | utf-8 | J Intell Inf Syst | 2,022 | 10.1007/s10844-022-00768-8 | oa_other |
==== Front
Transgenic Res
Transgenic Res
Transgenic Research
0962-8819
1573-9368
Springer International Publishing Cham
36441390
329
10.1007/s11248-022-00329-3
Meeting Report
TT2022 meeting report on the 17th Transgenic Technology meeting in Helsinki, Finland
http://orcid.org/0000-0002-4464-8744
Parker-Thornburg Jan [email protected]
1
http://orcid.org/0000-0003-1275-4320
Benavides Fernando [email protected]
2
Shawlot William [email protected]
3
http://orcid.org/0000-0002-5577-2230
McBeath Elena [email protected]
4
1 grid.240145.6 0000 0001 2291 4776 Professor Emeritus, MD Anderson Cancer Center, Houston, TX USA
2 grid.240145.6 0000 0001 2291 4776 Department of Genetics, MD Anderson Cancer Center, Houston, TX USA
3 grid.89336.37 0000 0004 1936 9924 Department of Molecular Biosciences, Center for Biomedical Research Support, University of Texas at Austin, Austin, TX USA
4 grid.240145.6 0000 0001 2291 4776 Department of Endocrine Neoplasia and Hormonal Disorders, MD Anderson Cancer Center, Houston, TX USA
28 11 2022
2022
31 6 677681
24 10 2022
2 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
http://dx.doi.org/10.13039/100006955 Office of Extramural Research, National Institutes of Health R50CA211121 Parker-Thornburg Jan issue-copyright-statement© Springer Nature Switzerland AG 2022
==== Body
pmcThe much-anticipated TT2022 conference was held with great success in mid-September of 2022 in Helsinki, Finland. After the long delay in in-person meetings, ISTT members greatly appreciated the opportunity to get together with new and old friends in person near the shore of the Baltic Sea. The Scandic Grand Marina Hotel and Conference Center was the site of the meeting from 17 to 19 September, an excellent venue that easily handled the 250 on-site delegates and the additional 250 virtual participants. Attendees could easily walk to the city center to see the magnificent white cathedral, or to the public library to see the miniscule sculpture of the literary mouse. Participants with extra time even took the ferry to Tallinn, Estonia or visited Lapland for sightings of reindeer and landscapes near the Arctic Circle.
Day 1 A warm Finnish welcome countered the autumn chill to begin the meeting on Saturday evening. Ernst-Martin Füchtbauer, the ISTT President, welcomed attendees in Finnish, eliciting a few chuckles from the Finnish members of the audience. Then the chairs of the local organizing committee, Satu Kuure and Reetta Hintala, welcomed the onsite delegates and introduced Kari Alitalo (Univ. Helsinki), who described his studies of the lymphatic system in the brain, and how various growth factors affect the lymphatic system. He has used numerous genetically modified mice to examine the pathways of this system and found unexpected connections among vascular growth factors that are used to maintain lymphangiogenesis. His research has led to a current clinical study in humans that examines the effects of VEGF-C/D in dry macular degeneration. Professor Alitalo’s lecture was followed by a welcome reception, where old friends had a chance to chat and catch up on several years of new developments in their labs.
Day 2 The Sunday 18 September sessions began with talks demonstrating how animal models allow researchers to take ideas from the bench to the clinic. Emma Andersson from the Karolinska Institute showed an amazing method of delivering DNA in situ in the early embryo using a nano-injection device developed specifically for her studies of gene expression that contribute to the development of the nervous system. She achieved temporal and spatial specificity in her system (titled NEPTUNE) using lentiviral vectors and avoiding ubiquitous promoters (e.g., PGK). NEPTUNE is now being tested for CRISPR/Cas9 targeting and the development of a barcode lentiviral library. The second plenary talk was given by Martin Bergö (Karolinska Inst.) where he shared his analysis of the antioxidant-stimulated metastasis pathway to show that BACH1, a critical protective factor in this pathway, stimulates genes that promote metastasis in response to antioxidants.
Two short talks followed the plenary talks, both describing animal models generated to mimic human conditions. Daniel Davis (Univ. Missouri) described three mouse models of human diseases that mimic the human causal variants, including spinal muscle atrophy (SMA) models that recapitulate hotspot mutations found in humans; KCNT1 models of mutations that have been shown to generate epilepsy; and a model of cystic fibrosis (CF) where exon 8 was replaced with a mutant form of human exon 8 found in CF patients. Doug Strathdee (CRUK Beatson Inst.) followed by presenting an animal model of Barth Syndrome where the Tafazzin gene (a mitochondrial lipid transferase) was floxed. Analysis of mice that were knocked out for the TAFAZZIN protein showed growth delay and changes in heart function as was expected based on human disease. Subsequently, his group replaced the entire mouse gene with the human gene that contains an extra exon. They are now using this gene to add in patient-specific point mutations.
The next session included both a talk on the auxin-inducible degron (AID) system (a system adapted to degrade specific proteins) in mice, and a lively round-table discussion on quality control of CRISPR/Cas-generated animals. Philip Jordan (Johns Hopkins Univ.) described his use of the AID system to degrade proteins that maintain the structure of chromosomes, including cohesins, condensins, and SMC5/6. He is currently using AID2 technology (Yesbolatova et al. 2020) and sees tight control of the system that is rapid, specific, inducible and reversible.
Next, a highly interactive round table was chaired by Lydia Teboul (MRC Harwell) where Søren Warming (Genentech), Lauryl Nutter (The Centre for Phenogenomics) and Marie-Christine Birlling (IGBMC) presented concerns and ideas for validating animal models produced using CRISPR techniques. There are no current standards required for ensuring the CRISPR model produced is not affected by off-target mutations, indels, or even mating procedures that can fix undesired/unknown mutations in a line. Some easily adaptable ideas for rodent models include regular backcrossing to the wild-type inbred strain, obtaining both experimental and control mice from the same colony, tracking and reporting generations, and cryopreservation of early generations. After these easy validations, though, additional criteria that could or should be checked provoked a lot of discussion leading to a consensus among all present that investigators and core directors would be well-served by an ISTT-generated white-paper that can be used as a blueprint for publishing validated CRISPR animal models.
This session was followed by a short interval to view posters, have lunch, and visit with vendors. In all, more than 80 posters were presented over six short sessions during the meeting, spanning from simple culture techniques to the generation of large animal models that mimic human diseases. In addition, two lunch-and-learn presentations were given. Day 2’s lunch session was given by Jean Cozzi of Charles River on methods that can be used to improve research reproducibility and animal welfare when breeding, genotyping and phenotyping animal models. He described several confounding factors commonly seen when problems occur, such as failure to properly clean instruments between taking samples (as one example).
The afternoon session began with five short, flash poster presentations. Elizabeth Bryda (Univ. of Missouri) presented the work of her group where they produced and phenotyped human ACE2 overexpression rat models generated to study SARS-CoV-2 and its impacts. Graham Duddy (Francis Crick Inst.) followed with a description of their workflow using AAV and electroporation (CRISPR-READI) to efficiently create mouse knock-in models. Next, Marie-Christine Birling (IGBMC) detailed methods to generate humanized mouse models by inserting large genomic DNAs produced by BAC engineering into mouse embryonic stem cells via CRISPR/Cas9. Roger Askew (Ozgene) then presented a new molecular screen to identify gene-targeted alleles using a fusion transcript that incorporates a polyA trap. Finally, Kathy Krentz (Univ. of Wisconsin) described their successful recovery of the Copenhagen rat line (a tumor-resistant inbred strain) many years after the initial freeze.
The 3Rs session started with a talk from Thorsten Buch (Univ. of Zurich) describing his attempt to develop a breeding calculation program to reduce mouse usage during complex breeding. His talk was followed by an overview by Vootele Voikar (Helsinki Institute of Life) of how EU regulations and outreach, in addition to adherence to the 3Rs, can contribute to ethical use of animals in research. Finally, the 3Rs Prize recipient, Ronald Naumann (MPI-CBG), presented his work showing that, by genotyping sperm samples from chimeras and looking for those with a high percentage of embryonic stem cell contribution, one can choose animals that are more likely to transmit through the germline.
After the second short poster session, a virtual plenary talk was given by Jihan Osborne (Univ. Texas-SWMC) that described her studies of the miRNAs Lin28a and Lin28b in lung development and branching morphogenesis. Using mouse embryos and embryonic stem cells, she found that the Lin28a/b miRNAs regulate the master regulators Sox2 and Sox9. She is currently investigating how these miRNAs are regulated. The following talk by Ritva Heljasvaara (Univ. of Oulu) described her use of numerous mouse transgenic lines to assess the role of the extracellular matrix in cancer. One of the genes studied, Col18a1, has isoforms that may act as tumor promoters. Finally in this session, an enthusiastic presentation was given by Andrew Syvyk (Texas A&M Univ.) to promote the use of rat spermatogonial stem cell lines to generate animal models. He demonstrated the ease of this method by showing how an overexpressing ErbB2 stem cell line could be introduced into the rete testis of Dazl transgenic rats to generate progeny that were wholly modified.
The final session of the first day was the highly anticipated presentation of the ISTT Prize to Drs. Elizabeth Lacy (Professor Emeritus, Sloan Kettering Cancer Center) and Frank Costantini (Professor Emeritus, Columbia University). A video of the presentation of the sculpture took place in a park in New York City, where Wojtek Auerbach, ISTT Past-President, gave the award to Drs. Lacy and Costantini. After viewing the presentation, we heard a live lecture that incorporated both awardees’ reminisces and discussions of their scientific careers which included some of the first methods described in the generation of transgenic animals. The audience was humbled to hear how these pioneers, at the very beginning of the age of molecular biology, overcame hurdles that included isolating genomic DNA for genotyping, determining how transgenes were incorporated, and figuring out how prokaryotic sequences inhibit the expression of transgenes. Their contribution to the methods now used to generate gene-modified animals has been immense and greatly appreciated within the community.
This presentation was followed by the conference banquet, held in the Scandic Marina Conference Center, where attendees were greeted with a flute of champagne and then treated to an excellent Finnish feast.
Day 3 The next full day of talks began with a session on non-mammalian models, and included a plenary lecture from Howy Jacobs (Univ. of Tampere) describing his use of Drosophila transgenics to study mitochondrial DNA (mtDNA). By generating mutants in nuclear DNA that function in mtDNA expression he has mimicked some human diseases. His most recent work includes the generation of cybrid flies (introducing mtDNA from one genotype into the cells from another genotype), thus establishing fruit flies as a first approach before generating the mutations in mice. In the second talk of this session, Annamarie Locasio (Stazione Zoologica Anton Dohm) described a unique animal model—Ciona robusta—a marine invertebrate chordate with embryonic development similar to other higher animals. Manipulation by electroporation of CRISPR/Cas9 components and comparison with similar mutations in Danio rerio (zebrafish) was used to identify enhancers and other means of regulating ONECUT and other HOX genes.
Short talks after the plenary lectures included one from Ralf Kühn (MDC-Mol.Med.) describing his attempts to generate the “perfect” C57BL/6 mouse by repairing known mutations and erasing endogenous retroviral sequences (ERVs), and a second talk by Mitra Cowan (McGill Univ.) describing the success she has had in generating large knock-in sequences using CRISPR/Cas9. She found that their best results were achieved using IVF to generate fertilized embryos followed by injection the following day at the 2-cell stage.
A fourth short poster session was followed by the much-anticipated “Running a Transgenic Service Facility” Roundtable chaired by Ben Davies (Univ. Oxford, Wellcome Trust). The panelists, Ben Lowe (The Jackson Laboratory), Peter Sipila (Univ. of Turku), Natalia Moncaut (CRUK-Manchester Inst.), and Pawel Pelczar (Univ. of Basel), described methods of embryo generation, reagent delivery, delivery of large DNAs, development of new technologies and how they implemented them, and took questions and comments from the audience. A very enlightening discussion surrounded how failed projects were managed: typically, most facilities perform multiple attempts to deliver the requested service. All agreed that it was difficult to determine when to stop and that communication with the contracting investigator is critical when projects run into difficulty.
The lunchtime break included a presentation by Gurpreet Balrey from Merck that described their new Cas9Plus, along with an enticing project in development to develop a chimeric Cas9 protein that will function efficiently within closed chromatin.
Jaan-Olle Andressoo (Univ. of Helsinki) began the afternoon session with a presentation of how their conditional 3’-UTR mutants of glial-derived neurotrophic factor (GDNF) were able to recapitulate aspects of schizophrenia in humans, but that the timing of Cre-excision was critical in the process. Interestingly, they noted that an interacting adenosine receptor, Adora2a, is inhibited by caffeine, and determined that blocking this receptor normalized the Gdnf mutant mice. The subsequent lecture by Tarja Malm (Univ. of Eastern Finland) examined how non-coding RNAs regulate protein expression in complex regulatory networks. She described studies of the interactions among lncRNAs, miRNAs and circulating RNAs in Alzheimer’s disease, Parkinson’s disease and ischemic stroke. The plenary lectures were followed by two short talks. The first, by Lev Federov (Oregon Health & Science Univ.), described his comparison of embryo development after culture in human tubal fluid (HTF), M16, Global Media from Cooper Surgical, or KSOM, finding that more embryos developed normally in HTF than the comparison medias. The second short talk was given by Yueh-Chiang Hu (Cincinnati Children’s Hospital), describing doxycycline-inducible models of the four core genotypes in mice (XX female, XX male, XY male and XY female) that were generated using CRISPR-interference and CRISPR-activation constructs targeted to the Col1a1 safe harbor site.
After a short poster break, the Orbis Pictus lecture, sponsored by the Czech Center for Phenogenomics (CCP) was given by Maria Lehtinen (Boston Children’s Hospital). Dr. Lehtinen’s talk was replete with beautiful images showing how the choroid plexus contributes to the cerebral spinal fluid (CSF) to control the development of the brain and subsequent neurogenesis in early mouse embryogenesis. During the response to calcium induction, unusual blebs were identified that appear to function in apocrine secretion of transthyretin (TTR), which was confirmed after generating TTR-GFP mice that secreted GFP into the CSF.
Day 4 The final day of TT2022 began with Qiangge Zhang (Massachusetts Inst. Technology) describing the process used at MIT for developing transgenic marmosets to serve as NHP models in neuroscience research. IVF-generated embryos were microinjected with transgenes to knock out SHANK3, PSEN1, MeCP2 or SETD1A. The second talk, given by Benjamin Schusser (Technical Univ. Munich), described his successful efforts to show that chickens could be protected from disease using gene editing to delete an amino acid that is critical for infection by avian leukosis virus. The short talks following these presentations included one from Stuart Newman (PetMedix LTD) where transgenic mice were generated to express canine therapeutic antibodies, and a second talk from Guillaume Pavlovic (IGBMC) where changes in the gut microbiome were assessed in response to different factors such as housing area, sex of the animal and genetic background of the animal (among others).
The final regular scientific session of the meeting focused on descriptions of emerging technologies. The plenary lecture was given by Bon-Kyong Koo (Inst. Of Molecular Biotechnology) where he presented a new artificial intron that can be used to generate conditional alleles. His short conditional intron (SCON) was used to target numerous genes, including single exon genes within specific parameters for use. The next short talk by Stephane Pelletier (Indiana Univ.) discussed his work with the “Degradation based on Cre-regulated Artificial Intron” (DECAI), where he has been able to target several genes to generate conditional alleles. The final short talk of the meeting was given by Xiaojun Xing (Yale Univ.), where he demonstrated his high-throughput method for mouse sperm cryopreservation using microstraws and batch labeling.
In the last sessions of TT2022, the Young Investigator Award was presented to Davide Seruggia (St. Anna Children’s Cancer Research Inst.) followed by his lecture, and, subsequently, an EMBO-sponsored lecture by Elizabeth Fisher (UCL Inst. Of Neurology). Dr.Seruggia described his work using CRISPR, base editing, degrons and the adeno-associated viral method to deliver components to the embryo to determine essential areas required for enhancer function and critical transcription factor sites.
Dr. Fisher then depicted the generation of mice with humanized genes used to study the genes implicated in amyotrophic lateral sclerosis (ALS), including fused in sarcoma (FUS) and transactive response DNA binding protein (TARDBP), which appear to be critical for most neurodegeneration.
The meeting ended with a joyous celebration of this successful meeting, beginning with awards, which included the Mammalian Genome-sponsored awards presented for the best oral presentations of a poster (Stephane Pelletier and Yueh-Chiang Hu), three prizes for best posters (sponsored by emeritus ISTT members) (awarded to Vanessa Chenouard from INSERM, Graham Duddy from Francis Crick Institute, and Pawel Pelczar from Univ. of Basel), and two awards from the International Mammalian Genome Society given as complimentary registration for one of their upcoming meetings (Lauri Lintott from the Hospital for Sick Children, and Tereza Nickl from the Czech Center for Phenogenomics). Jan Parker-Thornburg then issued an invitation to TT2023, to be held in Houston Texas, USA in November of 2023. Final thank-yous and closing remarks from President Füchtbauer ended this happy occasion and left delegates anticipating the next time we all could be together to present and celebrate our science.
Acknowledgements
Partial funding for JPT to attend TT2022 was supplied by funding from the NCI (R50CA211121).
Author contributions
JPT wrote the main manuscript text. FB, WS and EM contributed suggestions and edited the text for completeness and clarity. All authors reviewed the manuscript.
Funding
Office of Extramural Research, National Institutes of Health (R50CA211121).
Declarations
Competing interests
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
Reference
Yesbolatova A Saito Y Kitamoto N The auxin-inducible degron 2 technology provides sharp degradation control in yeast, mammalian cells, and mice Nat Commun 2020 11 5701 10.1038/s41467-020-19532-z 33177522
| 36441390 | PMC9707113 | NO-CC CODE | 2022-12-14 23:47:19 | no | Transgenic Res. 2022 Nov 28; 31(6):677-681 | utf-8 | Transgenic Res | 2,022 | 10.1007/s11248-022-00329-3 | oa_other |
==== Front
Transgenic Res
Transgenic Res
Transgenic Research
0962-8819
1573-9368
Springer International Publishing Cham
36441390
329
10.1007/s11248-022-00329-3
Meeting Report
TT2022 meeting report on the 17th Transgenic Technology meeting in Helsinki, Finland
http://orcid.org/0000-0002-4464-8744
Parker-Thornburg Jan [email protected]
1
http://orcid.org/0000-0003-1275-4320
Benavides Fernando [email protected]
2
Shawlot William [email protected]
3
http://orcid.org/0000-0002-5577-2230
McBeath Elena [email protected]
4
1 grid.240145.6 0000 0001 2291 4776 Professor Emeritus, MD Anderson Cancer Center, Houston, TX USA
2 grid.240145.6 0000 0001 2291 4776 Department of Genetics, MD Anderson Cancer Center, Houston, TX USA
3 grid.89336.37 0000 0004 1936 9924 Department of Molecular Biosciences, Center for Biomedical Research Support, University of Texas at Austin, Austin, TX USA
4 grid.240145.6 0000 0001 2291 4776 Department of Endocrine Neoplasia and Hormonal Disorders, MD Anderson Cancer Center, Houston, TX USA
28 11 2022
2022
31 6 677681
24 10 2022
2 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
http://dx.doi.org/10.13039/100006955 Office of Extramural Research, National Institutes of Health R50CA211121 Parker-Thornburg Jan issue-copyright-statement© Springer Nature Switzerland AG 2022
==== Body
pmcThe much-anticipated TT2022 conference was held with great success in mid-September of 2022 in Helsinki, Finland. After the long delay in in-person meetings, ISTT members greatly appreciated the opportunity to get together with new and old friends in person near the shore of the Baltic Sea. The Scandic Grand Marina Hotel and Conference Center was the site of the meeting from 17 to 19 September, an excellent venue that easily handled the 250 on-site delegates and the additional 250 virtual participants. Attendees could easily walk to the city center to see the magnificent white cathedral, or to the public library to see the miniscule sculpture of the literary mouse. Participants with extra time even took the ferry to Tallinn, Estonia or visited Lapland for sightings of reindeer and landscapes near the Arctic Circle.
Day 1 A warm Finnish welcome countered the autumn chill to begin the meeting on Saturday evening. Ernst-Martin Füchtbauer, the ISTT President, welcomed attendees in Finnish, eliciting a few chuckles from the Finnish members of the audience. Then the chairs of the local organizing committee, Satu Kuure and Reetta Hintala, welcomed the onsite delegates and introduced Kari Alitalo (Univ. Helsinki), who described his studies of the lymphatic system in the brain, and how various growth factors affect the lymphatic system. He has used numerous genetically modified mice to examine the pathways of this system and found unexpected connections among vascular growth factors that are used to maintain lymphangiogenesis. His research has led to a current clinical study in humans that examines the effects of VEGF-C/D in dry macular degeneration. Professor Alitalo’s lecture was followed by a welcome reception, where old friends had a chance to chat and catch up on several years of new developments in their labs.
Day 2 The Sunday 18 September sessions began with talks demonstrating how animal models allow researchers to take ideas from the bench to the clinic. Emma Andersson from the Karolinska Institute showed an amazing method of delivering DNA in situ in the early embryo using a nano-injection device developed specifically for her studies of gene expression that contribute to the development of the nervous system. She achieved temporal and spatial specificity in her system (titled NEPTUNE) using lentiviral vectors and avoiding ubiquitous promoters (e.g., PGK). NEPTUNE is now being tested for CRISPR/Cas9 targeting and the development of a barcode lentiviral library. The second plenary talk was given by Martin Bergö (Karolinska Inst.) where he shared his analysis of the antioxidant-stimulated metastasis pathway to show that BACH1, a critical protective factor in this pathway, stimulates genes that promote metastasis in response to antioxidants.
Two short talks followed the plenary talks, both describing animal models generated to mimic human conditions. Daniel Davis (Univ. Missouri) described three mouse models of human diseases that mimic the human causal variants, including spinal muscle atrophy (SMA) models that recapitulate hotspot mutations found in humans; KCNT1 models of mutations that have been shown to generate epilepsy; and a model of cystic fibrosis (CF) where exon 8 was replaced with a mutant form of human exon 8 found in CF patients. Doug Strathdee (CRUK Beatson Inst.) followed by presenting an animal model of Barth Syndrome where the Tafazzin gene (a mitochondrial lipid transferase) was floxed. Analysis of mice that were knocked out for the TAFAZZIN protein showed growth delay and changes in heart function as was expected based on human disease. Subsequently, his group replaced the entire mouse gene with the human gene that contains an extra exon. They are now using this gene to add in patient-specific point mutations.
The next session included both a talk on the auxin-inducible degron (AID) system (a system adapted to degrade specific proteins) in mice, and a lively round-table discussion on quality control of CRISPR/Cas-generated animals. Philip Jordan (Johns Hopkins Univ.) described his use of the AID system to degrade proteins that maintain the structure of chromosomes, including cohesins, condensins, and SMC5/6. He is currently using AID2 technology (Yesbolatova et al. 2020) and sees tight control of the system that is rapid, specific, inducible and reversible.
Next, a highly interactive round table was chaired by Lydia Teboul (MRC Harwell) where Søren Warming (Genentech), Lauryl Nutter (The Centre for Phenogenomics) and Marie-Christine Birlling (IGBMC) presented concerns and ideas for validating animal models produced using CRISPR techniques. There are no current standards required for ensuring the CRISPR model produced is not affected by off-target mutations, indels, or even mating procedures that can fix undesired/unknown mutations in a line. Some easily adaptable ideas for rodent models include regular backcrossing to the wild-type inbred strain, obtaining both experimental and control mice from the same colony, tracking and reporting generations, and cryopreservation of early generations. After these easy validations, though, additional criteria that could or should be checked provoked a lot of discussion leading to a consensus among all present that investigators and core directors would be well-served by an ISTT-generated white-paper that can be used as a blueprint for publishing validated CRISPR animal models.
This session was followed by a short interval to view posters, have lunch, and visit with vendors. In all, more than 80 posters were presented over six short sessions during the meeting, spanning from simple culture techniques to the generation of large animal models that mimic human diseases. In addition, two lunch-and-learn presentations were given. Day 2’s lunch session was given by Jean Cozzi of Charles River on methods that can be used to improve research reproducibility and animal welfare when breeding, genotyping and phenotyping animal models. He described several confounding factors commonly seen when problems occur, such as failure to properly clean instruments between taking samples (as one example).
The afternoon session began with five short, flash poster presentations. Elizabeth Bryda (Univ. of Missouri) presented the work of her group where they produced and phenotyped human ACE2 overexpression rat models generated to study SARS-CoV-2 and its impacts. Graham Duddy (Francis Crick Inst.) followed with a description of their workflow using AAV and electroporation (CRISPR-READI) to efficiently create mouse knock-in models. Next, Marie-Christine Birling (IGBMC) detailed methods to generate humanized mouse models by inserting large genomic DNAs produced by BAC engineering into mouse embryonic stem cells via CRISPR/Cas9. Roger Askew (Ozgene) then presented a new molecular screen to identify gene-targeted alleles using a fusion transcript that incorporates a polyA trap. Finally, Kathy Krentz (Univ. of Wisconsin) described their successful recovery of the Copenhagen rat line (a tumor-resistant inbred strain) many years after the initial freeze.
The 3Rs session started with a talk from Thorsten Buch (Univ. of Zurich) describing his attempt to develop a breeding calculation program to reduce mouse usage during complex breeding. His talk was followed by an overview by Vootele Voikar (Helsinki Institute of Life) of how EU regulations and outreach, in addition to adherence to the 3Rs, can contribute to ethical use of animals in research. Finally, the 3Rs Prize recipient, Ronald Naumann (MPI-CBG), presented his work showing that, by genotyping sperm samples from chimeras and looking for those with a high percentage of embryonic stem cell contribution, one can choose animals that are more likely to transmit through the germline.
After the second short poster session, a virtual plenary talk was given by Jihan Osborne (Univ. Texas-SWMC) that described her studies of the miRNAs Lin28a and Lin28b in lung development and branching morphogenesis. Using mouse embryos and embryonic stem cells, she found that the Lin28a/b miRNAs regulate the master regulators Sox2 and Sox9. She is currently investigating how these miRNAs are regulated. The following talk by Ritva Heljasvaara (Univ. of Oulu) described her use of numerous mouse transgenic lines to assess the role of the extracellular matrix in cancer. One of the genes studied, Col18a1, has isoforms that may act as tumor promoters. Finally in this session, an enthusiastic presentation was given by Andrew Syvyk (Texas A&M Univ.) to promote the use of rat spermatogonial stem cell lines to generate animal models. He demonstrated the ease of this method by showing how an overexpressing ErbB2 stem cell line could be introduced into the rete testis of Dazl transgenic rats to generate progeny that were wholly modified.
The final session of the first day was the highly anticipated presentation of the ISTT Prize to Drs. Elizabeth Lacy (Professor Emeritus, Sloan Kettering Cancer Center) and Frank Costantini (Professor Emeritus, Columbia University). A video of the presentation of the sculpture took place in a park in New York City, where Wojtek Auerbach, ISTT Past-President, gave the award to Drs. Lacy and Costantini. After viewing the presentation, we heard a live lecture that incorporated both awardees’ reminisces and discussions of their scientific careers which included some of the first methods described in the generation of transgenic animals. The audience was humbled to hear how these pioneers, at the very beginning of the age of molecular biology, overcame hurdles that included isolating genomic DNA for genotyping, determining how transgenes were incorporated, and figuring out how prokaryotic sequences inhibit the expression of transgenes. Their contribution to the methods now used to generate gene-modified animals has been immense and greatly appreciated within the community.
This presentation was followed by the conference banquet, held in the Scandic Marina Conference Center, where attendees were greeted with a flute of champagne and then treated to an excellent Finnish feast.
Day 3 The next full day of talks began with a session on non-mammalian models, and included a plenary lecture from Howy Jacobs (Univ. of Tampere) describing his use of Drosophila transgenics to study mitochondrial DNA (mtDNA). By generating mutants in nuclear DNA that function in mtDNA expression he has mimicked some human diseases. His most recent work includes the generation of cybrid flies (introducing mtDNA from one genotype into the cells from another genotype), thus establishing fruit flies as a first approach before generating the mutations in mice. In the second talk of this session, Annamarie Locasio (Stazione Zoologica Anton Dohm) described a unique animal model—Ciona robusta—a marine invertebrate chordate with embryonic development similar to other higher animals. Manipulation by electroporation of CRISPR/Cas9 components and comparison with similar mutations in Danio rerio (zebrafish) was used to identify enhancers and other means of regulating ONECUT and other HOX genes.
Short talks after the plenary lectures included one from Ralf Kühn (MDC-Mol.Med.) describing his attempts to generate the “perfect” C57BL/6 mouse by repairing known mutations and erasing endogenous retroviral sequences (ERVs), and a second talk by Mitra Cowan (McGill Univ.) describing the success she has had in generating large knock-in sequences using CRISPR/Cas9. She found that their best results were achieved using IVF to generate fertilized embryos followed by injection the following day at the 2-cell stage.
A fourth short poster session was followed by the much-anticipated “Running a Transgenic Service Facility” Roundtable chaired by Ben Davies (Univ. Oxford, Wellcome Trust). The panelists, Ben Lowe (The Jackson Laboratory), Peter Sipila (Univ. of Turku), Natalia Moncaut (CRUK-Manchester Inst.), and Pawel Pelczar (Univ. of Basel), described methods of embryo generation, reagent delivery, delivery of large DNAs, development of new technologies and how they implemented them, and took questions and comments from the audience. A very enlightening discussion surrounded how failed projects were managed: typically, most facilities perform multiple attempts to deliver the requested service. All agreed that it was difficult to determine when to stop and that communication with the contracting investigator is critical when projects run into difficulty.
The lunchtime break included a presentation by Gurpreet Balrey from Merck that described their new Cas9Plus, along with an enticing project in development to develop a chimeric Cas9 protein that will function efficiently within closed chromatin.
Jaan-Olle Andressoo (Univ. of Helsinki) began the afternoon session with a presentation of how their conditional 3’-UTR mutants of glial-derived neurotrophic factor (GDNF) were able to recapitulate aspects of schizophrenia in humans, but that the timing of Cre-excision was critical in the process. Interestingly, they noted that an interacting adenosine receptor, Adora2a, is inhibited by caffeine, and determined that blocking this receptor normalized the Gdnf mutant mice. The subsequent lecture by Tarja Malm (Univ. of Eastern Finland) examined how non-coding RNAs regulate protein expression in complex regulatory networks. She described studies of the interactions among lncRNAs, miRNAs and circulating RNAs in Alzheimer’s disease, Parkinson’s disease and ischemic stroke. The plenary lectures were followed by two short talks. The first, by Lev Federov (Oregon Health & Science Univ.), described his comparison of embryo development after culture in human tubal fluid (HTF), M16, Global Media from Cooper Surgical, or KSOM, finding that more embryos developed normally in HTF than the comparison medias. The second short talk was given by Yueh-Chiang Hu (Cincinnati Children’s Hospital), describing doxycycline-inducible models of the four core genotypes in mice (XX female, XX male, XY male and XY female) that were generated using CRISPR-interference and CRISPR-activation constructs targeted to the Col1a1 safe harbor site.
After a short poster break, the Orbis Pictus lecture, sponsored by the Czech Center for Phenogenomics (CCP) was given by Maria Lehtinen (Boston Children’s Hospital). Dr. Lehtinen’s talk was replete with beautiful images showing how the choroid plexus contributes to the cerebral spinal fluid (CSF) to control the development of the brain and subsequent neurogenesis in early mouse embryogenesis. During the response to calcium induction, unusual blebs were identified that appear to function in apocrine secretion of transthyretin (TTR), which was confirmed after generating TTR-GFP mice that secreted GFP into the CSF.
Day 4 The final day of TT2022 began with Qiangge Zhang (Massachusetts Inst. Technology) describing the process used at MIT for developing transgenic marmosets to serve as NHP models in neuroscience research. IVF-generated embryos were microinjected with transgenes to knock out SHANK3, PSEN1, MeCP2 or SETD1A. The second talk, given by Benjamin Schusser (Technical Univ. Munich), described his successful efforts to show that chickens could be protected from disease using gene editing to delete an amino acid that is critical for infection by avian leukosis virus. The short talks following these presentations included one from Stuart Newman (PetMedix LTD) where transgenic mice were generated to express canine therapeutic antibodies, and a second talk from Guillaume Pavlovic (IGBMC) where changes in the gut microbiome were assessed in response to different factors such as housing area, sex of the animal and genetic background of the animal (among others).
The final regular scientific session of the meeting focused on descriptions of emerging technologies. The plenary lecture was given by Bon-Kyong Koo (Inst. Of Molecular Biotechnology) where he presented a new artificial intron that can be used to generate conditional alleles. His short conditional intron (SCON) was used to target numerous genes, including single exon genes within specific parameters for use. The next short talk by Stephane Pelletier (Indiana Univ.) discussed his work with the “Degradation based on Cre-regulated Artificial Intron” (DECAI), where he has been able to target several genes to generate conditional alleles. The final short talk of the meeting was given by Xiaojun Xing (Yale Univ.), where he demonstrated his high-throughput method for mouse sperm cryopreservation using microstraws and batch labeling.
In the last sessions of TT2022, the Young Investigator Award was presented to Davide Seruggia (St. Anna Children’s Cancer Research Inst.) followed by his lecture, and, subsequently, an EMBO-sponsored lecture by Elizabeth Fisher (UCL Inst. Of Neurology). Dr.Seruggia described his work using CRISPR, base editing, degrons and the adeno-associated viral method to deliver components to the embryo to determine essential areas required for enhancer function and critical transcription factor sites.
Dr. Fisher then depicted the generation of mice with humanized genes used to study the genes implicated in amyotrophic lateral sclerosis (ALS), including fused in sarcoma (FUS) and transactive response DNA binding protein (TARDBP), which appear to be critical for most neurodegeneration.
The meeting ended with a joyous celebration of this successful meeting, beginning with awards, which included the Mammalian Genome-sponsored awards presented for the best oral presentations of a poster (Stephane Pelletier and Yueh-Chiang Hu), three prizes for best posters (sponsored by emeritus ISTT members) (awarded to Vanessa Chenouard from INSERM, Graham Duddy from Francis Crick Institute, and Pawel Pelczar from Univ. of Basel), and two awards from the International Mammalian Genome Society given as complimentary registration for one of their upcoming meetings (Lauri Lintott from the Hospital for Sick Children, and Tereza Nickl from the Czech Center for Phenogenomics). Jan Parker-Thornburg then issued an invitation to TT2023, to be held in Houston Texas, USA in November of 2023. Final thank-yous and closing remarks from President Füchtbauer ended this happy occasion and left delegates anticipating the next time we all could be together to present and celebrate our science.
Acknowledgements
Partial funding for JPT to attend TT2022 was supplied by funding from the NCI (R50CA211121).
Author contributions
JPT wrote the main manuscript text. FB, WS and EM contributed suggestions and edited the text for completeness and clarity. All authors reviewed the manuscript.
Funding
Office of Extramural Research, National Institutes of Health (R50CA211121).
Declarations
Competing interests
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
Reference
Yesbolatova A Saito Y Kitamoto N The auxin-inducible degron 2 technology provides sharp degradation control in yeast, mammalian cells, and mice Nat Commun 2020 11 5701 10.1038/s41467-020-19532-z 33177522
| 0 | PMC9707114 | NO-CC CODE | 2022-12-01 23:20:20 | no | Versicher Mag. 2022 Nov 29; 69(12):54-55 | latin-1 | null | null | null | oa_other |
==== Front
Int Ophthalmol
Int Ophthalmol
International Ophthalmology
0165-5701
1573-2630
Springer Netherlands Dordrecht
36443542
2576
10.1007/s10792-022-02576-z
Original Paper
Systemic biomarkers of retinopathy of prematurity in preterm babies
Sehgal Parrina 1
Narang Subina [email protected]
1
Chawla Deepak 2
Gupta Seema 3
Jain Suksham 2
Sharma Unnati 1
Katoch Deeksha 4
Kaur Jasbinder 3
1 grid.413220.6 0000 0004 1767 2831 Department of Ophthalmology, Government Medical College and Hospital, Sector32, Chandigarh, 160031 India
2 grid.413220.6 0000 0004 1767 2831 Department of Neonatology, Government Medical College and Hospital, Sector32, Chandigarh, 160031 India
3 grid.413220.6 0000 0004 1767 2831 Department of Biochemistry, Government Medical College and Hospital, Sector32, Chandigarh, 160031 India
4 grid.415131.3 0000 0004 1767 2903 Department of Ophthalmology, Post Graduate Institute of Medical Education and Research, Sector 12, Chandigarh, 160012 India
29 11 2022
19
22 4 2022
12 11 2022
© The Author(s), under exclusive licence to Springer Nature B.V. 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose
Retinopathy of prematurity (ROP) progression is an inter-play of various perinatal and neonatal angiogenic and inflammatory cytokines. A small subset of ROP progresses to ROP requiring treatment. The present study was conducted with the aim to determine whether levels of IL-6, IL-8 and VEGF in serum and urine at the time of first ROP screening visit could be a biomarker for the prediction of development of treatable ROP.
Method
Prospective single-center observational study of preterm babies screened for ROP. Blood and urine samples were collected as a part of routine sampling at initial ROP screening visit and stored at −80 °C for further processing. The babies were followed up and grouped into ‘Group A’ comprising of 35 babies who developed treatable ROP and ‘Group B’ comprising of 36 babies with regressed ROP or no ROP. The evaluation of blood and urine samples was done for IL6, IL8 and VEGF by solid-phase sandwich RayBio® Human ELISA kit.
Results
The median serum values for IL-6, IL-8 and VEGF in Group A and Group B were 5.8 pg/ml (IQR 1.5,128.5) and 8.7 pg/ml (IQR 1.5,30.5), 55.9 pg/ml (IQR 28.0, 392.9) and 27.0 pg/ml (IQR 20.5,444.9) and 26.6 pg/ml (IQR 6.3, 39.4) and 30.0 pg/ml (IQR9.2,70.3), respectively. Group A had significantly increased levels of IL-8 (p < 0.05). However, AUROC curve for serum IL-8 demonstrated suboptimal discriminating ability.
Conclusion
Babies developing ROP requiring treatment had significantly increased levels of IL-8 in the serum at the time of initial screening. However, it could not serve as predictor for treatable ROP.
Keywords
ROP inflammatory biomarkers
Biomarkers in blood
Biomarkers in urine
VEGF levels
IL-8 levels
IL-6 levels
==== Body
pmcIntroduction
Retinopathy of prematurity (ROP) is a vaso-proliferative disorder of the developing retina and is a major cause of acquired childhood blindness [1].The worldwide incidence of ROP varies from 20 to 51.9% [2, 3]. Up to 15% of screened babies progress to develop treatable ROP [4]. It is estimated that around 14.5 million children are born prematurely every year worldwide, out of which 1,84,700 children develop any stage of ROP. Of these, 20,000 suffer blindness or severe visual impairment.1In high-income countries where most of the babies have access to high-quality neonatal care, ROP-associated blindness is uncommon. On the other hand, developing countries have not only the highest number of preterm births but also suboptimal infrastructure and capacity to screen these babies.
ROP is a biphasic disease consisting of an initial phase of blunted vascular growth characterized by low vascular endothelial growth factor (VEGF) and insulin-like growth factor (IGF-1) levels followed by a second phase of vaso-proliferation with an increase in VEGF and other pro-angiogenic markers [5]. Angiogenesis, the fundamental process involved in the development of retinal vasculature, is tightly regulated by a complex network of various factors in a time-dependent fashion. Although inflammatory cytokines can modulate angiogenesis [6, 7], their role in triggering the dysregulated angiogenesis in ROP has not been fully understood. Recent evidence suggests that prenatal, perinatal, and post-natal inflammation might contribute to a gradual increase in the risk for ROP [7]. Upregulation of pro-inflammatory cytokines is seen in hypoxic neonatal retina in rats [8].Cytokines such as IL-1β, TNF-α, and IL-6 act as primary initiators of inflammation following infection or tissue damage with cytokine receptor activation leading to downstream upregulation of effector molecules such as chemokines. Some of the chemokines studied for ROP pathophysiology include IL-8, IL6, monocyte chemotactic protein-1, and interferon-inducible T-cell alpha chemoattractant (I-TAC).
IL-8 has been implicated in both inflammation and pathological neovascularization in the eye [7, 9]. In adults, IL-8 is seen to play a major role in inflammation and angiogenesis in cornea, conjunctiva, iris and retina [9].
There is the scant literature on the inflammatory cytokines in ROP. The present study was conducted to determine whether the levels of IL-6, IL-8 and VEGF levels, in the serum and/or urine, at initial ROP screening of babies [10], could predict the development of ROP requiring treatment as per ETROP criteria (Type1 ROP) [11]. We hypothesized that serum inflammatory interleukins and VEGF at initial screening could have a role in progression to treatable ROP. We also hypothesized that urine analysis could act as a surrogate marker for the serum levels of these markers.
Material and methods
This was a hospital-based, institution review board (IRB) approved, prospective observational study involving preterm babies done between January 2020 and October 2020. Eligible neonates were enrolled after obtaining a written informed consent from the parents. The preterm babies with birth weight ≤ 2000 g and gestational age ≤ 34 weeks who were screened for ROP [10] were eligible for enrollment. We excluded neonates with major congenital abnormalities, hydrocephalus, necrotizing enterocolitis, periventricular leukomalacia, any evidence of TORCH infection or babies born to a mother with gestational metabolic diseases or TORCH infection.
A detailed history was taken and gender, gestational age, post-conception age, birth weight, duration of ventilation, multiplicity, any existing co-morbidities or congenital abnormality were recorded. Enrolled babies were screened for ROP by an ophthalmologist (SN) with more than 20 years of experience in managing ROP. At the time of first screening for ROP, i.e., 3–4 weeks of post-natal age, three ml of blood samples was collected as part of routine sampling. After centrifugation, the serum was separated and stored at −80 °C. Five ml of urine sample was also collected into a sterile or disposable container and stored at −80 °C. The screening examination was conducted under topical anesthesia (2% proparacaine) using a + 20 D lens, an indirect ophthalmoscope, a pediatric eye speculum and a wire Vectis for globe rotation. The exact stage and zone of ROP along with the presence of plus disease were recorded as per ICROP classification [12]. Preterm neonates were screened and followed up till the development of treatable ROP as per the ETROP study [11] (Type1 ROP requiring treatment) or till complete vascularization till ora-serrata or till 45 weeks of post-menstrual age. The mean time gap between the sampling and treatment was 4 weeks.
Group A (n = 35) comprised of the neonates who developed ROP needing treatment in either eye (type 1 ROP) at any time during study. Group B or the control group (n = 36) comprised of gestational age-matched neonates from the remaining cohort, who did not develop ROP needing treatment in both the eyes (type 2 ROP or no ROP). For the purpose of classification as case or control, the worst stage of ROP in either eye was considered.
Outcome measurement
Quantitative analysis of systemic IL-6, IL-8 and VEGF was done in batches by solid-phase sandwich ELISA technique. The IL-6 and IL8 levels were measured using DIACLONE IL-6/IL8 ELISA kit. VEGF level was assessed using RayBio® Human VEGF-A ELISA kit. Readings were taken using Tecan’s Magellan™ universal ELISA reader. The minimum detectable concentration was 2 pg/ml, 29 pg/ml and 10 pg/ml for IL-6, IL-8 and VEGF, respectively. Intra-assay and inter-assay coefficient of variation (CV %) for all parameters was < 10% in all the kits. The data collected were analyzed, and expression of biomarkers in blood and urine samples was determined.
Statistical analysis
The sample size in the study was calculated to detect differences in the main outcome measure, that is, IL-6 in the study. In the study by Silveira et al., IL-6 levels were reported to be 72 ± 172 pg/ml in neonates without ROP and 302 ± 1583 in neonates with severe ROP[13]. To detect at least twofold increase in levels of IL-6 with 80% power and two-sided alpha error of 5%, 35 neonates were required to be enrolled in each group.
Data were collected in an Excel database (Microsoft office 2019; Microsoft Inc). Data analysis was performed using Stata 16.1 (College Station, TX). Depending on distribution, numerical data were analyzed using t test or Mann–Whitney U Test. Categorical data were analyzed using chi-square test or Fisher exact test. Correlation was analyzed using Spearman’s correlation coefficient. P value < 0.05 was considered as significant. Receiver operating characteristic (ROC) curves were analyzed for all cytokines studied. Once the final diagnosis of ROP was made by indirect ophthalmoscopy, the cutoff point for each cytokine was detected and the best cutoff point of each cytokine was then used to compare Group A and Group B.
Results
A total of 249 neonates were screened for ROP during the study period. Of these, 78 neonates fulfilled the inclusion criteria. Seven babies were excluded based on predefined criteria. (Fig. 1). Of the 35 babies with ROP, Zone 1 (A-ROP in 7) and Zone II (AROP in 6) ROP was seen in 7 and 28 eyes, respectively. Stage 2, Stage 3, Stage 4 A, and Stage 5 ROP was seen in 7, 11, 2, 2 babies, respectively. The babies with stage 4 and 5 were those who had missed their screening visits due to lock down during COVID-19 pandemic. All babies had symmetrical ROP. As neonates eligible to be enrolled as controls were not available as gestation matching pairs, we enrolled neonates from the available cohort of neonates without ROP. Thirty-six neonates who did not develop treatable ROP were included as controls. Of these, 24 developed type 2 ROP which regressed spontaneously and 12 did not develop any ROP. All the neonates with ROP had symmetrical disease in both eyes.Fig. 1 Flow diagram of the study
In comparison to controls, the cases had lower birth weight (1309 ± 244 g vs. 1524 ± 297 g) and smaller gestational age (30.2 ± 1.8 weeks vs. 32.0 ± 1.5 weeks). Cases were more likely to have respiratory distress syndrome (86% vs. 33%) and needed longer duration of respiratory support including oxygen, CPAP, and invasive ventilation (Table 1). The blood and urine samples were collected at the time of first screening examination. The post-conceptual age (PCA) at the time of sample collection in cases and controls was 34.8 ± 3.1 weeks and 36.0 ± 1.7 weeks, respectively.Table 1 Neonatal characteristics in cases developing treatable ROP and controls
S.No Factor Cases (n = 35) Controls (n = 36) p value Test
35 36
1 Birth weight, mean (SD) 1309 (241) 1524(297) 0.001 Two sample t test
2 Gestational age at birth, mean (SD) 30.2 (1.8) 32.0(1.5) < 0.001 Two sample t test
4 Male 21(60%) 19 (53%) 0.54 Pearson’s chi-squared
5 Appropriate for gestational age babies 27 (77%) 30 (83%) 0.51 Pearson’s chi-squared
6 Small for gestational age babies 6 (17%) 6 (17%) 0.96 Pearson’s chi-squared
7 Large for gestational age babies 2 (6%) 0 (0%) 0.24 Fisher’s exact
8 Respiratory Distress Syndrome 30 (86%) 12 (33%) < 0.001 Fisher’s exact
9 Transient tachypnoea of newborn 8 (23%) 2 (6%) 0.036 Pearson’s chi-square
10 Pneumothorax 1(3%) 0 (0%) 0.49 Fisher’s exact
11 Bacterial sepsis (culture positive) 7 (20%) 4 (11%) 0.30 Pearson’s chi-squared
12 Septic shock 3 (9%) 1 (3%) 0.36 Pearson’s chi-squared
13 Jaundice requiring treatment 33 (94%) 28 (78%) 0.05 Pearson’s chi-squared
14 Meningitis 3 (9%) 2 (6%) 0.67 Fisher’s exact
15 Necrotizing colitis stage 2 or more 1 (3%) 0 (0%) 0.49 Fisher’s exact
16 Broncho pulmonary dysplasia 35 (100%) 36 (100%) Fisher’s exact
17 Exchange transfusion 2 (6%) 4 (11%) 0.67 Fisher’s exact
18 PRBC transfusion 3 (9%) 0 (0%) 0.11 Fisher’s exact
19 Oxygen ventilation, median (IQR) 12.0 (10.0, 14.0) 1.5 (0.0, 8.0) < 0.001 Wilcoxon rank-sum
20 Continuous positive airway pressure ventilation, median (IQR) 4.0 (1.0, 8.0) 1.0 (0.0, 4.0) 0.004 Wilcoxon rank-sum
22 Total days of ventilation, median (IQR) 19.0 (13.0, 27.0) 3.0 (0.0, 13.0) < 0.001 Wilcoxon rank-sum
The details of the level of biomarkers are given in Table 2. Serum IL-6 and VEGF levels were similar in cases and controls. However, serum IL-8 levels were significantly higher in cases (median [IQR]: 55.9 [28.0 to 392.9] vs. 27.0 [20.5 to 441.9], P = 0.048). Levels of all the three markers were comparable in urine. The area under the ROC curve (AUROC) for serum IL-8 predicting case vs control was 0.636 (95% CI: 0.501–0.771), thus demonstrating poor diagnostic performance (Figs. 2, 3, 4).Table 2 The level of serum IL-6, IL-8 and VEGF in cases and controls
Factor GROUP A
Cases (n = 35) GROUP B
Controls (n = 36) p value
Serum IL-6 median (IQR) 5.8 (1.5, 128.5) 8.7 (1.5, 30.5) 0.43 Wilcoxon rank-sum
Serum IL-8 median (IQR) 55.9 (28.0, 392.9) 27.0 (20.5, 441.9) 0.048 Wilcoxon rank-sum
Serum VEGF median (IQR) 26.6 (6.3, 39.4) 30.0 (9.2, 70.3) 0.081 Wilcoxon rank-sum
Urine IL-6 median (IQR) 1.5 (1.0, 2.4) 2.2 (1.0,5.5) 0.14 Wilcoxon rank-sum
Urine IL-8 median (IQR) 26.0 (24.0, 45.5) 25.0(16.8, 31.8) 0.33 Wilcoxon rank-sum
Urine VEGF median (IQR) 20.7 (7.4, 46.0) 16.1 (6.8, 55.0) 0.78 Wilcoxon rank-sum
Fig. 2 AUROC curve for IL-8. The area under the ROC curve (AUROC) for serum IL-8 predicting case vs control was 0.636 (95% CI: 0.501–0.771), thus demonstrating suboptimal discriminating ability
Fig. 3 Fundus picture of baby of birthweight 1150 g, gestational age of 29 weeks with respiratory distress syndrome showing Zone II, Stage 3 ROP with initial screening serum IL-6 of 128 pg/ml, IL-8 239 pg/ml and VEGF 6.5 pg/ml
Fig. 4 Fundus picture of baby of birthweight 900 g, gestational age of 31 weeks with transient tachypnoea of newborn, showing Zone I A-ROP with initial screening serum IL-6 of 200 pg/ml, IL-8 2000 pg/ml and VEGF 31.5 pg/ml. Figure 4 inset fluorescein angiogram of the same eye showing large avascular area extending to zone I
IL-6 levels in serum showed positive correlation with IL-8 levels in serum, (p = 0.002). For every 1 unit increase in IL-6 (serum), the IL-8 (serum) increased by 2.32 units. Also, moderate positive correlation was seen in IL-8 in serum with IL-6 level in urine (p = 0.06). IL-8 in urine showed positive correlation with VEGF level in urine (p = 0.049). Both IL-8 and IL-6 levels had a positive effect on each other. If the level of one increased, levels of other parameters increased in concentration in both serum and urine.
In controls, positive correlation of IL-6 level in serum was seen in association with IL-8 level in serum (p = 0.075) and VEGF level in serum (p = 0.041). Urinary levels of IL-6 increased as IL-8 and VEGF in urine increased (p = 0.04 and p = 0.4). With rise in IL-8 in urine, rise in VEGF level in urine was seen in the control group (p = 0.046).
Association of stages and zones with biomarkers was also studied, and it was seen that higher levels of IL-6 and IL-8 in serum were present with increasing stages of the disease. VEGF level in serum and urinary levels of each biomarker showed no significant difference between various stages and zones, but values of IL-8 level in urine were higher as compared to other parameters.
Correlation between serum and urine levels of IL-6 (r = 0.07), IL-8 (r = −0.10) and VEGF in cases (r = 0.07) was poor.
Discussion
The pathogenesis of ROP remains a dilemma for the treating ophthalmologists. The role of vasculogenesis and angiogenesis has been vastly studied in ROP, but the inter-play of inflammatory biomarkers causing the disease is not well understood [7, 13]. Some studies have described the effect of multiple “hits” of exposure from various risk factors, like perinatal inflammation, oxygen supplementation, lower gestational birth or neonatal sepsis, on the development of severe form of ROP [7, 8]. Despite the documented affirmed role of various risk factors in ROP, etiology behind the inflammation is still not very clear. In a study by Rathi et al., the authors have done quantitative analysis of the cytokines in the tear film and the vitreous sample of eyes with severe ROP [14], while Vinekar A et al. studied the status of pro-angiogenic factors in the tear fluid of preterm infants with and without ROP [15]. However, in the present study, we studied the role of inflammatory cytokines in the blood and urine samples of babies undergoing ROP screening as a predictive tool for the development of severe ROP. The determination of systemic inflammatory biomarkers in babies with ROP could help in the better understanding of the disease.
In the present study, the mean birth weight of babies developing treatable ROP was 1309 ± 244 g and the mean gestational age was 30.2 ± 1.8 weeks which was significantly lower than the control group (p < 0.001). Birth weight and gestational age are major risk factors for ROP. However, the birth weight and gestational age in our cohort are higher than that reported by the Cryo-ROP study or the ETROP study [2, 11]. Lately, ROP has been reported in babies with higher birth weight and larger gestational age [16, 17]. Among the neonatal morbidity factors, respiratory distress syndrome (RDS), longer duration of respiratory support including oxygen, CPAP, and invasive ventilation were the only risk factors significantly associated (p = < 0.001) with the development of ROP which is in concurrence with previous studies [17, 18].
In the present study, IL-8 levels in blood were significantly raised in cases (55.9 pg/ml) in comparison to controls (27 pg/ml). The levels were much higher than the normal values as seen in healthy neonates where IL-8 in serum has been found to be in the 8- 21.1 pg/ml range [19]. There were other factors like respiratory distress syndrome, CPAP which could also be the contributing factors for IL8, but these are established risk factors for ROP.
IL-8 has broad spectrum effects on neutrophil recruitment in inflammation, cell adhesion, honing of neutrophils and lymphocytes, tumor growth, angiogenesis, neuronal protection and brain development. The raised levels of IL-8 have been seen in various vitreoretinal diseases like retinal vein occlusion, diabetic macular edema, age-related macular degeneration, suggesting a common pathway for angiogenesis in these diseases [9]. IL-8 has been correlated with raised IL-6 levels in numerous eye diseases which highlights the role of inflammation in these diseases. In the present study also, for every unit increase in IL-6 levels, there was 2.32 unit increase in IL-8 levels. This is consistent with earlier studies where the levels of IL-6 and IL-8 positively correlate with each other in anterior segment diseases with manifest inflammation or occult inflammation associated with hypoxia [20].
It was also seen that IL-8 levels in serum were more as compared to IL-6 levels which might be due to short half-life of IL-6 which remains active only for the first few days of inflammation. However, the levels could not be used as the predictor for disease progression and the area under the ROC curve demonstrated poor diagnostic performance.
Sood et al. have studied the role of cytokines with perinatal inflammatory response and have observed that although IL-6 was significantly associated with ROP, the association could only be seen at day 14 of birth and not thereafter making it a less reliable or less specific marker[21]. In the present study, the sample was withdrawn once at the time of initial screening, which is 3–4 weeks after birth. Sequential samples could have given us more information.
The mean value of VEGF in serum of cases (30 ± 33 pg/ml) and controls (43 ± 36 pg/ml) was comparable (p = 0.08). Lower levels of VEGF in infants with ROP compared with infants without ROP have been previously documented[15]. Hellgren et al. studied the median VEGF concentrations in cord blood and concluded that no correlation was present between VEGF levels at birth with later development of ROP [22]. It was found that serum VEGF levels peaked at the first detection of ROP and not at the time of laser treatment or proliferative stage of ROP. The authors concluded that no significant change occurs in the levels of serum VEGF when the samples were taken before the treatment or within 2 weeks after giving treatment. Pieh et al. investigated longitudinal VEGF concentrations where they have reported no differences in VEGF levels in serum in ROP and the control group in a subgroup of patients between 32 and 36 weeks of post-menstrual age [23]. It has been seen in different studies that various risk factors may play a role in keeping sustained high levels of plasma VEGF like bronchopulmonary dysplasia or supplemental oxygen therapy, in the absence of which VEGF levels in the blood usually come to low levels in a few weeks[24]. In the present study, we did not observe any difference between cases and controls in samples collected at 34.8 weeks of PCA which is consistent with the two above-mentioned studies. There are possibly other serum biomarkers or genetic factors which play a role in the development of ROP.
An alternative treatment to laser is desired as even in the present era with vast research work being done, adequate available treatment of ROP is still associated with unsuccessful visual outcome and recurrence of the disease. As an alternative treatment immune targeting of IL-8 by anti-IL-8 cytokines like IL-10 or medications like cholecalciferol or steroids early in life may be studied to prevent the occurrence of ROP[9]. Systemic biomarkers analysis can help us in targeting concerned biomarker in the early neonatal phase so as to avoid the future occurrence of Type 1 ROP even before screening and can help us in predicting the babies who are likely to develop a severe form of ROP in future.
The main limitation of our study was that we could not achieve the exact gestation age matching in the two groups as neonates with ROP were of significantly lower gestation and this could be the confounding factor. Moreover, in the present study, the PCA at the time of sample collection in cases and controls was 34.8 ± 3.1 weeks and 36.0 ± 1.7 weeks, respectively. Neonates of lower gestation who are more prone to develop ROP are screened for the first time by 3 weeks of post-natal age, while relatively mature neonates who would not have developed ROP were screened after completing 4 weeks of age. This accounts for a difference of about 1 week in the post-conceptional age. Circulating chemokine can bind to neutrophils, and therefore, plasma IL8 may not be a true indicator of total IL8 and plasma IL-8 reflects only a portion of the total IL-8 blood pool. Moreover, there was only one-time assessment of biomarkers. The serial longitudinal analysis could have given us a better understanding while studying the pattern of association.
To conclude, babies requiring treatment requiring ROP had significantly increased levels of IL-8 which are seen at the time of initial screening. In future, IL-8 may be used as a new biomarker or therapeutic tool for treatable ROP. More prospective studies are required in this direction. If validated by further studies, the incorporation of the quantitative assessment of IL-8 in blood and urine samples as a routine test in premature infants may help predict the course of ROP in these infants.
Author contributions
PS and SN wrote the main manuscript and evaluated babies for ROP stage and extent. DC and DK did the statistical analysis. SG and JK did the biomarker assay. SN, DC and SJ did the editing for the manuscript. All authors reviewed the manuscript.
Funding
This study was not funded by any other agency.
Declarations
Competing interests
The authors declare no competing interests.
Conflict of interest
None of the authors has any conflicts of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from parents of all individual participants included in the study.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Gilbert C Foster A Childhood blindness in the context of VISION 2020: the right to sight Bull World Health Organ 2001 79 227 232 11285667
2. Quinn GE Retinopathy of prematurity blindness worldwide: phenotypes in the third epidemic Eye brain 2016 8 31 36 10.2147/EB.S94436 28539799
3. Honavar SG Do we need India-specific retinopathy of prematurity screening guidelines? Indian J Ophthalmol 2019 67 711 716 10.4103/ijo.IJO_973_19 31124474
4. Jayadev C Vinekar A Bharamshetter R Mangalesh S Rao HL Dogra M Retinal immaturity at first screening and retinopathy of prematurity. Image-based validation of 1202 eyes of premature infants to predict disease progression Indian J Ophthalmol 2019 67 846 53 10.4103/ijo.IJO_469_19 31124500
5. Flynn JT Chan-Ling T Retinopathy of prematurity: two distinct mechanisms that underlie zone 1 and zone 2 disease Am J Ophthalmol 2006 142 46 59 10.1016/j.ajo.2006.02.018 16815250
6. Chen J Smith LEH Retinopathy of prematurity Angiogenesis 2007 10 133 140 10.1007/s10456-007-9066-0 17332988
7. Lee J Dammann O Perinatal infection, inflammation, and retinopathy of prematurity Semin Fetal Neonatal Med 2012 17 26 29 10.1016/j.siny.2011.08.007 21903492
8. Sivakumar V Foulds WS Luu CD Ling EA Kaur C Retinal ganglion cell death is induced by microglia derived pro-inflammatory cytokines in the hypoxic neonatal retina J Pathol 2011 224 245 260 10.1002/path.2858 21404274
9. Ghasemi H Ghazanfari T Yaraee R Faghihzadeh S Hassan ZM Roles of IL-8 in ocular inflammations: a review Ocul Immunol Inflamm 2011 19 401 412 10.3109/09273948.2011.618902 22106907
10. Pejawar R, Vinekar A, Bilagi A. National Neonatology Foundation’s Evidence-based Clinical Practise Guidelines, Retinopathy of Prematurity, NNF India, New Delhi 2010:253–62
11. Early Treatment for Retinopathy of Prematurity Cooperative Group Revised indications for the treatment of retinopathy of prematurity: results of the early treatment for retinopathy of prematurity randomized trial Arch Ophthalmol 2003 121 1684 1694 10.1001/archopht.121.12.1684 14662586
12. The International Committee for Classification of Retinopathy of Prematurity International classification of retinopathy of prematurity revisited Arch Ophthalmol 2005 123 991 9 10.1001/archopht.123.7.991 16009843
13. Silveira RC Fortes Filho JB Procianoy RS Assessment of the contribution of cytokine plasma levels to detect retinopathy of prematurity in very low birth weight infants Invest Ophthalmol Vis Sci 2011 52 1297 1301 10.1167/iovs.10-6279 21071735
14. Rathi S Jalali S Patnaik S Shahulhameed S Musada GR Balakrishnan D Rani PK Kekunnaya R Chhablani PP Swain S Giri L Chakrabarti S Kaur I Abnormal complement activation and inflammation in the pathogenesis of retinopathy of prematurity Front Immunol 2017 8 1868 10.3389/fimmu.2017.01868 29312345
15. Vinekar A Nair AP Sinha S Tear fluid angiogenic factors: Potential noninvasive biomarkers for retinopathy of prematurity in preterm infants Invest Ophthalmol Vis Sci 2021 62 2 10.1167/iovs.62.3.2 33646290
16. Dwivedi A Dwivedi D Lakhtakia S Chalisgaonkar C Jain S Prevalence, risk factors and pattern of severe retinopathy of prematurity in eastern Madhya Pradesh Indian J Ophthalmol 2019 67 819 823 10.4103/ijo.IJO_1789_18 31124494
17. Kumar P Sankar MJ Deorari A Azad R Chandra P Agarwal R Paul V Risk factors for severe retinopathy of prematurity in preterm low birth weight neonates Indian J Pediatr 2011 78 812 816 10.1007/s12098-011-0363-7 21340729
18. Dutta S Narang S Narang A Dogra M Gupta A Risk factors of threshold retinopathy of prematurity Indian Pediatr 2004 41 665 671 15297681
19. Boskabadi H Maamouri G TavakolAfshari J Evaluation of serum interleukins-6, 8 and 10 levels as diagnostic markers of neonatal infection and possibility of mortality Iran J Basic Med Sci 2013 16 1232 1237 24570828
20. Zenkel M Lewczuk P Junemann A Kruse FE Naumann GOH Schlotzer-Schrehardt U Pro-inflammatory cytokines are involved in the initiation of the abnormal matrix process in Pseudoexfoliation syndrome/glaucoma Am J Pathol 2010 176 2868 2879 10.2353/ajpath.2010.090914 20395431
21. Sood BG Madan A Saha S Schendel D Thorsen P Skogstrand K Perinatal systemic inflammatory response syndrome and retinopathy of prematurity Pediatr Res 2010 67 394 400 10.1203/PDR.0b013e3181d01a36 20032809
22. Hellgren G Löfqvist C Hård AL Hansen-Pupp I Gram M Ley D Smith LE Hellström A Serum concentrations of vascular endothelial growth factor in relation to retinopathy of prematurity Pediatr Res 2016 79 70 75 10.1038/pr.2015.181 26372519
23. Pieh C Agostini H Buschbeck C Kruger M Schulte-Monting J Zirrgiebel U VEGF-A, VEGFR-1, VEGFR-2 and Tie2 levels in plasma of premature infants: relationship to retinopathy of prematurity Br J Ophthalmol 2008 92 689 93 10.1136/bjo.2007.128371 18408080
24. Chen X Zhou L Zhang Q Xu Y Zhao P Xia H Serum vascular endothelial growth factor levels before and after intravenous ranibizumab injection for retinopathy of prematurity J Ophthalmol 2019 2019 20 2985161 31236289
| 36443542 | PMC9707116 | NO-CC CODE | 2022-12-01 23:20:20 | no | Int Ophthalmol. 2022 Nov 29;:1-9 | utf-8 | Int Ophthalmol | 2,022 | 10.1007/s10792-022-02576-z | oa_other |
==== Front
Public Organiz Rev
Public Organization Review
1566-7170
1573-7098
Springer US New York
688
10.1007/s11115-022-00688-9
Article
Innovative Business Models in the Context of Organizational Culture Transformation in the COVID-19 Pandemic
Bayramov Shahin Vaqif [email protected]
[email protected]
1Shahin Vaqif Bayramov
PhD. in Econ., Rector, Department of Economics, Mingachevir State University (MSU), Mingechevir, Azerbaijan. In 2005 he entered the doctoral studies of UNEC. In 2012 he defended his dissertation on the theme “EU Neighborhood Policy and Prospects for Integrating Azerbaijan into the European Economic Area” and in 2012, by the decision of the Higher Attestation Commission under the President of the Republic of Azerbaijan, he received PhD in economics. From January 2005 to September 2016, he worked as Senior Laboratory Assistant, Lecturer, Senior Lecturer and Associate Professor at the Department of International Economic Relations of Azerbaijan State University of Economics. In 2005-2009, he worked as a Manager in Master Business Administration (MBA) Program at UNEC, in 2009-2011 he worked as a head of department at the centre of Distant and Additional Education. From November 2012 to April 2017, he worked as a vice-rector for International Relations and Programs at Azerbaijan State University of Economics. In 2015, she was chosen as the winner of the Hubert H. Humphrey Program implemented by the Institute of International Education under the US State Department within the framework of Fulbright Scholarship and participated in professional development trainings in the United States. In the 2015/2016 academic year, she received professional education in Higher Education Administration and Academic Leadership at the College of Education of University of Pennsylvania . He also participated in a six-week internship at Cornell University. During his studies in the United States, he has participated in numerous professional trainings and forums on higher education management, academic leadership, university management, academic management in crisis situations, decision-making, human resource management, negotiating. From April 2017 to July 2019, he worked as a deputy head of the department of Science, Higher and Secondary Specialized Education at the Ministry of Education of the Republic of Azerbaijan. Since September 2018, he has been working as an Associate Professor (0.5) in the Department of Economics and Management at the Center for International Masters and Doctoral Studies at Azerbaijan State University of Economics. In 2009 he was chosen as a winner of the US State Department’s “Junior Faculty Development Program”, and studied economics at Indiana University in Bloomington (the USA). By the Decree of the President of the Republic of Azerbaijan dated July 31, 2019, he was appointed rector of Mingachevir State University.
Kookueva Viktoriya 2Viktoriya Vladimirovna Kookueva
PhD in Economics, Associate Professor, Department of Financial Management, Plekhanov Russian University of Economics, Moscow, Russian Federation. Research interests: economic resources in the digital society, innovative development in Russia; competitiveness in national bussiness. In 2004 she defended the thesis titled “Ways to strengthen the revenue base of local budgets”. From 2004 to 2013 - Assistant, Senior Lecturer, Associate Professor of the Department of Finance and Credit, Kalmyk State University. In 2010, he was awarded the title of Associate Professor in the Department of Finance and Credit. From 2013 to 2015 - Associate Professor of the Department of Banking and Financial Management, Moscow State University of Economics, Statistics and Informatics (MESI). Since 2015 - Associate Professor of the Department of Financial Management, Plekhanov Russian University of Economics. Since 2016, a member of the Union of Financial and Economic Forensic Experts.
Kolobov Alexander 3Alexander Vladimirovich Kolobov
PhD in Engineering Science (Baikov Institute of Metallurgy and Materials Science), Business System Development Director of Severgroup LLC, Cherepovets, Russian Federation.
Tsvetkova Liudmila 4Liudmila Ivanovna Tsvetkova
Candidate of Economic Sciences, Department of Risk Management and Insurance, MGIMO University. 2021 — defended her thesis for the degree of Doctor of Economics on the topic “Development and application of the theory of capital in the insurance market”. Research interests include risk management of business structures, the problems of public-private partnership in the field of insurance business, as well as the problems of forming an effective structure of insurance capital. As a developer, she participated in the project to create a risk management system for KAMAZ PTC, led and successfully completed a project to create a risk management system for PJSC “Nizhnekamskneftekhim”.
1 Department of Economics, Mingachevir State University (MSU), D. Alieva St., 21, Mingachevir , AZ4500 Azerbaijan
2 grid.446263.1 0000 0001 0434 3906 Department of Financial Management, Plekhanov Russian University of Economics, Stremyanny Lane, 36, Moscow, 108807 Russian Federation
3 Department of Business System Development, Severgroup LLC, Metallurgov St., 13a, 162606 Cherepovets, Russian Federation
4 grid.446171.1 0000 0001 2289 4349 Department of Risk Management and Insurance, MGIMO University, Vernadsky Avenue, 76, Moscow, 119435 Russian Federation
28 11 2022
123
3 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The purpose of this article was to establish the interrelations and mutual influence of organizational culture transformation, national culture, and innovative business models’ formation under the COVID-19 pandemic. To achieve the study goal, a multi-stage comprehensive exploratory study based on primary information obtained by expert interviewing was implemented. The developed methodological approach was tested on the example of small and medium-sized enterprises in Russia and China. As a result of the conducted testing, the previously formulated hypotheses were fully proved. H1: the negative shock of the pandemic and the ensuing quarantine restrictions caused critical changes in the business environment, cancelling the relevance of previously effective business models. H2: the new business environment required enterprises to significantly transform their organizational culture. H3: under the conditions of shock changes in the business environment and forced transformation of organizational culture the main condition for the sustainable development of enterprises is innovative business models. Н4: implementation of innovative business models depends on both the business environment and the influence of national culture. The main obstacles to the implementation of innovative business models in small and medium-sized enterprises were also identified. The developed methodological approach can be used in the real sector of the economy by state organizations and non-state funds in the development of programs to assist small and medium-sized businesses in the COVID-19 pandemic. In addition, the study results may be of interest to academic researchers, suggesting empirically determined stable relationships (hypotheses H1, H2, H3, H4) between the transformation of organizational culture, national culture, and innovative business models’ formation to ensure sustainable development of enterprises under the COVID-19 pandemic and opening new promising directions for further research.
Keywords
Cooperation
COVID-19 pandemic
Innovative business model
Integration
Licensing
==== Body
pmcIntroduction
In March 2020, humanity faced a new challenge: the Coronavirus (COVID-19) pandemic that forced governments to introduce unprecedented protective measures, such as self-isolation, social distance, mass public masking, and additional hygiene requirements. Under the COVID-19 outbreak, these efforts have become a part of daily life. The COVID-19 pandemic, unlike previous endogenous economic crises, has not just hit the world economy, but introduced new rules and behavior patterns for the population which will remain unchanged after the crisis (Choi & Sethi, 2021). The enterprises involved in the manufacturing of physical products, especially labor-intensive industries, have been forced to optimize their business processes, minimize operations or temporarily close their business (Akhmetzhanov et al., 2019; Seetharaman, 2020). The COVID-19 pandemic has affected all sectors of the global economy. For many enterprises, the transition to a remote format of work was the only alternative to complete closure, which undoubtedly entailed the need to change the business models of enterprises and their organizational culture as a whole. Thus, the primary motivation for this study was the surge in social demand for innovative business models that consider peculiarities of doing business and changed behaviors as a result of the COVID-19 pandemic.
Literature Review
The relevance of changing business models in conditions of organizational culture transformation, caused by unprecedented challenges of the external environment, led to increased interest in the research topic in both a highly specialized and interdisciplinary aspects.
Theoretical Aspects and Systematization of Prior Knowledge
Much of the research is devoted to general theoretical issues, rethinking of the conceptual apparatus, as well as systematization of the experience of predecessors (literature reviews). The research conducted by Piccarozzi et al. (2021) examined growth opportunities and challenges faced by enterprises under the COVID-19 pandemic. The researchers recommend new scenarios of doing business including an increasing role of online commerce and workplace and business operations digitalization. The main challenges faced by businesses are disruptions in supply chains that reduced consumer demand and caused the closure of enterprises. The research conducted by Wit (2016) explores strategic management including an innovative business model which helps to understand the essence of the organizational development process (Kreutzer, 2017), functionality, and enterprise survival strategies.
Problems of Transformation
Much of the existing research is devoted to the problem of business transformation, primarily - organizational culture, in a turbulent economy and significant challenges of the external environment. The article by Pwc (2021) considers an organization’s culture as a set of behavior patterns that influence business performance. The research highlights that the management culture is based on two key issues: employees’ values and behavior patterns. They argue that change in organizational behavior patterns and values of employees is the most effective approach to initiate change in management culture. Therefore, new behavior patterns force employees to think outside the box resulting in new values and culture change. The transformation of organizational culture should be seen as a continuous process that cannot be ignored (Singaraju, 2021). Furthermore, organizational culture should meet the strategic goals of an enterprise and improve business operations (Dobrotina, 2021). In many ways, organizational strategy and culture encourage each other to continue the further development of an enterprise. A motivated and enthusiastic workplace goes beyond the set goals and provides an enterprise with disruptive innovations which help to anticipate and overcome transformational challenges. Thus, organizational culture is closely connected with organizational values, assumptions, behavior patterns, and attitudes. It can be considered a driven factor that makes an enterprise unique reflecting its leadership models and creating the conditions for further cultural transformations. New organizational culture reshapes organizational behavior engaging all employees in the change process during day-to-day work (Eaton et al., 2021).
Lankhorst (2016) considered that in the digital age, organizational culture change plays a key role because a company’s architecture facilitates innovation development, flexibility, and scalability, learning by doing and experimentation. The organizational culture change means changing those behavior patterns that have a negative impact on the overall performance and are resistant to change and keeping unchanged those cultural aspects that have a positive impact on the enterprise management (Eaton et al., 2021; Lankhorst, 2016) argues that an enterprise should consider the key factors of the transformation process including but not limited to: leadership which implies the adoption of agile management to facilitate the assessment and implementation of new business models, sources of finance and organizational opportunities.
Innovative Business Models
In today’s economy, the business models are characterized as a new business culture (Abreu, 2021) that provides business opportunities through corporate culture management and contributes to operational effectiveness and new relationships with consumers. The research conducted by Seetharaman (2020) uses the three-dimensional approach to analyze the innovative business models of an organization:
information intensity of a product and service (the degree of customer interest in the product selection and buying),
information intensity of the process and value chain (investments in information technology, which benefit the enterprise, for example, access to online sites),
the need for a product and service in the national economy under the COVID-19 pandemic.
The research found that the digitalization of business processes met the needs of the new market ecosystem. Under the pandemic restrictions, the business must react accordingly to coming changes and implement digital tools into business operations. It will help to minimize physical contact and ensure a secure social environment, flexibility, and dynamic capabilities of an enterprise such as innovation, alternative digital products and services development, search for strategic solutions and partners. Thus, enterprises that adopted an innovative business model are more market-oriented, consumer-centered, and responsive to the new business ecosystem (Digital Reality, 2020), which creates a new socio-cultural environment and develops a management culture for further qualitative transformations (Bossert & Van der Wildt, 2021).
Digitalization (Digital Transformation) in the Context of Organizational Culture Change
The emergence of a very large part of innovative business models is closely related to digital transformation, which has led to considerable attention by contemporary researchers to various aspects of digital transformation.
Almeida et al. (2020) examined the impact of the COVID-19 pandemic on business systems and identified changes in three main areas: new employment relationships in the labor market and new social relations between employers and employees. Remote work modes forced employees to combine office responsibilities with household duties, which resulted in a decrease in average working and living space per each family member. Marketing and sales management have been affected by an increase in online commerce due to the closure of retail chains and the limited mobility of buyers. Technologies have been widely adopted by businesses and led to a transition from an old structure to a digital one including the Internet of Things, artificial intelligence, big data, and robotics development (Shestak et al., 2020).
According to the researchers, digital tools enhance business opportunities, help them to overcome the impact of the COVID-19 and accelerate companies’ digital transformation (Ting et al., 2020). The scholars (Richter et al., 2017; Seliverstov et al., 2020) underline that business digitalization is based on an innovative architecture characterized by nine aspects: digital content sharing; physical goods sharing; customers (prosumers) who produce and consume value; open mindset; changed living conditions; urban environment; win-win situation; added value; and the trustworthy business model.
Startseva et al. (2021) analyzed the business operations of digital companies in People’s Republic of China (China) and Russian Federation (Russia) and found significant differences between them. The Chinese market growth is ensured by domestic demand and introduction of the advanced approaches in business such as digital payment platforms, secure payment systems, and promotion strategies that reshape the initial economy structure, the traditional model of economic growth and business operations. Startseva et al. (2021) argued that digital transformation creates opportunities for business and facilitates integration within the distribution channels regardless of the level of market development. However, the questions and problems of the impact of the COVID-19 pandemic on an enterprise’s organizational culture transformation, including the implementation of innovative business models, have gained maximum popularity among modern researchers.
Impact of the COVID-19 on Organizational Culture Transformation Processes
In their research, Gregurec et al. (2021) explored business strategies that help to respond to the supply chain disruptions caused by the COVID-19 pandemic in Small and Medium-sized Enterprises (SMEs). The research identified three strategic approaches for enterprises to survive under the pandemic crisis. Enterprises should take into account the competitors’ performance, a value proposition that differs from others and renew the existing business processes. Moreover, the scholars recommend focusing on business architecture, making use of the transformation driven by the economy, reshaping business models and making them more adaptive to changing environments, and introducing digitalization in organizations.
Choi and Sethi (2021) examined the innovative business measures to combat the COVID-19 in China. The researchers concluded that the quarantine restrictions led to new innovative operations in business using simple technologies such as WhatsApp technology for service delivery (catering, clothing, leisure) and education (delivery of knowledge to a student outside the traditional classroom, improving organizational issues in education). Moreover, the WhatsApp technology is seen as an alternative solution to traditional call centers which is used for communication purposes, ordering of products or services, and delivery during a company’s working hours. One of the main benefits of WhatsApp is to keep physical stores open and avoid layoffs. WhatsApp provides consumers with an alternative way to get advice from real sellers, without visiting stores.
The scholars (Paters, 2020) suggested that the COVID-19 pandemic would foster online interactions in the near future. Under the quarantine restrictions, the new trends in business changed the practical (performance based on digital tools) and tactical strategies (customers relationships based on the internal aspects of corporate architecture and external influences) in the enterprise management aimed to foreshadow economic uncertainty and possible risks (Liguori & Pittz, 2020).
While paying tribute to the merits of predecessors in the development of innovative business models, it should be noted that at the time of the study a single definition of business model has not been adopted, so this study will use the definition of Baizhanova and Kunanbaeva. According to their definition, a business model is a set of ways of doing business in a company (its structure, products, ways of delivery and service of goods, increase in market value), as well as basic company strategy (Baizhanova & Kunanbaeva, 2019). A business ecosystem in this study should be understood as a network of interconnected business entities, vendors, suppliers, distributors, and other value chain participants who interact with each other in a complementary way within the key components of their business models (Venkatesh et al., 2019).
In addition, there are not enough studies of innovative business models in the midst of transforming organizational culture of enterprises under the COVID-19 pandemic. Thus, the relevance of this study is determined by the demand for effective innovative business models that ensure sustainable development of enterprises in the COVID-19 pandemic, and insufficient development of this issue by modern science.
The following hypotheses were formed. H1: The negative shock of the pandemic and the ensuing quarantine restrictions caused critical changes in the business environment, undoing the relevance of previously effective business models. H2: The new business environment has required enterprises to significantly transform organizational culture. H3: Under the conditions of shock changes in the business environment and forced transformation of organizational culture, the main condition for sustainable development of enterprises is innovative business models. H4: Implementation of innovative business models depends on both the business environment and the influence of national culture. The purpose of this article was to establish the relationships and mutual influences of organizational culture transformation, national culture, and innovative business models’ formation under the influence of the COVID-19 pandemic.
The research objectives include but not limited to the following:
study the modern scientific approach to understanding the category of business models,
develop a methodological study design, identify the main tools and sources of information,
develop a methodological approach to assessing the implementation of innovative business models in different markets,
test the study on the example of SMEs in Russia and China,
determine relationship between the transformation of organizational culture, national culture, and innovative business models’ formation under the COVID-19 pandemic.
The national statutory regulations for each country were used to define small and medium-sized enterprises, as follows:
In Russia, small and medium-sized enterprises are defined by the Federal Law (FL) of July 24, 2007 No. 209-FZ “On the development of small and medium-sized businesses in the Russian Federation” according to three basic criteria: (1.1) composition of founders: the share of the state, foreign legal entities and foreign citizens, public and religious organizations or associations, charitable and other funds in the authorized (share) capital (unit trust) of legal entities shall not exceed 25% (except for assets of joint-stock investment funds and closed unit trusts); (1.2) annual revenue: up to 120 million rubles - microenterprises; up to 800 million rubles - small businesses, up to 2 billion rubles - medium-sized businesses; (1.3) the number of employees: up to 15 - microenterprises, 15–100 - small businesses, 101–250 - medium-sized businesses (State Duma, 2007);
In China, small and medium-sized enterprises are defined by two main criteria: (2.1) up to 20 employees for micro enterprises, 20–300 for small enterprises and 300–1000 for medium-sized enterprises; (2.2) annual output: up to 3 million RMB for micro enterprises, up to 20 million RMB for small enterprises, up to 400 million RMB for medium-sized enterprises (Standing Committee of the National People’s Congress, 2017).
The research has the following structure. The second section describes the methodology including the analysis of the research design, materials, and database. Third section presents the main study results. Fourth section is about the discussion and comparison of the findings with previous studies. Fifth section provides research conclusions important for both academicians and practitioners. The section creates recommendations for business management and identifies what areas may be missing in the research.
Materials and Methods
The solution of these scientific problems was made possible during the implementation of a multi-stage comprehensive exploratory research endeavor involving two countries: Russia and China. The main study stages are presented in Fig. 1.
Fig. 1 The main study stages. Source: own development
Conducting a comprehensive literature review allowed not only clarifying the definitions used, but also enriching the methodology and tools with the experience of predecessors. The methodological study design was developed taking into account its main limitations. One of such limitations, which had a significant impact on the choice of methodology and the use of research tools, was the lack of reliable statistical information characterizing the dynamics of innovative business models. Given this limitation, it would be most appropriate to conduct a ‘field’ study phase to gather primary information. When selecting the instruments for the ‘field’ study stage, the limitations of the research budget, as well as the time constraints for conducting the study, significantly reduced the possibility of conducting multi-country quantitative research with a reliable sample. Given the study limitations, a partially structured expert interview method was chosen to gather information. At the same time, due to the study limitations caused by the continuation of the COVID-19 pandemic, expert interviews were conducted partially in a face-to-face format and partially using electronic means of communication (Viber, Skype, Telegram, etc.).
A roadmap for expert interviews was developed to conduct the study. When conducting expert interviews, there are two main ways of obtaining information: structured and unstructured expert interviews, with the main difference being in the structure of obtaining information. In an unstructured interview, the interview process takes the form of an almost free conversation. Individual topics and questions of an unstructured interview can be discussed in an arbitrary order, changing from expert to expert. An unstructured interview provides more in-depth information, especially nonverbal information, but requires not only more time compared to a structured interview, but also a much higher qualification of the interviewer and analysts. In structured and partially structured interviews, the main study stages are unified for all interviews, usually in the form of a roadmap, where each step means a certain interview topic, and the initiative to change the topic of the interview in this case belongs to the interviewer. Structured and partially structured interviews, as a rule, are less effective in terms of nonverbal information, but require less skill, and greatly facilitate the analysis and comparison of interviews, which is very important when using the developed methodology in the real economy. Thus, the roadmap step “Defining the Expert Profile” means, for example, the initial stage of the interview, the introduction of the interviewer and the expert, during which the basic parameters of the expert profile are established in the form of a free conversation. The step “Determining Awareness of Innovative Business Models” predetermines a conversation to determine the expert’s awareness of innovative models, their opinion on the advantages and disadvantages of this or that model, etc. Schematically the expert interview roadmap is presented in Fig. 2.
Fig. 2 Roadmap of expert interviews. Source: own development
The roadmap involves limited use of statistical analysis tools during interviews, including closed-ended, single-choice questions (in determining the respondent profile), as well as a 5-point Likert scale to obtain measurable and comparable information on certain questions of this study. The use of a 5-point Likert scale is due to the high level of understanding of the scale by a respondent, the ease of processing, as well as the unambiguity of interpretation of the information obtained (Vogel et al., 2020).
The Likert scale scores for expert interviews are interpreted as follows: completely disagree − 1 point; rather disagree − 2 points; unsure − 3 points; rather agree − 4 points; completely agree − 5 points. Reference question RQ22 is presented as a multiple-choice question. The interview roadmap was developed based on the experience of predecessors (Baizhanova & Kunanbaeva, 2019; Botelho et al., 2021; Venkatesh et al., 2019) in the issues of assessing the perception, implementation, and prospects for the development of innovative models, including to test the conclusions of predecessors regarding the popularity and prospects of certain models (questions RQ 10–13). The basic constructs of the expert interview roadmap are presented in Table 1.
Table 1 The main constructs of the expert interview roadmap
Question code Reference sheet question
RQ1 The COVID-19 pandemic has completely changed the business environment
RQ2 The methods we used to build our value chain suddenly became ineffective
RQ3 Our company had to switch to a remote working format to avoid closure and bankruptcy
RQ4 In order to ensure the efficiency of our company’s operations under the COVID-19 pandemic, we had to significantly change our organizational culture.
RQ5 The switch to a remote working format was purely a forced measure for the period of quarantine restrictions. We have no plans to maintain this format once quarantine restrictions are completely lifted.
RQ6 To ensure the sustainable development of the enterprise in the context of the pandemic, we consider the possibilities/realize the benefits of integration
RQ7 To ensure the sustainable development of the enterprise under the conditions of the pandemic, we consider/realize the advantages of cooperation
RQ8 To ensure the sustainable development of the enterprise in the conditions of the pandemic, we consider/realize the advantages of licensing
RQ9 To ensure the sustainable development of the enterprise in a pandemic, we consider/realize the benefits of using mixed innovative business models
RQ10 Our company is using/preparing to use the advantages of prosuming
RQ11 Our company is using/preparing to take advantage of crowdfunding
RQ12 Our company is using/preparing to take advantage of blockchain
RQ13 Our company is using/preparing to use the advantages of trading platforms 4.0.
RQ14 We believe that the current situation is a temporary market reaction to an external irritant (COVID-19 pandemic), and we are not planning any serious changes in our business organization
RQ15 The implementation of innovative business models at our company is possible only in case of state support
RQ16 We are confident that the market will never be the same, and we plan to continue working in a new format after the quarantine restrictions are lifted
RQ17 Our company is experiencing technical problems in implementing innovative business models.
RQ18 Our company has organizational problems implementing innovative business models.
RQ19 Our company has financial problems in implementing innovative business models.
RQ20 Our enterprise has personnel problems in implementing innovative business models.
Source: own development
Database
Russia and China were chosen for the pilot test. The criterion of scale was the main criterion for selecting countries for testing because the economies of China and Russia play an important role in the global economy, in terms of resources, production, and consumer market. The population of managers and/or owners of SMEs was used as the general population in the selection of experts, and SMEs were defined according to the regulatory and legal definition. Modern economic science recommends the analysis of 20 to 30 interviews when conducting expert interviews (Marshall et al., 2013). Taking into account the two-country approach, the maximum recommended sample size was chosen − 30 interviews, distributed equally between China and Russia. Enterprises were selected at random from official registries of legal entities: Unified State Register of Legal Entities - for Russia (Russian Unified State Register of Legal Entities, 2022), State System of Open Information on Enterprise Creditworthiness - for China (National Enterprise Credit Information Publicity System, 2022). All heads of enterprises were informed about participation in the study, its goals and objectives, and confirmed their consent to participate in the research. Since the interviews were completely anonymous, written confirmation of consent to process personal data was not required. The industry structure of the enterprises whose managers participated in the expert interviews is shown in Fig. 3.
Fig. 3 Sectoral structure of enterprises, %. Source: own development
An examination of the acceptability and normality of the sample is presented in Table 2.
Table 2 Main indicators of normality and acceptability assessment
Question Mean Standard deviation Item-total correlation
RQ1 4.67 0.55 0.25
RQ2 4.20 0.89 0.29
RQ3 3.37 0.96 0.22
RQ4 3.13 0.78 0.34
RQ5 3.37 0.96 0.22
RQ6 2.83 0.91 0.28
RQ7 3.17 0.99 0.42
RQ8 3.00 0.98 0.45
RQ9 3.00 0.91 0.31
RQ10 2.77 0.82 0.25
RQ11 2.67 0.92 0.50
RQ12 3.17 0.91 0.49
RQ13 3.73 0.98 0.34
RQ14 2.93 0.98 0.25
RQ15 3.43 0.86 0.24
RQ16 2.93 0.94 0.35
RQ17 3.20 0.71 0.26
RQ18 3.03 0.85 0.28
RQ19 3.20 0.89 0.20
RQ20 3.20 0.85 0.31
Source: own development
Thus, the mean value of each item is in the range from 2.67 to 4.67, the standard deviation does not exceed 0.99, which indicates the acceptability and normality of all items for analysis. The item-total correlation values in the range from 0 to 0.19 give grounds to characterize the correlation as low and the obtained data as undesirable for further use, but the study results of all values are greater than 0.19, which indicates the normality of the sample. The item-total correlation values in the range between 0.2 and 0.39 characterize the correlation as sufficient and testify to the advisability of using the obtained data for further analysis. In the range from 0.2 to 0.39 are the values of item-total correlation for RQ1, RQ2, RQ3, RQ4, RQ5, RQ6, RQ9, RQ10, RQ13, RQ14, RQ15, RQ16, RQ17, RQ18, RQ19, RQ20. Therefore, the data obtained can be used for further analysis. The item-total correlation values in the range above 0.4 indicate a very good correlation and minimal risk of information distortion. The item-total correlation values for RQ7, RQ8, RQ11, RQ12 are above 0.4, indicating high reliability of the sample.
Methods
Normality and reliability of the sample in the analysis of quantitative indicators (Likert scale results) were assessed using the mean and standard deviation, as well as item-total correlation.
The mean value of µ was calculated by formula 1:
1 μ=1n∑i=1nxi,
х value of the analyzed parameter;
n number of values in the time series.
Standard deviation (root mean square) σ is calculated by formula 2:
2 σ=1n∑i=1n(xi-μ)2,
х value of the analyzed parameter;
n number of values in the time series;
µ mean value.
Coefficient of variation cv is calculated by formula 3:
3 cv=σμ,
σ standard deviation;
µ mean value.
Pearson correlation coefficient to determine the normality and acceptability of item-total correlation questions was calculated by formula 4 using a Microsoft Excel spreadsheet processor.4 rxy=∑i=1mxi-x-yi-y-∑i=1mxi-x-2∑i=1myi-y-2=cov(x,y)sx2sy2
In addition, at various study stages, the method of encryption, the method of graphical representation of data, as well as a set of methods for the analysis of verbal and nonverbal information directly during the interview were used.
Results
The results of the conducted interviews show that the experts are almost unanimous in stating that the COVID-19 pandemic has completely changed the business environment. 70% of the experts said they strongly agreed, 27% agreed rather than disagreed, 3% of the experts were uncertain; none of the experts expressed any form of disagreement. During the interviews, there was a fairly high level of emotional reaction to this question, with a significant portion of experts noting that the greatest damage was caused by the suddenness of the changes (“like thunder from a clear sky”, “like snow in July”). The reaction of experts to the claim of a sudden loss of effectiveness of previously used value chain creation methods is presented in Fig. 4.
Fig. 4 Experts’ reactions to the statement about the sudden loss of effectiveness of previously used value chain creation methods. Source: own development
The vast majority of experts agree that the COVID-19 pandemic has caused a devastating blow to the value chain techniques previously used, with 43% of experts responding in strong agreement, 40% agreeing rather than disagreeing with this statement, and 10% having difficulty answering. Only 7% of experts rather disagree with this statement, with no expert disagreeing strongly. Structurally, experts’ reactions to the statement about the forced organizational culture changes are presented in Fig. 5.
Fig. 5 Experts’ reactions to the statement about the forced organizational culture changes. Source: own development
A significant proportion of experts (43%) acknowledged the forced nature of changes in the organizational structure, with only 13% expressing strong agreement. Not a single expert expressed strong disagreement, but 37% of experts found it difficult to answer. In particular, during interviews, the experts referred to difficulties in determining the significance of changes in the organizational culture, which indicates that additional research on this issue is advisable.
At the same time, none of the experts expressed a confident refusal to return to the previously used methods of work after the abolition of quarantine restrictions, although almost 1/3 of the experts (30%) rather agree than disagree with this statement. While 13% of the experts are definitely ready to return to the old methods and business models; 20% of the experts are more ready than not. Bivariate distribution shows significant differences in the opinions of Chinese and Russian experts: for example, only Russian experts expressed confident agreement with the return to the old methods, while no Chinese experts confirmed their intention to return to traditional business models, and five Chinese experts and only one Russian expert disagreed with this statement. Especially significant are the differences in the study of the main directions of developing innovative business models (Fig. 6).
Fig. 6 The main directions of developing innovative business models. Source: own development
While the structure of responses seems to be quite harmonious at first glance, the bivariate distribution by country shows significant differences in the opinions of experts from China and Russia regarding the prospects of innovative business models. The bivariate distribution (by country) of mixed business models’ assessments is shown in Fig. 7.
Fig. 7 Bivariate distribution (by country) of experts’ assessment of mixed business models. Source: own development
When assessing the prospects of using mixed business models, 9 experts from China and only one from Russia positively assessed the prospects for using these models, while 11 Russian experts and only one Chinese expert - negatively. At the same time, the number of experts who could offer no opinion is the same for China and Russia - three experts each. The results of experts’ evaluation of the main methods for implementing the advantages of innovative business models are presented in Fig. 8.
Fig. 8 Expert assessment of the main methods of realizing innovative business models’ benefits. Source: own development
At the same time, the bivariate distribution (by country) also confirms significant differences in the opinions of Chinese and Russian experts. Chinese experts constructively and positively perceive the most common methods of implementing innovative business models’ benefits. 10% of experts are sure that introduction of innovative business models at the enterprise is possible only in case of state support, more than 1/3 (37%) of experts rather agree than disagree with the necessity of state support of innovative business models introduction, 40% are undecided with the answer, and only 13% rather disagree than agree. When interviewed, the experts, who expressed uncertain disagreement, reported that “the development of an enterprise is the concern of its manager, not the state. And demanding another ‘help’ (‘help’ was expressed sarcastically) to increase the efficiency of one’s company is pure consumerism.”, “I will not refuse state assistance, especially in the development of technological infrastructure, training and consultations, but it is by no means a prerequisite.” The main problems of SMEs in the study of innovative business models are presented in Fig. 9.
Fig. 9 The main problems of SMEs in the implementation of innovative business models. Source: own development
Thus, the formulated hypotheses were proved. Expert opinions confirmed the assumption that the negative shock of the pandemic and the ensuing quarantine restrictions caused critical changes in the business environment, cancelling the relevance of previously effective business models (H1), while the new business environment required significant transformation of the organizational culture of enterprises (H2). In addition, the study results confirmed the assumption that the main condition for sustainable development of enterprises is innovative business models under the conditions of shock changes in the business environment and forced transformation of organizational culture (Н3). The results of bivariate distribution established significant conceptual differences in the opinions of Russian and Chinese experts, which gives the right to assert significant differences in the perception and implementation of innovative business models at the country level (H4). At the same time, the nature of their differences, the main factors that form these differences are beyond the scope of this study and may become promising directions for further research on innovative business models. The main study limitations were (1) the lack of reliable statistical information describing the dynamics of innovative business models; (2) research budget limitations; (3) research time constraints; (4) quarantine restrictions related to the COVID-19 pandemic.
Discussion
The analysis showed that in the COVID-19 pandemic, the traditional model became ineffective, and unemployment increased. Only businesses that implemented digital tools, architectural thinking and information technology were able to cope with the challenges (Tripathi, 2021). A study by Sethi (2021) confirmed the results of the current study. The researcher argued that key cultural traits should include internal management system policies, digitalization, team spirit, and employee health and safety.
Innovative solutions, introduced on the organizational level, helped enterprises to process, store and report financial information, as well as react to emergencies (Una et al., 2020). Burgos and Ivanov (2021) showed that the purchasing power and government interventions under the COVID-19 pandemic have been adapted to a new environment including changes in online sales channels, digital twins in supply chains, end-to-end transparency and transformation of the organizational culture. Richter et al. (2017) underlined that business based on digital technologies and a culture of innovation can be seen as a future-oriented one. Similar to the current research, Sneader and Singhal (2021) discussed the main trends that define socio-economic life and how this model affects the global economy, business and society. The researchers found that leisure travel recovers from the recession. Moreover, the research admits that there is an increasing monetization of ideas, an increase in entrepreneur activity, acceleration of Industry 4.0 through production automation and innovative management.
From the societal point of view, enterprises invest much more in public health. Moreover, companies experience financial problems after the pandemic. Bettiol et al. (2021) concluded that quarantine restrictions have influenced business digitalization and encouraged all market stakeholders to use digital tools to promote hygiene and physical safety in everyday life.
The researchers identified two main aspects that companies should focus on to cope with the crisis: (1) introduce a customer-centric strategy; (2) build all business processes based on the sharing economy. The research conducted by Bettiol et al. (2021) came to the same conclusions on agile innovation management for the sharing economy and a value proposition to customers. The research analyzes the innovative activities and business imperatives arising from the COVID-19 pandemic and their impact. Despite the positive trends in China and Russia (Garcia-Herrero & Xu, 2019) and the adoption of the centralized management model, there are significant differences in the business innovation activity in supply and demand. At the same time, Hoang et al. (2021) found that e-commerce helps to increase sales and ensure stable market operations under social distancing and other quarantine restrictions.
The scholars suggest that political leaders should encourage companies to introduce technology and foster an innovation culture to achieve flexibility. Shumakov (2021) underlined that enterprises with corporate culture align to the business strategy and overcome the challenges of the pandemic at a minimal cost. The research recommends transforming the management culture in the following ways: monitor the current state, introduce changes, and demonstrate new behavior patterns that support the innovative business models (Gribanov, 2019).
Conclusion
The study developed a methodological approach to assess the implementation of innovative business models in the context of transforming organizational structures in the COVID-19 pandemic. Testing the proposed approach on the example of SMEs in Russia and China allowed fully confirming the formulated hypotheses. The study was conducted using qualitative methods of analysis, which is due to the imperfection of the statistical information system, which limits the ability to analyze statistical data. The expert interviews fully confirmed hypothesis H1 that the negative shock of the pandemic and subsequent quarantine restrictions caused critical changes in the business environment and abolished the relevance of previously used business models. The study results also confirmed that the new business environment requires a significant transformation of enterprises’ organizational culture (hypothesis H2). At the same time, this issue suggests the possibility of further in-depth research, as part of the experts perceive this transformation as temporary, related exclusively to the impact of the pandemic, which requires additional clarification. The study results also proved hypothesis 3 (H3), according to which innovative business models become the main condition for the sustainable development of enterprises under the conditions of shock changes in the business environment and forced transformation of organizational culture. In addition, the bivariate distribution of the interview results revealed significant differences in the opinions of Russian and Chinese experts, which confirms hypothesis 4 (H4) about the relationship between the introduction of innovative business models, national business environment, and the influence of national culture. Assessing the impact of individual factors on the implementation of innovative business models at the national level is beyond the scope of this study but is a very promising area for future research.
The study assessed the main barriers to implementing innovative business models by SMEs to achieve sustainable development during the COVID-19 pandemic. This assessment can be useful in the development of SME assistance programs by governmental and non-governmental agencies at national and regional levels. The proposed assessment approach is easily scalable and adaptable to the specific conditions of individual regions or industries due to the chosen research format and does not require additional expensive software. In addition, the use of benchmark questions with quantitative assessment when conducting interviews makes it possible to use this approach for monitoring, which will make it possible to assess the effectiveness of programs to promote SMEs. The study may also be of interest to academic researchers as it suggests empirically determined relationships (hypotheses H1, H2, H3, H4) between organizational culture transformation, national culture, and the formation of innovative business models to ensure sustainable development of enterprises under the COVID-19 pandemic.
Author Contributions
Conceptualization, S.V.B. and L.T.; methodology, V.K.; software, A.K.; validation, S.V.B., V.K., A.K., and L.T.; formal analysis, A.K.; investigation, S.V.B.; resources, V.K. and L.T.; data curation, A.K.; writing—original draft preparation, S.V.B. and L.T.; writing—review and editing, V.K.; visualization, L.T.; supervision, A.K.; project administration, S.V.B.; funding acquisition, V.K. All authors have read and agreed to the published version of the manuscript.
Data Availability
Data will be available on request.
Declarations
Informed Consent
The research did not include any human and / or animal participants, so the informed consent was not required.
Ethical Approval
The authors declare that the work is written with due consideration of ethical standards. The study was conducted in accordance with the ethical principles approved by the Ethics Committee of Mingachevir State University.
Conflict of Interest
The authors declare no conflict of interest.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Abreu A Innovation ecosystems: A sustainability perspective Sustainability 2021 13 1675 10.3390/su13041675
Akhmetzhanov B Tazhibekova K Shametova A Urazbekov A Coal supply chain management and economic efficiency of using high-ash coking coal in ferroalloy manufacturing International Journal of Supply Chain Management 2019 8 2 624 632
Almeida F Santos JD Monteiro JA The challenges and opportunities in the digitalization of companies in a post-COVID-19 World IEEE Engineering Management Review 2020 48 3 97 103 10.1109/EMR.2020.3013206
Baizhanova L Kunanbaeva D Development of companies and innovative business models Qazaqstan Respwblïkası 2019 3 88 91
Bettiol, M., Capestro, M., Di Maria, E., & Micelli, S. (2021). Reacting to the COVID-19 pandemic through digital connectivity with customers: the Italian experience. Italian Journal of Marketing, 2021, 305–330. 10.1007/s43039-021-00031-y
Bossert, O., & Van der Wildt, N. (2021). How enterprise architects need to evolve to survive in a digital world. Retrieved May 12, 2022, from https://www.mckinsey.com/business-functions/mckinsey-digital/our-insights/how-enterprise-architects-need-to-evolve-to-survive-in-a-digital-world
Botelho DF Dias B de Oliveira L Soares T Rezende I Sousa T Innovative business models as drivers for prosumers integration - Enablers and barriers Renewable and Sustainable Energy Reviews 2021 144 111057 10.1016/j.rser.2021.111057
Burgos D Ivanov D Food retail supply chain resilience and the COVID-19 pandemic: A digital twin-based impact analysis and improvement directions Transportation Research Part E: Logistics and Transportation Review 2021 152 102412 10.1016/j.tre.2021.102412 34934397
Choi TM Sethi SP Innovative service operations for survivals of SMEs under COVID-19: Two cases in Hong Kong IEEE Engineering Management Review 2021 49 1 50 54 10.1109/EMR.2021.3050406
Digital Reality (2020). Digital transformation. A digital reality initiative white paper. Retrieved May 12, 2022, from https://digitalreality.ieee.org/images/files/pdf/DRI_White_Paper_-_Digital_Transformation_-_Final_11Nov.pdf
Dobrotina О Brand-oriented behaviour in the context of forming corporate culture: Introduction to the problem Science and Education Journal for Students and Educators 2021 4 6 721 727
Eaton, D., Hammerich, K., Kilby, G., Hazucha, J., & Thibodeau, J. (2021). The power of culture transformation. Retrieved May 12, 2022, from https://www.kornferry.com/content/dam/kornferry/docs/article-migration/The_Power-of-Culture-Transformation.pdf
Garcia-Herrero A Xu J How does China fare on the Russian market? Implications for the European Union Russian Journal of Economics 2019 8 1 13 10.2139/ssrn.3508490
Gregurec I Tomičić Furjan M Tomičić-Pupek K The impact of COVID-19 on sustainable business models in SMEs Sustainability 2021 13 3 1098 10.3390/su13031098
Gribanov, Y. (2019). Digital transformation of socio-economic systems based on the development of the service integration institute. Dissertation. Saint Petersburg State University of Economics.
Hoang TDL Nguyen HK Nguyen HT Towards an economic recovery after the COVID-19 pandemic: empirical study on electronic commerce adoption of small and medium enterprises in Vietnam Management & Marketing 2021 16 1 47 68 10.2478/mmcks-2021-0004
Kreutzer RT Treiber und Hintergründe der digitalen Transformation, in Digitale transformation von geschäftsmodellen Springer Gabler 2017 10.1007/978-3-658-12388-8_2
Lankhorst, М. (2016). Enterprise architecture and innovation: A cultural change. Retrieved May 12, 2022, from https://bizzdesign.com/blog/enterprise-architecture-and-innovation-a-cultural-change/
Liguori EW Pittz TG Strategies for small business: Surviving and thriving in the era of COVID-19 Journal of the International Council for Small Business 2020 1 2 106 110 10.1080/26437015.2020.1779538
Marshall B Cardon P Poddar A Fontenot R Does sample size matter in qualitative research? A review of qualitative interviews in IS research Journal of Computer Information Systems 2013 54 1 11 22 10.1080/08874417.2013.11645667
National Enterprise Credit Information Publicity System (2022). State system of open information about the creditworthiness of enterprises. Retrieved May 12, 2022, from https://www.gsxt.gov.cn/index.html
Paters, М. (2020). Global customer insights survey. Retrieved May 12, 2022, from https://www.pwc.ru/ru/retail-consumer/publications/assets/pwc-global-customer-insights-survey-2020-russia-ru.pdf
Piccarozzi M Silvestri C Morganti P COVID-19 in management studies: A systematic literature review Sustainability 2021 13 7 3791 10.3390/su13073791
Pwc (2021). Transformation of corporate culture. Retrieved May 12, 2022, from https://www.pwc.ru/ru/publications/transformatsiya-korporativnoy-kultury.html
Richter C Kraus S Brem A Durst S Giselbrecht C Digital entrepreneurship: Innovative business models for the sharing economy Creativity and Innovation Management 2017 26 3 300 310 10.1111/caim.12227
Russian Unified State Register of Legal Entities (2022). Official web site. Retrieved May 12, 2022, from https://egrul.nalog.ru/
Seetharaman P Business models shifts: Impact of Covid-19 International Journal of Information Management 2020 54 102173 10.1016/j.ijinfomgt.2020.102173 32834338
Seliverstov Y Rudychev A Dmitrieva Y Digital transformation of small and medium enterprises as a factor of competitive advantage Bulletin of the Altai Academy of Economics and Law 2020 11 3 531 539 10.17513/vaael.1459
Sethi, В. (2021). Organisational culture it`s time to take action. Global culture survey. Retrieved May 12, 2022, from https://www.pwc.com/gx/en/issues/upskilling/global-culture-survey-2021.html
Shestak, V., Gura, D., Khudyakova, N., Shaikh, Z. A., & Bokov, Y. (2020). Chatbot design issues: building intelligence with the Cartesian paradigm. Evolutionary Intelligence, in press. 10.1007/s12065-020-00358-z
Shumakov, V. (2021). Five steps to bridging the cultural divide. Retrieved May 12, 2022, from https://www.pwc.ru/ru/services/people-and-organization/culture-change.html
Singaraju, Р. (2021). Organizational culture is an overlooked factor in transformation. Retrieved May 12, 2022, from https://www.forbes.com/sites/forbestechcouncil/2021/08/31/organizational-culture-is-an-overlooked-factor-in-transformation/
Sneader, K., & Singhal, S. (2021). The next normal arrives Trends that will define 2021 – and beyond. Retrieved May 12, 2022, from https://www.mckinsey.com/featured-insights/leadership/the-next-normal-arrives-trends-that-will-define-2021-and-beyond
Standing Committee of the National People’s Congress (2017). Law of the People’s Republic of China on the promotion of small and medium-sized enterprises of September 1, 2017. Retrieved May 12, 2022, from https://www.6laws.net/6law/law-gb/%E4%B8%AD%E8%8F%AF%E4%BA%BA%E6%B0%91%E5%85%B1%E5%92%8C%E5%9C%8B%E4%B8%AD%E5%B0%8F%E4%BC%81%E6%A5%AD%E4%BF%83%E9%80%B2%E6%B3%95.htm
Startseva E Zasng HX Rathnayake DN E-Commerce developments and strategies for value creation: the case of Russia and China Journal of Contemporary Issues in Business and Government 2021 27 3 1231 1242 10.47750/cibg.2021.27.03.165
State Duma (2007). Federal Law (FL) of July 24, 2007, No. 209-FZ “On the development of small and medium-sized businesses in the Russian Federation”. Retrieved May 12, 2022, from http://www.consultant.ru/document/cons_doc_LAW_52144/
Ting D Wei S Carin L Dzau V Wong TY Digital technology and COVID-19 Nature Medicine 2020 26 4 459 461 10.1038/s41591-020-0824-5
Tripathi, S. (2021). A year since Covid – what can businesses continue to do? Institute for Human Rights and Business. Retrieved May 12, 2022, from https://www.ihrb.org/focus-areas/covid-19/commentary-a-year-since-covid-19
Una, G., van Eden, H., Singh, A., Bardella, F., & Verma, A. (2020). Enhancing digital solutions to implement emergency responses Retrieved May 12, 2022, from https://www.imf.org/~/media/Files/Publications/covid19-special-notes/en-special-series-on-covid-19-enhancing-digital-solutions-to-implement-emergency-responses.ashx?la=en
Venkatesh R Mathew L Singhal TK Imperatives of business models and digital transformation for digital services providers International Journal of Business Data Communications and Networking 2019 15 1 105 124 10.4018/ijbdcn.2019010107
Vogel, C., Koutsombogera, M., & Costello, R. (2020). Analyzing Likert scale inter-annotator disagreement. In Neural approaches to dynamics of signal exchanges (pp. 383–393). Springer. 10.1007/978-981-13-8950-4_34
Wit, B. (2016). Ekologistyka w Systemie Zarządzania Odpadami Niebezpiecznymi. Towarzystwo Naukowe Organizacji i Kierowania.
| 0 | PMC9707119 | NO-CC CODE | 2022-12-01 23:20:20 | no | Public Organiz Rev. 2022 Nov 28;:1-23 | utf-8 | null | null | null | oa_other |
==== Front
Stud Comp Int Dev
Stud Comp Int Dev
Studies in Comparative International Development
0039-3606
1936-6167
Springer US New York
9368
10.1007/s12116-022-09368-7
Article
Puzzling Partnerships: Overseas Infrastructure Development by Chinese State-Owned Enterprises and Humanitarian Organizations
http://orcid.org/0000-0003-3692-9244
Leutert Wendy [email protected]
1
Plantan Elizabeth 2
Strange Austin 3
1 grid.411377.7 0000 0001 0790 959X Indiana University, Bloomington, IN 47405 USA
2 grid.264307.4 0000 0000 9688 1551 Stetson University, DeLand, FL 32723 USA
3 grid.194645.b 0000000121742757 University of Hong Kong, Hong Kong, Hong Kong SAR
28 11 2022
129
26 7 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Chinese state-owned enterprises (SOEs) are now working with humanitarian organizations to develop infrastructure abroad. This emerging phenomenon is puzzling: when, where, and why do Chinese SOEs, best known for constructing massive overseas infrastructure projects for commercial and political gain, execute smaller, lower-profile humanitarian projects? Similarly, why would humanitarian organizations––often with minimal experience in infrastructure contracting––select partners criticized for lack of emphasis on the international standards and best practices that they seek to promote? We address these questions through qualitative case studies of Chinese SOE-humanitarian organization collaboration in the Democratic Republic of Congo, Nigeria, and the Republic of Congo. These cases suggest that such partnerships are more likely when a humanitarian organization has previous experience working in China or with Chinese actors, when Chinese SOEs aim to enter new markets, or when these firms operate in dangerous or politically unstable environments. This study contributes to scholarship on China’s evolving role in international development by providing the first empirical analysis of Chinese SOE-humanitarian organization partnerships.
Keywords
State-owned enterprises
Humanitarian organizations
International development
China
Africa
==== Body
pmcIntroduction
The 40,000-MW Grand Inga Dam along the Congo River in the western Democratic Republic of Congo (DRC) will be one the world’s largest dam systems if completed (Bax and Kavanaugh 2020). Construction of the dam, power stations, and associated transmission lines across six phases would cost approximately $80 billion and generate power equivalent to 40% of Africa’s electricity demand (Currie and Cropley 2021). A long-planned––and long-delayed––third phase, known as Inga III, would cost $14 billion.1 The project remains a priority for the DRC government, even though numerous civil society organizations have raised concerns about corruption and the proposed dam’s environmental and social effects (International Rivers 2021). DRC leaders are in ongoing negotiations to finalize a consortium potentially including multiple Chinese state-owned enterprises (SOEs): China Three Gorges Corporation, Sinohydro, and others (Holland and Burton 2021).2
Chinese SOEs are often associated with “megaprojects”: multi-billion dollar projects like the Grand Inga Dam (Flyvbjerg et al. 2003, 1).3 Their commercial involvement in the global infrastructure industry has grown steadily since the 1980s. Two major policy initiatives—the “going out” strategy and Belt and Road Initiative (BRI), formally adopted in 2002 and 2013, respectively—have further accelerated their global expansion (Ye 2020; Zhang 2020).4 Completed megaprojects––such as China Road and Bridge Corporation (CRBC)’s Standard Gauge Rail in Kenya and China Harbour’s Hambantota Port in Sri Lanka––elevate Chinese SOEs’ international profile. Megaprojects also position SOEs as important players in China’s international relations (Gill and Reilly 2007; Lee 2017).
Yet on the eastern side of the DRC, some 1700 km away from Grand Inga, the Chinese SOEs Sinohydro and China Geo-engineering Corporation (CGC) are building a much smaller and less flashy water project. In Goma and Bukavu, the capital cities of the country’s North and South Kivu provinces, the two firms have been constructing water taps and treatment facilities for the IMAGINE Program, a multi-year initiative to improve water access, sanitation, and hygiene and thereby reduce child mortality (Mercy Corps 2019). IMAGINE is a flagship project for the international non-governmental organization (INGO) Mercy Corps and the project’s funder, the UK’s Foreign, Commonwealth, and Development Office (FCDO).5 Although the project is critical for Mercy Corps, with a contract valued at £5.5 million (approximately $7.6 million), IMAGINE is merely a fraction of Sinohydro’s bottom line in DRC and Africa overall.6
Compared to megaprojects like Inga III, Chinese SOEs’ motivation for involvement in projects like IMAGINE is less clear. As a relatively small-scale, localized infrastructure project, IMAGINE is unlikely to be highly lucrative or serve as a springboard to national or regional commercial expansion. The operating environment is technically challenging and potentially dangerous, due to glass-like volcanic soil, recurrent volcanic eruptions, the continued Ebola outbreak, and ongoing deadly political unrest.7 IMAGINE is not branded as a Chinese-led development project, nor is it a megaproject with high levels of national or international media coverage. Neither profits nor China’s political interests appear to strongly drive the two Chinese contractors’ participation.
Mercy Corps’ decision to partner with Chinese SOEs appears equally puzzling. Principles like humanity, neutrality, impartiality, and independence motivate humanitarian organizations like Mercy Corps. Such organizations and their funders strive to uphold stringent ethical, labor, and environmental standards in their overseas projects. These standards are internally set8 and often subject to external scrutiny.9 Chinese SOEs have a reputation among policymakers and analysts for their ability to build quickly, not only reducing bureaucratic red tape but also potentially neglecting social, environmental, and other safeguards (Mwase and Yang 2012). In the DRC, for example, the Sicomines copper and cobalt mining agreement involving Chinese SOEs has suffered delays, inadequate social and environmental impact assessments, and reports of poor infrastructure quality (Maiza-Larrarte and Claudio-Quiroga 2019). Yet despite apparent incentive misalignment, Chinese SOEs and humanitarian organizations are now collaborating to develop infrastructure abroad.
When, where, and why are Chinese SOEs partnering with humanitarian organizations on overseas infrastructure development? We address these questions using three qualitative case studies: (1) Sinohydro and China Geo-engineering Corporation’s work with Mercy Corps on the IMAGINE Program in the DRC; (2) China Geo-engineering Corporation and the ICRC’s partnership on the Alhamduri Urban Water Works in Nigeria; and (3) Weihai International Economic & Technical Cooperative (WIETC)’s collaboration with the World Food Programme on a cassava plant in the Republic of Congo. Our research draws on 36 interviews conducted between December 2019 and August 2021 with respondents in the USA, Canada, China, the DRC, the Republic of Congo, Nigeria, Kenya, the UK, France, and Switzerland.
Analysis of these cases suggests that several conditions make Chinese SOE-humanitarian organization partnerships more likely. Humanitarian organizations appear more likely to collaborate with Chinese SOEs if they have previous experience working in China and/or with Chinese actors. Chinese SOEs may also pursue projects with humanitarian organizations to enter new national markets or sectors within existing ones. Dangerous or politically unstable environments can also motivate Chinese SOEs to look for humanitarian organization partners. As Chinese SOEs confront scandals and pushback abroad in countries from Myanmar to Peru to Zambia, humanitarian organizations can help manage risk and navigate operational challenges by providing strong local knowledge, networks, and experience.
The emerging phenomenon of Chinese SOE-humanitarian organization partnerships is modest in scale but worthy of scholarly and policy attention. Overseas infrastructure projects are highly consequential for the millions of people whose lives they affect daily. They also offer important insight into how Chinese SOEs and humanitarian organizations are changing their strategies over time. Chinese SOEs are expanding beyond mega-projects to work with a wider array of international and domestic actors on smaller-scale projects abroad. Some humanitarian organizations are now also pursuing infrastructure development as a more direct approach to humanitarianism. Understanding how Chinese SOEs and humanitarian organizations interact enables more nuanced assessment of the potential externalities of China’s overseas investment impact and of values-driven humanitarian organizations’ ability to moderate it.
This paper proceeds as follows. First, we introduce Chinese SOEs and humanitarian organizations and outline their interests and incentives for collaboration on overseas infrastructure development. Then, we introduce our argument and describe the study’s case selection, data and methods. Next, we present our three case studies and discuss our findings, including the potential advantages and challenges of such partnerships. We conclude by considering implications for international development and directions for future research.
Puzzling Partnerships: Chinese SOE and Humanitarian Organization Incentives and Interests
Chinese SOEs
State-owned enterprises (SOEs) are companies owned in whole or part by central or sub-national governments. In China, the State-owned Assets Supervision and Administration Commission (SASAC), a special commission of the State Council, administers a portfolio of 97 SOEs owned by the central government (“central SOEs” hereafter); it also manages state-owned assets below the central level through local SASACs.10 In addition to the largest and best-known central SOEs, such as China Communications and Construction Corporation, Sinohydro, and China Geo-Engineering Corporation, provincial SOEs like Weihai International Economic & Technical Cooperative (WIETC) and Shenzhen Energy have also become major global industry players.
Today’s Chinese SOEs have operated in Africa and other developing regions for decades, albeit not initially as commercial entities. After China’s government under Mao Zedong began providing foreign aid in the early 1950s, its revolutionary approach to financing aid and diplomacy in newly independent countries across Africa and Asia peaked during the 1960s and early 1970s (Strange 2019). China’s government had nearly 100 turn-key projects under construction worldwide in 1972, and it had committed aid to 29 African countries by 1973 (Brautigam 2011, 41). Most of this aid was highly concessional in nature and focused primarily on political rather than economic considerations. In the 1980s, Deng Xiaoping reformed the foreign aid bureaucracy to focus China’s overseas aid on economic development. China launched a Contract Responsibility Mechanism under which newly formed subsidiary enterprises of central and provincial government institutions—which would eventually become SOEs—became implementing parties for Chinese-financed aid projects (Bach et al. 2017, 39–42). Although these projects were not initially designed to be profit-making, together with Deng’s reforms, they laid the groundwork for SOEs to become commercial contractors in the 1980s.
Since then, Chinese SOEs have evolved into major global economic actors. Across the developing world, their infrastructure contracting, greenfield investments, and mergers and acquisitions (M&A), supported by Chinese policy bank financing, have increased exponentially. Available data suggests that China’s government has committed hundreds of billions in development finance worldwide between 2000 and 2017, much of which has been financed with loans issued by China’s two major policy banks, China Eximbank and China Development Bank (Custer et al. 2021; Dreher et al. 2022). In 2016, China’s outward foreign direct investment similarly reached a historic high of approximately $196 billion (Statista 2020). Chinese firms are also increasingly active in global markets, accounting for approximately 16% of all global M&A activity by value in 2017 (Gordon and Milhaupt 2019, 209).
SOEs’ international infrastructure business has strategic value for China’s government both at home and abroad. It furthers multiple state policy goals: solving domestic challenges of excess industrial capacity and surplus labor, developing external sources of demand, exporting technologies like high-speed railways and ultra-high voltage power grids, and supporting Beijing’s broader push to move up global value chains in high-tech and strategic sectors (Zhang 2021). SOE participation and performance on large overseas infrastructure projects is inherently political, because these works are often linchpins in China’s bilateral relations with foreign governments (Mohan and Tan-Mullins 2019; Ye 2019; Ho 2020). Host countries and other international actors also frequently scrutinize such projects as indicators about the nature of China’s economic intentions and actions abroad.
At the project level, Chinese SOEs formally adhere to international governance and health, safety, and environment (HSE) standards. However, they also seek to make profits—even if not necessarily to maximize those profits—and their contractor status decreases their stakes in long-term project operations and local impact (Lee 2017). SOEs aim to serve domestic Chinese development by achieving commercial success while simultaneously responding to the state’s policies and priorities. Chinese SOEs must meet annual construction contract targets set by the government and support strategic state aims like maintaining employment and exporting indigenously developed technologies. Given these factors, SOEs should be most interested in major infrastructure projects that can enhance their own commercial expansion while also serving larger Chinese policy goals.
Humanitarian Organizations
Value-driven international humanitarian organizations differ starkly from politically and profit-motivated Chinese SOEs. Humanitarian organizations primarily consist of inter-governmental organizations (IGOs) such as the World Food Programme (WFP) and UN International Children’s Emergency Fund (UNICEF), INGOs such as Mercy Corps and Oxfam, and non-governmental humanitarian agencies like the International Red Cross and Red Crescent Movement. These actors subscribe to the four basic humanitarian principles of impartiality, neutrality, independence, and humanity derived from the Fundamental Principles of the Red Cross and Red Crescent Movement proclaimed in Vienna in 1965.11 Two UN General Assembly resolutions (46/182 and 58/114) have formally adopted these four core principles, and the UN has tasked its Office for the Coordination of Humanitarian Affairs (OCHA) with leading international humanitarian response to complex emergencies and natural disasters.12
The humanitarian space is a site of “principled aid” (Holhorst and Jansen 2010) and part of a broader postwar, ideological movement to realize moral and ethical norm-driven global development (Lumsdaine 1993). In addition to the four basic humanitarian principles detailed above, many humanitarian organizations have developed international codes of conduct and guidelines for best practices.13 Individual humanitarian organizations also adhere to their own strict standards and codes of conduct.14
Humanitarian organizations work worldwide with international donor support, particularly in places experiencing disasters, complex emergencies, and protracted humanitarian crises. In 2019, before the COVID-19 pandemic began, the top destinations for international humanitarian assistance included Yemen, Syria, Turkey, and South Sudan. In 2019, most of the $24.81 billion in international humanitarian aid was channeled through UN agencies, such as the WFP, UNHCR, and UNICEF, followed by international NGOs like Save the Children and Mercy Corps. Together, UN agencies and INGOs are thus two of the most important global distributors of humanitarian assistance. Top bilateral donors include the USA, Germany, the UK, the European Commission, and Saudi Arabia.15 In sum, a small number of countries are the primary donors supporting the work of humanitarian organizations across the globe.
The Chinese government has historically given minimal support to international humanitarian organizations’ overseas work. China has allocated some foreign aid to poverty alleviation, a key factor in many humanitarian crises, but this remains a very small share of its overall development finance (Hirono 2018). When the Chinese government does give humanitarian aid, it prefers to use a bilateral approach. In 2018, China channeled more than 60% of its humanitarian assistance directly to national governments, 37% through UN agencies, and none through INGOs.16 Moreover, China’s humanitarian assistance typically follows major disasters, like the 2014 Ebola crisis or the 2015 Nepal earthquake, yielding fluctuating annual levels of humanitarian spending (Hirono 2018). In short, the Chinese government provides limited amounts of humanitarian aid, distributes it largely through bilateral channels and on an ad hoc basis in response to short-term emergencies, and does not usually involve Chinese SOEs. Unlike commercial infrastructure projects, Chinese-financed humanitarian projects have thus not been a springboard for Chinese SOEs in international development.
For their part, humanitarian organizations rarely partner directly with Chinese companies––or firms of any nationality––on infrastructure development abroad. While studies have increasingly documented a role for business in humanitarian supply chains and logistics, humanitarian organization-foreign company partnerships to develop overseas infrastructure works remain infrequent (Nurmala et al. 2017). This makes the collaboration now emerging between leading humanitarian organizations (both INGOs and IGOs) and Chinese SOEs all the more novel and surprising.
Local Actors
Local actors, while not the main focus of this study, are also important players in Chinese SOE-humanitarian organization partnerships on infrastructure projects abroad. These collaborations do not occur in a vacuum—they are embedded within local geographic, historical, and social contexts. Humanitarian organizations have long endeavored to engage local governments, civil society organizations, and communities near projects (Hickey and Mohan 2004). Chinese infrastructure contractors operating abroad have increasingly emphasized corporate social responsibility (CSR) activities and local engagement, including with civil society organizations and community members (Tan Mullins 2020). However, firms’ actions do not always match their rhetoric; overall, their efforts remain significantly less than those of humanitarian organizations.
Local actors affect the short-term success and the long-term viability of infrastructure projects involving Chinese SOEs and humanitarian organizations. Numerous studies show how host country governments, politicians, civil society organizations, and individuals at both local and national levels shape project initiation, terms, effects, and reception (Leslie 2016; Soulé-Kohndou 2018; Tritto 2020; Wang 2022). Even when local actors are initially excluded from high-level negotiations, subnational governments, civil society organizations, and local communities can influence implementation—in some cases even causing infrastructure projects to be halted or cancelled altogether (Kirchherr et al. 2017; Lampton et al. 2020). Recent analysis suggests that local citizens may view projects involving international actors with distinct reputations and comparative advantages, such as Chinese SOEs and INGOs, as complementary and therefore more positively than single-actor projects (Strange, Plantan, and Leutert 2022).
Conditions
Chinese SOEs and humanitarian appear to have weak, if any, incentives to partner on overseas infrastructure development. Where, when, and why are these actors working jointly to build infrastructure projects abroad? We argue that three conditions may help to explain the emergence of partnerships between humanitarian organizations and Chinese SOEs. Specifically, these include humanitarian organizations’ track records working in China or with Chinese actors, SOEs’ market presence in host countries, and host country political instability.
First, we expect that these partnerships are more likely when a humanitarian organization has an established presence in China or a record of working with Chinese actors. A track record of engaging Chinese actors within China helps humanitarian organizations to extend these relationships beyond China’s borders in multiple ways. Although the host country office of a humanitarian organization typically manages the infrastructure contractor, China country offices can still mitigate language or cultural barriers remotely or through staff visits. Their multinational and bilingual staff can also help to identify collaboration opportunities. As China itself has rapidly developed, many major INGOs and IGOs have reimagined their China country offices to focus on working with Chinese actors in third countries (Farid and Li 2021). For this reason, we expect to find evidence of growing ties between humanitarian organizations’ China offices and offices in other countries on how to engage with Chinese actors, including SOE contractors.
Second, from Chinese SOEs’ perspective, partnering with humanitarian organizations may be attractive if it facilitates market access or provides reputational benefits. Although infrastructure contracts with humanitarian organizations are typically far smaller than those for other types of project owners, they can still aid Chinese SOEs to establish a larger market presence in the future. Smaller projects can help firms to grow their business nationally or regionally. They can also signal companies’ ability to operate in new business areas in existing markets. “Investing up,” by voluntarily adhering to the stricter international standards associated with humanitarian organization partnerships, may also provide a reputational benefit for Chinese SOEs that could help secure other, larger contracts in a given industry, country, or region (Zeng and Eastin 2012). This is because working with INGOs or IGOs requires companies to adhere to international best practices related to environmental or labor safeguards. The value of such reputational benefits to a given firm may vary depending on its international business experience and project portfolio.17 Thus, we expect that Chinese firms will be more interested in working with humanitarian organizations if a project would aid market access or confer reputational benefits.
A third factor that may encourage such partnerships is host country instability. For SOEs, partnering with humanitarian organizations may be especially valuable in dangerous or politically unstable host country contexts. Late arrival to international markets and pursuit of natural resources have pushed many SOEs into riskier environments. Over one quarter of China’s overseas direct investment is located in recipient countries with high levels of political risk (Zou and Jones 2020, 93). After high-profile missteps, such as the Myitsone Dam in Myanmar and the Kamchay Dam in Cambodia, Chinese companies are increasingly willing to work with non-state actors to mitigate risk (Maurin and Yeophantong 2013; Zou and Jones 2020; Yeophantong 2020).18 Working directly with humanitarian organizations, which often have rich operational experience and knowledge of these contexts, as well as strong local networks, can provide greater assurance that projects will proceed despite high political risk. These same factors may also increase the likelihood that humanitarian organizations will partner with Chinese SOEs for these projects, since other foreign firms may be less willing to bid and host country companies may lack sufficient capacity or technical expertise.
Case Selection, Data, and Methods
We analyze three cases of Chinese SOE collaboration with humanitarian organizations on overseas infrastructure projects. Specifically, they include the following: (1) Sinohydro and China Geo-engineering Corporation’s work with Mercy Corps on the IMAGINE Program in the DRC; (2) China Geo-engineering Corporation and the International Committee of the Red Cross’s partnership on the Alhamduri Urban Water Works in Nigeria; and (3) Weihai International Economic and Technical Cooperative (WIETC)’s collaboration with the World Food Programme on a cassava plant in the Republic of Congo. Figure 1 below illustrates the location of the three cases.Fig. 1 Map of project locations
We select these three cases because they are the only instances to date of direct overseas infrastructure cooperation between humanitarian organizations and Chinese SOEs that we identified. Our case selection is thus a census of the universe of all known cases, rather than a sample drawn from a larger population. In a parallel study, we systematically collected data on all publicly known instances of INGO cooperation with Chinese actors in developing countries: out of a total of 133 projects involving collaborations with Chinese actors, eight projects involved Chinese SOEs and only three—each of which is examined in this article—were direct partnerships for infrastructure project implementation (Plantan, Leutert, and Strange 2022).19 This larger data collection exercise offers reassurance that we have identified the vast majority—or even all—current cases of this phenomenon. However, we cannot rule out the possibility of other additional cases, particularly those that have begun since we completed data collection in 2021 or those involving other actor types such as IGOs. The three cases analyzed here constitute a plausibility probe providing initial evidence for an emerging phenomenon of Chinese SOE-humanitarian organization partnerships on overseas infrastructure development.
All three cases involve humanitarian organizations that are developing in-country infrastructure to achieve their missions. From clean water infrastructure to domestic food production, each case demonstrates how humanitarian work, driven by values-based and altruistic motivations, can include opportunities to work jointly with Chinese SOEs to build or facilitate on-the-ground projects. The cases share several key features: (1) involvement of INGOs or IGOs operating in the humanitarian space; (2) collaboration with Chinese SOEs for infrastructure development critical to their mandates; (3) location in sub-Saharan Africa; (4) location in fragile states. Table 1 compares the three cases.Table 1 Case comparison
Case 1:
IMAGINE Program Case 2:
Alhamduri Urban Water Works Case 3:
Cassava plant
Location Democratic Republic of Congo Nigeria Republic of Congo
Level of state stability* FSI score = 109.4
Rank = 5th
Dysfunctional
FSI score = 97.3
Rank = 14th
Low-capacity
FSI score = 92.1
Rank = 25th
Low-capacity
Project type Water infrastructure Water infrastructure Cassava flour plant
Humanitarian organization Mercy Corps ICRC WFP
Funder UK Foreign, Commonwealth and Development Office (FCDO) [formerly the Department for International Development [DFID]) ICRC Zhengwei Technique, subsidiary of Weihai International Economic and Technical Cooperative (WIETC)**
Chinese SOE(s) Sinohydro; China Geo-engineering Corporation (CGC) CGC Zhengwei Technique**
State ownership type Central SOEs Central SOE Local SOE
Cooperation type Direct contracting Direct contracting Facilitation
Contract value £9,500,000 ($13,000,000) $3,000,000 No formal engagement, no contract value
Project duration 2013–present 2017–2019 2019–present
*2020 Fragile States Index data (https://fragilestatesindex.org/country-data/); classifications based on Ziaja et al. (2019)
**This company is in talks for the project, but the final plans and funding structure have not been confirmed as of late 2021.
These cases’ shared attributes may provide insight about the external validity of our findings and where similar projects might emerge in the future. All three cases are located in underdeveloped, fragile states in sub-Saharan Africa. According to World Bank classifications, Nigeria and the Republic of Congo are lower-middle income countries and the DRC is a low-income country. All three countries have historically experienced high levels of civil conflict and instability, and the infrastructure projects analyzed in the DRC (Eastern DRC) and Nigeria (northeastern Nigeria) are situated in particularly volatile subnational regions. These shared case attributes could be potential scope conditions; they provide suggestive evidence that humanitarian organizations and Chinese SOEs might find it more valuable to cooperate on infrastructure projects in underdeveloped, conflict-prone settings. Furthermore, large humanitarian organizations with resources sufficient to support infrastructure contracts have spearheaded these projects. Smaller humanitarian organizations may not have comparable capacity, and thus, our findings are potentially less generalizable to those cases. Additional research is needed to track this phenomenon over time and to further assess our findings’ external validity in other geographic and cooperative contexts.
Our research draws on 36 interviews conducted between December 2019 and August 2021 with key stakeholders in the USA, Canada, China, the DRC, the Republic of Congo, Nigeria, Kenya, the UK, France, and Switzerland.20 We identified respondents through snowball sampling. One author conducted interviews with China-based individuals in-person in December 2019. After the outbreak of the COVID-19 pandemic, the authors did subsequent interviews remotely as a team using Zoom, Skype, WeChat, and other online platforms. The pandemic disrupted the authors’ plans to conduct in-person fieldwork in China, the DRC, and Rwanda in 2020 and 2021. This constrained our ability to collect data and triangulate inferences by interviewing local workers and community members. However, we have sought to mitigate this limitation by integrating additional qualitative data from online research and secondary sources. We used all of the data collected to analyze the projects, including the advantages and disadvantages of collaboration as each stakeholder perceived them.
Case Studies
Mercy Corps and the IMAGINE Program in the DRC
Mercy Corps’ IMAGINE Program in the Democratic Republic of Congo (DRC) is one example of Chinese SOE-humanitarian organization cooperation on overseas infrastructure development. Mercy Corps is an INGO headquartered in the USA that works in more than 40 countries worldwide to combat poverty, disaster, violent conflict, and climate change. The Mercy Corps DRC program, the organization’s largest country program, employs nearly 500 national and international staff.21 The DRC program conducts several multi-year and shorter-term projects focused on clean water access and emergency humanitarian response. Rebuilding the physical infrastructure for public services like water and sanitation is an urgent imperative: only 52% of the population in the DRC has access to an improved water source, and only 29% have access to improved sanitation facilities—even though the country has over 50% of Africa’s water reserves.22
The IMAGINE program is a multi-year project funded by the FCDO of the UK to improve water access, sanitation, and hygiene and thereby reduce child mortality in the eastern DRC. Established through collaboration with REGIDESO, the national state-owned water utility, it is the first public–private partnership in the country’s water sector. This status and IMAGINE’s potential public health impact make it a priority initiative for Mercy Corps. As the DRC Country Director Whitney Elmer explained: “It’s a program that’s well known within [Mercy Corps] and one that has [the] attention of senior leadership even outside the country team.”23 The IMAGINE Program primarily involves the construction of water tap and treatment facilities in the cities of Goma and Bukavu and community hygiene promotion activities. Construction is still underway and is expected to be completed in 2022.
Mercy Corps recruited companies to build IMAGINE Program infrastructure through a competitive international tender. In addition to issuing a call for expressions of interest, Mercy Corps staff traveled to Kenya and Uganda to promote the project and reach out to potential bidders.24 After a pre-qualification process in which Mercy Corps provided the engineering design for the project, companies from multiple countries submitted bids. Ultimately, Mercy Corps awarded contracts to two Chinese SOEs: Sinohydro and China Geo-Engineering Corporation (CGC). Sinohydro was responsible for building the infrastructure in Goma––two water reservoirs on Mount Goma and in Bushara––and rehabilitating and installing distribution pipes and other infrastructure linking the reservoirs with tap stands in neighborhoods. CGC was tasked with the project facilities in Bukavu, which included the additional work of rehabilitating the REGIDESO water treatment plant in Murhundu.
Mercy Corps cited several reasons for awarding contracts to the two Chinese SOEs. First, they met or exceeded the bid selection criteria. “Our selection criteria [were] pretty much driven by administrative qualifications, technical qualifications, and price,” said Elmer. While Mercy Corps did not deliberately target Chinese companies, they expected them to submit bids because of their established presence in the global infrastructure industry and African markets in particular. As Jeff Bosa explained: “They have the knowledge of the region. They have the financial muscle and resources.”25 Among all the different stakeholders involved in the bidding process––including Mercy Corps’ staff in the DRC, headquarters employees in Portland, Oregon, its legal team, REGIDESO and other DRC government representatives––there was a consensus to select the two Chinese contractors.
Sinohydro and CGC, both companies owned by China’s central government, signed contracts with Mercy Corps in 2018 to construct the IMAGINE Program infrastructure.26 The first, Sinohydro, is the name of a former central SOE that was merged with another central SOE in 2011 to create a new conglomerate—PowerChina.27 Although the company is popularly known as “Sinohydro” in international markets, the legal entity partnering with Mercy Corps on the IMAGINE Program is PowerChina International, PowerChina’s flagship subsidiary. China Geo-Engineering Corporation (CGC) is also a Chinese central SOE.28 First established in 1982, today, it is a comprehensive contractor with business activities ranging from the construction of water works and transportation infrastructure to mineral surveying and mining.
Multiple factors motivated Sinohydro and CGC to submit bids for the Mercy Corps contracts. First, although the contracts were much smaller than the companies’ other international projects, the IMAGINE Program helped them expand into new markets and business areas. The IMAGINE Program was CGC’s first project in the DRC, and it was Sinohydro’s first project in the country’s water sector. In addition, the SOEs viewed INGOs and IGOs as reliable partners. As a representative of Sinohydro’s West Africa department explained: “We care not only about the project, but also about the employer. NGOs we think are good employers. Why do we think so? One important way in which we judge potential projects is the funding issue. How will we be able to issue construction payments? If we assess the funding for the project to be stable and reliable, then we take this into consideration when bidding.”29 In sum, the Chinese companies perceived Mercy Corps to be a reliable partner that could help them break into new markets and navigate an operationally challenging business environment.
International Committee of the Red Cross and the Alhamduri Urban Water Works in Nigeria
Our second case involves CGC’s work with the International Committee of the Red Cross (ICRC) on the Alhamduri Urban Water Works in Nigeria. The ICRC is one of the world’s oldest humanitarian organizations, founded in Switzerland in 1863. Its mandate, formalized under the Geneva Convention of 1949, is to assist the victims of conflict and violence and promote laws protecting victims of war worldwide. It employs over 20,000 people and operates in more than 80 countries.30 In Nigeria, the ICRC’s activities center on responding to the humanitarian consequences of ongoing armed conflict in the North East, as well as insecurity in the North West, North Central, and South South regions. Specifically, it assists with missing persons investigations, communications, and counseling; public health assistance, including improved access to water and sanitation; and the provision of food as well as other forms of material aid.31
The Alhamduri Urban Water Works is one of the ICRC’s recent initiatives in the North East in Nigeria. The ICRC jointly commissioned the project with the Borno State Ministry of Water Resources to improve water quality and access for approximately 150,000 people, including those displaced by armed conflict, in an area of limited water supplies.32 Built on pre-existing infrastructure, the project involved drilling three new boreholes, building production and treatment facilities, and expanding water stands and taps in the local water distribution network.33 The project launched at the end of 2017, contractor work on the main facility began in January 2018, and construction was completed in January 2019. According to ICRC Water and Habitat Coordinator Fernando Resta, the ICRC is transitioning to a position of systemic support for the project, while transferring primary implementation to the Borno State Ministry of Water Resources.34
Like Mercy Corps’ IMAGINE project, the ICRC ultimately selected CGC through a competitive international tender due to its combination of technology, expertise, and experience.35 The company had demonstrated experience building nearby road projects, equipment available nearby as well as a steel fabricating plant elsewhere in the country, an expressed willingness to follow ICRC rules and procedures concerning HSE, and neutrality with respect to the ongoing conflict in northeast Borno state.36 Although the contract value of the Alhamduri Urban Water Works was only $3 million, it was still a worthwhile commercial opportunity for CGC to utilize its existing equipment in northern Nigeria and add to its bottom line for business in the country. There were also fewer competitive bids for mid-sized infrastructure projects: “European construction companies would hardly bother for anything below $10 or even $20 million contract. Chinese companies showed here that they could be interested in smaller amounts, like a couple of million.”37 In this way, Chinese companies like CGC emerged as highly competitive bidders for small and mid-sized contracts, which were too small for European companies to be interested in bidding on––but also too big for local companies to be competitive.
To the authors’ knowledge, ICRC has not yet partnered again with CGC or another Chinese SOE, although further collaboration in Nigeria or elsewhere remains possible. As ICRC Urban Water Program Manager David Lee Zziwa described the incentives for such a future partnership: “[Chinese companies] have the equipment, they have the expertise, but they lacked supervision of local staff on the ground. So if those elements are rectified, I would recommend working with them again.”38 ICRC Head of Operations for West Africa Doris El Doueihy also expressed the possibility of future cooperation, underscoring that contractor capability rather than nationality is of primary concern: “If the project is managed well, we [are] fine with it. It [the partnership decision] will be dependent on the project itself and the standard or the offer that they are capable to give us.”39 Those at the ICRC therefore expect partnerships with Chinese companies to continue and expand in the future.
Similarly, multiple factors incentivized CGC to work with the ICRC on the Alhamduri Urban Water Works. The company had previously completed road construction projects in northern Nigeria and already had equipment and materials in the area that it could use. Like the Chinese SOE Sinohydro that partnered with Mercy Corps on the IMAGINE Program in the DRC, CGC also viewed the ICRC as a reliable financier. Zziwa recalled that it was clear to CGC that “ICRC was in a position to pay them.”40 The ability to use existing equipment and materials on the ground, coupled with ICRC’s reliable funding, made the contract attractive to CGC despite its smaller scale.
Overall, the Alhmanduri Water Works again illustrates the opportunities and challenges for joint projects involving humanitarian organizations and Chinese SOEs. For ICRC, CGC brought expertise, a competitive bid, and access to a broader supply chain in an operationally challenging geography. In return, CGC was able to lower its costs by using existing materials and equipment in Nigeria, while ensuring on-time payment from a reliable project owner. However, the personnel challenges of not having Chinese managers on-site raised concerns for ICRC about future Chinese SOE partners again adopting a sub-contracting approach and thus creating additional logistical and supervision challenges. Although Chinese SOEs will likely bid for future ICRC infrastructure projects in Nigeria and globally––with bids for other projects in Nigeria already submitted––any decision to partner again would depend on project needs and contractor expertise, price, and project delivery.41
World Food Programme and Cassava Flour Production in Congo
Our third case examines collaboration between the World Food Programme (WFP), an IGO, and several Chinese commercial actors––including an SOE––in support of cassava value chain development in the Republic of Congo. The WFP is the food assistance branch of the UN and the world’s largest humanitarian organization. As an IGO, it works closely with host-country governments to achieve their national priorities. In the Republic of Congo, the WFP works with the government on its three main goals of its National Development Plan: (1) improving governance; (2) increasing human capital; and (3) economic diversification through agriculture.42 Efforts to realize this third priority, economic diversification through agriculture, have created an opportunity for WFP Congo to work with Chinese companies.
Despite its wealth of available arable land, the Republic of Congo depends on imported food to meet its population’s needs. Over 90% of Congo’s arable land is uncultivated, and 70% of its food is imported from overseas.43 At the same time, the country relies extremely heavily on export revenue which makes up 98% of GDP (World Bank, 2018). The country’s top exports are natural resources like petroleum, copper, and timber—the majority of which are bound for China.44 The need to diversify the Congolese economy and feed its population with domestically produced food is a major government priority.
Facing this challenge, the WFP began to think of ways to support Congo’s agriculture and to diversify its economy. Jean-Martin Bauer, the country director for the WFP in Congo-Brazzaville between 2017 and 2020, viewed smallholder farmers as a key part of the solution. Instead of continuing a decades-long practice of importing rice for its aid programs, the WFP started purchasing food from local smallholder farmers to feed vulnerable populations like schoolchildren, refugees, and internally displaced persons. While the WFP has been able to increase the amount of food purchased from local smallholder farmers over time, Bauer saw a need for small-scale industrial production of flour from Congo’s major local crop: cassava.
As plans for local cassava production began to take shape, WFP considered how to engage Chinese partners. The Republic of Congo and China have historically had a close diplomatic relationship and China is a major investor in the country’s economy. Because of these strong connections to China, Brazzaville has a local branch of the Chinese Academy for Tropical Agriculture and Sciences (CATAS),45 which helps to share China’s expertise about cassava production. The WFP initially worked with the Brazzaville-branch of CATAS to provide training for local smallholder farmers about cassava farming techniques and how to reduce post-harvest losses. That positive experience with Chinese experts provided an impetus for WFP Congo to consider working with a Chinese company to build and operate the cassava flour plant. In fact, the initial suggestion to work with a Chinese company arose through an informal conversation between WFP Congo representatives and the Chinese embassy in Brazzaville.46
After that conversation, the WFP began to communicate with several Chinese companies. At first, the WFP discussed plans with a Chinese central SOE, but ultimately moved forward with Zhengwei Technique. Weihai International Economic and Technical Cooperative (WIETC), a local SOE headquartered in Weihai in Shandong Province, established Zhengwei Technique in 2001 as its locally incorporated subsidiary. The company has since completed multiple large infrastructure projects in Congo, including the country’s largest international airport, and has also engaged in potash extraction.47
In September 2019, the Investing in Africa Forum in Brazzaville convened public and private sector actors from China, Africa, and the international community to promote collaboration on international infrastructure development (World Bank 2019). Bauer describes attending this forum and meeting WIETC representatives, who were featured at a high-level event. After this, the WFP began facilitating conversations between the company and the Congolese government. In October 2019, the company toured southern regions of the country to evaluate their agricultural potential. The company ultimately signed an MOU with the Ministry of Agriculture stating their intent to consider investment in the sector.
Once built, the WFP envisions that the cassava plant would provide locally sourced food and help stimulate the local economy. According to Bauer, the factory would produce 10,000 to 20,000 tons of fortified cassava flour per year. In addition to building and operating the plant, the WFP is also encouraging the Chinese company to source raw cassava tubers from local smallholder farmers.48 In this way, the plant would help to diversify the Republic of Congo’s economy through domestic production of agriculture without displacing small farmers.
Rather than partnering on projects directly as in the previous two cases, the WFP’s role has primarily been to facilitate negotiations between the Congolese government and the Chinese company. As WFP Congo interim country director Ali Outtara explained: “We are not directly contracting with the company, [but] we are trying to facilitate the implementation of this factory because it will help us implement our activities.”49 WFP Congo plans to help smallholder farmers increase their cassava production in order to sell excess supply to the plant. The WFP could then buy the finished product from the cassava mill to use in their local food aid programs, such as the school feeding program. If this first project goes well, the WFP and the Congolese government envision spreading the model to other parts of the country.
From Chinese companies’ perspective, the WFP’s role in the project can uniquely help alleviate concerns about political and commercial risk. The Republic of Congo has a risky business environment due to political instability and civil unrest. In 2016, allegations of electoral fraud in the presidential election led to protests and violent conflict involving insurgent groups that lasted until December 2017, causing massive internal displacement.50 The fact that the WFP could purchase the cassava flour provides much-needed stability for Chinese investors. “We’re here to advise, but we’re also here to de-risk. As long as the WFP is here to say, ‘look, we’re interested in potentially buying the product,’ that really matters to an investor,” Bauer added.51
Although the ongoing COVID-19 pandemic has delayed the cassava plant project, it is still moving forward. In February 2020, a joint proposal between the WFP, Food and Agricultural Organization (FAO), and the International Fund for Agricultural Development (IFAD) for cassava value chain development, which includes plans for the flour mill, received $492,437 in funds from the China-IFAD South-South and Triangular Cooperation Facility financed by the Chinese Ministry of Finance.52 Then, in December 2020, WFP Congo jointly hosted a webinar on cassava value chain development together with the WFP’s Regional Excellence Centre Against Hunger and Malnutrition (CERFAM) in Cote d’Ivoire and the WFP Center of Excellence in China. The webinar also convened other actors from WFP offices in Africa, Brazil, China, and WFP headquarters in Rome.53 The planned Congolese cassava plant and the prospect of similar projects in other countries was a key topic of discussion.54 As of May 2021, the plans for the project have moved forward with a market analysis and social and environmental assessments.55
If realized, the cassava plant project could benefit all three actors — the Congolese government, the Chinese company, and WFP Congo. It would advance the government’s economic diversification goals, allow the Chinese company to pivot from infrastructure and extraction to production while minimizing risks, and increase the WFP’s ability to locally source food for its programs. The plant, once completed, could also become a model for other similar initiatives within the WFP’s operations in Africa and beyond.
Chinese SOE-Humanitarian Organization Partnerships: Opportunities and Challenges
The above cases illustrate several opportunities and challenges for SOE-humanitarian partnerships. From humanitarian organizations’ perspective, one advantage is Chinese companies’ willingness and ability to work in politically and economically challenging contexts. The WFP’s Bauer described with appreciation the prospect of collaborating with a company willing to work “in a country that has virtually no industrial processing capacity.”56 Besides this, humanitarian partners have been impressed with the quality of the work produced at a competitive price. Mercy Corps’ Elmer explained: “In a context like Eastern Congo, again, historically local companies don’t have the capacity, the quality standards aren’t there. …The issues that we have encountered, it’s not as much on the actual quality of the work that’s being done. …the product that’s being left behind is still of good quality.”57 In low capacity and politically risky contexts, Chinese companies’ growing international experience and technical expertise positions them as viable infrastructure partners for humanitarian organizations.
Similarly, Chinese SOEs can derive unique benefits from these partnerships despite their limited scale. In particular, Chinese SOEs’ commercial interests can align with humanitarian organizations’ goals. As Bosa of Mercy Corps, a Process Analysis Specialist responsible for overseeing all DRC-based programs, explained regarding the IMAGINE Program: “[The DRC is a] big country [with] a lot of resources, [and] the infrastructure is almost nothing. If they stabilize, then they need to put up infrastructure. Being in the [DR] Congo will give [Chinese companies] opportunities to do future work.”58 This observation further underscores that these partnerships are more likely to occur in contexts where Chinese companies are breaking into new markets or sectors.
However, the three cases also provide evidence that Chinese SOE-humanitarian organization partnerships also present challenges. First, small humanitarian projects have limited attractiveness to Chinese SOEs which routinely bid on contracts worth tens or hundreds of millions of dollars. For many of these firms, INGO and IGO projects are too few and small-scale to be a priority. As a Chinese SOE representative observed: “NGO cooperation could be popular all around the world if we can find these projects, but the size of the project can’t be big – there will be an upper bound. So the potential cooperation between SOEs and NGOs will be ‘small and popular’ [widespread]. We can’t do big projects with NGOs.”59 Unless some of the conditions discussed above are present––such as a humanitarian organization’s strong links to China or a firm’s desire to enter new markets––it is difficult to see how these partnerships would otherwise emerge.
Another frequently cited obstacle is language barriers and other communication-related issues. INGO and IGO respondents describe these as typical “transaction costs” of partnering with Chinese companies. Communication challenges were greater for projects involving actors of multiple nationalities and linguistic backgrounds, because translators may have varying command of all the languages required to effectively facilitate communication among diverse stakeholders. For example, communication between Mercy Corps and the Chinese SOE contractors on the IMAGINE Program routinely involved translating among three languages: French, English, and Chinese.
Humanitarian organizations’ China offices can play an important role by facilitating communication and alleviating coordination difficulties. In the case of the WFP, Bauer observed: “Thankfully, there was a very constructive role played by Chinese staff members at WFP who happen to know me and speak both English and Chinese. They played a very constructive role in making this happen.”60 When Bauer visited China, the WFP Centre of Excellence in China helped to facilitate his meetings. For the IMAGINE program, a Mercy Corps China staff member flew to the DRC for a week to assess and assist with communication. This person also attended meetings with Mercy Corps DRC staff and Sinohydro representatives in Goma and with CGC representatives at the company’s regional headquarters in Kigali, Rwanda. They helped directly address challenges including communication, reporting, labor issues, and technical and governance standards. As they recalled: “One successful outcome [of my visit] was that both sides had an adjustment of expectations about working together. They learned together about their respective strengths and capabilities.”61
Humanitarian organizations’ strict standards for environmental, labor, and gender safeguarding can also be a formidable hurdle. While such standards are a hallmark of many humanitarian organizations’ development activities, they may pose problems for Chinese SOEs less steeped in such practices. In the IMAGINE Program, concerns about safeguarding and labor contracts arose quickly. Mercy Corps has a hotline for workers and community members to report violations, and a large volume of community complaints were recorded after the project broke ground. As complaints came in, Mercy Corps informed the Chinese SOEs. Typically, the contractor should follow up to investigate the complaints, but this did not happen as Mercy Corps expected.62 While many early issues have improved, there have also been significant violations, including a health and safety issue that led to a fatal accident at one of the construction sites in March 2020. In response to these more serious issues, Mercy Corps DRC suspended works at different times at both sites, Goma and Bukavu.
A related challenge pertains to project monitoring and compliance. When implementing the IMAGINE Program, for example, Mercy Corps felt that CGC fell short of the high standards required by Mercy Corps, FCDO, and the project’s International Federation of Consulting Engineers (FIDIC) contract.63 Elmer, the country director for Mercy Corps DRC, explained: “We’ve had to adapt the way that we work … ensuring that our minimum standards are being met and that the workers do have a mechanism for providing feedback and complaints. Whereas the initial understanding, …was that the company would put in place that mechanism on their own and would have the capacity to do their own follow up of complaints or investigations.”64 Bosa, who handled the complaints at Mercy Corps, observed: “I also understand that safeguarding and things like that, these are new things for them and they are aligning. I’m sure they’ll adapt to it.”65 From the perspective of the Chinese SOE contractors, a Sinohydro representative shared: “For the IMAGINE Program, it’s not that we don’t have the reporting that they want. In Chinese we have tons of paperwork! [Rather it’s that] they want it written, in French or in English. We follow their book.”66 There has been a learning curve for both Mercy Corps and the Chinese SOE contractors as they acknowledge areas for future improvement and adjust to working with one another.
Corruption also remains a serious concern in Chinese SOEs’ international business and can occur at any stage, from project identification to tendering to implementation. Chinese leader Xi Jinping’s ongoing anti-corruption campaign has underscored the persistence and prevalence of graft in SOEs.67 Both at home and abroad, Chinese firms use “speed money” (petty bribes given to non-elites to surmount bureaucratic hurdles or accelerate exchanges) or “access money” (rewards to powerful officials to access exclusive, valuable privileges) (Ang, 2020). Graft involving international infrastructure projects has resulted in the censure and even blacklisting of Chinese SOEs.68 In this study’s three cases, corruption did not appear to be a primary driver or obstacle in Chinese SOE-humanitarian organization partnerships. This may be due in part to humanitarian organizations’ stringent standards and competitive international tenders for the projects.69 However, corruption remains an important concern in Chinese SOEs—and other foreign firms’—overseas infrastructure projects, and it should not be dismissed as a factor in any future Chinese SOE-humanitarian organization collaborations.
Growing pains in Chinese SOE-humanitarian organization partnerships also derive from fundamental differences in their perspectives on development. For humanitarian organizations, host country actors and needs are foremost, and their top priority is maximizing local impact by delivering quality infrastructure built to high governance and HSE standards. Chinese SOEs do view the infrastructure they build with humanitarian organizations as serving host country needs, but they prioritize commercial benefits. As a CGC employee stated: “The nature of our work in infrastructure and technical training is already promoting development.”70 On paper, many Chinese SOEs also share the HSE and governance standards of their INGO and IGO partners, but their focus is on making a profit from the projects they build. As Zziwa described from the ICRC’s experience working with CGC: “Because [the company has] a profit-making motivation, whatever knowledge they give comes with a cost …One of the advantages is that you will be working with a contractor who is experienced. …But of course, you have to be very careful because the contractor will guide you in a direction where he can make profit.”71
These different perspectives also underpin the division of labor between SOEs and humanitarian organizations within such partnerships. Chinese SOEs’ role as contractors limits their long-term stake in overseas infrastructure projects. Over the past decade, Chinese SOEs have been actively diversifying their partners and project modes abroad through public–private partnerships (PPP) with host country governments, taking equity shares in the projects they build, establishing joint-ventures with foreign firms, and even financing and constructing smaller works themselves. However, Chinese SOEs’ traditional and primary role remains as contractors. As such, Chinese SOEs’ physical presence and stakes in the project end when they complete project construction. They therefore have limited long-term stakes in humanitarian projects overseas—and in the local communities and environments where they are located. This contrasts with humanitarian organizations, who are often committed to long-term local development outcomes.
Fundamentally, the top priority for Chinese SOEs remains serving commercial objectives and China’s interests. Chinese SOEs’ business—including their international business—supports the state at home by providing employment for workers at home and abroad, exporting excess industrial capacity, and providing tax revenues and dividends (Leutert 2020). With international projects making up a growing proportion of many Chinese contractors’ business portfolios, these firms’ contributions to domestic development in China increasingly depend on their ability to win and execute profitable projects abroad. As a Sinohydro employee in Beijing explained: “The company belongs to the [Chinese] government. We need to do business so that the company can survive. In this way SOEs can help local people and places [in China] to develop. We are a company and we run its property for all people in China and the government.”72 Citing the IMAGINE Program in the DRC, they added: “We like CSR [corporate social responsibility], but to be frank that is not our motivation for this project.”73
Conclusion
As Chinese SOEs expand their operations worldwide, surprising new partnerships for overseas infrastructure development are emerging with humanitarian organizations like Mercy Corps, the ICRC, and the WFP. For Chinese SOEs, such small, low-profile works seem commercially and politically unattractive compared to the international megaprojects for which these firms are best known. Humanitarian organizations, which rarely contract for infrastructure, also emphasize the international standards and best practices Chinese SOEs are often critiqued for failing to uphold. The three cases analyzed here suggest that such partnerships are more likely to occur when a humanitarian organization has previous experience working in China or with Chinese actors, when Chinese SOEs seek to enter new markets or sectors, or when these firms operate in dangerous or politically unstable environments.
While currently modest in number, such partnerships may proliferate in the coming years. As the Chinese government combines goals for overseas development with humanitarian aid, it may even help to fund them. In September 2020, for example, the Chinese government and the WFP signed an agreement to provide food aid to Ethiopia and Guinea under the framework of the South-South Cooperation Assistance Fund.74 Also in 2020, the Chinese government gave $100,000 of humanitarian aid directly to Mercy Corps — a relatively rare instance of China channeling its humanitarian assistance through an NGO instead of transferring it directly to another national government.75 These examples illustrate that humanitarian organizations have recently been moderately successful in attracting Chinese funding for their work. Perhaps more importantly, to the extent that SOE-humanitarian partnerships further the interests of Chinese firms overseas, China’s government has a strong incentive to support the expansion of these projects. But apart from high-level agreements negotiated between Chinese government officials and international humanitarian organizations, humanitarian aid organizations and Chinese SOEs are also forging everyday partnerships more quietly––and organically––on the ground.
Indeed, all three of this paper’s cases illustrate a bottom-up approach to Chinese SOE-humanitarian organization partnerships. Mercy Corps DRC and ICRC Nigeria both held open international tenders for their water infrastructure projects and Chinese SOEs’ bids were the most competitive. An informal conversation between WFP Congo’s staff in Brazzaville with interlocutors at the Chinese embassy led to plans to build a cassava flour factory that would be mutually beneficial for a Chinese company and for WFP Congo’s food aid programs. Although top-down agreements and partnerships between the Chinese government and humanitarian organizations might make headlines and eventually provide additional funding, relationships on the ground reveal how these seemingly unlikely partners can align their divergent interests and initiate collaboration from the bottom up.
Future partnerships between “principled” actors and Chinese SOEs may extend beyond the humanitarian space. International NGOs are already engaging with a diverse cast of Chinese actors, including SOEs, around the globe (Plantan, Leutert, and Strange 2022). These INGOs work in diverse areas, from health to environmental protection to disaster relief. While cooperation on infrastructure development appears most puzzling, Chinese SOEs are also joining with INGOs to open spaces for dialogue, research, and training related to development work overseas. Additional research is needed to assess whether the same conditions affecting the likelihood of Chinese SOE-humanitarian organization partnerships on infrastructure projects also apply to other types of collaboration, such as joint research or training, or whether they apply to partnerships with privately owned Chinese companies.
In the short term, Chinese SOE-humanitarian organization partnerships face logistical challenges and mounting uncertainty. During the ongoing COVID-19 pandemic, China has restricted border entry for foreigners, discouraged non-essential travel for its citizens, and imposed stringent quarantine and testing requirements for international travelers of all nationalities.76 Repeated outbreaks of the Omicron variant inside China during 2022 have prompted large-scale lockdowns, including in the major commercial centers of Shanghai and Beijing, thereby inhibiting Chinese SOEs and other companies’ ability to pursue and conduct international business. And although Chinese SOEs in the construction sector have so far avoided regulatory crackdowns like those the Xi Jinping administration has executed in the technology and real estate sectors, this could change in the future.77 Operational difficulties and growing uncertainty could hamper SOEs’ ability to work with humanitarian organizations in the near future.
Nevertheless, several features of Chinese SOE-humanitarian organization partnerships may facilitate their continuation and possible expansion. First, Chinese SOEs have the technical expertise and equipment necessary to execute infrastructure projects while keeping costs competitive, even in politically unstable environments. As some countries slash overseas aid to mitigate the financial effects of the COVID-19 pandemic and Russia’s war against Ukraine,78 humanitarian organizations may be incentivized to keep project costs low in order to stretch limited budgets. Second, many humanitarian organizations have China operations that are explicitly pivoting to support South-South cooperation, such as the WFP’s Centre of Excellence in China or Ford Foundation Beijing’s new “China in the World” strategy.s79 China-based staff can apply years of accumulated knowledge and relationships to help identify new opportunities to collaborate with Chinese actors overseas, while China offices can help enhance communication and coordination. Third, Chinese SOEs view humanitarian organizations as reliable business partners that could provide opportunities to expand into new markets or business areas, further incentivizing partnerships. Each of these trends suggests that overseas development projects involving humanitarian organizations and Chinese SOEs may become more common in the future.
Acknowledgements
The authors gratefully acknowledge generous research support from Mercy Corps and the Tobias Center for Innovation in International Development at Indiana University. We especially thank Shahana Fedele, Florian Preusser, and Moritz Schuberth for invaluable support throughout the project. We also thank Serge Musubao and Ruoyi Song for excellent research assistance. Finally, thanks to Stephen Kaplan and participants at the 2020 American Political Science Association annual meeting, as well as three anonymous reviewers, for their helpful comments.
1 The first two phases, commissioned in 1972 and 1982, were notoriously unsuccessful and resulted in mass displacement, corruption, and debt stress (International Rivers 2021).
2 Final concession and consortium member details remain contested and uncertain.
3 It is also an example of what Strange (forthcoming) considers a “high-profile development project”: a project financed by foreign actors that is both highly visible and nationally salient for the host country.
4 Jiang Zemin formally inaugurated “going out” as a “new stage of China’s reform and opening movement” at the 16th National Party Congress in November 2002. Jiang Zemin, Report to the National Party Congress, November 8, 2002. Xi Jinping introduced the BRI during a visit to Kazakhstan in September 2013. “Chronology of China’s Belt and Road Initiative,” Xinhua, March 28, 2015.
5 The FCDO was the Department for International Development (DFID) prior to September 2020.
6 The 2019 annual report of PowerChina’s flagship publicly listed subsidiary reported total contract values of approximately 15.6 billion in eastern and southern Africa, $10.5 billion in central and western Africa, and $13.6 billion in the Middle East and northern Africa. 《中国电力建设股份有限公司 2019 年年度报告》 [China Power Construction Company, Limited 2019 Annual Report], http://www.powerchina.cn/art/2020/4/24/art_7640_863040.html.
7 Djafar Al Katanty and Hereward Holland, “Volcano Erupts in Eastern Congo, Thousands Flee Goma,” Reuters, May 22, 2021; United Nations Office of the High Commissioner for Human Rights, “UN Human Rights Chief and Special Adviser on Prevention of Genocide Disturbed by Increase in Violence and Hate Speech in Eastern DRC,” June 20, 2022.
8 See, for example, the internal policies of international NGOs like Mercy Corps (https://www.mercycorps.org/who-we-are/ethics-policies) or Oxfam (https://www.oxfam.org/en/what-we-do/about/safeguarding).
9 For example, many humanitarian organizations have committed to adhere to global standards through third-party accountability watchdogs like Accountable Now (https://accountablenow.org/) or EthicsPoint (https://www.oxfamamerica.org/about/contact-us/ethics-violations-fraud-or-other-complaints/).
10 Official list of the 97 central SOEs that SASAC manages (in Chinese): http://www.sasac.gov.cn/n4422011/n14158800/n14158998/c14159097/content.html.
11 For more information on the fundamental principles, see: https://www.icrc.org/en/fundamental-principles.
12 OCHA, “OCHA on Message: Humanitarian Principles,” 2012, https://www.unocha.org/sites/dms/Documents/OOM-humanitarianprinciples_eng_June12.pdf.
13 For example, the Good Humanitarian Donorship (GHD) Initiative establishes 24 principles and good practices for humanitarian donors to guide assistance and create greater accountability across its 42 members. For all 24 principles, see: https://www.ghdinitiative.org/ghd/gns/principles-good-practice-of-ghd/principles-good-practice-ghd.html.
14 This includes the Code of Conduct, which the ICRC and eight other sponsors established in 1994 to provide guidelines for all NGOs involved in humanitarian assistance (ICRC 1994). As an additional example, see the WFP’s Code of Conduct: https://www.wfp.org/publications/wfp-code-conduct.
15 This data is drawn from 2019 OCHA FTS database: https://fts.unocha.org/global-funding/overview/2019.
16 By comparison, UN agencies and INGOs were the top two recipients of U.S. humanitarian aid that same year, receiving 66% and 17%, respectively. These figures were calculated by the authors using 2018 global OCHA FTS data: https://fts.unocha.org/global-funding/countries/2018
17 Such reputational benefits are likely to be of greater value to Chinese SOEs that have less international experience and few if any projects abroad. Expressing the perspective of a large central SOE with an extensive record of international projects, a Sinohydro representative stated: “We don’t [need to] partner with INGOs to prove that we can meet international standards. Our cooperation with the World Bank proves that we have reached international standards.” In-person interview, Sinohydro Investment Department representative, Beijing, December 2019.
18 More generally, recent evidence suggests that Chinese-financed infrastructure and other development projects tend to increase the probability of civil protests in host countries (Iacoella et al. 2021).
19 A fourth case of a Chinese SOE implementing an infrastructure project was excluded from the analysis because in that instance the Chinese SOE was part of a larger consortium with other foreign companies.
20 Human subjects research protocols for this study have been approved by the institutional review boards of Harvard University (IRB19-1937; pertains to data collection through July 31, 2020); Indiana University (1,911,897,594; pertains to data collection beginning November 22, 2019); Stetson University (Protocol #856; pertains to data collection beginning August 11, 2020); and the University of Hong Kong (EA200060; pertains to data collection beginning August 14, 2020).
21 Remote interview with Whitney Elmer, Mercy Corps DRC Country Director, August 14, 2020.
22 UNICEF, “Democratic Republic of Congo, ‘What We Do;’ ‘Water Sanitation and Hygiene’”, https://www.unicef.org/drcongo/en/what-we-do/water-sanitation-and-hygiene.
23 Ibid.
24 Remote interview with Mort Anoushiravani, Mercy Corps Infrastructure Director, January 27, 2020.
25 Remote interview with Jeff Bosa, Process Analysis Specialist at Mercy Corps DRC, July 30, 2020.
26 Mercy Corps, “Communiqué de presse sur la Construction d'infrastructures d'eau afin.
d'améliorer l'accès à l'eau pour plus de 1 million d'habitants de Goma et Bukavu” [Press release on the Construction of water infrastructure to improve access to water for more than 1 million inhabitants of Goma and Bukavu], November 23, 2018.
27 PowerChina, “About Us,” https://en.powerchina.cn/2019-03/13/content_36436636.htm.
28 CGC, “Corporate Information,” http://www.en.cecep.cn/g12829.aspx.
29 In-person interview, Sinohydro West Africa Department representative, Beijing, December 2019.
30 ICRC, “Who We Are,” https://www.icrc.org/en/who-we-are.
31 ICRC, “ICRC in Nigeria Operational Facts & Figs. 2020,” March 15, 2021, https://www.icrc.org/en/document/nigeria-infographics-operational-facts-figures-2020.
32 ICRC, “Alhamduri Urban Water Project,” https://www.arcgis.com/apps/Cascade/index.html?appid=939340facfa2435b945b01f14dfee962.
33 Ibid.
34 Remote interview with Fernando Resta, Water and Habitat Coordinator at ICRC, May 20, 2021.
35 Remote interview with David Lee Zziwa, Urban Water Program Manager at ICRC, February 19, 2021.
36 Ibid.
37 Remote interview with a representative of an INGO working with Chinese companies in Africa, December 15, 2020.
38 Due to safety concerns after incidents of violence and kidnapping involving Chinese citizens in Borno state, CGC sent Nigerian staff from other parts of the country who they trained and monitored remotely from Abuja. The only time that the Chinese expatriate staff were on-site for construction was during the most technically complicated stages, such as the electromechanical works and the erecting of the above-ground water tank, approximately 8 months into the project. Remote interview with Zziwa, Urban Water Program Manager at ICRC, February 19, 2021.
39 Remote interview with Doris El Doueihy, Head of Operations for West Africa at ICRC, May 20, 2021.
40 Remote interview with Zziwa, Urban Water Program Manager at ICRC, February 19, 2021.
41 Ibid.
42 Remote interview with Jean-Martin Bauer, Country Director for WFP Congo in Brazzaville, July 13, 2020. If not otherwise noted, much of the background information in this section is informed by this interview.
43 World Bank, “Congo,” 2021, https://www.wfp.org/countries/congo.
44 OEC World, “Republic of the Congo,” 2021, https://oec.world/en/profile/country/cog/.
45 For more information, see: Chinese Academy of Tropical Agricultural Sciences, “Introduction,” December 22, 2017, http://www.catas.cn/EN/contents/1262/109161.html.
46 Remote interview with Ali Ouattara, Interim Country Director for WFP Congo, March 25, 2021.
47 For more information, see the timeline on WIETC’s website (Chinese and English): http://www.wietc.com/m/content.aspx?id=157.
48 Remote interview with Bauer, Country Director for WFP Congo in Brazzaville, July 13, 2020.
49 Remote interview with Ali Ouattara, Interim Country Director for WFP Congo, March 25, 2021.
50 Reuters, “Congo Republic Signs Peace Accord with ‘Ninja’ Rebels,” December 23, 2017.
51 Remote interview with Bauer, Country Director for WFP Congo in Brazzaville, July 13, 2020.
52 China-IFAD South-South and Triangular Cooperation Facility. “Approved proposals: second call.” https://www.ifad.org/documents/38714170/41790143/china-ifad-sstcf-2ndcall.pdf/00417b02-b582-6f51-efa4-045b299b382d?t=1581501509000.
53 WFP, “WFP Republic of Congo Country Brief – December 2020,” January 15, 2021, https://reliefweb.int/report/congo/wfp-republic-congo-country-brief-december-2020.
54 Remote interview with Ali Ouattara, Interim Country Director for WFP Congo, March 25, 2021.
55 WFP, “WFP Republic of Congo Country Brief – May 2021,” May 31, 2021, https://reliefweb.int/report/congo/wfp-republic-congo-country-brief-may-2021.
56 Remote interview with Jean-Martin Bauer, Country Director for WFP Congo in Brazzaville, July 13, 2020.
57 Remote interview with Whitney Elmer, Mercy Corps DRC Country Director, August 14, 2020.
58 Remote interview with Jeff Bosa, Process Analysis Specialist at Mercy Corps DRC, July 30, 2020.
59 In-person interview, Sinohydro West Africa Department representative, Beijing, December 2019.
60 Remote interview with Bauer, Country Director for WFP Congo in Brazzaville, July 13, 2020.
61 In-person interview with Mercy Corps China Office representative, Beijing, December 2019.
62 Information in this paragraph is largely gleaned from a remote interview with Jeff Bosa, Process Analysis Specialist at Mercy Corps DRC, July 30, 2020.
63 FIDIC is a French language acronym for Fédération Internationale Des Ingénieurs-Conseils.
64 Remote interview with Whitney Elmer, Mercy Corps DRC Country Director, August 14, 2020.
65 Remote interview with Jeff Bosa, Process Analysis Specialist at Mercy Corps DRC, July 30, 2020.
66 Interview with Sinohydro West Africa Department Representative, Beijing, December 2019.
67 SOE investigations during Xi’s anti-corruption campaign have uncovered executive embezzlement, bribes, and lavish perks like 25,000 RMB (3,600 USD) lunches (Zhou 2015).
68 For example, the World Bank and African Development Bank have censured Chinese SOEs like Sinohydro for corruption compliance concerns. African Development Bank Group, “Integrity in Development Projects: African Development Bank and Sinohydro Reach Settlement Agreement on Fraudulent Practice,” May 24, 2018; World Bank Group, “Annual Update: Integrity Vice Presidency (Int)—Fiscal Year 2016”:
http://documents1.worldbank.org/curated/en/330521476191334505/pdf/INT-FY16-Annual-Update-10062016.pdf. Multiple WIETC subsidiary entities are currently debarred by the African Development Bank for engaging in “coercive, collusive, corrupt, fraudulent or obstructive practices”; among them, the World Bank has also cross-debarred five. See https://www.afdb.org/en/projects-operations/debarment-and-sanctions-procedures and https://www.worldbank.org/en/projects-operations/procurement/debarred-firms.
69 As cited above, numerous humanitarian organization representatives expressed in interviews that Chinese SOEs were well-qualified and met tender requirements on the basis of their engineering expertise and operational experience.
70 Remote interview with CGC employee, DRC, December 2020.
71 Remote interview with Zziwa, Urban Water Program Manager at ICRC, February 19, 2021.
72 In-person interview with Sinohydro West Africa Department Representative, Beijing, December 2019.
73 Ibid.
74 WFP, “China Supports WFP’s Assistance to Vulnerable People in East and West Africa,” September 30, 2020.
75 Drawn from 2020 OCHA FTS data: https://fts.unocha.org/donors/2976/recipients/2020.
76 Thomas Hale, “China Says It Will ‘Strictly Limit’ Citizens From Going Abroad,” Financial Times, May 12, 2022.
77 “Factbox: How China’s Regulatory Crackdown has Reshaped its Tech, Property Sectors,” Reuters, April 29, 2022.
78 Kaamil Ahmed, “‘A Hammer Blow’: How UK Overseas Aid Cuts Affect the World’s Most Vulnerable,” The Guardian, June 3, 2021; Thalif Deen, “War in Ukraine & Rise in Arms Spending Undermine Development Aid to the World’s Poor,” Inter Press Service, April 15, 2022.
79 Ford Foundation, “China in the World,” July 2020, https://www.fordfoundation.org/media/5474/fordfoundation-china-july2020.pdf.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Ang Yuen Yuen China’s gilded age: The paradox of economic boom and vast corruption 2020 Cambridge Cambridge University Press
Bach Daniel C Cheng Zhangxi Taylor Ian China’s aid to Africa: Does friendship really matter? 2017 London Routledge
Bax, Pauline and Michael Kavanaugh. 2020. China dominates bid for Africa’s largest dam in new pact. Bloomberg, August 7, 2020. https://www.bloomberg.com/news/articles/2020-08-07/chinese-firms-dominate-bid-to-build-congo-s-inga-dam-in-new-pact#xj4y7vzkg
Brautigam Deborah The dragon’s gift: The real story of China in Africa 2011 Oxford Oxford University Press
Custer, Samantha, Axel Dreher, Thai-Binh Elston, Andreas Fuchs, Siddharta Ghose, Joyce Jiahui Lin, Ammar A. Malik, Bradley C. Parks, Brooke Russell, Kyra Solomon, Austin Strange, Michael J. Tierney, Katherine Walsh, Lincoln Zaleski, and Sheng Zhang. 2021. Tracking Chinese development finance: an application of AidData’s TUFF 2.0 methodology. Williamsburg, VA: AidData at William & Mary.
Mercy Corps. 2019. IMAGINE: improving access to safe water in DR Congo through public-private partnerships. October 7, 2019. https://www.mercycorps.org/research-resources/imagine-DRC
Currie, Antony and Ed Cropley. 2021. Green megalomania renews $80 billion Congo hydro dream. Reuters, June 17, 2021. https://www.reuters.com/breakingviews/green-megalomania-renews-80-bln-congo-hydro-dream-2021-06-18/
Dreher Axel Fuchs Andreas Parks Bradley C Strange Austin Tierney Michael J Banking on Beijing: The aims and impacts of China’s overseas development program 2022 Cambridge Cambridge University Press
Farid May Li Hui International NGOs as intermediaries in China’s ‘going out’ strategy International Affairs 2021 97 6 1945 1962 10.1093/ia/iiab183
Flyvbjerg Bent Bruzelius Nils Rothengatter Werner Megaprojects and risk: An anatomy of ambition 2003 Cambridge Cambridge University Press
Gill Bates Reilly James The tenuous hold of China Inc Africa the Washington Quarterly 2007 30 3 37 52 10.1162/wash.2007.30.3.37
Gordon Jeffrey N Milhaupt Curtis J China as a ‘national strategic buyer’: Toward a multilateral regime for cross-border M&A Columbia Business Law Review 2019 2019 192 251
Hickey Samuel Mohan Giles Participation: From tyranny to transformation: Exploring new approaches to participation in development 2004 London Zed Books
Hirono, Miwa. 2018. Exploring the links between Chinese foreign policy and humanitarian action. Humanitarian Policy Group Working Paper. Overseas Development Institute. https://cdn.odi.org/media/documents/12015.pdf
Ho Selina Infrastructure and Chinese power International Affairs 2020 96 6 1461 1485 10.1093/ia/iiaa171
Holhorst Dorothea Jansen Bram J Humanitarian space as arena: A perspective on the everyday politics of aid Development and Change 2010 41 6 1117 1139 10.1111/j.1467-7660.2010.01673.x
Holland, Hereward and Melanie Burton. 2021. Congo picks Australia’s fortescue to develop giant hydro project. Reuters, June 15, 2021. https://www.reuters.com/business/energy/australias-fortescue-talks-worlds-biggest-hydropower-project-congo-2021-06-15/#:~:text=MELBOURNE%2C%20June%2015%20(Reuters),to%20Chinese%20and%20Spanish%20developers.
Iacoella Francesco Martorano Bruno Metzger Laura Sanfilippo Marco Chinese official finance and political participation in Africa European Economic Review 2021 136 103741 10.1016/j.euroecorev.2021.103741
ICRC. 1994. Code of conduct for the International Red Cross and Red Crescent Movement and non-governmental organizations (NGOs) in Disaster Relief. https://www.icrc.org/en/doc/resources/documents/publication/p1067.htm
International Rivers. 2021. Grand Inga Dam, DR Congo. https://archive.internationalrivers.org/campaigns/grand-inga-dam-dr-congo
Kirchherr Julian Charles Katrina J Walton Matthew J The interplay of activists and dam developers: The case of Myanmar’s mega-dams International Journal of Water Resources Development 2017 33 1 111 131 10.1080/07900627.2016.1179176
Lampton David M Ho Selina Kuik Cheng-Chwee Rivers of iron: Railroads and chinese power in Southeast Asia 2020 Berkeley University of California Press
Lee Ching Kwan The specter of global China: Politics, labor, and foreign investment in Africa 2017 Chicago University of Chicago Press
Leslie Agnes Ngoma Zambia and China: Workers’ protest, civil society and the role of opposition politics in elevating state engagement African Studies Quarterly 2016 16 3–4 89 106
Leutert, Wendy. 2020. State-owned enterprises in contemporary China. In Luc Bernier, Florio Massimo, and Philippe Bance, eds. The Routledge handbook of state-owned enterprises. New York: Routledge.
Lumsdaine David Halloran Moral vision in international politics: The foreign aid regime, 1949–1989 1993 Princeton, NJ Princeton University Press
Maiza-Larrarte Andoni Claudio-Quiroga Gloria The impact of Sicomines on development in the Democratic Republic of Congo International Affairs 2019 95 2 423 446 10.1093/ia/iiz001
Maurin Cristelle Yeophantong Pichamon Going global responsibly? China’s strategies towards ‘sustainable’ overseas investments Pacific Affairs 2013 86 2 281 303 10.5509/2013862281
Mohan Giles Tan-Mullins May The geopolitics of south–south infrastructure development: Chinese-financed energy projects in the Global South Urban Studies 2019 56 7 1368 1385 10.1177/0042098018794351
Mullins Tan May. Smoothing the silk road through successful Chinese corporate social responsibility practices: Evidence from East Africa Journal of Contemporary China 2020 29 122 207 220 10.1080/10670564.2019.1637568
Mwase, Nkunde and Yongzheng Yang. 2012. BRICs’ philosophies for development financing and their implications for LICs. International Monetary Fund. Working Paper No. 2012/074.
Nurmala N de Leeuw Sander Dullaert Wout Humanitarian-business partnerships in managing humanitarian logistics Supply Chain Management 2017 22 1 82 94 10.1108/SCM-07-2016-0262
Plantan, Elizabeth, Wendy Leutert, and Austin Strange. 2022. Pivoting to overseas development: INGOs’ changing engagement with China. Working Paper.
Soulé-Kohndou, Folashadé. 2018. Bureaucratic agency and power asymmetry in Benin–China relations. In Chris Alden and Daniel Large, eds., New directions in Africa–China studies. New York: Routledge.
Statista. 2020. Annual outflow of foreign direct investment (FDI) from China between 2009 and 2019. https://www.statista.com/statistics/858019/china-outward-foreign-direct-investment-flows/
Strange Austin Seven decades of Chinese state financing in Africa: Tempering current debates Economic History of Developing Regions 2019 34 3 259 279 10.1080/20780389.2019.1618183
Strange, Austin. Forthcoming. Game changers? China’s global development projects. Cambridge: Cambridge University Press, Elements in Global China; under contract.
Strange, Austin, Elizabeth Plantan, and Wendy Leutert. 2022. Complementary partners? Attitudes toward multi-actor development projects in the Democratic Republic of Congo. Working paper.
Tritto Angela Contentious embeddedness: Chinese state capital and the Belt and Road Initiative in Indonesia Made in China Journal 2020 5 1 182 187 10.22459/MIC.05.01.2020.24
Wang Yuan Executive agency and state capacity in development: Comparing Sino-African railways in Kenya and Ethiopia Comparative Politics 2022 54 2 349 377 10.5129/001041522X16225661565091
World Bank. 2018. World Development Indicators Database. https://databank.worldbank.org/source/world-development-indicators
World Bank. 2019. The fifth investing in Africa forum: leveraging partnerships for economic diversification and jobs creation in African economies, Brazzaville, Congo, September 10–12, 2019. https://www.worldbank.org/en/events/2019/07/23/fifth-investing-in-africa-forum.
Ye Min Fragmentation and mobilization: Domestic politics of the belt and road in China Journal of Contemporary China 2019 28 119 696 711 10.1080/10670564.2019.1580428
Ye Min The belt road and beyond: State-mobilized globalization in China: 1998–2018 2020 Cambridge Cambridge University Press
Yeophantong Pichamon China and the accountability politics of hydropower development: How effective are transnational advocacy networks in the Mekong Region? Contemporary Southeast Asia 2020 42 1 85 117 10.1355/cs42-1d
Zeng Ka Eastin Joshua Do developing countries invest up? The environmental effects of foreign direct investment from less-developed countries World Development 2012 40 11 2221 2233 10.1016/j.worlddev.2012.03.008
Zhang Hong The aid-contracting nexus: The role of the international contracting industry in China’s overseas development engagements China Perspectives 2020 4 17 27 10.4000/chinaperspectives.11124
Zhang, Hong. 2021. Chinese international contractors in Africa: structure and agency. Working Paper No. 2021/47. China Africa Research Initiative, School of Advanced International Studies, Johns Hopkins University, Washington, DC.
Zhou Laura Senior executives of power utility state grid punished over lavish lunch South China Morning Post 2015 25 2015
Ziaja Sebastian Grävingholt Jörn Kreibaum Merle Constellations of fragility: An empirical typology of states Studies in Comparative International Development 2019 54 2 299 321 10.1007/s12116-019-09284-3
Zou Yizheng Jones Lee China’s response to threats to its overseas economic interests: Softening non-interference and cultivating hegemony Journal of Contemporary China 2020 29 121 92 108 10.1080/10670564.2019.1621532
| 36467127 | PMC9707120 | NO-CC CODE | 2022-12-01 23:20:20 | no | Stud Comp Int Dev. 2022 Nov 28;:1-29 | utf-8 | Stud Comp Int Dev | 2,022 | 10.1007/s12116-022-09368-7 | oa_other |
==== Front
J Happiness Stud
J Happiness Stud
Journal of Happiness Studies
1389-4978
1573-7780
Springer Netherlands Dordrecht
603
10.1007/s10902-022-00603-5
Research Paper
The Effect of Consumer Confidence and Subjective Well-being on Consumers’ Spending Behavior
http://orcid.org/0000-0001-5303-0594
Mynaříková Lenka [email protected]
1
Pošta Vít [email protected]
2
1 grid.448072.d 0000 0004 0635 6059 Department of Economics and Management, University of Chemistry and Technology in Prague, Jankovcova 23, 170 00 Prague, Czech Republic
2 grid.266283.b 0000 0001 1956 7785 Department of Managerial Economics, Prague University of Economics and Business, W. Churchill sq. 4, 130 67 Prague, Czech Republic
29 11 2022
125
13 11 2022
© The Author(s), under exclusive licence to Springer Nature B.V. 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The paper focuses on the role of consumer confidence and selected well-being measures in aggregate consumption and in subsets of aggregate consumption on a broad set of 22 OECD countries. Consumer confidence played a positive and statistically significant role in the development of expenditures especially on durable and semi-durable goods and services. The increase in cognitive, affective and eudaimonic measures of well-being, measured by the Cantril ladder, positive and negative affect and freedom to make life choices variables, had negative impact on total consumption and expenditures on semi-durable goods and services. Possible explanations for these estimates are provided in the paper. Based on the purpose of expenditure, consumer confidence was a significant determinant of all expenditures except for unavoidable spending such as food, health, housing, water, energy, and fuel. The subjective well-being indicators showed a negative impact on expenditures on clothing and footwear, recreation and culture, and restaurants and hotels. Possible explanations for the positive and negative effects of subjective well-being measures on consumption, benefits of including the freedom of choice variable, and directions for future research regarding the introduction of understudied variables are discussed.
Keywords
Consumer behavior
Consumer confidence
Locus of control
Freedom of choice
Subjective well-being
Cognitive appraisal theory
==== Body
pmcIntroduction
Psychology and economics can contribute to a better understanding of consumer behavior. While purchasing power (i.e., the ability) depends on real income and wealth, willingness stems from the consumers’ confidence, i.e., subjective perceptions and expectations about national and personal economic conditions (Katona, 1968). Studies have shown that consumers deviate from the rationality assumption proposed by neo-classical economic theory, and do not always attempt to maximize utility, nor do they behave in a predictable way (Gintis, 2000). In reality, individuals make economic decisions under the influence of various psychological, political, social, environmental, and other factors. Since confidence indexes measure perception and expectation, focusing on psychological factors can improve understanding of consumer behavior. However, the research (Dragouni et al., 2016; Gabriele et al., 2009; Kemp et al., 2014; Larson & Shin, 2018; Malgarini & Margani, 2007) suggests that psychological motives may affect some groups of households and some particular categories of expenditure more than others. One of the factors that mediate the effects of consumer confidence in spending behavior can be subjective well-being (SWB) or its subsets such as happiness. SWB is defined as the experience of positive emotions such as happiness and contentment, the development of one’s potential, having some control over one’s life, having a sense of purpose, and experiencing positive relationships (Huppert, 2009). As Jaunky et al. (2020) describe, the impact of consumption on happiness, which may be considered as one aspect of the psychological construct of SWB, has been introduced as a new and imperative domain of concern for economists, psychologists, sociologists, and diverse policy makers all over the world. Although the effect of short-term emotions has been studied, e.g., regarding their effect on less crucial buying decisions (Wood & Rettman, 2007), long-term emotions, such as SWB, have been rather ignored by consumer researchers (Lyubomirsky et al., 2005). Most of the research so far has investigated the effect of consumption behavior on SWB (Carter & Gilovich, 2010; Dunn et al., 2011; Nicolao et al., 2009; Xiao & Kim, 2009; Zhong & Mitchell, 2010). With the exception of Zhong and Mitchell (2012), who showed that SWB influences spending on hedonic products, studies did not examine the impact of well-being on consumers’ consumption.
The standard economic theory based on rational behaviour of households is unsatisfactory given the empirical results. What is needed is an extension of the factors that might affect consumption, and the present paper goes in this direction since it does not stem from the standard economic assumptions. Using psychological concepts described below, the paper looks at how the human psychology drives consumption expenditure as important aspects of economic behavior. The goal of this paper is to extend existing knowledge on consumption behavior in two ways. First, we directly evaluate the role of selected well-being measures in the dynamics of both aggregate consumption and its subsets defined by durability and purpose. Second, we assess the effects on a broad set of countries using dynamic panel techniques. The variables used in the analysis were a cognitive measure of SWB (Cantril ladder), affective measures of SWB (positive and negative emotions), and freedom of choice, which is considered a predictor of happiness, and an eudaimonic measure of SWB (see Inglehart et al., 2008; Veenhoven, 2000; Verme, 2009).
Consumer Confidence and its Measurement
Consumer confidence measures the perceptions of customers about their recent and future financial situation and economic climate (Ou et al., 2014). Views on the possibility of using the Consumer Confidence Index (CCI), a measure of the prevailing consumer sentiment, to analyze private consumption, vary, and the efficacy of the CCI in analyzing and forecasting economic phenomena is not universally accepted. Ludvigson (2004), Cotsomitis and Kwan (2006), or Karasoy Can and Yüncüler (2018) suggest that when the consumer confidence is combined with other variables, it makes only a small improvement for forecasting purposes. Contrary to that, Caleiro (2021) considers the CCI relevant to solving a variety of economic problems. Acuña et al. (2020) conclude that the CCI provides a more accurate prediction of consumption even after controlling for macroeconomic variables. Singal (2012), Dees and Soares Brinca (2013), Dragouni et al. (2016), Lahiri et al. (2016) or Benhabib and Spiegel (2019) show that the CCI can successfully predict consumer spending. Mourougane and Roma (2003), Chua and Tsiaplias (2009), and Utaka (2014) consider the CCI a relevant predictor of the short-term growth in gross domestic product. Lemmon and Portniaguina (2006), Chen (2011), Hsu et al. (2011), and Rojo-Suárez and Alonso-Conde (2020) indicate that changes in the consumer confidence Granger-cause stock returns. Vuchelen (2004), Kwan and Cotsomitis (2006), Mandal and McCollum (2013), Kłopocka (2017), Pan (2018), Karasoy Can and Yüncüler (2018), Claveria (2019), and Acuña et al. (2020) found a negative causality between the consumer confidence and the unemployment rate both in the short run and long run. On the other hand, Demirel and Artan (2017) found no relation between economic confidence and unemployment in 13 EU countries. Mermod and Dudzevičiūtė (2011) detected causality from the consumer confidence to consumption expenditure for developed countries, and causality from economic growth to the consumer confidence for developing countries. Klopocka and Gorska (2021) analyzed data for 14 EU countries and found that the combination of subjective and objective indicators offers a more reliable basis for predictions and policy assessments. Most studies cover one country (mostly the US or countries within the EU region) or a selected group of EU countries, while evidence for emerging economies remains scarce. Exceptions are represented, for instance, by Khumalo (2014) who found a positive effect of the consumer confidence on consumers’ behavior in South Africa. Campelo et al. (2020) and Ghosh (2021) looked at consumer spending in Brazil and found that unemployment changes, stock market fluctuations, interest rate variation, and alterations in the CCI affect consumer spending, making the CCI a good predictor of spending in Brazil. Most studies confirm that the CCI has predictive power on consumption expenditures under certain circumstances (e.g. Çelik & Özerkek, 2009; Dées & Brinca, 2013 ; Gelper, Lemmens & Croux, 2007; Van Aarle & Kappler, 2012). As suggested by Ou et al. (2014) or Hampson et al. (2020), a distinction between the national consumer confidence (i.e., an individual’s perception about the recent and future macroeconomic climate), and personal consumer confidence (an individual’s perception about changes in their recent and future personal financial situation) improves understanding of the index. Vanlaer et al. (2020) and Malovaná et al. (2021) consider the personal consumer confidence more important than the national consumer confidence in determining household savings decisions.
The predictive power of the consumer confidence is more obvious during exceptional events such as political or economic shocks, environmental crisis, etc. (Batchelor & Dua, 1998; Fuhrer, 1993; Garner, 2002; Katona, 1968; Throop, 1992). Malgarini and Margani (2007) suggest the effect of the shocks is higher for inactive people and for dependent workers who may react less rationally. The uncertain economic situation may lower the personal consumer confidence, increase perceived financial vulnerability, and affect future consumer behavior (Brüggen et al., 2017; Hampson et al., 2020; Lowe, 2018), independent of how probable the loss of a job or household income is. Vanlaer et al. (2020) looked at the relationship between the decrease in the consumer confidence after the global financial crisis and the household spending behavior in 18 EU countries and found that the impact of the consumer confidence on saving had increased after the crisis. Teresiene et al. (2021) studied the impact of COVID-19 on the consumer confidence and found a rapid and robust effect of the pandemic in a short period. In the longer period, the spread of the pandemic did not affect the CCI in the EU countries, while negatively affecting the CCI in the US or China (Teresiene et al., 2021). A negative tone of economic news and high media coverage of exceptional events have a strong mediating effect on the consumer confidence and subsequent consumption behavior (e.g. Boomgaarden et al., 2010; Garz, 2018; Hollanders & Vliegenthart, 2011; Kleijnnijenhuis et al., 2015; Sorić et al., 2019; Soroka et al., 2015; Svensson et al., 2017; Vliegenthart & Damstra, 2019). However, changes in consumer behavior can occur for other reasons than turbulent economic or political changes (El Alaoui et al., 2020; Mandal & McCollum, 2013).
Subjective Well-being, Emotions, and Consumers´ Consumption Expenditure
Economic research has confirmed the importance of studying consumers' feelings and emotions when predicting consumer behavior (Ahmad & Rangaraju, 2017; Johnson & Naka, 2014; Nyman & Ormerod, 2014). Regarding the relationship between consumption and SWB or its components, studies (Bertram-Hümmer & Baliki, 2015; Dumludag, 2015; Gokdemir, 2015; Guillen-Royo, 2008; Noll & Weick, 2015; Zhang & Xiong, 2015) focus mainly on levels of consumption. Specific findings depend on what a given study actually measures–e.g., some studies consider happiness a synonym for SWB, others draw a difference between these two. For instance, Schmutte and Ryff (1997) describe psychological well-being as a general feeling of happiness. The Organisation for Economic Cooperation and Development (2013) describes three main types of SWB measures: cognitive measures related to the evaluation of life experiences or life as a whole (e.g., life satisfaction, Cantril ladder); affective measures related to “emotional well-being” (positive and negative affects or affect balance at or during a certain point or period in time); and measures related to the concept of “eudaimonia,” which capture individual happiness or welfare (Deci & Ryan, 2008; Heintzelman, 2018; Huta & Waterman, 2014). Most studies regard life satisfaction as a cognitive self-evaluation of happiness (Diener, 1984; Veenhoven, 1994) or SWB (Andrews & Withey, 1976; Campbell, 1976; Michalos, 1980). Tsurumi et al. (2021) found that total consumption contributes primarily to cognitive and eudaimonic measures of SWB. Duesenberry’s (1949) theory of consumer behavior suggests that various types of consumption enhance happiness and therefore also SWB in at least three ways. Increased consumption of durable goods, food, or housing may enhance happiness by alleviating material hardship or making life easier, serving as a form of coping mechanism against increased stress (Cheng et al., 2016). Conspicuous consumption of visible goods such as expensive vehicles, holidays, clothes, cosmetics, or jewelry may enhance happiness by increasing social status (Chao & Shor, 1998; Dutt, 2006; Johanson-Stenman & Martinsson, 2006; Kaus, 2013; Perez-Truglia, 2013). Finally, spending on leisure or charitable activities may enhance happiness by positively affecting social relationships (Pugno, 2009).
Similar to the consumer confidence, the impact of SWB on consumer behavior may be more profound during exceptional situations that lead to more dramatic behavioral reactions and changes, as we see in cases of panic buying and other herd behavior phenomena during disease outbreaks, national disasters, wars, or terrorist attacks (Leach, 1994; Lins & Aquino, 2020). Terror management theory (Arndt et al., 2004; Kennett-Hensel et al., 2012) explains how exceptional events motivate compensatory behavior to alleviate negative emotions. The compensatory behavior can take the form of purchasing unnecessary products or products of daily need in extensive quantity to regain the sense of control, security, or comfort. The threatening situation can make these purchases look necessary for survival (Arafat et al., 2020; Chua et al., 2021; Dodgson, 2020; Fairfield et al., 2015; Hendrix & Brinkman, 2013). Studies (Bentall et al., 2021; Burroughs & Rindfleisch, 2002) show that fear, anxiety, depressive mood, or elevated stress levels can lead to an active response such as over-purchasing or impulse spending behavior. These responses serve as a self-protective mechanism to manage negative emotions and restore a positive sense of self (Sneath et al., 2009). However, Landau et al. (2011) point out that some individuals may react passively and decrease their spending.
According to cognitive behavioral theories, cognitive evaluations (i.e., appraisals) influence emotions, while, at the same time, emotions influence the cognitive evaluation, and this interaction leads to a behavioral response (Ellsworth, 2013; Moors et al., 2013). The consumer confidence may function as an appraisal tied to specific emotions. Studies (Hampson et al., 2020; Kursan Milaković, 2021; Ng & Russell-Bennett, 2015) show that psychological mechanisms through which the consumer confidence leads to changes in spending behavior have a cognitive and affective dimension, but the affective component has not been sufficiently explored, with few exceptions (e.g., Sekizawa et al., 2021; Van Giesen & Pieters, 2019). Sekizawa et al. (2021) found that the level of the CCI and its fluctuation in Japan are associated with anxiety and positive affects; therefore, when the consumer confidence is higher, people tend to be happier and less anxious. The affect-as-information model suggests that emotions provide information related to one’s current available tendencies and cognitions (Schwarz & Bohner, 1996); therefore, people use emotional information to make judgments that influence their attitudes and behavior (e.g., Gino & Shea, 2012; Gino et al., 2012; Higgins, 2006). Two types of current emotions affect our decisions. We experience integral emotions when we make decisions, but we happen to have incidental emotions, unrelated to the appraisal (Brooks & Schweitzer, 2011; Olekalns & Smith, 2009; Tsay & Bazerman, 2009). Anxiety aroused prior to a decision may lead to perceiving certain behavior as worse, while positive emotions may lead to overvaluing benefits, undervaluing losses, and being more open to risk-taking (e.g., Barry et al., 2004; Friedman et al., 2004; Steinel et al., 2008; Van Kleef et al., 2004). Our decisions are also affected by anticipated emotions that we expect to have post-decision and that influence our risk estimation, intentions, and expectations (see e.g., Bagozzi et al., 2016; Carrera et al., 2011; Kotabe et al., 2019; Riquelme & Alqallaf, 2020; Zampetakis et al., 2016). Dread (i.e., extent of perceived lack of control, feelings of dread, and perceived catastrophic potential) is one of the anticipated emotions with a significant effect on our decisions and behavior (Senik, 2008). Together with the uncertainty of the situation, they create two psychological dimensions of the “risk” (Peters & Slovic, 1996) that influence the cognitive evaluation of risk and determine behavior, as explained by the risk-as-feelings hypothesis (Loewenstein et al., 2001). Perceived risk and uncertainty can strengthen fear and anxiety (Mi et al., 2019; Zheng et al., 2019), which can result in higher pessimism, more pessimistic risk estimates, and make consumers more risk-averse (Kuhnen & Knutson, 2011; Miu et al., 2008; Patt & Weber, 2014; Peng et al., 2014; Smithson, 2008; Stanton et al., 2014). Therefore, uncertainty and perceived risk can increase saving behavior (Bande & Riveiro, 2013; Carroll et al., 2012; Ceritoglu, 2013; Chamon et al., 2013; Mastrogiacomo & Alessie, 2014; Mody et al., 2012). However, this precautionary motive to build up a financial reserve is not universally supported (Fossen & Rostam-Afschar, 2013).
Both the consumer confidence and SWB are closely related to expectations, which can be defined as the assumptions individuals uphold about their future (Augusto-Landa et al., 2011; Conversano et al., 2010; Diener et al., 2003; Eid & Diener, 2004; Mäkikangas & Kinnunen, 2003; Pleeging & Burger, 2020). Optimistic people are generally happier, more resilient to negative economic or political shocks, and have a greater SWB (Arampatzi et al., 2015, 2020; Ekici & Koydemir 2016; Frijter et al., 2012). Optimism represents a psychological capital that serves as a buffer against misfortune (Youssef & Luthans, 2007). This corresponds to the economical view of confidence related to predictability (Malovaná et al., 2021). As explained by Akerlof and Shiller (2010), high confidence can lead to increased optimism about the future, while low confidence leads to higher pessimism. Similarly, a high trait of optimism can lead to higher confidence and thus spending more/saving less, while a high trait of pessimism leads to lower confidence and thus spending less/saving more. A distinction between optimism and pessimism as stable personality traits and as states that are more changeable may be necessary. The self-regulatory model talks about “dispositional optimism “ as a global expectation that good things will be plentiful in the future and bad things sparse and is associated with less distress, more active coping, and lower engagement in avoidance or denial (Scheier et al., 2001). Buchanan and Seligman (1995) describe that pessimists explain away bad events with internal, stable, and global causes, while optimists focus on external, unstable, and specific causes. Both theories suggest that optimism and pessimism involve cognitive, emotional, and motivational components, and thus influence our judgments, decisions, and behaviors. Since optimists can be pessimistic under certain conditions and vice versa, optimism and pessimism probably have both a trait and a state component (Luthans & Youssef, 2007). Although traits are more related to overall well-being, states relate more to specific outcomes such as educational or work-related goals and success (Kluemper et al., 2009; Peterson, 2000). While the trait may be important in explaining consumers´ habits, the state may help explain changes in the consumer confidence. Katona (1968) hypothesized that spending would increase when people became optimistic, and precautionary savings would rise when they became pessimistic. Kahneman and Tversky (1982) describe the forecast error as a tendency to overestimate the likelihood of positive events, and underestimate the likelihood of negative events, which can be explained by psychological biases such as the law of small numbers (Rabin, 2002) or the hindsight bias (Shiller, 2003), which make situations seem more predictable and more probable. Shiller (2003) explains that people make forecasts in uncertain situations by looking for familiar patterns and assuming that future favorable patterns will resemble past ones. Finally, due to the illusion of control, people optimistically distort the reality, believing that their own situation will be consistently better than the general one, which makes them expect a personal success with a probability inappropriately higher than the objective probability warrants (DeBondt & Thaler, 1995). The biases affect both the subjective probability of future economic events and their retrospective interpretation and may create the illusion that we can control the external factors to create an optimistic future. Especially during critical events, people amplify the forecast error and perceive their personal and future conditions better than the aggregate and past ones (Bovi, 2009), so that they can be individual optimists and social pessimists at the same time (Rosner & Nagdy, 2014). In line with Buchanan and Seligman (1995), if we feel we are in control of our lives, we feel more optimistic about our situation (regardless of the objective factors) than about the national situation, which we cannot control directly.
One of the important variables explaining differences in dealing with uncertain situations, making decisions, and coping with emotions is therefore the perceived control we have over our life. The locus of control (Rotter, 1954) reflects individual differences in beliefs about the degree to which we can control the outcomes of events in our life (Galvin et al., 2018). It moderates the effects of external stressors on affective and behavioral responses (Debus et al., 2014; Jiang et al., 2020; Reknes et al., 2019). Similar to optimism, it can serve as a buffer against economic, psychological, political, and other shocks by giving the individual a sense of control and freedom to decide what to do. Individuals with an internal locus of control believe that they have control over the outcomes in their life (Twenge et al., 2004). They have a greater appreciation of freedom of choice, represented by the size of an opportunity set with mutually exclusive alternatives (Verme, 2009). People with an external locus of control believe that things happening in their lives are beyond their control and have no power in affecting them, since they happen due to chance, fate, luck, or are the result of the control by powerful others (Fong et al., 2017). They have a lower appreciation of freedom of choice, since it is regulated by the degree of perceived control, which shapes the expectations we have about the outcome of our choices (Verme, 2009). The external locus of control leads to more problems in dealing with stress and uncertainty (Debus et al., 2014; Reknes et al., 2019). Externals often blame others for their problems and adopt the victim mentality to protect their self against shame, guilt, or regret we may feel when we accept that things went wrong because of our actions (Ng et al., 2006; Twenge et al., 2004). This mentality may lead the externals to be more passive because they do not believe they can actively cope with the situation (Ng et al., 2006). Veenhoven (2000), Inglehart et al. (2008), and Verme (2009) in their analyses of relationship between happiness and other psychological variables used a “perceived fate control” variable represented by a question: “Please use this scale where 1 means ‘none at all’ and 10 means ‘a great deal’ to indicate how much freedom of choice and control you feel you have over the way your life turns out.” The question combines information on both the freedom of choice and the locus of control. Based on their studies, we may consider these variables interrelated. Since the locus of control is not measured internationally, but data on the freedom of choice are available, it may improve our understanding of the psychological variables behind consumer behavior. Hampson et al. (2020) show that the effects of the consumer confidence depend on the locus of control, with the influence of national consumer confidence significantly stronger for consumers with an external locus of control, who are more susceptible to lowered well-being in response to external stressors (Debus et al., 2014). As suggested by Sekizawa et al. (2021), cognitive evaluations of the national economy lead to behavioral changes based on whether an individual feels personally financially affected by the situation and whether the level of affectedness is strong enough to evoke emotional feelings of financial vulnerability. In Hampson et al. (2020), the locus of control served as a moderator of the relationship between the national consumer confidence and perceived financial vulnerability, defined as the probability that an individual will experience financial hardship, i.e., will not be able to maintain the current standard of living (O’Connor et al., 2019). When individuals experience higher perceived financial vulnerability, they become more price-conscious when making new purchases, as this helps to conserve financial resources (Hampson & McGoldrick, 2017). As the financial vulnerability is psychologically very taxing, it can lead to reduced well-being, physical and mental problems, or material deprivation (O’Loughlin et al., 2017). Understanding its role in consumers’ behavior and its relation to the national and personal consumer confidence and psychological variables of overall well-being, locus of control, negative or positive affect is of both theoretical and practical interest (O’Loughlin et al., 2017; Treanor, 2016).
Studies (Demirel & Artan, 2017; Kłopocka, 2017; Matošec & Obuljen Zoričić, 2019; Taylor & McNabb, 2007) agree that macroeconomic variables alone explain only a small proportion of consumer behavior. Therefore, we expect that other factors play an important role, though they may affect different consumers differently, and probably influence especially discretionary, infrequent, and planned purchases, not strictly necessary for life. Their effect may be more visible during exceptional circumstances (Desroches & Gosselin, 2002), as these result in a strong emotional reaction and affect the perception of uncertainty. The psychological concepts described above can be understood through the lenses of the cognitive appraisal theory, which shows how the cognitive evaluation of stressors (for instance, economic recession) interacts with emotions, potentially resulting in a behavioral change (Moschis, 2007). The reaction to a stressor follows an appraisal-emotion-behavior sequence (Folkman & Lazarus, 1984), where individuals evaluate to what extent a stressor potentially affects them. The cognitive appraisal leads to a positive or negative emotional response that affects our expectations about behavioral outcomes, while the locus of control or freedom of choice gives the individual a sense of how the stressor is controllable. This results in a behavioral response, which may include active or passive coping strategies and behavioral adaptations to deal with the stressor and accompanying emotions, such as hedonic shopping, over-purchasing, or saving (Hampson et al., 2020; O’Loughlin et al., 2017; Sekizawa et al., 2021; Treanor, 2016).
Data and Statistical Approach
We used yearly data for a panel of 22 OECD countries. The sample starts in 2008 and ends in 2020. The countries were chosen so that the final number of observations was maximized. The reason for limiting the number of countries was missing data on the confidence indicators. The estimates are therefore based on a total of 286 observations. The selected countries are Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Mexico, Netherlands, Poland, Portugal, South Korea, Spain, Sweden, United Kingdom, and the United States.
To capture the development of consumption, we make use of the real total consumption of households. We also employ consumption in two disaggregated ways. The first approach disaggregates total consumption (CONSUMPTION) into four types of expenditures according to durability: expenditures on durable goods (DURABLE), semi-durable goods (SEMIDURABLE), non-durable goods (NONDURABLE), and expenditures on services (SERVICES). The second approach differentiates total consumption according to purpose, which is embodied in the COICOP methodology (The Classification of Individual Consumption according to Purpose), into 12 divisions: food and non-alcoholic beverages (COICOP1), alcoholic beverages, tobacco (COICOP2), Clothing, and footwear (COICOP3), Housing, water, energy, fuel (COICOP4), Furnishings, household equipment and maintenance (COICOP5), Health (COICOP6), Transport (COICOP7), Post and telecommunication (COICOP8), Recreation and culture (COICOP9), Education (COICOP10), Restaurants and hotels (COICOP11) and Miscellaneous goods and services (COICOP12). For the purposes explained below, we also employ the gross domestic product (GDP) in the analysis. The notation in brackets is the designation of the series, which will be used further below, especially in the tables.
The sentiments of consumers are captured by four indicators: consumer confidence indicator published by the OECD (CONFIDENCE), and Life ladder (LADDER), Freedom to make life choices (FREEDOM), positive affect (POSITIVE), and negative affect (NEGATIVE) indicators derived from the World Happiness Report (Helliwell et al., 2021). The Life ladder score, which represents the cognitive measure of SWB, is measured by answering a question: “Please imagine a ladder, with steps numbered from 0 at the bottom to 10 at the top. The top of the ladder represents the best possible life for you and the bottom of the ladder represents the worst possible life for you. On which step of the ladder would you say you personally feel you stand at this time?” Freedom to make life choices, as a eudaimonic measure of SWB, is the national average of responses to the question “Are you satisfied or dissatisfied with your freedom to choose what you do with your life?” Affective measures of SWB are measured by positive and negative affect. The positive affect represents the average of three positive affects (happiness, laughter, and enjoyment) measured through a question: “Did you experience the following feelings during A LOT OF THE DAY yesterday? The negative affect is defined as the average of three negative affects (worry, sadness, and anger) measured by responses to the question: “Did you experience the following feelings during A LOT OF THE DAY yesterday?” Although the last two variables are called “affects”, they correspond to what we generally consider “emotions”. For a full explanation of the methodology of the specific measurements, see the Statistical Appendix 1 of the World Happiness Report (Helliwell et al., 2021).
All consumption variables and the gross domestic product series are measured using constant prices of the national base year. The national base year differs slightly among the countries; usually, it is 2015. The CCI published by the OECD is expressed as an index with a long-term average set to 100. The published series are first transformed by natural logarithm and then detrended. The SWB of consumers captured by the variables of freedom to make life choices, positive affect, and negative affect are measured as the average response for each person, with values ranging from 0 to 1, while the life ladder is measured on a scale from 0 to 10.
The initial inspections of the data showed that consumption data was all nonstationary, which is to be expected, while the sentiment indicators were all stationary, which in the case of the CCI is by definition since it is published as detrended.
To express all the variables in a common measure, the consumption series were transformed into growth rates expressed as decimal numbers. The CCI was divided by 100 and the life ladder indicator was divided by 10. This means that all the SWB indicators assume values from the interval 0 to 1. Table 1 summarizes some statistical properties of the adjusted data.Table 1 Statistical properties of the data
Series/Statistics Mean Standard deviation Normality Levin, Lin & Chu t-statistic Fisher-ADF Chi-square
CONSUMPTION 0.00 0.04 509.51*** −7.06*** 100.43***
DURABLE 0.03 0.10 197.57*** −12.49*** 150.50***
SEMIDURABLE 0.01 0.06 233.96*** −2.99*** 80.56***
NONDURABLE 0.00 0.03 611.31*** −12.57*** 136.91***
SERVICES 0.00 0.05 884.46*** −7.45*** 100.79***
COICOP1 0.01 0.03 713.49*** −7.92*** 104.80***
COICOP2 −0.01 0.04 77.18*** −7.94*** 122.60***
COICOP3 0.00 0.07 286.48*** −6.02*** 102.78***
COICOP4 0.01 0.02 94.55*** −15.52*** 190.05***
COICOP5 0.01 0.06 229.52*** −8.43*** 108.97***
COICOP6 0.02 0.06 285.12*** −7.04*** 110.87***
COICOP7 −0.00 0.08 223.28*** −8.68*** 135.96***
COICOP8 0.03 0.06 219.25*** −12.60*** 159.23***
COICOP9 0.01 0.07 528.88*** −7.43*** 114.21***
COICOP10 0.00 0.06 442.84*** −6.99*** 118.68***
COICOP11 −0.01 0.11 755.24*** −7.29*** 101.18***
COICOP12 0.01 0.05 867.69*** −8.83*** 116.70***
CONFIDENCE 0.99 0.02 311.65*** −5.09*** 82.92***
LADDER 0.65 0.08 15.61*** −2.93*** 71.06***
FREEDOM 0.79 0.14 39.97*** −6.33*** 111.51***
POSITIVE 0.74 0.08 20.15*** −1.92** 58.04*
NEGATIVE 0.24 0.06 57.42*** −5.36*** 77.44***
GDP 0.01 0.04 639.60*** −12.86*** 124.65***
COMPONENT 0.00 1.58 23.12*** −3.31*** 83.14***
Normality is tested by the Jarque–Bera test under the null hypothesis of normal distribution, the unit root is tested by Levin-Lin-Chu, and Fisher-ADF tests under the null hypothesis of unit root *, **, *** signifies rejection of the null hypothesis at 10%, 5%, 1% level of significance, respectively.
We test for the presence of unit roots by utilizing two tests: the Levin-Lin-Chu unit root test and the Fisher-ADF test. Suppose the following representation of the panel:1 Δyi,t=αyi,t-1+∑j=1niβi,jΔyi,t-j+δxi,t+εi,t
where y represents the endogenous variable, x stands for exogenous variables, ε represents errors, i denotes cross-sections, j = 1 … ni signifies possible different lag orders for the cross-sections. The null hypothesis is related to the coefficient α. More specifically, the null hypothesis is that α is equal to zero. If the null hypothesis is rejected, then the series is considered stationary. The Levin-Lin-Chu test assumes a common unit root process across the cross-sections. On the other hand, the Fisher-ADF unit root test allows for individual unit root processes across cross-sections. The number of lags in Eq. (1) was selected by the Schwarz information criterion, which in most cases meant one lag. The vector x includes only the intercept, no trends were assumed. Since the property of stationarity is crucial for the subsequent analysis, we employ these two different tests to check for the presence of unit roots. As the results in Table 1 show, all the series may be considered stationary after the adjustments mentioned above.
The variables that capture the SWB of consumers should be suspected to exhibit significant correlations among them. Table 2 presents the correlation matrix between the five variables.Table 2 Correlation matrix
CONFIDENCE LADDER FREEDOM POSITIVE NEGATIVE
CONFIDENCE 1.00
LADDER 0.26** 1.00
FREEDOM 0.30* 0.78*** 1.00
POSITIVE 0.15*** 0.73*** 0.74*** 1.00
NEGATIVE −0.30** −0.49** −0.48** −0.38*** 1.00
*, **, *** means rejection of the null of no correlation at 10%, 5%, and 1% level of significance, respectively
The correlation matrix in Table 2 shows that the correlations between the CCI (CONFIDENCE) and the SWB indicators are statistically significant, but they are relatively small with respect to their magnitude. Therefore, the CCI should be considered as a separate variable. On the other hand, the correlations between the variables of life ladder (LADDER), freedom to make life choices (FREEDOM), and positive affect (POSITIVE) are quite considerable. The correlations between any of these three variables and the negative affect variable (NEGATIVE) are then again significantly smaller. Therefore, we consider the CCI and the variable of negative effect separately and use principal components analysis for the other three variables to extract a principal component that would capture most of their original variability.
The principal components analysis starts with the variance matrix of the original variables, here life ladder, freedom to make life choices, and positive affect. The principal components are obtained by eigenvalue decomposition of the original variance matrix. The principal components analysis was performed using the Pearson correlation matrix. Table 3 shows the eigenvalues associated with the decomposition of the original correlation matrix.Table 3 Principal components analysis based on the following three variables: life ladder, freedom to make life choices and positive affect
Eigenvalue Value Cumulative value Cumulative proportion
1 2.50 2.50 0.83
2 0.28 2.78 0.93
3 0.22 3.00 1.00
Since there are three variables, there are three eigenvalues. The sum of the three eigenvalues must be three because the analysis is based on a correlation matrix. The last column shows that the first principal component accounts for 83% of the total variance. There are no strict rules as to what the threshold should be. In practice, threshold of 80% or 85% are common. We conclude that it is sufficient to use the first component only to capture the vast majority of the variance produced by the three variables. This variable is named COMPONENT and its properties are shown in Table 1.
The objective of the empirical analysis is to verify the impacts of the three SWB indicators on consumption. The empirical analysis faces the wide-spread problem of endogeneity. Endogeneity arises from dependence of the variable assumed as independent on the variable designated as dependent, omitted variables and measurement errors. In a vast majority of economic modelling all these problems should be expected to be present. The problem of omitted variables stems from the fact that the true economic model is never known in practice. Mutual relationships between the confidence and SWB indicators on one hand and consumption on the other hand cannot be ruled. The frequent approach employed to tackle this issue is to make use of instrumental variables, which means variables correlated with independent variables but uncorrelated with the error term. We apply the estimation based on instrumental variables in the form of the generalized method of moments.
The general structure of the empirical model we estimate is:2 yi,t=ci+δxi,t+εi,t
where c represents constants, y contains a measure of consumption, x contains the three SWB measures. In Table 1, we showed that the variables are already stationary; therefore, we do not apply the typical transformation in the form of differencing Eq. (2) as it would lead to over-differencing.
The possible cross-section fixed effects are accounted for by using orthogonal deviations. Arellano-Bond 2-step estimation is used. To compute coefficient covariances, White period method is used to account for within cross-section heteroskedasticity and serial correlation. This means that the computed coefficient covariances are robust to within cross-section heteroskedasticity and serial correlation.
Since three coefficients are estimated, it is necessary to employ at least three instruments. Lagged regressors by one period are used as instruments, and contemporary real GDP growth is used as the fourth instrument. Real GDP growths are correlated with the regressors while the correlation with the error of (2) is small. The validity of instruments is tested by Sargan-Hansen J-test, which tests the null hypothesis that the over-identifying restrictions are valid.
Results and Discussion
The results presented in Table 4 indicate that the CCI always plays a positive and statistically significant role in the development of consumption. This means that an increase in the CCI increases the growth of total consumption and expenditures on durable and semidurable goods and services, which is similar to findings by Gelper et al. (2007) and Adrangi and Macri (2011) for durable goods or Malgarini and Margani (2007) for services. However, both Garner (2002) and Malgarini and Margani (2007) show that the CCI is not statistically significant in the case of durable goods. The positive estimated coefficient associated with CCI is the lowest in the case of expenditures on nondurable goods, which can be explained by the fact that a relatively high proportion of these expenditures is directed on goods whose consumption is necessary. Nondurable goods mainly consist of food and non-alcoholic beverages and alcoholic beverages and tobacco. We may expect that the consumer confidence influences spending on nondurable goods especially during exceptional events, when consumer priorities are centered on basic needs (Di Crosta et al., 2021). The uncertainty of the external situation may lead to panic buying and over purchasing of nondurables to restore the sense of security by creating a supply of food and beverages to be prepared for the possible bad times. The effect of CCI on consumption spending is estimated to be the strongest in the case of consumption of durable goods. We plot this relationship in Fig. 1.Table 4 Estimates for total consumption, consumption according to durability, and services
CONSUMPTION DURABLE SEMIDURABLE NONDURABLE SERVICES
CONFIDENCE 1.60*** 3.50*** 2.02*** 0.84*** 1.51***
COMPONENT −0.04*** −0.05 −0.07*** −0.00 −0.006***
NEGATIVE −0.76** 0.05 −1.34** −0.06 −0.120**
J-statistic 1.02 1.11 0.94 1.50 1.51
Instruments consist of lagged regressors and contemporary real GDP growth. J-statistic refers to the test of the validity of over-identifying restrictions with the null of the restrictions being valid. *, **, *** means rejection of the null at 10%, 5%, and 1% level of significance, respectively
Fig. 1 Relationship between consumption of durable goods and consumer confidence
The SWB indicators have statistically significant effects in the case of total consumption and expenditures on semidurable goods and services. The component variable represents the joint impact of the variables of the life ladder, the freedom to make life choices, and the positive affect. The impacts of the component variable as well as that of negative affect are negative. While the positive impact of the CCI on the total consumption was expected, the previous theoretical discussion showed that the impacts of the other variables might be positive or negative. Our results show that within the total consumption they tend to have an impact on those expenditures, which are more easily postponed. Expenditures on semidurable goods include especially expenditures on clothing and footwear.
It is difficult to compare the magnitudes of the impact in our study with other studies since there are no comparable studies. When interpreting the results only from the point of view of this study it should be noted that the median values of consumer confidence, component, and negative affect are 0.998, 0.483, and 0.229, respectively. Given the values of the estimated coefficients it may be, therefore, concluded that given these three variables the impact of consumer confidence is on average the highest. Although the estimated coefficients of the component and negative affect are lower than that of consumer confidence, given the median values mentioned above, the effects of both the component and the negative affect on consumption dynamics are non-negligible.
We may interpret the results in line with Larson and Shin (2018) or Zwanka and Buff (2020) who suggest that buying non-necessities may also include pursuing freedom or with Kemp et al. (2014) who suggest that buying non-necessities serves as a coping mechanism to deal with negative emotions. Therefore, in the case of the component variable, a higher sense of freedom, higher cognitive well-being, and positive affect (therefore, higher overall SWB) may prevent consumers from spending. As stated by Guven (2009), happy people tend to save more and spend less with respect to their future decisions and expectations. Ozari (2007) and Zhong and Mitchell (2012) suggest that high well-being consumers prefer lower-cost, everyday pleasures, such as going swimming or engaging in their favorite hobby, to expensive indulgences. Other studies (Carter & Gilovich, 2010; Dunn et al., 2011; Nicolao et al., 2009; Xiao & Kim, 2009; Zhong & Mitchell, 2010) also suggest the opposite direction, i.e., that consumption of certain goods leads to increased happiness or SWB. We may expect that respondents who evaluate their SWB positively (in this case, all three aspects of the SWB) do not feel the need to use consumption as a mechanism to enhance their happiness through any of the described channels such as coping with distress, improving social status, or social relationships. In case of the negative emotions, they can make consumers more passive, negatively affect the decision-making and lead to more pessimistic estimates of the outcomes of their choices, which may be important in purchasing semidurable goods or services, as suggested by Miu et al. (2008) or Peng et al. (2014). Landau et al. (2011) and Degli Esposti et al. (2021) point out that individuals may react passively and decrease spending on non-necessities such as nonbasic clothes when feeling negative emotions.
Table 5 presents the estimated impacts of the SWB indicators on consumption classified according to purpose. The CCI is not a statistically significant determinant of consumption in the case of expenditures on housing, water, energy and fuel, which is an expected result since these expenditures are practically unavoidable. Its impact on expenditures on health is also rather limited for the very same reason.Table 5 Estimates for consumption according to purpose
COICOP1 (food, beverages) COICOP2 (alcohol, tobacco) COICOP3 (clothing) COICOP4 (housing) COICOP5 (furnishings) COICOP6 (health)
CONFIDENCE 0.67*** 0.80*** 2.30*** 0.06 1.67*** 0.78*
COMPONENT 0.01 −0.02 −0.10*** 0.01 −0.01 −0.02
NEGATIVE 0.20 −0.23 −1.87** −0.07 −0.08 −0.67
J-statistic 1.27 0.00 0.96 0.37 1.44 1.00
COICOP7 (transport) COICOP8 (telecommunication) COICOP9 (recreation) COICOP10 (education) COICOP11 (restaurants) COICOP12 (other)
CONFIDENCE 3.34*** 0.89*** 2.49*** 1.52*** 3.50*** 1.50***
COMPONENT −0.08** 0.01 −0.10*** −0.03* −0.16*** −0.04
NEGATIVE −0.52 0.27 −2.14*** −0.52 −2.37** −0.73
J-statistic 1.07 0.07 0.98 2.54 1.02 0.99
Food and non-alcoholic beverages (COICOP1), alcoholic beverages, tobacco (COICOP2), Clothing, and footwear (COICOP3), Housing, water, energy, fuel (COICOP4), Furnishings, households equipment and maintenance (COICOP5), Health (COICOP6), Transport (COICOP7), Post and telecommunication (COICOP8), Recreation and culture (COICOP9), Education (COICOP10), Restaurants and hotels (COICOP11) and Miscellaneous goods and services (COICOP12). Instruments consist of lagged regressors and contemporary real GDP growth. J-statistic refers to the test of the validity of over-identifying restrictions with the null of the restrictions being valid. *, **, *** means rejection of the null at 10%, 5%, and 1% level of significance, respectively
The statistically significant and negative estimated coefficients associated with SWB indicators are found in the cases of expenditures on clothing and footwear, recreation and culture, and restaurants and hotels. These results confirm the estimates presented in Table 4 where it was shown that these indicators have a negative impact on expenditures on semidurable goods and services. Furthermore, the variables of life ladder, the freedom to make life choices, and the positive affect (COMPONENT) jointly negatively influence the expenditures on transport, which may be just a reflection of their negative impact on the previously mentioned expenditures since the use of various means of transport often accompanies the consumption of services in the divisions of restaurants and hotels, and recreation and culture. These three variables also have a slight negative impact on the growth of expenditures on education. Individuals who are currently feeling high level of well-being, experiencing happiness and control over their life may not be feeling the pressure to invest in their education to improve their future. In line with the forecasting error (Bovi, 2009; Kahneman & Tversky, 1982), but also due to the current satisfactory state, they may expect their future to be similarly satisfying as their current life.
Conclusions
As Veenhoven et al. (2021) state, there is a growing demand for information about how our choices will affect our happiness. However, equally important is understanding how our level of happiness or overall well-being affects our choices in various aspects of our life, consumers´ behavior included. Our study contributes to the area of research on the relationship between various aspects of well-being and consumption by adding the variable of consumer confidence, which is considered an important psychological factor affecting consumers´ spending. Indeed, our results suggest that the consumer confidence is a useful concept in predicting the consumer behavior for durable and semidurable goods and services in selected OECD countries, i.e., not only for countries within the EU or for the US. Since the consumer confidence is based on our expectations about the national and personal future economic situation, it plays a more important role in deciding whether to buy durable goods or spend money on clothes or services, i.e., on goods that require more planning or that may be associated with hedonic shopping, than on other types of goods. Individuals may purchase them more when evaluating their economic situation as good and expecting it will stay the same (or get better), so that they can either afford or deserve the purchase. On the other hand, the nondurables are bought frequently and spending on them may be a result of a habit more than of the level of the consumer confidence. The effect of the consumer confidence on purchasing nondurables may increase during exceptional events such as political, economical, or environmental crisis, when individuals may prepare for worse times by making stocks of food, and may be affected by the negative picture created by media resulting in panic buying and overpurchasing. The results in several countries suggest that the consumer confidence correlates positively with happiness and negatively with worry or anxiety (Barazzetta, 2015; Ekici & Koydemir, 2016; Frijters et al., 2012).
To analyze the effect of SWB on consumption, we looked at three aspects of SWB as suggested by the OECD (2013). The cognitive measure represented by the Cantril ladder, the affective measure represented by the positive and negative affect, and the eudaiomonic measure represented by the freedom of choice show a negative effect on total consumption and expenditures on clothing and footwear, recreation and culture, and restaurants and hotels. Our results suggest that the freedom of choice may have a moderating effect on spending behavior similarly to the locus of control concept, helping consumers cope with positive or negative affect in ways that do not result in an increase in spending behavior. Since the locus of control is not measured repeatedly and on an international basis, the use of freedom of choice as an alternative eudaimonic measure of SWB, as suggested also by Verme (2009), may be a helpful addition for studies on consumer behavior. Finally, our study can contribute to the understanding of how SWB influences consumption, contrary to most of research that studied the opposite direction of relationship, i.e. how consumption affects SWB.
In Sekizawa et al.´s (2021) study, the variables we analyzed, except for the freedom of choice that was not included, were associated with higher levels of the consumer confidence. These results were not replicated in our study. One possible explanation, for instance, may be that the negative affect variable retrieved from the World Happiness Report includes both inhibiting and disinhibiting negative emotions, which may have opposing effects on consumer behavior. Also, we combined two data sets, one from the World Happiness Report and one from the OECD, however, the combination allowed us to compare larger data sets and variables that would be difficult to measure to such an extent for one specific study. For future research, including more types of emotions and dividing them based on their impact on behavior may provide additional information, as well as adding understudied variables such as hope or a combination of the optimism and pessimism trait and state measurement.
Our study used data from countries based on the availability of complete data sets. As research in countries outside the EU region or the US is sparse, future research should also focus more on developing countries and countries outside the mentioned geographical areas. In our study, we did not focus on the effect of possible exceptional events on changes in consumer confidence, although, as explained in the theoretical part of the paper, consumer confidence may serve as an important predictor of consumer behavior especially during exceptional and turbulent situations. Therefore, another possible direction of future research, together with the focus on the understudied variables, may include a more extensive analysis of the impact of national and international exceptional events.
Author Contributions
Both authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Vít Pošta. The first draft of the manuscript was prepared and written by Lenka Mynaříková. Both authors commented on previous versions of the manuscript and read and approved the final manuscript.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declarations
Conflict of interest
The authors have no competing interests to declare that are relevant to the content of this article.
Informed consent
This paper does not contain any studies involving human participants performed by any of the authors.
Data and Materials
Data for the analysis were retrieved from public databases. Data on the consumer confidence and consumption can be found in the OECD database here: https://stats.oecd.org/, data on the tested psychological variables (named Data for Fig. 2.1) can be found in the World Happiness Report 2021 available here: https://worldhappiness.report/ed/2021/#appendices-and-data. We used data on four variables from the World Happiness Report 2021. The transcript of the relevant survey questions is here https://happiness-report.s3.amazonaws.com/2021/Appendix1WHR2021C2.pdf.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Acuña G Echeverría C Pinto-Gutiérrez C Consumer confidence and consumption: Empirical evidence from Chile International Review of Applied Economics 2020 34 75 93 10.1080/02692171.2019.1645816
Adrangi, B., & Macri, J. (2011). Consumer Confidence and Aggregate Consumption Expenditures in the United States, Review of Economics & Finance, Better Advances Press, Canada, 1, 1–18.
Ahmad N Rangaraju SK Impact of consumer confidence during good times and bad times 2017 Social Science Electronic Publishing
Akerlof GA Shiller RJ Animal spirits: How human psychology drives the economy, and why it matters for global capitalism 2010 Princeton University Press
Andrews FM Withey SB Social indicators of well-being: America’s perception of life quality 1976 Plenum Press
Arafat SMY Kar SK Marthoenis M Sharma P Hoque Apu E Kabir R Psychological underpinning of panic buying during pandemic (COVID-19) Psychiatry Research 2020 289 e113061 10.1016/j.psychres.2020.113061
Arampatzi, E., Burger, M., Stavropoulos, S., & Tay, L. (2020). The role of positive expectations for resilience to adverse events: Subjective well-being before, during and after the Greek bailout referendum. Journal of Happiness Studies, 21, 965–995. 10.1007/s10902-019-00115-9
Arampatzi E Burger MJ Veenhoven R Financial distress and happiness of employees in times of economic crisis Applied Economic Letters 2015 22 3 173 179 10.1080/13504851.2014.931916
Arndt J Solomon S Kasser T Sheldon KM The urge to splurge: A terror management account of materialism and consumer behavior Journal of Consumer Psychology 2004 14 198 212 10.1207/s15327663jcp1403_2
Augusto-Landa JM Pulido-Martos M Lopez-Zafra E Does perceived emotional intelligence and optimism/pessimism predict psychological well-being? Journal of Happiness Studies 2011 12 463 474 10.1007/s10902-010-9209-7
Bagozzi RP Belanche D Casaló LV Flavián C The role of anticipated emotions in purchase intentions Psychology of Marketing 2016 33 629 645 10.1002/mar.20905
Bande R Riveiro D Private saving rates and macroeconomic uncertainty: Evidence from Spanish regional data Economic and Social Review 2013 44 3 323 349
Barazzetta, M. (2015). The asymmetric effect of expectations on subjective well-being. Working Papers 374, ECINEQ, Society for the Study of Economic Inequality. Available at http://www.ecineq.org/milano/WP/ECINEQ2015-374.pdf
Barry B Fulmer IS Van Kleef GA Gelfand MJ Brett JM I laughed, I cried, I settled: The role of emotion in negotiation The handbook of negotiation and culture 2004 Stanford University Press 71 94
Batchelor R Dua P Improving macro-economic forecasts: The role of consumer confidence International Journal of Forecasting 1998 14 1 71 81 10.1016/S0169-2070(97)00052-6
Benhabib J Spiegel MM Sentiments and economic activity: Evidence from US states The Economic Journal 2019 129 618 715 733 10.1111/ecoj.12605
Bentall RP Lloyd A Bennett K McKay R Mason L Murphy J McBride O Hartman TK Gibson-Miller J Levita L Martinez AP Stocks T Butter S Vallières F Hyland P Karatzias T Shevlin M Pandemic buying: Testing a psychological model of over-purchasing and panic buying using data from the United Kingdom and the Republic of Ireland during the early phase of the COVID-19 pandemic PLoS ONE 2021 16 1 e0246339 10.1371/journal.pone.0246339 33503049
Bertram-Hümmer V Baliki G The role of visible wealth for deprivation Social Indicators Research 2015 124 765 783 10.1007/s11205-014-0824-2
Boomgaarden HG Vliegenthart R De Vreese CH Schuck ART News on the move: Exogenous events and news coverage of the European Union Journal of European Public Policy 2010 17 4 506 526 10.1080/13501761003673294
Bovi M Economic versus psychological forecasting. Evidence from consumer confidence surveys Journal of Economic Psychology 2009 30 4 563 574 10.1016/j.joep.2009.04.001
Brooks AW Schweitzer M Can nervous nelly negotiate? How anxiety causes negotiators to make low first offers, exit early, and earn less profit Organizational Behavior and Human Decision Processes 2011 115 1 43 54 10.1016/j.obhdp.2011.01.008
Brüggen E Hogreve J Holmlund M Kabadayi S Löfgren M Financial well-being: A conceptualization and research agenda Journal of Business Research 2017 79 228 237 10.1016/j.jbusres.2017.03.013
Buchanan G Seligman MEP Explanatory style 1995 Lawrence Erlbaum Associates
Burroughs JE Rindfleisch A Materialism and well-being: A conflicting values perspective Journal of Consumer Research 2002 29 348 370 10.1086/344429
Caleiro AB Learning to classify the consumer confidence indicator (in Portugal) Economies 2021 9 3 125 10.3390/economies9030125
Campbell AC Subjective measures of well-being American Psychologist 1976 31 117 124 10.1037/0003-066X.31.2.117 1267244
Campelo A Seda Bittencourt V Malgarini M Consumers confidence and households consumption in Brazil: Evidence from the FGV survey Journal of Business Cycle Research 2020 16 1 19 34 10.1007/s41549-020-00042-2
Carrera P Caballero A Muñoz D Oceja L Anticipated emotions and personal experience for predicting behavioral intentions and behavioral expectations The Spanish Journal of Psychology 2011 14 2 535 547 10.5209/rev_SJOP.2011.v14.n2.2 22059300
Carroll, C. D., Slacalek, J., & Sommer, M. (2012). Dissecting saving dynamics: measuring wealth, precautionary and credit effects. ECB working paper series, 1474, 1–50.
Carter, T. J., & Gilovich, T. (2010). The relative relativity of material and experiential purchases. Journal of personality and social psychology, 98(1), 146–159. 10.1037/a0017145
Carter TJ Gilovich T I am what I do, not what I have: The differential centrality of experiential and material purchases to the self Journal of Personality and Social Psychology 2012 102 6 1304 1317 10.1037/a0027407 22369046
Çelik S Özerkek Y Panel cointegration analysis of consumer confidence and personal consumption in the European Union Journal of Business Economics and Management 2009 10 2 161 168 10.3846/1611-1699.2009.10.161-168
Ceritoglu E The impact of labour income risk on household saving decisions in Turkey Review of Economics of the Household 2013 11 1 109 129 10.1007/s11150-011-9137-2
Chamon M Liu K Prasad E Income uncertainty and household savings in China Journal of Development Economics 2013 105 C 164 177 10.1016/j.jdeveco.2013.07.014
Chao A Schor JB Empirical tests of status consumption: Evidence from women’s cosmetics Journal of Economic Psychology 1998 19 107 131 10.1016/S0167-4870(97)00038-X
Chen S-S Lack of consumer confidence and stock returns Journal of Empirical Finance 2011 18 2 225 236 10.1016/j.jempfin.2010.12.004
Cheng Z King S Smyth R Wang H Housing property rights and subjective wellbeing in Urban China European Journal of Political Economy 2016 45 Suppl 160 174 10.1016/j.ejpoleco.2016.08.002
Chua CL Tsiaplias S Can consumer sentiment and its components forecast Australian GDP and consumption? Journal of Forecasting 2009 28 698 711 10.1002/for.1120
Chua G Yuen KF Wang X Wong YD The determinants of panic buying during COVID-19 International Journal of Environmental Research and Public Health 2021 18 3247 10.3390/ijerph18063247 33801077
Claveria O Forecasting the unemployment rate using the degree of agreement in consumer unemployment expectations Journal for Labour Market Research 2019 53 1 10 10.1186/s12651-019-0253-4
Conversano C Rotondo A Lensi E Della Vista O Arpone F Reda MA Optimism and its impact on mental and physical well-being Clinical Practice and Epidemiology in Mental Health : CP & EMH 2010 6 25 29 10.2174/1745017901006010025 20592964
Cotsomitis JA Kwan ACC Can consumer confidence forecast household spending? Evidence from the European commission business and consumer surveys Southern Economic Journal 2006 72 3 597 610
De Grauwe P Booms and busts in economic activity: A behavioral explanation Journal of Economic Behavior & Organization 2012 83 3 481 501 10.1016/j.jebo.2012.02.013
DeBondt WFM Thaler RH Jarrow R Maksimovic V Ziemba WT Financial decision making in markets and firms: A behavioral perspective Finance 1995 Handbooks in operations research and management science 385 410
Debus ME Konig CJ Kleinmann M The building blocks of job insecurity: The impact of environmental and person-related variables on job insecurity perceptions Journal of Occupational and Organizational Psychology 2014 87 2 329 351 10.1111/joop.12049
Deci EL Ryan RM Hedonia, eudaimonia, and well-being: An introduction Journal of Happiness Studies 2008 9 1 1 11 10.1007/s10902-006-9018-1
Dees S Brinca P Consumer confidence as a predictor of consumption spending: Evidence for the United States and the Euro area International Economics 2013 134 1 14 10.1016/j.inteco.2013.05.001
Degli Esposti P Mortara A Roberti G Sharing and sustainable consumption in the era of COVID-19 Sustainability 2021 13 1903 10.3390/su13041903
Demirel SK Artan S The causality relationships between economic confidence and fundamental macroeconomic indicators: Empirical evidence from selected European Union countries International Journal of Economics and Financial Issues 2017 7 417 424
Desroches, B., & Gosselin, M.-A. (2002). The Usefulness of Consumer Confidence Indexes in the United States. Staff Working Papers, Bank of Canada.
Di Crosta A Ceccato I Marchetti D La Malva P Maiella R Cannito L Psychological factors and consumer behavior during the COVID-19 pandemic PLoS ONE 2021 16 8 0256095 10.1371/journal.pone.0256095
Di Domenico A Palumbo R Fairfield B Mammarella N Fighting apathy in Alzheimer’s dementia: A brief emotional-based intervention Psychiatry Research 2016 242 331 335 10.1016/j.psychres.2016.06.009 27336799
Diener E Subjective well-being Psychological Bulletin 1984 75 542 575 10.1037/0033-2909.95.3.542
Diener E Oishi S Lucas RE Personality, culture, and subjective well-being: Emotional and cognitive evaluations of life Annual Review of Psychology 2003 54 403 425 10.1146/annurev.psych.54.101601.145056
Dodgson, L. (2020). A human behavior expert explains 4 psychological reasons why people are panic buying items in bulk during the coronavirus pandemic. Retrieved from: https://www.insider.com/why-people-are-panic-bulk-buying-during-the-coronavirus-pandemic-2020-3 . Accessed on 14/12/2021.
Dragouni M Filis G Gavriilidis K Santamaria D Sentiment, mood and outbound tourism demand Annals of Tourism Research 2016 60 80 96 10.1016/j.annals.2016.06.004
Duesenberry J Income, savings, and the theory of human behavior 1949 Harvard University Press
Dumludag D Consumption and life satisfaction at different levels of economic development International Economic Review 2015 62 163 182 10.1007/s12232-015-0226-z
Dunn EW Gilbert DT Wilson TD If money doesn't make you happy, then you probably aren't spending it right Journal of Consumer Psychology 2011 21 2 115 125 10.1016/j.jcps.2011.02.002
Dutt, A. K. (2006). Consumption and happiness: Alternative approaches, rough draft prepared for the conference on new directions in the study of happiness. Notre Dame: University of Notre Dame.
Eid M Diener E Global judgments of subjective well-being: Situational variability and long-term stability Social Indicators Research 2004 65 3 245 277 10.1023/B:SOCI.0000003801.89195.bc
Ekici T Koydemir S Income expectations and happiness: Evidence from British panel data Applied Research in Quality of Life 2016 11 2 539 552 10.1007/s11482-014-9380-9
El Alaoui M Bouri E Azoury N The determinants of the US consumer sentiment: Linear and nonlinear models International Journal of Financial Studies 2020 8 38 10.3390/ijfs8030038
Ellsworth PC Appraisal theory: Old and new questions Emotion Review 2013 5 2 125 131 10.1177/1754073912463617
Fairfield B Mammarella N Palumbo R Di Domenico A Emotional meta-memories: A review Brain Sciences 2015 5 509 520 10.3390/brainsci5040509 26569320
Folkman S Lazarus RS Stress, appraisal, and coping 1984 Springer
Fong LHN Lam LW Law R How locus of control shapes intention to reuse mobile apps for making hotel reservations: Evidence from Chinese consumers Tourism Management 2017 61 331 342 10.1016/j.tourman.2017.03.002
Fossen FM Rostam-Afschar D Precautionary and entrepreneurial savings: New evidence from German households Oxford Bulletin of Economics and Statistics 2013 75 4 528 555 10.1111/j.1468-0084.2012.00702.x
Friedman R Anderson C Brett J Olekalns M Goates N Lisco CC The positive and negative effects of anger on dispute resolution: Evidence from electronically mediated disputes Journal of Applied Psychology 2004 89 2 369 376 10.1037/0021-9010.89.2.369 15065982
Frijters P Liu AYC Meng X Are optimistic expectations keeping the Chinese happy? Journal of Economic Behavior and Organization 2012 81 159 171 10.1016/j.jebo.2011.09.007
Fuhrer, J. (1993). What role does consumer sentiment play in the U.S. macroeconomy? New England Economic Review, 32–44. Available at: http://www.bostonfed.org/economic/neer/neer1993/neer193b.pdf
Gabriele S Martelli BM Raitano M The consumers' mood coping with the crisis. Evidences from isae consumer survey Journal of Applied Economic Sciences 2009 4 372 386
Galvin BM Randel AE Collins BJ Johnson RE Changing the focus of locus (of control): A targeted review of the locus of control literature and agenda for future research Journal of Organizational Behavior 2018 39 7 820 833 10.1002/job.2275
Garner CA Consumer confidence after September 11 Federal Reserve Bank of Kansas City Economic Review 2002 87 5 26
Garz M Effects of unemployment news on economic perceptions—evidence from German Federal States Regional Science and Urban Economics 2018 68 172 190 10.1016/j.regsciurbeco.2017.11.006
Gelper S Lemmens A Croux C Consumer sentiment and consumer spending: Decomposing the Granger causal relationship in the time domain Applied Economics 2007 39 1 1 11 10.1080/00036840500427791
Ghosh S Consumer confidence and consumer spending in Brazil: A Nonlinear autoregressive distributed lag model analysis Arthaniti Journal of Economic Theory and Practice 2021 20 1 53 85 10.1177/0976747919898906
Gino F Brooks A Schweitzer M Anxiety, advice, and the ability to discern: Feeling anxious motivates individuals to seek and use advice Journal of Personality and Social Psychology 2012 102 3 497 512 10.1037/a0026413 22121890
Gino F Shea C Deception in negotiations: The role of emotions 2012 Oxford University Press
Gintis H Strong reciprocity and human sociality Journal of Theoretical Biology 2000 206 2 169 179 10.1006/jtbi.2000.2111 10966755
Gokdemir O Consumption, savings and life satisfaction: The Turkish case International Review of Economics 2015 62 2 183 196 10.1007/s12232-015-0227-y
Guillen-Royo M Consumption and subjective wellbeing: Exploring basic needs, social comparison, social integration and hedonism in Peru Social Indicators Research 2008 89 3 535 555 10.1007/s11205-008-9248-1
Guven, C. (2009). Weather and financial risk-taking: Is happiness the channel? Available here: https://ideas.repec.org/p/diw/diwsop/diw_sp218.html. Accessed on 20/12/2021.
Hampson DP Gong S Xie Y How consumer confidence affects price conscious behavior: The roles of financial vulnerability and locus of control Journal of Business Research 2020 10.1016/j.jbusres.2020.10.032.Advanceonlinepublication
Hampson DP McGoldrick PJ Antecedents of consumer price consciousness in a turbulent economy International Journal of Consumer Studies 2017 41 4 404 414 10.1111/ijcs.12344
Heintzelman SJ Diener E Oishi S Tay L Eudaimonia in the contemporary science of subjective well-being Psychological well-being, self-determination, and meaning in life Handbook of well-being 2018 DEF Publishers
Helliwell, J. F., Layard, R., Sachs, J., & De Neve, J.-E. (Eds.) (2021). World Happiness Report 2021. New York: Sustainable Development Solutions Network. Available here: https://worldhappiness.report/ed/2021/ . Accessed on 16/12/2021.
Hendrix C Brinkman H-J Food insecurity and conflict dynamics: Causal linkages and complex feedbacks Stability International Journal of Security and Development 2013 2 26 10.5334/sta.bm
Higgins ET Value from hedonic experience and engagement Psychological Review 2006 113 439 460 10.1037/0033-295X.113.3.439 16802877
Hollanders D Vliegenthart R The influence of negative newspaper coverage on consumer confidence: The Dutch case Journal of Economic Psychology 2011 32 3 367 373 10.1016/j.joep.2011.01.003
Hsu C-C Lin H-Y Wu J-Y Consumer confidence and stock markets: The panel causality evidence International Journal of Economics and Finance 2011 3 6 91 10.5539/ijef.v3n6p91
Huppert FA Psychological well-being: Evidence regarding its causes and consequences Applied Psychology: Health and Well Being 2009 1 2 137 164
Huta V Waterman AS Eudaimonia and its distinction from hedonia: Developing a classification and terminology for understanding conceptual and operational definitions Journal of Happiness Studies 2014 15 6 1425 1456 10.1007/s10902-013-9485-0
Inglehart R Foa R Peterson C Welzel C Development, freedom, and rising happiness: A global perspective (1981–2007) Perspectives on Psychological Science 2008 3 264 285 10.1111/j.1745-6924.2008.00078.x 26158947
Jaunky VC Jeetoo J Rampersad S Happiness and consumption in Mauritius: An exploratory study of socio-economic dimensions, basic needs, luxuries and personality traits Journal of Happiness Studies 2020 21 2377 2403 10.1007/s10902-019-00178-8
Jiang X Hu X Liu Z Sun X Xue G Greed as an adaptation to anomie: The mediating role of belief in a zero-sum game and the buffering effect of internal locus of control Personality and Individual Differences 2020 152 109566 10.1016/j.paid.2019.109566
Johansson-Stenman O Martinsson P Honestly, why are you driving a BMW? Journal of Economic Behavior & Organization 2006 60 129 146 10.1016/j.jebo.2004.08.006
Johnson MA Naka A Downside risk: What the consumer sentiment index reveals Financial Services Review 2014 23 1 45
Kahneman D Tversky A The psychology of preferences Scientific American 1982 246 1 160 173 10.1038/scientificamerican0182-160
Karasoy Can HG Yüncüler Ç The explanatory power and the forecast performance of consumer confidence indices for private consumption growth in Turkey Emerging Markets Finance and Trade 2018 54 2136 2152 10.1080/1540496X.2017.1358608
Katona G On the function of behavioral theory and behavioral research in economics American Economic Review 1968 58 146 149
Kaus W Conspicuous consumption and “race”: Evidence from South Africa Journal of Development Economics 2013 100 63 73 10.1016/j.jdeveco.2012.07.004
Kemp E Kennett-Hensel PA Williams KH The calm before the storm: Examining emotion regulation consumption in the face of an impending disaster Psychology & Marketing 2014 31 933 945 10.1002/mar.20744
Kennett-Hensel PA Sneath JZ Lacey R Liminality and consumption in the aftermath of a natural disaster Journal of Consumer Marketing 2012 29 52 63 10.1108/07363761211193046
Khumalo J Consumer spending and consumer confidence in South Africa: Cointegration analysis Journal of Economics and Behavioral Studies 2014 6 2 95 104 10.22610/jebs.v6i2.473
Kleijnnijenhuis J Schultz F Oegema D Frame complexity and the financial crisis: A comparison of the United States, the United Kingdom and Germany in the period 2007–2012 Journal of Communication 2015 65 1 23 10.1111/jcom.12141
Kłopocka AM Does consumer confidence forecast household saving and borrowing behavior? Evidence for Poland Social Indicators Research 2017 133 693 717 10.1007/s11205-016-1376-4 28890595
Klopocka, A.M., & Gorska, R. (2021). Forecasting household saving rate with consumer confidence indicator and its components: Panel data analysis of 14 European countries. European Research Studies Journal, 24(3), 874–898.
Kluemper DH Little LM Degroot T State or trait: Effects of state optimism on job-related outcomes Journal of Organizational Behavior 2009 30 2 209 231 10.1002/job.591
Kotabe HP Righetti F Hofmann W How anticipated emotions guide self-control judgments Frontiers in Psychology 2019 10 1614 10.3389/fpsyg.2019.01614 31396124
Kuhnen CM Knutson B The influence of affect on beliefs, preferences, and financial decisions Journal of Financial and Quantitative Analysis 2011 46 3 605 626 10.1017/S0022109011000123
Kursan Milaković I Purchase experience during the COVID-19 pandemic and social cognitive theory: The relevance of consumer vulnerability, resilience, and adaptability for purchase satisfaction and repurchase International Journal of Consumer Studies 2021 10.1111/ijcs.12672.Advanceonlinepublication
Kwan ACC Cotsomitis JA The usefulness of consumer confidence in forecasting household spending in Canada: A national and regional analysis Economic Inquiry 2006 44 185 197 10.1093/ei/cbi064
Lahiri K Monokroussos G Zhao Y Forecasting consumption: The role of consumer confidence in real time with many predictors Journal of Applied Econometrics 2016 31 7 1254 1275 10.1002/jae.2494
Landau D Iervolino AC Pertusa A Santo S Singh S Mataix-Cols D Stressful life events and material deprivation in hoarding disorder Journal of Anxiety Disorders 2011 25 192 202 10.1016/j.janxdis.2010.09.002 20934847
Larson LRL Shin H Fear during natural disaster: Its impact on perceptions of shopping convenience and shopping behavior Services Marketing Quarterly 2018 39 293 309 10.1080/15332969.2018.1514795
Leach J Survival psychology 1994 New York University Press
Lemmon M Portniaguina E Consumer confidence and asset prices: Some empirical evidence The Review of Financial Studies 2006 19 4 1499 1529 10.1093/rfs/hhj038
Lins S Aquino S Development and initial psychometric properties of a panic buying scale during COVID-19 pandemic Heliyon 2020 6 04746 10.1016/j.heliyon.2020.e04746
Loewenstein GF Weber EU Hsee CK Welch N Risk as feelings Psychological Bulletin 2001 127 2 267 286 10.1037/0033-2909.127.2.267 11316014
Lowe TS Perceived job and labor market insecurity in the United States: An assessment of workers’ attitudes from 2002 to 2014 Work and Occupations 2018 45 3 313 345 10.1177/0730888418758381
Ludvigson SC Consumer confidence and consumer spending Journal of Economic Perspectives 2004 18 2 29 50 10.1257/0895330041371222
Luthans F Youssef CM Emerging positive organizational behavior Journal of Management 2007 33 321 349 10.1177/0149206307300814
Lyubomirsky S King L Diener E The benefits of frequent positive affect: Does happiness lead to success? Psychological Bulletin 2005 131 6 803 855 10.1037/0033-2909.131.6.803 16351326
Mäkikangas A Kinnunen U Psychosocial work stressors and well-being: Self-esteem and optimism as moderators in a one-year longitudinal sample Personality and Individual Differences 2003 35 3 537 557 10.1016/S0191-8869(02)00217-9
Malgarini M Margani P Psychology, consumer sentiment and household expenditures Applied Economics 2007 39 13 1719 1729 10.1080/00036840600606351
Malovaná S Hodula M Frait J What does really drive consumer confidence? Social Indicators Research 2021 155 885 913 10.1007/s11205-021-02626-6 33612916
Mandal A McCollum J Consumer confidence and the unemployment rate in New York State: A panel study New York Economic Review 2013 44 3 19
Mastrogiacomo M Alessie R The precautionary savings motive and household savings Oxford Economic Papers-New Series 2014 66 1 164 187 10.1093/oep/gpt028
Matošec M Obuljen Zoričić Z Identifying the interdependence between consumer confidence and macroeconomic developments in Croatia Interdisciplinary Description of Complex Systems 2019 17 2-B 345 354 10.7906/indecs.17.2.10
Mermod AY Dudzevičiūtė G Frequency domain analysis of consumer confidence, industrial production and retail sales for selected European countries Journal of Business Economics and Management 2011 12 4 589 602 10.3846/16111699.2011.599406
Mi LH Ryong KJ Han JS The effect of perceived risk on negative emotions, trust, and investment intentions: Focused on investors and potential investors International Journal of Tourism and Hospitality Research 2019 33 19 31 10.21298/IJTHR.2019.8.33.8.19
Michalos AC Satisfaction and happiness Social Indicators Research 1980 8 385 422 10.1007/BF00461152
Miu AC Heilman RM Houser D Anxiety impairs decision-making: Psychophysiological evidence from an Iowa gambling task Biological Psychology 2008 77 3 353 358 10.1016/j.biopsycho.2007.11.010 18191013
Mody A Ohnsorge F Sandri D Precautionary savings in the great recession IMF Economic Review 2012 60 1 114 138 10.1057/imfer.2012.5
Moors A Ellsworth PC Scherer KR Frijda NH Appraisal theories of emotion: State of the art and future development Emotion Review 2013 5 2 119 124 10.1177/1754073912468165
Moschis GP Stress and consumer behavior Journal of the Academy of Marketing Science 2007 35 3 430 444 10.1007/s11747-007-0035-3
Mourougane A Roma M Can confidence indicators be useful to predict short term real GDP growth? Applied Economics Letters 2003 10 519 522 10.1080/1350485032000100305
Ng S Russell-Bennett R The role of affect in consumer evaluation of health care services Health Marketing Quarterly 2015 32 1 31 47 10.1080/07359683.2015.1000708 25751317
Ng TWH Sorensen KL Eby LT Locus of control at work: A meta-analysis Journal of Organizational Behavior 2006 27 8 1057 1087 10.1002/job.416
Nicolao L Irwin JR Goodman JK Happiness for sale: Do experiential purchases make consumers happier than material purchases? Journal of Consumer Research 2009 36 2 188 198 10.1086/597049
Noll H-H Weick S Consumption expenditures and subjective well-being: Empirical evidence from Germany International Review of Economics 2015 62 2 101 119 10.1007/s12232-014-0219-3
Nyman R Ormerod P Big data, socio-psychological theory, algorithmic text analysis and predicting the Michigan consumer sentiment index 2014 Quantitative Finance Publishing
O’Loughlin DM Szmigin I McEachern MG Barbosa B Karantinou K Fernández-Moya ME Man thou art dust: Rites of passage in austere times Sociology 2017 51 5 1050 1066 10.1177/0038038516633037 28989200
O'Connor GE Newmeyer CE Wong NYC Bayuk JB Cook LA Komarova Y Loibl C Lin O Warmath D Conceptualizing the multiple dimensions of consumer financial vulnerability Journal of Business Research 2019 100 C 421 430 10.1016/j.jbusres.2018.12.033
OECD OECD guidelines on measuring subjective well-being 2013 OECD Publishing
Olekalns M Smith PL Mutually dependent: Power, trust, affect and the use of deception in negotiation Journal of Business Ethics 2009 85 3 347 365 10.1007/s10551-008-9774-4
Ou YC de Vries L Wiesel T Verhoef PC The role of consumer confidence in creating customer loyalty Journal of Service Research 2014 17 3 339 354 10.1177/1094670513513925
Ozari, G. (2007). Cheap thrills. Retrieved from: http://www.demodirt.com/happiness_simple_pleasures.html . Accessed on 25/01/2022
Pan W-F Unemployment and confidence in Canada: Evidence from national and regional level data Economics Bulletin 2018 38 1111 1123
Patt A Weber EU Perceiving and communicating climate change uncertainty Wires Climate Change 2013 5 219 232 10.1002/wcc.259
Patt, A., & Weber, E.U. (2014). Perceptions and communication strategies for the many uncertainties relevant for climate policy. Wiley Interdisciplinary Reviews: Climate Change, 5.
Peng J Xiao W Yang Y Wu S Miao D The impact of trait anxiety on self-frame and decision making Journal of Behavioral Decision Making 2014 27 1 11 19 10.1002/bdm.1783
Perez-Truglia R A test of the conspicuous-consumption model using subjective well-being data The Journal of Socio-Economics 2013 45 146 154 10.1016/j.socec.2013.05.012
Peters E Slovic P The role of affect and worldviews as orienting dispositions in the perception and acceptance of nuclear power Journal of Applied Social Psychology 1996 26 1427 1453 10.1111/j.1559-1816.1996.tb00079.x
Peterson C The future of optimism American Psychologist 2000 55 1 44 55 10.1037/0003-066X.55.1.44 11392864
Pleeging E Burger M van den Heuvel SC Hope in economics Historical and multidisciplinary perspectives on hope 2020 Springer 165 178
Pugno M The Easterlin paradox and the decline of social capital: An integrated explanation The Journal of Socio-Economics 2009 38 590 600 10.1016/j.socec.2009.03.012
Rabin M Inference by believers in the law of small numbers Quarterly Journal of Economics 2002 117 775 816 10.1162/003355302760193896
Reknes I Visockaite G Liefooghe A Lovakov A Einarsen SV Locus of control moderates the relationship between exposure to bullying behaviors and psychological strain Frontiers in Psychology 2019 10 1323 10.3389/fpsyg.2019.01323 31244725
Riquelme, H., & AlQallaf, A. (2020). Anticipated emotions and their effects on risk and opportunity evaluations. Journal of International Entrepreneurship, 18(3), 312–335. 10.1007/s10843-019-00262-3
Rojo-Suárez J Alonso-Conde AB Impact of consumer confidence on the expected returns of the Tokyo stock exchange: A comparative analysis of consumption and production-based asset pricing models PLoS ONE 2020 15 11 0241318 10.1371/journal.pone.0241318
Roser, M., & Nagdy, M. (2014). Optimism and Pessimism. Retrieved from: https://ourworldindata.org/optimism-pessimism. Accessed on 16/12/2021.
Rotter JB Social learning and clinical psychology 1954 Prentice Hall
Sangnier M Does trust favor macroeconomic stability? Journal of Comparative Economics 2013 41 3 653 668 10.1016/j.jce.2012.10.002
Scheier MF Carver CS Bridges MW Chang EC Optimism, pessimism, and psychological well-being Optimism & pessimism: Implications for theory, research, and practice 2001 American Psychological Association 189 216
Schmutte PS Ryff DC Personality and well-being: What is the connection? Journal of Personality and Social Psychology 1997 73 549 559 10.1037/0022-3514.73.3.549 9294901
Schwarz, N., & Bohner, G. (1996). Feelings and their motivational implications: Moods and the action sequence. In P. M. Gollwitzer & J. A. Bargh (Eds.), The psychology of action: Linking cognition and motivation to behavior (pp. 119–145). The Guilford Press.
Schwarz N Clore G Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states Journal of Personality and Social Psychology 1983 45 513 523 10.1037/0022-3514.45.3.513
Sekizawa Y Goto Y Noguchi R So M Shimizu E Associations between Japan’s consumer confidence index and emotion-related variables such as depression, anxiety, and positive affect Cogent Psychology 2021 8 1 1897199 10.1080/23311908.2021.1897199
Senik C Is man doomed to progress? Journal of Economic Behavior and Organization 2008 68 140 152 10.1016/j.jebo.2008.03.004
Shiller RJ From efficient markets theory to behavioral finance The Journal of Economic Perspectives 2003 17 1 83 104 10.1257/089533003321164967
Siegrist M Gutscher H Earle TC Perception of risk: The influence of general trust, and general confidence Journal of Risk Research 2005 8 2 145 156 10.1080/1366987032000105315
Singal M Effect of consumer sentiment on hospitality expenditures and stock returns International Journal of Hospitality Management 2012 31 2 511 521 10.1016/j.ijhm.2011.07.009
Smithson M Bammer G Smithson M The many faces and masks of uncertainty Uncertainty and risk: Multidisciplinary perspectives 2008 Earthscan 13 25
Sneath JZ Lacey R Kennett-Hensel PA Coping with a natural disaster: Losses, emotions, and impulsive and compulsive buying Marketing Letters 2009 20 45 60 10.1007/s11002-008-9049-y
Sorić P Lolić I Claveria O Monte E Torra S Unemployment expectations: A socio-demographic analysis of the effect of news Labour Economics 2019 60 64 74 10.1016/j.labeco.2019.06.002
Soroka SN Stecula DA Wlezien Ch It’s (change) in the (future) economy, stupid: Economic indicators, the media, and public opinion American Journal of Political Science 2015 59 457 474 10.1111/ajps.12145
Stanton SJ Reeck C Huettel SA LaBar KS Effects of induced moods on economic choices Judgment and Decision Making 2014 9 2 167
Steinel W Van Kleef GA Harinck F Are you talking to me?! Separating the people from the problem when expressing emotions in negotiation Journal of Experimental Social Psychology 2008 44 2 362 369 10.1016/j.jesp.2006.12.002
Svensson HM Albæk E van Dalen A de Vreese CH The impact of ambiguous economic news on uncertainty and consumer confidence European Journal of Communication 2017 32 2 85 99 10.1177/0267323116677205
Taylor K McNabb R Business cycles and the role of confidence: Evidence for Europe Oxford Bulletin of Economics and Statistics 2007 69 185 208 10.1111/j.1468-0084.2007.00472.x
Teresiene D Keliuotyte-Staniuleniene G Liao Y Kanapickiene R Pu R Hu S Yue X-G The impact of the COVID-19 pandemic on consumer and business confidence indicators Journal of Risk and Financial Management 2021 14 4 159 10.3390/jrfm14040159
Throop, A. W. (1992). Consumer sentiment: its causes and effects. Economic Review, Federal Reserve Bank of San Francisco, 3, 35–59. Available at: https://fraser.stlouisfed.org/scribd/?toc_id=514282&filepath=/docs/publications/frbsfreview/rev_frbsf_1992no3.pdf&start_page=54
Treanor M The effects of financial vulnerability and mothers’ emotional distress on child social, emotional and behavioral well-being: A structural equation model Sociology 2016 50 4 673 694 10.1177/0038038515570144
Tsay C-J Bazerman MH A decision-making perspective to negotiation: A review of the past and a look to the future Negotiation Journal 2009 25 467 480 10.1111/j.1571-9979.2009.00239.x
Tsurumi T Yamaguchi R Kagohashi K Managi S Are cognitive, affective, and eudaimonic dimensions of subjective well-being differently related to consumption? Evidence from Japan Journal of Happiness Studies 2021 22 2499 2522 10.1007/s10902-020-00327-4
Twenge JM Zhang L Im C It's beyond my control: A cross-temporal meta-analysis of increasing externality in locus of control, 1960–2002 Personality and Social Psychology Review 2004 8 3 308 319 10.1207/s15327957pspr0803_5 15454351
Utaka A Consumer confidence and the Japanese economy. Comparison of pre- and post-bubble period Economics Bulletin 2014 34 2 1165 1173
van Aarle B Kappler M Economic sentiment shocks and fluctuations in economic activity in the euro area and the USA Intereconomics 2012 47 44 51 10.1007/s10272-012-0405-z
Van Giesen RI Pieters R Climbing out of an economic crisis: A cycle of consumer sentiment and personal stress Journal of Economic Psychology 2019 70 109 124 10.1016/j.joep.2018.12.004
Van Kleef GA De Dreu CKW Manstead ASR The interpersonal effects of emotions in negotiations: A motivated information processing approach Journal of Personality and Social Psychology 2004 87 4 510 528 10.1037/0022-3514.87.4.510 15491275
Vanlaer W Bielen S Marneffe W Consumer confidence and household saving behaviors: A cross-country empirical analysis Social Indicators Research 2020 147 677 721 10.1007/s11205-019-02170-4
Veenhoven, R., Chiperi, F., Kang, X., & Burger, M. (2021). Happiness and consumption: A research synthesis using an online finding archive. SAGE Open. Retrieved from: 10.1177/2158244020986239. Accessed on 25/01/2022.
Veenhoven R World database of happiness: Correlates of happiness 1994 Erasmus University
Veenhoven R Diener E Suh E Freedom and happiness: A comparative study in 46 nations in the early 1990’s Culture and Subjective Well-being 2000 MIT press 257 288
Verme P Happiness, freedom and control Journal of Economic Behavior and Organization 2009 71 2 146 161 10.1016/j.jebo.2009.04.008
Vliegenthart R Damstra A Parliamentary questions, newspaper coverage, and consumer confidence in times of crisis: A cross-national comparison Political Communication 2019 36 17 35 10.1080/10584609.2018.1478472
Vuchelen J Consumer sentiment and macroeconomic forecasts Journal of Economic Psychology 2004 25 493 506 10.1016/S0167-4870(03)00031-X
Wood SL Rettman JR Predicting happiness: How normative feeling rules influence (and even reverse) durability bias Journal of Consumer Psychology 2007 17 188 201 10.1016/S1057-7408(07)70028-1
Xiao G Kim JO The investigation of Chinese consumer values, consumption values, life satisfaction, and consumption behaviors Psychology & Marketing 2009 26 7 610 624 10.1002/mar.20291
Youssef CM Luthans F Positive organizational behavior in the workplace: The impact of hope, optimism, and resilience Journal of Management 2007 33 5 774 800 10.1177/0149206307305562
Zampetakis LA Lerakis M Kafetsios K Moustakis VS The moderating role of anticipated affective ambivalence in the formation of entrepreneurial intentions International Entrepreneurship and Management Journal 2016 12 3 815 838 10.1007/s11365-015-0367-2
Zhang J Xiong Y Effects of multifaceted consumption on happiness in life: A case study in Japan based on an integrated approach International Review of Economics 2015 62 143 162 10.1007/s12232-015-0225-0
Zheng CH Zhang J Guo YR Zhang YL Qian LL Disruption and reestablishment of place attachment after large-scale disasters: The role of perceived risk, negative emotions, and coping International Journal of Disaster Risk Reduction 2019 40 101273 10.1016/j.ijdrr.2019.101273
Zhong JY Mitchell VW A mechanism model of the effect of hedonic product consumption on well-being Journal of Consumer Psychology 2010 20 152 162 10.1016/j.jcps.2010.01.001
Zhong JY Mitchell VW Does consumer well-being affect hedonic consumption? Psychology & Marketing 2012 29 583 594 10.1002/mar.20545
Zwanka RJ Buff C COVID-19 generation: A conceptual framework of the consumer behavioral shifts to be caused by the COVID-19 pandemic Journal of International Consumer Marketing 2020 33 1 58 67 10.1080/08961530.2020.1771646
| 36467538 | PMC9707121 | NO-CC CODE | 2022-12-01 23:20:20 | no | J Happiness Stud. 2022 Nov 29;:1-25 | utf-8 | J Happiness Stud | 2,022 | 10.1007/s10902-022-00603-5 | oa_other |
==== Front
Curr Nutr Rep
Curr Nutr Rep
Current Nutrition Reports
2161-3311
Springer US New York
36443548
446
10.1007/s13668-022-00446-9
Diabetes and Obesity (M Dalamaga and F Magkos, Section Editors)
Dietary Interventions for Type 2 Diabetes in South Asian Populations—A Systematic Review
Subhan Fatheema B. [email protected]
1
Fernando Dineli N. [email protected]
2
Thorlakson Jessica [email protected]
3
http://orcid.org/0000-0003-3882-0592
Chan Catherine B. [email protected]
4
1 grid.253547.2 000000012222461X Department of Nutrition and Food Science, California Polytechnic State University at Pomona, Pomona, CA 91768 USA
2 grid.17089.37 0000 0001 2190 316X Department of Cell Biology, University of Alberta, Edmonton, AB T6G 2H7 Canada
3 grid.17089.37 0000 0001 2190 316X University of Alberta Libraries, Edmonton, AB T6G 2J8 Canada
4 grid.17089.37 0000 0001 2190 316X Department of Agricultural, Food and Nutritional Science & Department of Physiology, 4-126A Li Ka Shing Centre for Health Innovation Research, University of Alberta, Edmonton, AB T6G 2E1 Canada
29 11 2022
117
23 9 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose
South Asians face a high burden of type 2 diabetes (T2D). We systematically summarized current research on the efficacy, cultural relevance, and research gaps of nutrition interventions that could be used for treatment in this population.
Findings
We identified 18 articles published since 2010. Dietary pattern interventions have focused on low-glycemic index (GI) solutions and consistently reported improvement in glycemic management. Trials of nutrition education and counselling had diverse approaches, with those utilizing more intensive interventions generally eliciting better glycemic outcomes. Many studies developed interventions with cultural relevance by including traditional foods, providing materials in the local language, and acknowledging important food-related customs. These adaptations were seen in South Asian countries as well as Western countries hosting immigrants.
Summary
Data from South Asian countries support low-GI and intensive counselling approaches for the treatment of T2D. Given the high prevalence of T2D in these populous countries, approaches that can reach large numbers of people are needed. In Western countries, more emphasis on providing culturally relevant nutrition therapy is needed.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13668-022-00446-9.
Keywords
Type 2 diabetes
Nutrition therapy
Nutrition education
Nutrition counseling
Cultural relevance
South Asia
India
Pakistan
Sri Lanka
Bangladesh
Immigrant
==== Body
pmcIntroduction
Nearly 122 million people in the region of South Asia had diagnosed with diabetes in 2021. This represents a prevalence of 8.7%, 9.6%, 10.4%, 11.3%, 14.2%, and 30.8% in Nepal, India, Bhutan, Sri Lanka, Bangladesh, and Pakistan, respectively [1] and is a crushing health burden individually and for society. Millions more South Asians who have immigrated to other countries are at high risk of developing T2D. South Asians living in Canada have 3–5 times higher risk of diabetes than white people [2], with prevalence ranging from 16.5–26.8% [3]. Similar high risk is documented in the USA [4] and the UK [5]. This may be related, in part, to the adoption of new dietary patterns with low nutritional quality. South Asians living in Alberta, Canada, consume 30% of calories from foods high in sodium, sugar, and saturated fat [6].
Together with pharmacotherapy, nutrition therapy is a fundamental component of diabetes care. Although aggressive pharmaceutical management is recommended because of higher glycated hemoglobin (HbA1C) in South Asians at diagnosis [7], the cost of newer generation antiglycemic agents is a barrier in South Asian countries [8]. A recent consensus document highlights the lack of consistent delivery of medical nutrition therapy in South Asian countries [9]. The majority of people with diabetes are not prescribed specific diabetes diets [10]. In Western countries, South Asian immigrants experience barriers related to language and culturally tailored T2D therapy [11] and the majority do not meet treatment targets for HbA1C, blood pressure, or lipids [12]. A recent systematic review found 4 behavioral interventions incorporating culturally relevant content conducted in the UK and the Netherlands. Overall, they did not lower HbA1C. Methods to integrate culturally congruent care were lacking as was the necessity to consider the heterogeneity of South Asian culture [13]. The efficacy of diabetes interventions in South Asian populations needs to be documented to identify promising strategies for more widespread implementation. Sohal et al. [11] found that many barriers and facilitators to improved diabetes care were similar whether the South Asians lived in India or Western countries, highlighting the enduring importance of culture.
The purpose of this systematic review is to evaluate the efficacy to improve glycemic outcomes in recent medical nutrition therapy interventions carried out in South Asia, North America, or Europe. The primary research question was: In South Asian people with T2D, living in their home countries or immigrants to North America/Europe, what nutrition interventions improve health outcomes (glycemia, lipidemia, blood pressure, anthropometrics) or nutrition behaviors or knowledge? Additional research questions included:What measures were taken to make these interventions culturally relevant to the target audience?
What are the gaps in research that could improve nutrition intervention strategies?
Methods
Eligibility Criteria
Eligible articles were original studies focused on people diagnosed with T2D, an outcome of glycemic management and/or nutrition behavior and a nutrition intervention conducted in India, Pakistan, Sri Lanka, Bhutan, Bangladesh, Nepal, or in a North American or European country specifically for immigrants from a South Asian country. We included case–control, quasi-experimental and randomized controlled trials (RCT). Secondary outcomes were blood pressure, lipid panel, and body weight or body mass index (BMI). Exclusion criteria were as follows: studies published before 2010, not English language, and studies focused on a single food, nutrient, or supplement. Abstracts, letters, editorials, general or systematic reviews and meta-analyses, practice guidelines, and study protocols were excluded.
Information Sources and Search Strategy
Six databases (CINAHL, Medline, Web of Science, Cochrane, Psych Info, Proquest Dissertations) were searched in October 2021 by JT and the output was uploaded into Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia. Available at www.covidence.org). The MEDLINE search strategy, which documents the keywords used, is provided in Supplemental Table 1.
Selection Process
Duplicate studies were removed automatically by the Covidence software. Title and abstract screening was conducted by DNF and CBC. Potentially eligible articles were retrieved, and full-text review was conducted by DNF and CBC.
Data Collection Process
Data were extracted independently by DNF and FBS. Data included study title, authors, year and DOI; the study objective and population (country, urban/rural, sample size and gender, age range; study design, and details of the dietary intervention. Results were recorded in Excel and included clinical outcomes of glycemic management and cardiovascular risk factors, behavioral outcomes of dietary intake, self-efficacy, nutrition knowledge; cultural relevance of the intervention, barriers, and facilitators to uptake (by patients or by the health system).
Effect Measures and Synthesis
Where possible, quantitative data such as glycemic or nutritional outcomes were reported as intergroup mean differences for RCT or changes from baseline for quasi-experimental designs ± standard deviation or confidence intervals. A narrative synthesis was conducted and organized to address each research question.
Results
Study Selection
The study selection details are shown in Supplemental Fig. 1. Eighteen articles met the inclusion criteria.
Study Characteristics (Table 1)
Table 1 Study characteristics by study design
Study Country Sample size, age Design Intervention /control group Duration Intervention and cultural relevance
Non-randomized trials with or without a control group
Pande et al. [14] India (urban) N = 20
Age 35–58y
Case–control Both T2D (n = 10) and healthy group (n = 10) received 1 meal/day Acute Provision of low-GI mixed meals using traditional Indian foods and spices in a medical setting.
Pande et al. [15] India (urban) N = 15
Age 42–58 y
Quasi-experimental, pre-post Provided meals and snacks
No control group
4 wk Provision of low-GI mixed meals and snacks using traditional Indian foods and spices to cook at home.
Jain et al. [12] India (urban) N = 180
Mean age 47 y
Quasi-experimental, pre-post Education
No control group
3 mon In-person education supported with a booklet describing the principles of diabetes management.
Singh et al. [16] India (urban) N = 120
Age 35–65 y
Quasi-experimental, pre-post Education
No-intervention control
3 mon In-person nutrition education from dietitians. Type of PA recommended to participants not reported.
Krishnan et al. [13] India (urban) N = 134
Mean age 50 y
Quasi-experimental pre-post Counselling–
I – attended 1 session, no follow-up
II – attended dietary counselling with periodic follow-up
III – attended dietary and exercise counselling with periodic follow-up
No control group
3 and 6 mon In-person counselling to participant and caregiver/family, home visits and telephone counselling for those unable to attend in person. Type of physical activity recommended to participants not reported.
Bairy et al. [11] India (urban) N = 101
Median age 55 y
Quasi-experimental, pre-post Provision of meals + education as part of an integrated naturopathy and yoga program
No control group
3 mon Meals were low-GI, plant-based, low in salt, oil, sugar. Also engaged in yoga, mild aerobic activity (walking, swimming, boat-pedalling in river), rest, “therapeutic fasting” (1–3 d) + 2 lectures/d on naturopathy, personal development + 1:1 physician counselling + 1-h cooking classes daily. Based on traditional practices. Patients were admitted to a naturopathy hospital for 15- or 30-d treatment.
Kumari et al. [17] India
(urban)
N = 202
Mean age 53 y
Quasi-experimental Holistic lifestyle intervention
Usual care controls
3, 6, and 12 mon Interdisciplinary team provided monthly counselling for 6 mon in local language. Supported with pictures, videos, 1:1, and group sessions. Included diet, physical activity (brisk walking, yoga), tobacco cessation, stress management, self-management adherence.
St. John [18] UK N = 34 Quasi-experimental pre-post Education
No control group
2 mon In-person education supported with a booklet developed for Pakistani immigrants to teach carbohydrate content of traditional foods.
Randomized controlled trials
Johansen et al. [19] Norway N = 198
Age 26–63 y
RCT Education
No-intervention control
7 mon In-person, 2-h group sessions on managing blood glucose with diet and exercise (walking). Used culturally adapted resources, translated written materials, focused on traditional foods.
Raberg Kjollesdal [20] Norway N = 82
Age 28–62 y
RCT Education
No-intervention control
7 mon Same as Johansen et al. [19]
Myers et al. [21] India N = 239
Age 25–69 y
Cluster RCT Education and self-management support
Usual care controls
6 and 12 mon Dietitians received standardized training prior to implementation. Education provided in person by dietitian based on evidence-based nutrition practice guidelines (EBNPG) (Diabetes in India Nutrition Guidelines), motivational interviewing, goal setting, self-monitoring. Participants received handouts including meal plans, exchange lists.
Pavithran et al. [22] India N = 40
Age 35–65 y
RCT Nutrition intervention
Controls maintain usual diet
24 wk Educational in-person interview focused on consuming low-GI whole grain cereals using traditional foods, periodically reinforced by a dietitian.
Pavithran et al. [23] India N = 80
Age 35–65 y
RCT Nutrition intervention
Controls maintain usual diet
24 wk Educational interview focused on consuming low-GI traditional foods, periodically reinforced by a dietitian.
Yasmin et al. [24] Bangladesh N = 320
Age 30–85 y
RCT Interactive voice calls and telephone-based physician access
Usual care controls
6 mon Regular personalized advice regarding diet, physical activity, medications, clinic visits. Call service provided physician contact 24/7.
Devi [25] India (urban) N = 340
Age 20–80 y
Pre-post, with randomization of participants I—Education with self-learning module, paper-based
II – Education with a powerpoint presentation
III – usual care control group
6 mon I—Diet and nutrition self-learning module available in Hindi and English (participant choice) + in-person 90-min discussion in the participant’s home.
II – The same materials provided via laptop with an educator present, and DVD and handout for viewing independently at the participant’s home.
Islam et al. [26] Bangladesh N = 200
Mean age 48 y
RCT Text messages
Usual care controls
6 mon Daily text messages (20 out of 90 related to diet) over 6 mon, guided by behavior change theory and diabetes guidelines (Bangladesh + UK and ADA).
Thadchanamoorthy et al. [27] Sri Lanka N = 135
Mean age 58 y
RCT I – usual care controls
II – general diabetes management education
III – nutrition education
3 and 6 mon II – Standard diet based on Dietary Guidelines and Nutrition Therapy for Specific Diseases, Sri Lanka; 4 sessions.
III—Nutrition education focused on low carbohydrate diet; 4 sessions.
Varadaraj et al. [28] India N = 98
Age 18 + y
RCT Education comparing 2 intensities
Usual care control
3 and 6 mon In-person nutrition education provided by a dietitian either once monthly or every 3 mon after an initial session (also provided to the usual care control).
ADA American Diabetes Association, DVD digital video disk, GI glycemic index, PA physical activity, RCT randomized controlled trial, UK United Kingdom
Country
Ten studies (eliciting 12 articles) were conducted in India [14–17, 21–28], 2 in Bangladesh [18, 20], and 1 study in each of Sri Lanka [19], the UK [29], and Norway (2 articles) [30, 31].
Design
Seven studies were quasi-experimental [15, 16, 22–26, 29], 9 (10 articles) were RCT [14, 16–21, 27, 28, 30] and 1 was a case–control study [26].
Sample Size, Participant Age, and Study Duration
The trial size ranged from 15 to 340 participants and most were of middle-aged adults. One pilot study was an acute evaluation of low-GI foods. All other trials ranged from 4 weeks to 12 months.
Effects of Nutrition Interventions on Health Outcomes
Low Glycemic Index Dietary Interventions
Five studies focused on intervening with a low-GI diet (Table 1) with results reported in Table 2 [17, 21, 23, 24, 26].Table 2 Study outcomes by study design
Study Biological outcomes Behavioral outcomes Barriers
Pande et al. [14] GI range of 7 meals 28–45
Acute BG AUC response: Healthy group – 960–1719 mg/dl × h vs Diabetes group—1677–1959 mg/dl × h
Sensory and satiety responses similar between healthy and diabetes groups Low patient motivation
Lack of access to low GI foods
Pande et al. [15] After 4 wk vs baseline:
FBG decreased (− 36 mg/dl) (p < 0.001)
HbA1C decreased (− 0.9%) (p < 0.001)
TG decreased (− 80 mg%) (p < 0.001)
TC decreased (− 39 mg%) (p < 0.001)
HDL-C increased (+ 6.8 mg%) (p < 0.001)
LDL-C decreased (− 25.8 mg%) (p < 0.001)
Patients indicated low acceptability of foods, low motivation
Jain et al. [12] After 3 mon vs baseline:
FBG decreased in people diagnosed with diabetes for ≥ 5 y (− 52 mg/dL) (p < 0.001) and ≥ 10 y (− 58 mg/dL) (p < 0.001)
– Low motivation, which was mitigated by individual counseling and education
Singh et al. [16] After 3 mon vs baseline:
Weight and BMI decreased in all 3 intervention groups
Improvements in body composition (upper arm circumference, triceps skinfold thickness) after intervention
Increased nutrition knowledge, attitudes, and practices after the intervention. PA results not reported –
Krishan et al. [13] After 3 mon vs baseline:
FBG (mg/dL): Group I + 4; Group II − 24; Group III -14
PPG (mg/dL): Group I + 4; Group II − 11; Group III -8.3
SBP (mmHg): Group I − 4; Group II + 1; Group III -1.5
DBP (mmHg): Group I + 1; Group II + 0.7; Group III -0.7
BMI (kg/m2): Group I + 0.2; Group II -0.2; Group III − 0.8
After 6 mon vs baseline:
HbA1C (%): Group I 0; Group II + 0.3; Group III − 0.2
FBG (mg/dL): Group I − 6.3; Group II -21; Group III − 16
PPG (mg/dL): Group I − 10; Group II -10; Group III − 19
TG (mg/dL): Group I − 1; Group II + 18; Group III − 19
TC (mg/dL): Group I − 5; Group II -14; Group III − 15
LDL-C (mg/dL): Group I − 7; Group II -12; Group III − 14
HDL-C (mg/dL): Group 1 0; Group II -2; Group III − 0.1
SBP (mmHg): Group I − 2; Group II -0.4; Group III − 1
DBP (mmHg): Group I − 0.5; Group II -1; Group III − 0.4
BMI (kg/m2): Group I − 0.6: Group II -0.1; Group III − 0.3
Noted frequent consumption of rice, pulse grains, egg, sheep, and chicken; milk and buttermilk, and sunflower oil by all groups. Higher consumption of leafy green vegetables by Groups II and III. PA results not reported Patients reported difficulty adhering to diet during festivals; reported various taboos related to vegetable consumption. These issues were addressed in counselling sessions
Bairy et al. [11] After 3 months vs baseline:
HbA1C − 0.9% overall; decrease was greater (− 1.7%) with excellent compared with poor compliance (− 0.4%)
FBG (mg/dL) decreased from 149 to 109
PPG (mg/dL) decreased from 188 to 152
Increased compliance to diet, increased knowledge of Integrated Naturopathy and yoga-recommended (low-GI) diet, meal spacing, meal preparation methods Patients reported low motivation and cost as obstacles; increased follow-up support would be beneficial
HCP reported lack of time and capacity to provide lifestyle interventions; indicated some responsibilities could be delegated
Kumari et al. [17] After 3 mon vs baseline in intervention (I) and control (C):
HbA1C (%): I − 0.5, C − 0.2 (p < 0.001 between groups)
After 6 mon from baseline:
HbA1C (%): I − 0.9, C − 0.2 (p < 0.001 between groups)
After 12 mon from baseline:
HbA1C (%): I − 0.9, C + 0.1 (p < 0.001 between groups)
Significant benefit (p < 0.001) of the intervention also seen for TG, TC, LDL-C, HDL-C, SBP at 12 mon relative to baseline. No effect of body weight between groups over follow-up period
Self-report of diet: Intervention group improved overall diet, intakes of fibrous foods, reduced sugar-rich and junk foods
Self report of PA: Intervention improved adherence to recommendations
Adherence tapered off by 12 months, recommend continued counselling. Patient motivation was a barrier
St. John [18] – Increased knowledge, understanding, and carbohydrate estimating ability for traditional and commonly eaten foods; awareness of carbohydrate’s impact on blood glucose Used simple messages and pictures for people with low literacy
Raberg Kjollesdal et al. [20] – 76% reported personal dietary changes
68% reported dietary habits of their families changed
No PA results reported
About half of women reported difficulty in changing dietary habits of self and family; barrier was likes/dislikes of family members, especially children
Johansen et al. [19] – Intervention increased the stage of change for sugar (p = 0.01), fat (p < 0.001) but vegetables and pulse grains NS. Women in intervention reported dietary changes for self and their families See above
Myers et al. [21] After 6 mon vs baseline in intervention (I) and control (C):
HbA1C (%): I − 1.7; C − 1.3 (NS between groups)
TG (mg/dL): I − 87; C − 3 (p < 0.01)
NS intergroup differences for TC, LDL-C, HDL-C
BMI (kg/m2): I − 0.8; C − 0.3 (NS)
After 12 mon vs baseline in intervention (I) and control (C):
HbA1C (%): I − 1.0; C − 1.0 (NS between groups)
TG (mg/dL): I − 74; C − 4 (p < 0.01)
BMI (kg/m2): I − 0.95; C − 0.5 (NS)
NS intergroup differences for TC, LDL-C, HDL-C
Noted reductions in excessive carbohydrate intake, increase in fiber intake, better carbohydrate spacing in at least 5 participants; noted improved physical activity, nutrition knowledge, energy intake, self-management practices, adherence to recommendations, food choices, involuntary weight gain, carbohydrate quality, fluid intake Dietitian-reported barriers included complexity of nutrition recommendations, inconsistency with Indian systems, norms and practices, time required to teach patients, low familiarity of dietitians with teaching lifestyle modifications. Enhanced support for dietitians (technical, expertise) could mitigate difficulties
Pavithran et al. [23] After 24 wk vs baseline in intervention (I) and control (C):
HbA1C (%): I − 1.0; C + 0.1 (p < 0.001 intergroup difference)
BMI (kg/m2): I − 0.75; C + 0.07 (p = 0.007)
TC, TG, HDL-C, LDL-C and VLDL all were NS
– Patients wanted familiar foods to increase compliance and acceptability
Yasmin et al. [24] After 6 mon vs baseline in intervention (I) and control (C):
FBG (mmol/L): I − 1.9 (p < 0.001); C − 0.4 (NS)
2-h PPG (mmol/l): I − 2.5 (p < 0.001); C − 0.9 (p = 0.035)
Intervention – adherence to recommended carbohydrate (p < 0.001), total energy (p < 0.001), and fruit (p < 0.001) improved but protein and fat did not (NS)
Control group – NS changes in diet
NS difference in medication adherence in both groups. PA volume (h/wk) was unchanged in both groups although intervention increased PA frequency
Patients reported low motivation, high cost were barriers. Patients could benefit from subsidies, more counselling
Pavithran et al. [22] After 24 wk vs baseline in intervention (I) and control (C):
HbA1C (%): I − 0.9; C + 0.1 (p = 0.003 intergroup difference)
TG (mg/dL): I − 27; control + 22 (p = 0.025 between groups)
BMI, SBP, DBP, TC (NS)
After 24 wk vs baseline:
Macronutrients as % energy intake
Carbohydrate (%): I 61.6, C 65.9 (p < 0.001 between groups)
Fat (%): I 24.4, C 21.1 (p = 0.003)
Protein—NS
Devi [25] After 6 mon vs baseline:
Meeting blood glucose targets: SLM increased from 52 to 95% of participants (p < 0.01); PPP increased from 42 to 98% of participants (p < 0.01); control increased from 49 to 54% (NS)
Body weight maintenance target: SLM increased from 4.7% to 32% (p < 0.01); PPP increased from 4.7% to 40% (p < 0.01); control increased from 6.5% to 10.6% (NS)
SLM and PPP increased overall nutrition knowledge similarly –
SLM increased score from 6.3 to 25.1 (p < 0.001); PPP increased score from 6.1 to 28.0 (p < 0.001); control increased score from 6.0 to 6.8 (NS)
Knowledge of GI of foods increased in both SLM and PPP (with PPP superior for some foods)
Islam et al. [26] After 6 mon:
HbA1C (intergroup difference: − 0.66%, 95% CI: − 0.97, − 0.35)*
Adjusted mean difference between groups –
Vegetables – NS
Fruit – NS
Sugar beverages – decreased in both groups (NS)
Sugar in tea—NS
Thadchanamoorthy et al. [27] After 3 mon vs baseline:
HbA1C: Control − 0.31%, standard diet + 0.03%, LCD: + 1.07% (p < 0.001)
BMI: Control + 0.21 kg/m2, standard diet + 0.04 kg/m2, LCD − 1.17 kg/m2 (p = 0.007)
SBP, DBP, TC, TG – NS
After 6 months vs baseline:
SBP: Control + 9.2 mmHg, Standard diet + 1.6 mmHg, LCD − 5.8 mmHg (p = 0.001)
DBP: Control + 2.4 mmHg, standard diet 0.0 mmHg, LCD + 5.4 mmHg (p = 0.022)
All other outcomes NS
– No benefit of standard diet based on Sri Lankan guidelines
Varadaraj et al. [28] Results at baseline, 3 and 6 months:
HbA1C: Intensive 7.6%, 7.0%, 6.4% (p < 0.001), intermittent 7.7%, 7%, 7.5% (p < 0.001), control 7.2%, 7.5%, 7.9% (p < 0.001)
BMI: intensive 27.1, 26.6, 26.0 kg/m2 (p < 0.001), intermittent 27.5, 27.0, 27.4 kg/m2 (p < 0.001), control 26.7, 27.0, 27.5 kg/m2 (p < 0.001)
– –
AUC area under the curve, BG blood glucose, BMI body mass index, C control group, DBP diastolic blood pressure, FBG fasting blood glucose, GI glycemic index, HbA1C glycosylated hemoglobin, HDL-C high-density lipoprotein cholesterol, I intervention group, LCD low carbohydrate diet, LDL-C low-density lipoprotein cholesterol, NS not significant, PA physical activity, PPG post-prandial glucose, PPP powerpoint presentation intervention, SBP systolic blood pressure, SLM self-learning module intervention, TC total cholesterol, TG triglyceride
*These data appeared in a letter, which was excluded by our search strategy. Shariful Islam SM, Niessen LW, Ferrari U, et al. Effects of Mobile Phone SMS to Improve Glycemic Control Among Patients With Type 2 Diabetes in Bangladesh: A Prospective, Parallel-Group, Randomized Controlled Trial. Diabetes Care 2015;38:e112-3
Glycemic Outcomes
In 2 RCT of 40 and 80 T2D participants that were well-matched at baseline [17, 21], people in the intervention arm were advised to include low-GI whole grain cereals and traditional foods and compared to usual-diet controls for 24 weeks, with periodic support from a dietitian. In both trials, the low-GI group significantly reduced FBG and PPG. Compared with controls, low-GI participants exhibited a significant lowering of HbA1C (− 0.9%). In a pilot case–control trial, researchers developed Indian-style low-GI foods that enabled blood glucose to return to baseline within 2 h of consumption in people with T2D [26]. They then reported that consuming low-GI meals and snacks for 4 weeks in T2D participants lowered HbA1C by − 0.9% compared to baseline [24]; however, this could not be conclusively attributed to the intervention as the study duration was < 12 weeks. FBG was improved [24]. Another non-randomized study used diet, naturopathy and yoga in participants admitted to a naturopathy hospital for 15–30 days. They were provided plant-based low-GI meals, performed yoga and other exercises, received comprehensive education, and one-on-one counselling with their physician daily [23]. At 3-month follow-up, participants showed significant declines in HbA1C, FBG, and PPG compared with baseline; however, 55% of those enrolled were not followed up. A dose–response relationship between compliance to dietary practices and HbA1C was observed with mean reductions of 0.4%, 1.1%, and 1.7% among those with poor, moderate, and excellent compliance to dietary practices, respectively. A majority of participants (65%) achieved a reduction in HbA1C with 19% stopping medication [23]. However, given the comprehensive lifestyle intervention, biological benefits cannot be fully attributed to the low-GI diet.
Lipid and Blood Pressure Management
Evidence from the above studies also indicates some beneficial effects of a low-GI diet on lipid profile (Table 2). An RCT of 80 participants found a sustained significant reduction in triglyceride (TG) compared with controls [21] although the smaller (n = 40) trial did not [17]. No effect on blood pressure (BP) was detected [21]. In a single-arm intervention, a low-GI diet elicited significant improvement in TG, total cholesterol, high-density lipoprotein-cholesterol (HDL-C), and low-density lipoprotein-cholesterol (LDL-C) compared with baseline [24].
Anthropometrics
One RCT (n = 40) [17] reported a significant reduction in BMI in the low-GI group but this was not replicated in the larger trial [21].
Summary
Overall, the 4 studies of a low-GI diet lasting 4 weeks to 6 months found significant reductions in HbA1C or other glycemic outcomes [17, 21, 23, 24]. Other cardiovascular risk factors were inconsistently measured, precluding conclusions. While 2 were RCT, they were conducted by the same group and had a total enrolment of 120 participants [17, 21].
Education and Counselling Interventions
Thirteen articles focused on nutrition education or counselling as the primary intervention (Table 1) with the outcomes reported in Table 2. Group education sessions were the primary intervention in 4 articles (3 studies) [19, 25, 30, 31] whereas individual counselling methods were employed in 5 interventions [14–16, 22, 27]. Two articles investigated the use of mobile phone-based interventions [18, 20] while 1 study each focused on self-learning modules [28] and the development of educational materials [29].
Glycemic Outcomes
Nine articles reported on glycemic management in participants. In an RCT employing individual nutrition counselling, the HbA1C of participants receiving education on low carbohydrate diet for 3 months was not different from controls at 3- or 6-month follow-up despite excellent retention (84–96%) in the intervention. However, no measure of compliance was reported [19]. An RCT with a pre-post evaluation of self-learning modules versus educator-led instruction supported by PowerPoint presentations reported that the proportion of individuals meeting blood sugar targets increased similarly in both groups after 6 months [28]. In a single-arm, 3-month trial, individual counseling on diet, exercise, and medication elicited improvement in FBG in participants compared with baseline but no information on dietary intake was provided [22].
The intensity (frequency) of educational intervention was compared in a 3-arm trial [27]. In one group, participants received intensive medical nutrition therapy from dietitians once per month for 6 months and experienced a sustained decrease in HbA1C up to 6 months. Participants who received intermittent nutritional therapy (2 sessions over 6 months) had improvements in HbA1C only at 3 months, with benefits attenuating by 6 months. Conversely, a control group taken from a different clinic receiving usual care recorded an increase in HbA1C [27].
The importance of providing comprehensive education (nutrition plus other health behavior information) was unclear in a pre-post study in which one-third of participants received periodic, ongoing, in-person counselling on diet and physical activity. Although declines in FBG and PPG relative to baseline were reported, HbA1C was unchanged. Participants receiving periodic ongoing counselling on diet alone showed improvements only in FPG [16]. A control group receiving only 1 education session had similar HbA1C values at 6 months as the intervention groups [16].
Approaches combining both intensive counselling and comprehensive education on diabetes management were explored in 2 RCTs. An intensive individual counseling intervention led by dietitians and grounded in evidence-based Indian nutrition practice guidelines (EBNPG) included education, motivational interviewing, goal setting, and self-monitoring but did not improve HbA1C compared with controls [14]. In another, monthly sessions for 6 months led by content experts from an urban hospital covering diet, physical activity, and other self-management practices resulted in an intergroup difference in HbA1C favoring the intervention up to 1 year of follow-up [15].
Technology was used to deliver content in 2 RCTs. The first examined comprehensive messaging on various topics including nutrition (20 out of 90 messages in total) as well as physical activity and medications delivered via text messaging and resulted in greater improvement in HbA1C among intervention than controls at 6 months [18]. Another 6-month RCT employing both text messaging and interactive voice calls, plus telephone-based access to a physician 7 days/week, 24 h/day showed improvements in FBG and PPG in the intervention group whereas in the control group only PPG was lower than baseline. However, the dropout was high, possibly due to low motivation or high cost of intervention-recommended foods [20].
Lipid and Blood Pressure Management
Among participants in an RCT who received intermittent nutritional therapy, improvement in lipids was observed only at 3 months. However, more intensive nutritional therapy showed decreases in total cholesterol and LDL-C up to 6 months [27]. One RCT intervening with either general diabetes management or nutrition education focused on a low-carbohydrate diet did not report significant intergroup differences in lipid profile. However, BP decreased in the intervention group [19]. Periodic intensive counselling on diet and exercise elicited significant decreases in total cholesterol and LDL-C compared to baseline but less intensive interventions did not [16]. Participants in a single-arm trial of a holistic lifestyle modification experienced significant reduction in TG, LDL-C, and total cholesterol, an increase in HDL-C, and reduced systolic BP [15]. Similarly, an improved lipid profile was also observed among participants who followed EBNPG recommendations [14].
Anthropometrics
Participants in a low-carbohydrate diet health education RCT had a significant reduction in BMI compared with controls [19]. In another RCT [14, 15], no intergroup differences were found whereas Varadaraj et al. [27] reported decreases in BMI compared with baseline in both more- and less-intense intervention groups but an increase in BMI in controls. In a non-randomized trial, significant improvements were observed in BMI in participants who received individual and group nutrition education but only men were included [25]. Similar results were observed in participants receiving periodic intensive counselling on either diet alone or diet and exercise [16]. However, these results might not be generalizable to the broader population, as most enrollees were of moderate-high socioeconomic status.
Summary
Four out of 9 studies of 3–6 months duration reported improvements in glycemic outcomes and 1 RCT intervention yielded sustained benefit at 12 months. Successful interventions tended to have a higher frequency of sessions along with comprehensive education encompassing dietary and other health behaviors. Where benefits on lipid or blood pressure outcomes were reported, it was usually in higher intensity interventions. Of the 6 trials that measured BMI, 2 randomized and 2 non-randomized interventions found a benefit.
Effects of Nutrition Interventions on Behavior Outcomes
Fifteen studies assessed behavior change outcomes in participants as described in Table 1 and reported in Table 2. Most behaviors were assessed subjectively using participant self-report and included parameters such as nutrition knowledge, dietary intake (some of which was objectively measured using validated questionnaires), dietary compliance, stage of change for certain behaviors, or diabetes self-management practices.
Low-GI Dietary Interventions with Educational Support
In an RCT with objectively measured dietary intake, the intervention was low-GI foods based on local, traditional whole grains with support provided by dietitians, who evaluated dietary compliance using repeated 24-h dietary recall. Compliance and acceptability of the low-GI diet were higher than the usual diet. The low-GI group consumed significantly less carbohydrate and fat as % energy at the trial conclusion [21]. A holistic diet and yoga program held in a naturopathy hospital included a variety of strategies to improve dietary intake, twice daily yoga practice plus optional physical activities and diabetes management skills in addition to the provision of low GI meals, meditation, and a variety of education sessions throughout the study period [23]. Overall, 80% of patients showed moderate to excellent dietary compliance and increased knowledge of the educational topics presented, meal preparation, and spacing. However, self-reported better adherence to yoga practice was not associated with favorable glycemic outcomes [23]. To assess the sense of satisfaction with low-GI versions of traditional Indian mixed meals, sensory evaluation of the low-GI meals with healthy and T2D participants was conducted. Satiety and hedonic ratings for flavor, appearance, texture, taste, and acceptability were similar between low-GI and original foods [26]. The low-GI meals were prescribed to T2D participants in a follow-up study for 4 weeks with good dietary compliance by 13 of 15 participants [24].
Culturally Adapted Dietary Education
An RCT of Norwegian-Pakistani women [31] demonstrated that culturally adapted group nutrition education sessions changed the participants’ intention to eat healthier. Women in the intervention group received 6 education sessions over 6–8 months. There was a significant shift from the pre-action (pre-contemplation, contemplation, and preparation) stages to the action stage in the intervention group for intentions to consume a healthier diet (reduced sugar and white flour, healthier fats, increased vegetables, fruits, and legumes). At the follow-up, all food intake variables, except fruit intake, were improved in the intervention group compared with the control group [31].
Summary
Compliance to dietary recommendations was assessed by a variety of techniques. The trials utilizing traditional foods with low GI or a focus on healthier eating patterns were generally successful in improving dietary outcomes. Other behaviors, such as physical activity, were inconsistently reported.
Barriers to Uptake of Interventions and Culturally Responsive Mitigation Strategies
Here, we present results for 5 of the most prominent barriers and culturally relevant solutions. Figure 1 depicts factors that influence dietary behaviors in the study population, and summarizes which were most important in people with T2D in their home country vs immigrants to North America/Europe. The conceptual model is adapted from Zeng et al. [32].Fig. 1 Factors influencing dietary behaviors in South Asians living in their home country or immigrants to North America or Europe. Legend—clear boxes depict factors identified both in South Asia and in immigrant settings; stippled boxes depict factors only identified in South Asia; blue boxes depict factors identified only in immigrant settings. As indicated, there is considerable overlap in the barriers and challenges faced by South Asian people with T2D, independent of where they live. Mentioned in other literature; not based on an exhaustive interview.
Adapted from the conceptual model of Zeng et al. [32] originally developed for Chinese immigrants. https://doi.org/10.3390/ijerph110706727
Diverse Languages
Several languages and dialects are spoken among people in South Asia. Five studies created educational resources, text messages, or conducted counselling in local South Asian languages. In an RCT implementing EBNPG, participants’ local dietitians provided individualized counselling in 1 of 11 Indian languages across India, focusing on dietary recommendations and behavioral strategies such as goal setting. Outcomes up to 1 year included reduced energy and carbohydrate quantity and quality and improved nutrition knowledge, increased physical activity, self-management practices, and adherence to recommendations [14]. Another intervention based on holistic lifestyle practices in a naturopathy clinic provided the intervention in the local language, supported by pictures, videos, and interviews and discussions with individuals or groups. Self-reported diet of the intervention group indicated increased intake of fiber while decreasing sugary and junk foods [15]. In a study comparing the effectiveness of education delivery by self-learning modules and educator-led PowerPoint presentations, the content was available in both English and Hindi. Other translated resources such as handouts and DVDs were also provided to participants, with both self-learning and educator-led interventions yielding increased nutrition knowledge [28].
Low Literacy
In Pakistani women living in Norway, the authors identified that the immigrant women may have lower literacy, education, and poorer language skills than natives of the host country, creating barriers to accessing healthcare services and health information [31]. Also, some women were reluctant to go out alone or speak to healthcare staff of the opposite sex without an accompanying family member. Culturally adapted and translated audiovisual materials were used in the education program. Dietary assessment was conducted in Urdu and/or Punjabi using a culturally adapted Food Frequency Questionnaire [31]. The outcomes included participants advancing through the stages of change for reducing sugar and fat intake, although intentions to increase vegetable and pulse grain intake were not increased [30].
The authors of a UK-based study created an educational resource for ethnic minority communities including Gujarati and Pakistani people. To address language and literacy barriers, it included simple messages, culturally relevant information in the form of photographs of foods commonly eaten in the selected communities, alongside their carbohydrate content. This enabled participants to understand the carbohydrate content of traditional foods and how they might affect their blood glucose [29]. In a pilot trial, participants reported increased knowledge and understanding of carbohydrate counting for traditional and commonly eaten foods while gaining an improved understanding of the impact of carbohydrates on blood glucose [29].
Personal Preferences and Family Influences
Participants in a 6-month counselling intervention indicated that adherence to the prescribed dietary recommendations was most difficult during festivals and special occasions when the consumption of certain traditional foods is essential. To manage this situation, study participants were encouraged to substitute low-fat snacks and sweets prepared with artificial sweeteners [16].
In a study among Pakistani immigrants in Norway, preferences of the children and husbands were identified as a significant barrier to dietary changes, as were social expectations to consume sweets and soft drinks [30]. Barriers limiting the intake of fruits and vegetables included high prices, taste and appearance of the food, lower quality and freshness, and decreased availability of certain foods compared to Pakistan. To mitigate the barriers, the intervention was based on the Pakistani lifestyle in Pakistan and in Norway. Acknowledging the women’s roles as mothers, wives, and as providers of food for their family [31] was successfully navigated, as the women reported improved dietary habits of themselves and their families [30].
Some participants’ diets were influenced by taboos regarding fruit and vegetable consumption. For example, “Gourd vegetables and citrus fruits were avoided during rainy season as they were believed to be cold producing foods. Many of the subjects did not consume any fruit as they believed fruits were not permitted for subjects with diabetes” [16]. Diet counselling encouraged increased intake of vegetables and allaying misconceptions about fruit. In addition, participants following a non-vegetarian diet were encouraged to include more vegetable dishes. Overall, the intake of leafy green vegetables increased after the counselling intervention [16].
Responsive Programming to Increase Uptake
In 1 study [16], family members and caregivers were invited to participate in the counselling sessions. When participants were unable to attend in person, home visits or telephone calls were provided. In the Norwegian study of Pakistani women, walking groups were created as part of the study and, to encourage participation, childcare was provided and stroller-friendly walking paths were identified [30].
Traditional Foods and Exercise
Prescriptions for a dietary recommendation that are culturally unfamiliar may be daunting to participants and prevent sustained adherence. To improve adherence to and acceptability of the dietary recommendations some studies used local, traditional foods and recipes in their recommendations. In a 24-week RCT, participants in the intervention arm were provided low-GI, local, traditional whole grains such as red rice puttu (steamed rice flour logs), whole wheat flour, barley, and rolled oats, while the control arm followed usual dietary practice [21]. Compliance to dietary recommendations was excellent in the intervention group versus the control group, but could not be attributed solely to the adoption of traditional recipes because of the meticulous phone follow-up by dietitians with the intervention group. Pande et. al. [26] reported that the use of modified meals and snacks prepared with Indian foods was associated with improved satiety, acceptance, and compliance to the recommendations in 13 out of 15 participants. Others reported that 80% of participants were compliant with dietary intake recommendations based on low-GI plant-based local foods. In some interventions, traditional yoga exercises and Pranayama (breathing exercises) to improve physical activity were utilized [15, 23] while others recommended walking [23, 31]. Other studies did not report the type of physical activity recommended, nor outcomes [16, 20].
Discussion
Medical nutrition therapy is a cornerstone of the management of T2D and, when successfully implemented, is proven to improve health outcomes and prevent comorbidities. The current literature provides consistent evidence that low-GI diets are efficacious in the management of hyperglycemia in South Asians in RCT [17, 21] and single-arm trials [23, 24]. Low-GI diets can be made culturally relevant by the inclusion of familiar whole grains, pulses, vegetables, and fruits, and the pattern is adaptable to both vegetarian and omnivorous diets. One limitation might be South Asians’ preference for refined carbohydrates including white rice and rice-based products that are consumed several times daily (i.e., food acceptability in Fig. 1) and can contribute two-thirds of the daily caloric needs [33]. However, alternative traditional whole grains, millets, and whole wheat meal preparations are well received [17]. Adapting cooking methods to reduce the available starch content may also work [26]; for example, rice cooked in excess water drained after cooking has a GI of 19, compared to white rice (GI 72) and parboiled rice (GI 57). The high GI of rice is primarily due to its low content of amylose (a resistant starch), which improves its softness and palatability [34] but also enables more rapid digestion, resulting in high blood sugar following consumption [35]. The GI of rice is also affected by other properties of starch and its interactions with other components during processing and cooking [36]. Innovative technology and research are needed to fully understand the GI properties of rice and create new palatable rice varieties with low-GI content [34].
Interventions of nutrition education and counselling sessions had mixed results on glycemic control. As expected, interventions that were more intensive (more frequent visits) and/or comprehensive had sustained improvements in HbA1C for 6–12 months [15, 27] as shown elsewhere [37]. Less intense or shorter interventions were less effective [19, 27]. Consistent with these findings in South Asian countries, a systematic review and meta-analysis of 4 nutrition education interventions conducted in low/middle-income countries reported significant benefits on HbA1C or fasting glucose at 3 but not 6 months [38]. Globally, trials including at least 10 contact hours of education produced the greatest HbA1C reduction [39]. However, participants receiving intensive counselling and education sessions may benefit from changes in dietary intake behaviors that improve adherence to the dietary recommendations [31] that may reduce the risk of complications if there is sufficient follow-up. A meta-analysis of combined dietary and physical activity interventions for type 2 diabetes reports that efficacy is generally not sustained beyond 6 months [40].
Given the substantial number of people in South Asia who have T2D [1], individual counselling will not reach everyone who needs treatment. We found no studies of public health measures to educate and raise awareness about healthier diets for T2D. Most South Asian countries have a limited capacity of government-funded hospitals to serve large populations, thus medical treatment is mostly obtained through privatized health care providers for which there is a limited medical insurance coverage (Fig. 1, healthcare system support) [41]. We see an immediate need to develop and implement large-scale public education initiatives to reach those diagnosed with T2D also applicable to those who are undiagnosed or have pre-diabetes. The use of technology offers a promising solution for addressing several barriers. The increase in web connectivity and 5G services [42], the use of cellular/smartphones and technological advances in both urban and rural areas make technology-based services a viable solution. These services are cost-effective, convenient, and can reach large audiences with a small time commitment [43]. Two articles found that the use of technology improved glycemic control and evoked positive behavior changes in South Asian people with T2D [18, 20].
We also highlighted several approaches to reducing barriers related to cultural relevance (Fig. 1, diverse languages, cultural characteristics, social support) such as the use of local foods, creating educational materials in local languages, tailoring interventions to support the lifestyle needs of families, offering supportive services such as home visits, and inclusion of family members [14, 16–21, 27, 28, 30]. Offering services like these and others to make interventions culturally relevant and meaningful to participants is likely to increase participation and compliance to program requirements. In Western countries, assessing the degree of acculturation can improve the tailoring of healthcare interventions for immigrant populations [44, 45]. Given recent adaptations to healthcare necessitated by the COVID-19 pandemic, future interventions may be able to make greater use of tele- or video-conferencing or social media [46].
Limitations and Gaps
Many studies were small, lacked control groups, or did not provide intergroup statistical analysis but several well-done RCT have been published recently. Single-arm trial outcomes need to be interpreted with caution in terms of the magnitude of results and generalizability. Many studies relied on self-report of dietary intakes (not using standardized questionnaires) and other behaviors. Low GI is the only dietary pattern reported; we were unable to find studies examining other dietary patterns. Since there are regional variations in dietary patterns within South Asian countries, other interventions such as healthy vegetarian diets, lacto-ovo vegetarian diet, pescatarian diet, or Mediterranean diet need could be adapted for segments of this heterogeneous population. One of the biggest challenges faced is the huge population affected by T2D, which also makes it a limitation because the healthcare systems of most South Asian countries are not equipped to offer universal diabetes self-management, education, and support initiatives. Few studies reported barriers from the healthcare system perspective, although lack of dietitians’ and other experts’ availability to deliver lifestyle interventions was mentioned [14, 23]. Possibly, low-GI diet messaging could be integrated into public health initiatives and broadly promoted to reach large populations. Moreover, all of the trials reported were conducted in urban settings. Approaches to support people living in rural areas need to be developed.
Conclusions
Studies from South Asian countries support approaches using low-GI dietary pattern adapted to use local foods, along with intensive counselling approaches for the treatment of T2D. Practical, public health approaches are needed in order to provide access to care because of the high prevalence of T2D in South Asian countries. In Western countries, more research is required on best practices for providing culturally relevant nutrition therapy.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 204 KB)
Funding
Work in the laboratory of CBC is funded by Dairy Farmers of Canada, Alberta Health Services, Alberta Agriculture and Forestry, Alberta Canola. DNF received a stipend from the Alberta Diabetes Institute.
Data Availability
Data are available upon reasonable request to the corresponding author.
Declarations
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
Conflict of Interest
The authors declare no competing interests.
This article is part of the Topical Collection on Diabetes and Obesity
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. International Diabetes Federation (2021) Diabetes Atlas. https://diabetesatlas.org/atlas/tenth-edition/. Accessed 4 September 2022.
2. Sohal PS Prevention and management of diabetes in South Asians Can J Diabetes 2008 32 3 206 210 10.1016/S1499-2671(08)23011-X
3. Banerjee AT Shah BR Differences in prevalence of diabetes among immigrants to Canada from South Asian countries Diabet Med 2018 35 7 937 943 10.1111/dme.13647 29663510
4. Lee JW Brancati F Yeh H-C Trends in the prevalence of type 2 diabetes in Asians versus whites: results from the United States National Health Interview Survey, 1997–2008 Diabetes Care 2011 34 2 353 357 10.2337/dc10-0746 21216863
5. Hanif W Susaria R Diabetes and cardiovascular risk in UK South Asians: an overview Br J Cardiol 2018 25 S8 S13 10.5837/bjc.2018.s08
6. Subhan FB Chan CB Diet quality and risk factors for cardiovascular disease among South Asians in Alberta Appl Physiol Nutr Metab 2019 44 8 886 893 10.1139/apnm-2018-0868 31172794
7. Hanif W Ali SN Bellary S Patel V Farooqi A Karamat MA Saeed M Sivaprasad S Patel K Khunti K Pharmacological management of South Asians with type 2 diabetes: consensus recommendations from the South Asian Health Foundation Diabet Med 2021 38 4 e14497 10.1111/dme.14497 33301625
8. Misra A Ramchandran A Jayawardena R Shrivastava U Snehalatha C Diabetes in South Asians Diabet Med 2014 31 10 1153 1162 10.1111/dme.12540 24975549
9. Kapoor N Sahay R Kalra S Bajaj S Dasgupta A Shrestha D Dhakal G Tiwaskar M Sahay M Somasundaram N Reddy R Bhattacharya S Reddy VB Viswanathan V Krishnan D Baruah M Das AK Consensus on medical nutrition therapy for diabesity (CoMeND) in adults: a South Asian perspective Diabetes Metab Syndr Obes 2021 14 1703 1728 10.2147/DMSO.S278928 33889005
10. Kasturia S Ali MK Narayan VKM Tandon N Shivashankar R Garg V Kapoor D Mohanasundaram A Mohan D Kadir MM Prabhakaran D Mohan V Jaacks L Diets for South Asians with diabetes: recommendations, adherence, and outcomes Asia Pac J Clin Nutr 2018 27 4 823 831 10.6133/apjcn.112017.03 30045427
11. Sohal T Sohal P King-Shier KM Khan NA Barriers and facilitators for type-2 diabetes management in South Asians: a systematic review PLoS ONE 2015 10 9 e0136202 10.1371/journal.pone.0136202 26383535
12. Bhurji N Javer J Gasevic D Khan NA Improving management of type 2 diabetes in South Asian patients: a systematic review of intervention studies BMJ Open 2016 6 4 e008986 10.1136/bmjopen-2015-008986 27098819
13. Navodia N Wahoush O Tang T Yost J Ibrahim SM Sherifali D Culturally tailored self-management interventions for South Asians with type 2 diabetes: a systematic review Can J Diabetes 2019 43 445 452 10.1016/j.jcjd.2019.04.010 31375180
14. Myers EF Trostler N Varsha V Voet H Insights from the Diabetes in India Nutrition Guidelines Study: adopting innovations using a knowledge transfer model Top Clin Nutr 2017 32 1 69 86 10.1097/TIN.0000000000000089 28232773
15. Kumari G Singh V Jhingan AK Chhajer B Dahiya S Effectiveness of lifestyle modification counseling on glycemic control in type 2 diabetes mellitus patients Curr Res Nutr Food Sci 2018 6 1 70 82 10.12944/CRNFSJ.6.1.07
16. Krishnan D Gururajan R Hafez-Baig A Kondalsamy-Chennakesavan S Wickramasinghe N Gururajan R The impact of diet counselling on type 2 diabetes mellitus: an Indian case study J Diabetes Metab 2015 6 10 1000610 10.4172/2155-6156.1000610
17. Pavithran N, Kumar H, Menon AS, Pillai GK, Sundaram KR, Ojo O. The effect of a low GI diet on truncal fat mass and glycated hemoglobin in South Indians with type 2 diabetes-a single centre randomized prospective study. Nutrients. 2020;12:179. 10.3390/nu12010179.
18. Islam SMS George ES Maddison R Effectiveness of a mobile phone text messaging intervention on dietary behaviour in patients with type 2 diabetes: a post-hoc analysis of a randomised controlled trial Mhealth 2021 7 10 10.21037/mhealth-2020-3 33634193
19. Thadchanamoorthy S Gnanaselvam K Somasuriyam K Ibrahim MS Dietary intervention for glycaemic control among patients with type 2 diabetes mellitus at the medical clinic, teaching hospital, Batticaloa, Sri Lanka J Res Med Dent Sci 2021 9 8 343 350
20. Yasmin F Nahar N Banu B Ali L Sauerborn R Souares A The influence of mobile phone-based health reminders on patient adherence to medications and healthy lifestyle recommendations for effective management of diabetes type 2: a randomized control trial in Dhaka Bangladesh BMC Health Serv Res 2020 20 1 520 10.1186/s12913-020-05387-z 32513164
21. Pavithran N, Kumar H, Menon AS, Pillai GK, Sundaram KR, Ojo O. South Indian cuisine with low glycemic index ingredients reduces cardiovascular risk factors in subjects with type 2 diabetes. Int J Environ Res Public Health 17. 2020;17:6232. 10.3390/ijerph17176232.
22. Jain B Kuvera D Effect of nutritional counselling on glucose level of middle aged non-insulin dependent diabetics Asian J Home Sci 2012 7 2 472 478
23. Bairy S Kumar AM Raju M Achanta S Naik B Tripathy JP Zachariah R Is adjunctive naturopathy associated with improved glycaemic control and a reduction in need for medications among type 2 diabetes patients? A prospective cohort study from India BMC Complement Altern Med 2016 16 1 290 10.1186/s12906-016-1264-0 27534941
24. Pande A Krishnamoorthy G Moulick ND Hypoglycaemic and hypolipidaemic effects of low GI and medium GL Indian diets in type 2 diabetics for a period of 4 weeks: a prospective study Int J Food Sci Nutr 2012 63 6 649 658 10.3109/09637486.2011.649247 22229934
25. Singh U Kochhar A Impact of nutrition education on anthropometric and knowledge, attitudes and practices (KAP) score on nutritional status of non-insulin dependent diabetics J Hum Ecol 2013 41 2 157 163 10.1080/09709274.2013.11906563
26. Pande A Krishnamoorthy G Moulick N Effect of redesigned Indian mixed meals on blood glucose and insulin levels in normal versus type 2 diabetic subjects–a comparative study Int J Food Sci Nutr 2011 62 8 881 892 10.3109/09637486.2011.591368 21696299
27. Varadaraj G Sangeetha B Nithiya DR Dixit PS Effectiveness of medical nutritional therapy in the management of type 2 diabetes mellitus J Assoc Physicians India 2021 69 1 11
28. Devi R Comparison of two educational approaches on knowledge regarding diet among type 2 diabetes patients in East Delhi Indian J Publ Health Res Devel 2021 12 2 312 321
29. St. John J, on behalf of the World Foods team The journey towards building engaging dietary resources for BAME communities Diabetes Prim Care 2019 21 13 16
30. RabergKjollesdal MK Telle Hjellset V Bjorge B Holmboe-Ottesen G Wandel M Barriers to healthy eating among Norwegian-Pakistani women participating in a culturally adapted intervention Scand J Public Health 2010 38 5 Suppl 52 59 10.1177/1403494810378923 21062839
31. Johansen KS Bjorge B Hjellset VT Holmboe-Ottesen G Raberg M Wandel M Changes in food habits and motivation for healthy eating among Pakistani women living in Norway: results from the InnovaDiab-DEPLAN study Publ Health Nutr 2010 13 6 858 867 10.1017/S1368980009992047
32. Zeng B Sun W Gary RA Li C Liu T Towards a conceptual model of diabetes self-management among Chinese immigrants in the United States Int J Environ Res Public Health 2014 11 6727 6742 10.3390/ijerph110706727 24978878
33. Radhika G Sathya RM Ganesan A Saroja R Vijayalakshmi P Sudha V Mohan V Dietary profile of urban adult population in South India in the context of chronic disease epidemiology (CURES-68) Public Health Nutr 2011 14 4 591 598 10.1017/S136898001000203X 20701818
34. Huang M Hu L-Q Low glycemic index: the next target for rice production in China? J Integr Agric 2021 20 6 1727 1729 10.1016/s2095-3119(20)63299-3
35. Ohtsubo K Nakamura S Maeda S Kobayashi A Yamazaki A Watanabe S Possibility of diabetes prevention by high-amylose rice and super hard rice J Diabetes Obes 2016 3 1 1 7 10.15436/2376-0949.16.728
36. Kaur B Ranawana V Henry J The glycemic index of rice and rice products: a review, and table of GI values Crit Rev Food Sci Nutr 2016 56 2 215 236 10.1080/10408398.2012.717976 25590950
37. Pimentel GD Portero-McLellan KC Oliveira EP Spada AP Oshiiwa M Zemdegs JC Barbalho SM Long-term nutrition education reduces several risk factors for type 2 diabetes mellitus in Brazilians with impaired glucose tolerance Nutr Res 2010 30 3 186 190 10.1016/j.nutres.2010.03.003 20417879
38. Guilbert E Perry R Whitmarsh A Sauchelli S Short-term effectiveness of nutrition therapy to treat type 2 diabetes in low-income and middle-income countries: systematic review and meta-analysis of randomised controlled trials BMJ Open 2022 12 3 e056108 10.1136/bmjopen-2021-056108 35273056
39. Pillay J Armstrong MJ Butalia S Donovan LE Sigal RJ Vandermeer B Chordiya P Dhakal S Hartling L Nuspl M Featherstone R Dryden DM Behavioral programs for type 2 diabetes mellitus: a systematic review and network meta-analysis Ann Intern Med 2015 163 11 848 860 10.7326/M15-1400 26414227
40. Cradock KA OLaighin G Finucane F Gainforth HL Quinlan LR Martin Ginis KA Behaviour change techniques targeting both diet and physical activity in type 2 diabetes: A systematic review and meta-analysis Int J Behav Nutr Phys Act 2017 14 1 18 10.1186/s12966-016-0436-0 28178985
41. Tikkanen R, Osborn R, Mossialos E, Djordjevic A, Wharton GA. International health care system profiles: India. International Commonwealth Fund. 2020. https://www.commonwealthfund.org/sites/default/files/2020-12/2020_IntlOverview_INDIA.pdf. Accessed 4 September 2022.
42. Corner S. The state of 5G in South Asia 2022, country-by-country guide. NetworkWorld. 2022. https://www.networkworld.com/article/3652228/the-state-of-5g-in-south-asia-2022-country-by-country-guide.html. Accessed 4 September 2022.
43. Li D 5G and intelligence medicine-how the next generation of wireless technology will reconstruct healthcare? Precis Clin Med 2019 2 4 205 208 10.1093/pcmedi/pbz020 31886033
44. Deng F, Zhang A, Chan CB. Acculturation, diet acceptability, and diabetes management among Chinese in North America. Front Endocrinol (Lausanne) 2013;4(Aug 27):108. 10.3389/fendo.2013.00108.
45. Deng F Chan CB Nutrition interventions for type 2 diabetes in Chinese populations: a scoping review J Immigr Minor Health 2019 21 6 1416 1431 10.1007/s10903-018-0845-z 30515634
46. Leong CM Lee T-I Chien Y-M Kuo L-N Kuo Y-F Chen H-S Social media-delivered patient education to enhance self-management and attitudes of patients with type 2 diabetes during the COVID-19 pandemic: randomized controlled trial J Med Internet Res 2022 24 3 e31449 10.2196/31449 35319478
| 36443548 | PMC9707124 | NO-CC CODE | 2022-12-01 23:20:20 | no | Curr Nutr Rep. 2022 Nov 29;:1-17 | utf-8 | Curr Nutr Rep | 2,022 | 10.1007/s13668-022-00446-9 | oa_other |
==== Front
Neurosci Bull
Neurosci Bull
Neuroscience Bulletin
1673-7067
1995-8218
Springer Nature Singapore Singapore
986
10.1007/s12264-022-00986-8
Review
Triggering of Major Brain Disorders by Protons and ATP: The Role of ASICs and P2X Receptors
http://orcid.org/0000-0003-4058-1834
Cherninskyi Andrii [email protected]
http://orcid.org/0000-0001-5525-4137
Storozhuk Maksim
http://orcid.org/0000-0003-3288-4974
Maximyuk Oleksandr
http://orcid.org/0000-0003-2300-2431
Kulyk Vyacheslav
http://orcid.org/0000-0003-3342-9930
Krishtal Oleg
grid.418751.e 0000 0004 0385 8977 Bogomoletz Institute of Physiology of National Academy of Sciences of Ukraine, Kyiv, 01024 Ukraine
29 11 2022
118
15 2 2022
14 8 2022
© Center for Excellence in Brain Science and Intelligence Technology, Chinese Academy of Sciences 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Adenosine triphosphate (ATP) is well-known as a universal source of energy in living cells. Less known is that this molecule has a variety of important signaling functions: it activates a variety of specific metabotropic (P2Y) and ionotropic (P2X) receptors in neuronal and non-neuronal cell membranes. So, a wide variety of signaling functions well fits the ubiquitous presence of ATP in the tissues. Even more ubiquitous are protons. Apart from the unspecific interaction of protons with any protein, many physiological processes are affected by protons acting on specific ionotropic receptors—acid-sensing ion channels (ASICs). Both protons (acidification) and ATP are locally elevated in various pathological states. Using these fundamentally important molecules as agonists, ASICs and P2X receptors signal a variety of major brain pathologies. Here we briefly outline the physiological roles of ASICs and P2X receptors, focusing on the brain pathologies involving these receptors.
Keywords
ASICs
P2X receptors
Stroke
Chronic pain
Epilepsy
Neurodegenerative
Psychiatric diseases
==== Body
pmcIntroduction
Since water is an essential attribute of life, protons are essential ions for the most fundamental biological processes such as pH homeostasis, synaptic transmission, and the mitochondrial respiratory chain [1]. Slight fluctuations in intracellular or extracellular pH can have marked effects on protein function, synaptic vesicle trafficking, and the electrical machinery of neuronal and glial cells [1]. Apart from nonspecific interactions with biological molecules, numerous physiological processes are triggered by protons acting on their highly specific effectors—acid-sensing ion channels (ASICs) [2, 3].
Adenosine triphosphate (ATP) is well known to be a universal source of energy in cells. Less known, but also a very important function of this molecule, is the activation of specific receptors on the cell membrane, in particular, ionotropic receptors such as P2X (ATP-gated receptor cation channel family).
Both ASICs and P2X receptors are involved in numerous physiological functions [2–5]. Although the amino-acid sequences of ASICs and P2X are substantially different, both channels are trimeric complexes (Fig. 1) sharing a similar pore architecture and conducting monovalent cation currents [6].Fig. 1 Structures of ASIC1 in complex with psalmotoxin at a high pH (A) and the P2X4 channel in the closed state (B). Different colors correspond to different subunits of the channels. TM, transmembrane domain; EC, extracellular domain. Images 4fz1 and 3h9v from the Orientations of Proteins in Membranes database of the University of Michigan (CC BY 3.0) with changes.
Both ASICs and P2X receptors mediate a variety of pathologies [5, 7], the majority of which have a considerable social impact [8–11] (Table 1). Several recent reviews have summarized findings on the role of ASICs in the peripheral nervous system, in particular in nociception and proprioception [2, 12]. The structure–function relationship of ASICs [4, 13] and their evolution [14] has been described as well as developing pharmacology [15]. Similarly, the importance of P2X receptors in the peripheral nervous system, smooth muscle, vascular physiology, and pathology has been the subject of several recent reviews [16–19].Table 1 Some brain pathologies involving ASICs and P2X have a considerable social impact
Pathology Number of people affected worldwide
Ischemic stroke Stroke is ranked as the second leading cause of death worldwide with an annual mortality rate of about 5.5 million [8]
Chronic pain Globally, it is estimated that about one in five [9], or about 1.5 billion people, suffer from chronic pain
Degenerative disorders, including multiple sclerosis (MS) and Parkinson’s disease A total of 2.8 million people was estimated to live with MS worldwide [10], In 2016, 6.1 million had Parkinson's disease globally
Epilepsy Around 50 million people worldwide have epilepsy, making it one of the most common neurological diseases globally [11]
The roles of ASICs and P2X receptors in the brain have also been summarized in several reviews devoted to each of these receptors [20–25]. On the other hand, it is reasonable to review their roles in the framework of the same work. Protons and ATP serve as co-transmitters in specific brain areas, and consequently, both ASICs [26–30] and P2X receptors [31, 32] mediate some components of the postsynaptic response. ASICs [26, 28, 33, 34] and P2X receptors [35] are also involved in synaptic plasticity, which makes them important players in this brain function. Furthermore, ASICs and P2X receptors have been shown to form molecular complexes with one another, suggesting the possibility of their functional interaction(s) [36, 37]. In this review, we briefly outline the physiological roles of ASICs and P2X receptors and the brain pathologies involving these receptors.
General Information on ASICs and P2X Receptors
ASICs
ASICs are proton-gated voltage-independent ion channels that are ubiquitous in the nervous system. The existence of receptors for protons was first revealed in 1980 as currents activated by extracellular acidification of the neuronal membrane [38]. They were named acid-sensing ion channels after being cloned in 1997 [39]. ASICs are members of the epithelial Na+ channel/degeneration family of amiloride-sensitive ion channels [3].
Currently, at least eight subunits encoded by five genes (ASIC1-5) are known. Splice variants of the ASIC1 gene give ASIC1a, ASIC1b, and ASIC1b2 subunits; and those of ASIC2 give ASIC2a and ASIC2b subunits [40–43]. There are no known splice variants of the ASIC3–5 genes. Almost all subunits of these channels are expressed in the mammalian brain, except for ASIC1b [44] and ASIC1b2 [45]. Of them, the most common are ASIC1a, ASIC2a, and ASIC2b [46]. The subunits form homo- or heterotrimeric complexes working as channels with pH-sensing properties [46].
ASIC1a homomers are more sensitive to extracellular protons than other subunits and their combinations [41, 47, 48]. ASIC2b, ASIC4, and ASIC5 are not activated by protons at all [40, 49, 50].
ASICs are found in most regions of the mammalian brain: cerebral cortex, cerebellum, brain stem, amygdala, hippocampus, diencephalic structures, striatum, and olfactory bulb, among other brain structures [24].
Under physiological conditions, the main stimulus known so far to activate ASICs is a rapid drop in pH. ASIC1a is of particular importance since it can be activated by acidosis in the physiological range. In a laboratory setting, the activity of ASICs can be altered by several natural or synthetic substances: amiloride and A-317567 (non-discriminative low-affinity blockers), 2-guanidine-4-methylquinazoline (a pH-independent activator), ammonium ions (an activator), the spider peptide psalmotoxin-1 (PcTx1, complex state-dependent effects on ASIC1), the snake venom components mambalgins (blockers), novel compound 5b (an orthosteric antagonist of ASIC1a), and others [7, 51–53]. Several endogenous substances such as spermine [54], histamine, or corticosterone [55] also modulate the activity of ASICs.
At the periphery, ASICs are primarily expressed in sensory neurons that innervate tissues and organs, and participate in sensory processes, including mechanotransduction, chemoreception, and nociception [2]. Within the brain, ASICs appear to play important roles in synaptic transmission, and its modulation and plasticity [24]. Briefly, the rapid acidification occurring during synaptic transmission [26, 56] activates ASICs on postsynaptic neurons in several brain structures [26–30, 57], implying that some component of postsynaptic current is mediated by ASICs with protons acting as co-transmitters. Before this component was identified, it was already known that ASICs play a functional role in central synapses. Their activation regulates the synaptic strength of glutamatergic and GABAergic synapses and/or contributes to use-dependent synaptic plasticity [24]. Depending on the pattern of neuronal activity, the structure of a given brain region, and downstream signaling pathways, ASICs may contribute to both short-term [29, 58, 59], and long-term [26, 33, 34, 52, 58, 60, 61] plasticity. The involvement of ASICs in the functioning of synapses in the amygdala, hippocampus, striatum, and cingulate cortex determines their role in higher-order functions like learning, memory, and fear conditioning. A detailed review of the behavioral effects of ASIC activation or inactivation can be found in [24]; here we describe the main known effects.
Pharmacological inhibition or genetic elimination of ASICs negatively affects motor performance in tests of high difficulty [57], and, interestingly, this effect decreases after training. Also, ASICs contribute to the freezing component of innate fear reactions. Disruption of ASICs in the amygdala decreases the ability to form long-term potentiation (LTP), which manifests as deficits in both cue and context fear conditioning [33, 44, 62]. In the ventral hippocampus, inhibition of ASIC1a leads to impairment of fear extinction [63], which is mediated by the alteration of NMDA receptors. ASICs influence not only the negative reinforcement system but also the positive. Global (genetic) disruption of ASIC1a leads to a decrease in learning performance in the classical conditioning paradigm based on positive stimuli [64]. In the nucleus accumbens, ASIC1a emphasizes cocaine-seeking behavior [65, 66].
The pH-sensing properties of ASICs determine their role in controlling brain homeostasis related to CO2 concentration. Systemic blockade or deletion of ASIC1a decreases hypercarbia-induced freezing [62, 67]. Reconsolidation of related memories also depends on the proper functioning of ASICs [68]. Interestingly, suppressing the activity of ASIC2 and ASIC3 in the locus coeruleus disturbs sleep structure [69]. This region is known to be a CO2-dependent modulator of respiration and sleep-wakefulness cycles.
In general, the contribution of ASICs to complex forms of behavior and brain homeostasis is essential, modulatory, and region-dependent.
P2X
The idea of ATP-mediated intercellular signaling appeared almost immediately after the discovery of this molecule in 1929 [70]. The first information on the release of ATP from sensory nerves was obtained by Pamela Holton in the 1950s [71, 72]. In 1972, Geoffrey Burnstock proposed the existence of "purinergic" nerves [73], thus founding a new research field. We now know that extracellular ATP signaling is widespread among different physiological systems [74]. Much of the early data on purinergic signaling were obtained from experiments on smooth muscles. In 1983, it was shown that extracellular ATP causes depolarizing ionic current in mammalian sensory neurons [75, 76] due to the opening of the membrane ion channels [77, 78] and can mediate rapid synaptic transmission in both peripheral and central parts of the nervous system [74]. Extracellular ATP activates both inotropic P2X and metabotropic P2Y receptors [74]; the latter are not reviewed here.
P2X receptors are ligand-gated ion channels that trigger their open state by binding ATP or related agonists. In mammals, seven genes encoding P2X receptor subunits are known [79]. Their amino-acid sequences are 40%–50% identical. The length of P2X1–6 subunits is 388–471 amino-acids, while the P2X7 receptor is larger (595 amino-acids). Each subunit contains two hydrophobic transmembrane domains separated by an ectodomain (~ 280 amino-acids). Functional channels are formed as trimers. The pore is formed by six transmembrane domains (two from each subunit). The cytoplasmic part of the channel is formed by the N- and C- termini of the subunits [79]. Due to the larger size of the P2X7 subunit, its intracellular part contains additional domains [80]. The trimers can be homomeric (P2X1, P2X2, P2X3, P2X4, P2X5, or P2X7, but not P2X6) or heteromeric (except P2X7). Common heteromeric variants are P2X1/2, P2X1/4, P2X1/5, P2X2/3, P2X2/6, and P2X4/6 [81]. P2X receptors are widely expressed in the cells of various tissues: smooth (P2X1) and skeletal (P2X5) muscles, immune cells (P2X7), as well as, critical for this review, neurons, glial cells, and brain microvessels (P2X2, P2X3, P2X4, and P2X6) [79, 82].
All P2X receptors are permeable to small monovalent cations, while some have significant permeability to Ca2+ (the highest reported for P2X4) [79, 83]. Homomeric P2X7 receptors can transmit even relatively large organic cations such as some fluorescent dyes [83]. However, P2X5 receptors are permeable to Cl– anions too [84].
After activation, P2X receptors undergo desensitization. Based on the speed of this process, we distinguish two groups: with fast (< 1 s; P2X1 and P2X3) and slow (> 20 s; P2X2, P2X4, P2X5, and P2X7) onset [79, 83, 85]. The molecular mechanisms controlling desensitization are insufficiently studied. The available data indicate that this process is determined mainly by the properties of the channel's cytoplasmic cup [80, 86].
The main physiological agonist of P2X receptors is ATP. The efficient concentration range to activate most P2X receptors is 0.5–12 μmol/L. The exception is P2X7, which requires a significantly higher concentration of up to 100 μmol/L [79]. Other efficient agonists of P2X receptors are ATP derivatives: 2-meSATP, ATPγS, α,β-meATP, βγ-meATP, BzATP, and the like. Many novel compounds act as non-selective (suramin, PPADS, and RB-2) or selective antagonists. A detailed review of the pharmacological properties of P2X receptors can be found in [79].
The wide expression of P2X receptors, in particular their presence on smooth muscle cells (including blood vessels), immune cells, neurons, and glia, determines the broad variety of their physiological effects. Numerous ATP effects are carried out in both the efferent and sensory parts of the peripheral nervous system. ATP is released as the co-transmitter in the terminals of autonomic fibers (along with acetylcholine or norepinephrine), where it acts mainly through the activation of P2X1 receptors. P2X2 and P2X3 receptors of gustatory afferents detect the ATP released from taste receptor cells [87]; modulate the activity of the auditory system [88]; and induce the excitation of nociceptors [89]. In the central nervous system (CNS), P2X receptors contribute to excitatory intercellular communication in the locus coeruleus [90, 91], hippocampus [31, 35, 92], spinal cord [93], hypothalamus [94, 95], and cortex [32, 96]. Of particular interest are P2X4-containing receptors, which have the highest permeability to Ca2+. In the presynapse, their activation may trigger Ca2+-dependent signaling, enhancing, for example, the release of glutamate [97]. The release of ATP under physiological conditions facilitates LTP in the hippocampus, and this facilitation is mediated by Ca2+ entry through P2X4 receptors followed by the downregulation of postsynaptic NMDA receptors [98]. Also, P2X4 receptors are significantly expressed in microglial cells and are involved in immune function [99].
P2X receptors are involved in the interaction of neurons and glial cells and hence may play a role in the development, maintenance, and plasticity of neuronal circuits [100]. In particular, astrocytic P2X7 receptors have the potential to cause rapid non-vesicular glutamate release [101]. Microglial P2X7 receptors are important for immune responses, particularly in the CNS. The role of P2X receptors in pathological processes is discussed in more detail in the following section.
ASICs in Brain Pathology
As outlined in the previous section, ASICs are widely expressed in the mammalian nervous system in general and the brain in particular. The list of established functions of this receptor family is already long and is rapidly increasing (for instance, reviewed in [102, 103]). ASICs are involved in many important physiological functions, as well as contributing to a variety of pathologies. This is not surprising given that tissue acidosis is a key contributor to neuronal cell death in many neurological diseases, whereas ASICs, namely subtype ASIC1a, are pivotal in acidosis-induced damage to brain neurons. [104, 105]. Chronic pain [106], ischemic stroke [4], epileptic seizures [106, 107], degenerative disorders including multiple sclerosis (MS), Parkinson’s disease (PD), and even neonatal hyperbilirubinemia [108] are just a few examples of pathologies involving ASICs. Interestingly, apart from the well-established role of peripheral ASICs in pain, there is emerging evidence that ASIC1a located in the brain is an important player in pain processing [28].
Some of these pathologies affect millions of people worldwide and are still awaiting adequate pharmacological tools for their cure (for instance up to 40% of epilepsy cases are resistant to currently available drug therapy). Emerging evidence from intensive research strongly suggests that the pharmacological modulation of ASICs is a very promising strategy for the treatment of these pathologies.
In this section, we briefly review evidence regarding the role of ASICs in acid-induced toxicity in relation to some brain pathologies. In addition, we outline a potential explanation for the following paradox. ASICs, especially ASIC1a, are typically activated by a rapid drop in pH, whereas slow acidification results in steady-state desensitization without inducing detectable currents [109, 110] (Fig. 2). On the other hand, acidification associated with pathological states is prolonged and typically slow, and thus should not activate ASICs. Still, inhibition of ASICs (especially ASIC1a), is protective in multiple models of pathological states reported in numerous studies. How to account for this contradiction?Fig. 2 Slow acidification results in the steady-state desensitization of ASIC currents. ASIC-currents are evoked by acidification with similar amplitudes (from pH 7.4 to pH 6.0) but with different rise times. Upper panel, the pH of the applied solution was measured using pH-sensitive fluorescent dye; the 10–90% fall-time of the fluorescence change corresponds to the interval between the two squares (corresponding values are shown above the traces). Lower panel, ASIC currents evoked by these acidifications. Notice the striking reduction of the amplitudes of ASIC currents evoked by slower acidification. Adapted from [110] (part of Figure 4) published under the terms of the Creative Commons Attribution License (CC BY).
Copyright © 2020 Alijevic, Bignucolo, Hichri, Peng, Kucera and Kellenberger.
Ischemic Stroke
Acidosis is a common feature of ischemia and is assumed to play a critical role in brain injury [111]. Important evidence regarding the role of Ca2+-permeable ASICs in acidosis-induced brain injury was reported in 2004 by Xiong et al. [104]. In particular, it was found that in focal ischemia, intracerebroventricular injection of ASIC1a blockers or knockout of the ASIC1a gene protects the brain from ischemic injury more potently than glutamate antagonism [104]. Since then, the pathological role of ASIC1a in stroke has been confirmed in many studies and extensively reviewed [7, 111, 112]. Still, the exact mechanism of involvement of ASIC1a in stroke requires further investigation because at least two studies [105, 113] that do confirm the important role of ASIC1a in an in vitro model of acidotoxicity and an in vivo model of ischemic stroke, suggesting that this role involves an ion-conducting independent function of ASIC1a [105]. Authors [105] have reported that extracellular protons trigger a novel form of neuronal necroptosis via ASIC1a. Indeed, they identified serine/threonine kinase receptor interaction protein 1 (RIP1) as a critical component of this form of neuronal necroptosis. Acid stimulation recruits RIP1 to the ASIC1a C-terminus, causing RIP1 phosphorylation and subsequent neuronal death [105]. Arguments used to support the point that the role of ASIC1a is independent of the ion-conducting function include the following [105]:Pretreatment of cultured mouse cortical neurons with mild acidification (pH 6.8) greatly suppresses the ASIC currents evoked by greater acidification (pH 6); however, this fails to prevent acid-induced neuronal death [105].
Removing Ca2+ from the treatment solution has no clear protective effect on acid-induced neuronal death [105].
Although co-administration of the non-specific ASIC blocker amiloride with acid significantly inhibits ASIC-currents, it does not show neuroprotection [105].
This study has been recently expanded by revealing an auto-inhibitory mechanism that regulates interactions between the N-terminus (NT) and C-terminus (CT) of the ASIC1a protein, in particular, by using fluorescence resonance energy transfer [113]. Disruption of this auto-inhibition underlies conformational signaling of ASIC1a to induce neuronal necroptosis; a synthetic membrane-penetrating peptide representing the distal NT of ASIC1a has been identified as protective in both acidosis-induced necroptosis in vitro and a mouse model of ischemic stroke [113]. These results suggest a therapeutic potential of targeting the auto-inhibition of ASIC1a for neuroprotection against acidotoxicity [113]. This study also provided additional evidence that the role of ASIC1a in necroptosis is independent of the ion-conducting activity. Indeed, acid treatment mediates cell death in mutants with non-conducting ASIC1a just as in wild-type animals [113].
Chronic Pain
Chronic pain is a challenging clinical problem affecting millions of people worldwide. It is well established that ASICs play important roles in nociception at peripheral and spinal neurons [2, 12]. However, the role of these channels in pain perception at the supraspinal level is less studied. Still, recent findings [28] indicate that brain ASIC1 does contribute to pain perception, making it an attractive therapeutic target for chronic pain management. Several lines of evidence support an important role of cortically-located ASIC1a channels in pain hypersensitivity by promoting long-term synaptic potentiation in the anterior cingulate cortex [28]. In particular, region-specific genetic deletion or pharmacological blockade of ASIC1a results in attenuated heat and mechanical hypersensitivity in the CFA model of inflammatory pain, while at the cellular level it reduces the probability of LTP induction in this brain area [28]. It has been concluded that ASIC1a critically gates pain sensitivity by regulating the probability of cingulate LTP induction, in which PKCλ serves as a crucial mediator [28].
Epilepsy
Epilepsy, a disorder of unprovoked seizures, is a multifaceted disease affecting individuals of all ages with a particular predilection for the very young and old [114]. In addition to seizures, many patients often report cognitive and psychiatric problems associated with both the seizures themselves and their therapy [114]. About 1% of the human population suffers from chronic epilepsy disorder and 30–40% of patients develop drug-resistant epilepsy. Therefore, finding new targets for pharmacological intervention to treat patients with epilepsy is very important.
During and following seizures, lactic and glutamic acids are released into the extracellular space, causing a significant fall in pH that activates ASICs [115–117]. These results imply that ASICs may be involved in the regulation of epileptic discharges. On the other hand, the exact role of ASICs in such discharges is rather controversial. Some authors suggest that acidification diminishes seizure severity due to the more pronounced ASIC-dependent activation of inhibitory GABAergic interneurons by acidification as compared to excitatory glutamatergic cells [118]. Consistent with this idea, ASIC density on hippocampal inhibitory neurons is higher than on principal neurons, and the loss of ASIC1a reduces postictal depression [118]. Other studies have demonstrated that inhibition of ASIC activity reduces epileptic manifestations in in vitro models of epilepsy: perfusion of hippocampal slices with amiloride or PcTx1 reduces the epileptic discharges evoked by electrical stimulation or removal of extracellular Mg2+ and hippocampal slices from ASIC1a-knockout mice are less susceptible to seizures [119]. Similarly, the application of the novel ASIC1 antagonist compound 5b to rat hippocampal slices during epileptic activity significantly and reversibly reduces the frequency of seizure discharges [107]. In addition, in an in vivo model of epilepsy, selective ASIC antagonists reduce the seizure-like activity evoked by the injection of kainic acid into the amygdala or hippocampus [107, 119].
Neurodegenerative Diseases
In spite of many differences, neurodegenerative diseases have several similarities including accumulation of misfolded proteins, damage to specific neuronal populations, neuroinflammation, and persistent acidification [120]. Neuroinflammation is a complex process that occurs in various neurodegenerative diseases such as Alzheimer's disease (AD), PD, MS, and Huntington's disease (HD) [120]. Moreover, there is growing evidence suggesting that neuroinflammation is among the important causes of neurodegenerative diseases [120–122]. Inflammatory stimuli, such as ATP, lactic acid, arachidonic acid (AA), agmatine, and hypertonicity, are able to enhance the proton-induced ASIC current [121], whereas persistent acidification accompanies neurodegenerative diseases.
In addition, it has been demonstrated that ASICs: (1) mediate a component of synaptic current at glutamatergic synapses as postsynaptic ionotropic receptors for protons [26–29]; (2) play a role in the regulation of synaptic transmission at glutamatergic [59, 123] and GABAergic [107, 124], synapses; and (3) are involved in short- and long-term plasticity [26, 28, 52, 59].
Therefore, ASICs are considered to be potential targets for the treatment of neurodegenerative diseases, and promising results have been obtained as briefly outlined below.
Alzheimer’s Disease
AD is a progressive neurodegenerative disorder, the most common form of dementia, accounting for 60–70% of all cases [125]. In spite of many attempts, currently there are still only symptomatic treatments available for this disease; no drugs capable of arresting or reversing it have so far been identified [125, 126].
While the question regarding the primary cause of the disease is still debated, it is widely accepted that the progression of AD occurs due to multiple factors. There is emerging evidence that the activation of ASICs may be one of these factors (reviewed recently [22, 120]). Here, we only note that observational studies in humans, though indirect, corroborate this view. Amiloride, a non-specific blocker of ASICs, is a well-known K+-sparing diuretic used in medical practice. The use of diuretics, including amiloride, is associated with a reduced risk of AD [127, 128]. Moreover, while all antihypertensive drugs are associated with a decrease in AD risk, K+-sparing diuretics (including amiloride) are associated with the greatest reduction of this risk [127].
A selective antagonist of ASIC1a, PcTx1, inhibits the mGlu receptor-dependent long-term depression recorded in hippocampal slices, which were either treated with Aβ oligomers or obtained from the Tg2576 mouse model of AD [129]. Curcuminoids, considered a treatment for AD [126], block ASICs [130]. In this regard, it is also interesting to note that memantine, one of the very few drugs approved for the amelioration of AD, apart from its blocking action on NMDA receptors (believed to be its primary mechanism of action) also affects ASICs [131].
Multiple Sclerosis
MS is a chronic autoimmune inflammatory disease of the CNS that involves demyelination and axonal degeneration [132]. ASIC1a is upregulated in the brain tissue of patients with MS [133]. Similarly, ASIC1a is upregulated in oligodendrocytes and axons in a mouse model of MS (acute autoimmune encephalomyelitis, EAE) [133]. The enhanced expression of ASIC1 is associated with axonal damage while blocking these channels with amiloride protects both myelin and neurons from damage in this model [133]. These results are in line with evidence obtained elsewhere [134]. After induction of EAE, ASIC1-/- mice show both a markedly reduced clinical deficit and reduced axonal degeneration compared to wild-type animals [134]. ASIC1 disruption is also protective against acidosis-mediated injury in nerve explants in vitro [134]. Importantly, evidence supporting the neuroprotective effect of amiloride has been obtained from a cohort of MS patients [135].
Parkinson’s Disease
PD is a progressive neurodegenerative disorder resulting from a pathological loss or degeneration of dopaminergic neurons in the substantia nigra of the midbrain and the development of neuronal Lewy bodies [136]. The potential role of ASICs in PD has been studied in a mouse model of degeneration induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) [137]. Amiloride has been found to protect substantia nigra neurons from MPTP-induced degeneration. Administration of the selective ASIC1a antagonist PcTx1 is also protective, resulting, however, in a much more modest effect [137].
Huntington’s Disease
HD is a neurodegenerative disease caused by CAG (cytosine, adenine, and guanine) repeat expansion in the huntingtin gene and involves multiple pathogenic mechanisms [138]. There is evidence indicating that acidification-induced activation of ASIC1a may be one of these mechanisms [139]. In particular, in the R6/2 mouse model of HD, it has been shown that the amiloride derivative benzamil increases the activity of the ubiquitin–proteasome system, resulting in enhanced degradation of soluble huntingtin–polyglutamine aggregation [139]. Similarly, blocking the expression of ASIC1a leads to an enhancement of ubiquitin–proteasome system activity and this blockade also decreases huntingtin–polyglutamine aggregation in the striatum of R6/2 mice [139].
Psychiatric Disorders
There is not much data on the role of ASICs in psychiatric disorders. Nonetheless, ASIC1a inhibition (genetic or pharmacological) reduces depression-like behavior in mice [140] and rats [141]. Thus, ASIC1a has been proposed as a potential target to combat depression. At the same time, there is no evidence of the role of ASICs in anxiety or depressive behavior in patients. Despite this fact, using the ASIC blocker amiloride in the form of a nasal spray is considered to be a treatment for anxiety disorders [142].
ASIC1/2 inhibition reduces the freezing response to hypercarbia [62, 67]. An increase in CO2 in the inhaled air can induce a panic attack. Interestingly, the genetic variation of ASIC1a and ASIC2 structure is related to the individual differences in the panic response to an increased CO2 concentration [143]. In addition, a specific ASIC1 polymorphism (rs685012) is more frequent in patients with panic disorder than in healthy controls [144].
Protective Effects of Blocking ASICs in Spite of Slow Acidification in Pathologies: Potential Explanations
The roles of ASICs in several fundamental physiological processes as well as in numerous pathologies have been well documented. However, in contrast to pathologies, the roles of ASICs in physiological processes are easier to explain because rapid (though brief) changes in pH occur under physiological conditions. In particular, the latter is applicable to synapses ([56, 145, 146], but see [147] for conflicting evidence), where rapid acidification occurring during synaptic vesicle release can activate ASICs on both pre- and postsynaptic neurons [24]. Pathologies, on the other hand, are typically associated with slow and prolonged acidification, raising the question of how rapidly inactivating ASICs can contribute substantially to these pathologies.
Evidence regarding potential mechanisms relevant to this question can be subdivided into two broad categories: (1) conformational signaling of the ASIC independent of its ion-conducting function (as suggested for the role of ASIC1a in necroptosis [105, 113]), and (2) activation and decreased steady-state desensitization of ASICs by pro-inflammatory and algesic endogenous mediators.
Regarding conformational signaling via ASIC1a, the above results of Wang and colleagues [105, 113] indicate that conformational signaling involving ASIC1a induces neuronal necroptosis in in vitro and in vivo models of ischemic stroke. This signaling recruits RIP1 and is independent of the ion-conducting function of ASIC1a [113]. At the same time, these studies do not exclude the therapeutic potential of targeting ASIC1a in ischemic stroke with known blockers of this channel. Indeed, in contrast to co-administration, pretreatment with amiloride did show neuroprotection in the same study [105]. PcTx1, which is believed to inhibit ASIC1a activation by causing steady-state channel desensitization at neutral and alkaline pH (in contrast to mild acidification) is also neuroprotective [105]. An explanation of these results taken altogether is not straightforward. In addition, it should be noted that, while no clear protective effect of removing Ca2+ from the treatment solution on acid-induced neuronal death was reported by Wang et al. [105], the opposite results were reported in another study [104]. Thus, further studies are required to address these controversial issues.
As for pro-inflammatory and algesic endogenous mediators that enhance the activity of ASICs, there is growing evidence that numerous endogenous substances (such as lactate, AA, FMRFamide-related peptides—peptides related to mammalian RFamide transmitters, dynorphins, nitric oxide, and histamine), up-modulate ASICs, and the extracellular levels of these substances are increased in pathological states (see [121, 148, 149] for reviews). For instance, endogenous cationic polyamine spermine enhances ASIC1a and ASIC1b currents by decreasing the pH sensitivity of channel steady-state desensitization [54]. Moreover, via its effect on ASIC1a, increased extracellular spermine exacerbates neuronal damage following ischemic stroke [54]. AA, a polyunsaturated fatty acid, involved in cellular signaling activities as a lipid second messenger, plays a critical role in pathological conditions such as inflammation and neurological disorders [149]. AA regulates the activity of ASICs [150]. Indeed, in the cerebellum, AA significantly enhances the sustained component of current induced by acidification and increases the fraction of cells exhibiting a sustained component [150]. It should be noted, however, that this sustained component of current is insensitive to amiloride [150], raising the question of whether it is mediated by ASICs. The ASIC currents are also potentiated by lactate, and released in ischemia, and by FMRFamide [150]. Another endogenous mediator, histamine, directly and selectively potentiates ASIC1a homomers, most effectively when applied at modest levels of acidification [151].
P2X Receptors in Brain Pathologies
P2X receptors are widely expressed by both neurons and glia in various brain regions, indicating their crucial role in physiological and pathophysiological brain function. Following world pharma’s patent applications [152, 153] and recent numerous studies [154, 155], the P2X7 receptor is the most prominent drug target for numerous CNS diseases, especially those accompanied by inflammation [155, 156]. Brain P2X7 and P2X3 receptors participate in central pain transmission. P2X7 receptors are involved in the modulation of diabetes-induced neuropathic pain [157], whereas P2X3 (traditional primary nociception receptor) expressed in the midbrain periaqueductal gray is involved in the modulation of central post-stroke pain [158]. One could think that the P2X7 subunit is the sole one involved in almost all brain disorders. However, we suggest that the role of other subtypes of these receptors in brain disorders, particularly P2X4, which is more sensitive to ATP than P2X7 [79], is still underestimated.
Neuroinflammation, Ischemia, Stroke, Epilepsy, and COVID-19
Under healthy conditions, the concentration of extracellular ATP in tissues is in the nanomolar range [159]. However, under stress, hyperexcitability, ischemia, and cellular damage, the ATP concentration increases considerably, resulting in the activation of P2X receptors. In high concentrations, extracellular ATP activates P2X7 receptors that are highly expressed in microglial cells [159–161]. This results in the activation of transcription factor NF-κB, which consequently promotes the upregulation of the pro-inflammatory cytokines, pro-IL-1β and pro-IL-18, and NLRP3 protein. It has been shown that the P2X7 receptor is one of the most potent activators of the NRLP3-associated inflammasome [160]. The latter triggers the activation of caspase-1, which causes the maturation of IL-1β and IL-18 and, consequently, increases pro-inflammatory cytokine release [160, 162]. A recent study demonstrated that inhibition of P2X7 by the selective antagonist, JNJ-55308942, suppresses the development of neuroinflammation induced in different rodent models by lipopolysaccharide, Bacillus Calmette–Guerin, or chronic stress [162].
Brain ischemia induces ATP release and the subsequent activation of P2X receptors that conduct inward cationic current through the plasma membrane. However, despite their higher sensitivity to ATP than P2X7 [79], the contribution of P2X4 receptors to this current is virtually undetermined. Acute ischemia promotes the upregulation of P2X4 receptors in microglia or infiltrating macrophages [163, 164]. Activation of these receptors modulates the inflammatory response after stroke [165]. Correspondingly, a potent inhibitor of inflammation, minocycline, attenuates the upregulation of P2X4 receptors induced by hypoxia-ischemia [166]. Specific deletion of P2X4 is acutely neuroprotective and leads to a substantial reduction of infarct volume [165]. On the other hand, it predisposes depression-like behavior chronically after stroke due reduced expression of brain-derived neurotrophic factor (BDNF) [165]. Recent data show that P2X4 receptors are required for ischemic tolerance following middle cerebral artery occlusion in mice [167]. In contrast, blocking P2X4 receptors by TNP-ATP (but not PPADS that presumably inhibits P2X1-3 and P2X5-7 receptors) in a rat model of neonatal hypoxia shows a profound neuroprotective effect, pointing to the differences in P2X receptor function in ontogenesis [168].
There is a growing body of evidence for the role of ATP as well as P2X receptors in status epilepticus (SE) and epilepsy. Some studies report that extracellular ATP increases during SE and seizures, but this effect requires further examination [169]. Mice lacking P2X4 receptors display a decrease in neuronal death after SE, although seizure activity is not altered by the knockout of the corresponding gene [170]. In addition, the administration of P2X7 antagonists has been shown to have anticonvulsive effects in several experimental models of SE and acute seizures [171–175]. Genetic deletion of P2X7 receptors and inhibition of their activity by antibodies also suppress seizures in the kainic acid model of epilepsy [171]. Following SE, there is a prominent increase in P2X7 receptors and downregulation of P2X2 receptors in hippocampal neurons, particularly granule neurons, and microglia [171, 176]. It has also been reported that P2X4 receptors are upregulated in hippocampal microglia after SE in rats [170]. Upregulation of P2X3 receptor expression has also been shown in the LiCl-pilocarpine rat model of chronic epilepsy, as well as in patients with temporal lobe epilepsy [177]. The authors found that activation of P2X3 receptors accelerates sustained repetitive firing, while their inhibition results in relatively low-frequency discharges [177]. These data were later confirmed using the pentylenetetrazole-induced animal model of epilepsy. Administration of the P2X3 antagonist NF110 significantly and dose-dependently ameliorates pentylenetetrazole-induced impairments in behavior, learning, memory, locomotion, motor activity, discrimination, neuronal damage, hippocampal inflammation, oxidative stress, and mitochondrial dysfunction. These effects of NF110 are notably abolished by administration of the selective P2X3 agonist α,β-methylene ATP [178]. Growing data indicate that certain P2X receptors may be novel drug targets for seizure control.
The ATP release induced by the SARS-CoV-2 virus has recently been proposed to play a key role in the genesis of both the main symptoms and sequelae of the novel coronavirus disease COVID-19 [179]. According to recent findings, COVID-19 is associated with a number of severe consequences, including "brain fog", depression, confusion, stroke, and more [180–182]. The metallopeptidase named angiotensin-converting enzyme 2 has been identified as the functional receptor for SARS-CoV-2. It is highly expressed in lung alveolar epithelial cells and enterocytes of the small intestine, being also present in the brain neurons, glial cells, and microglia [181–184]. P2X7 receptors, which are known to induce the secretion of pro-inflammatory cytokines in response to elevated ATP levels, are abundantly expressed in both macrophages and microglia, suggesting that these receptors can be a potential target for treating COVID-19 [185].
Neurodegenerative Disorders
Alzheimer’s Disease
Some authors have proposed that ATP participates in the inflammatory process widely described in AD [186, 187] while others suggest that ATP, leaking through cell membrane pores made by soluble amyloid-β (Aβ) peptide, could participate directly through some P2X receptors [188, 189]. Numerous reports indicate that the P2X7 receptor plays a crucial role in AD, and its inhibition ameliorates the neuronal damage induced by both neuroimmune response activation [190–192] and reactive oxygen species production [193, 194]. There has been some conflicting information on the effect of these receptors on α-secretase activity, which cleaves amyloid precursor protein within the Aβ domain, thereby preventing the generation of Aβ. Two studies have shown that inhibition of P2X7 receptors leads to an increase in α-secretase activity through inhibition of glycogen synthase kinase 3 [195, 196]. However, results from Delarasse et al. revealed the opposite effect: P2X7 receptor stimulation may enhance α-secretase activity [197]. It has been reported that systemic administration of a P2X7 receptor antagonist, Brilliant Blue G, diminishes the spatial memory impairment and cognitive deficits induced by the injection of soluble Aβ1-42 into the hippocampal CA1 region of mice, which is an animal model of AD [198]. P2X4 overexpression in the neurons enhances the toxic effect of Aβ1-42, whereas its silencing decreases cellular death after exposure to Aβ1-42, indicating that this P2X subtype also contributes to the neuronal cell death induced by Aβ [199]. A recent paper suggests that P2X2 receptor expression increases in the brain of patients with AD, so it may participate in the toxic cellular and molecular events induced by Aβ [200]. In addition, the overexpression of P2X2 increases Aβ levels in a cellular model of AD [200]. We can conclude that the P2X4 and P2X7 receptors are important prospective targets in the treatment of AD.
Parkinson’s Disease
The cause and pathogenesis of PD remain elusive. The neurotoxin 6-hydroxydopamine (6-OHDA) is widely used to induce models of PD. Recently it has been shown that 6-OHDA increases ATP release and its extracellular conversion into adenosine through CD73 upregulation in SH-SY5Y cells [201]. P2X7 receptor inhibition has neuroprotective and neuroregenerative effects in various models of PD through anti-inflammatory actions along with modulation of the microglial activation state and cytokine release [202–205]. Also, overexpression of P2X4 receptors upregulates IL-6 and exacerbates 6-OHDA-induced dopaminergic degeneration in the rat model of PD [206]. A very recent paper suggests that P2X4 receptor activation might inhibit neuronal autophagy through regulation of the BDNF/TrkB signaling pathway, leading to dopaminergic neuron damage in the substantia nigra and the further inhibition of P2X4-mediated autophagy [207]. A role for P2X1 receptors in PD pathogenesis has also been reported, perhaps identifying a new therapeutic target [208, 209]. Consequently, P2X7, P2X4, and P2X1 receptors are interesting research subjects and might be possible targets for PD treatment.
Huntington’s Disease
We found only a few reports on the role of P2X receptors in HD. The first study demonstrated that P2X7 receptor expression is substantially increased in two HD animal models, also its channel pore has augmented Ca2+ permeability, and finally, its inhibition with Brilliant Blue G lessens the motor coordination deficits and cachexia and decreases neuronal loss [210]. The most recent paper reported elevated expression of the P2X7 receptor in the human postmortem striatum as well as alteration of its splicing [211], pointing to the important role of this receptor in HD.
Amyotrophic Lateral Sclerosis
The first experimental findings on the role of P2X receptors in ALS demonstrated a significant increase in P2X7 immunoreactivity in active microglia from postmortem spinal cords of patients with sporadic ALS [212]. Similarly, increased immunoreactivity for P2X7 has been found in an animal model of familial ALS, the SOD1-G93A mouse [213], while increased P2X4 receptor immunoreactivity is associated with degenerating neurons [214]. In addition, it has been shown that the allosteric modulator of the P2X4 receptor ivermectin extends the lifespan of SOD1-G93A mice by almost 10% [215]. P2X7 is known to play a complex role in ALS [216]: while knockout of P2X7 accelerates the disease in SOD1-G93A mice [217], antagonism immediately preceding the onset of symptoms partially ameliorates the motor neuron loss and inflammation [217]. Later it has been reported that P2X7 participates in ALS pathogenesis by directly modulating autophagy and particularly the expression of autophagosome component LC3-II and the autophagy receptor p62 in SOD1-G93A microglia [218]. These findings support the idea that P2X7 receptors might be one of the top targets in a multidrug strategy for ALS treatment.
Multiple Sclerosis
MS is an autoimmune disease with an unknown etiology. It is characterized by massive infiltration of immune cells, demyelination, axonal loss, and chronic inflammation that induces ATP release in lesions. It has been reported that both P2X4 [219] and P2X7 [220] receptors are strongly overexpressed in MS lesions in humans. Recently-found genetic mutations of P2X receptors support their role in MS: a rare P2X7 variant, Arg307Gln, with absent pore formation and function protects against neuroinflammation in MS [221]; and a rare P2X7 Gly150Arg–P2X4 Tyr315Cys haplotype which inhibits P2X7-mediated phagocytosis has been suggested as a risk factor for MS [222]. EAE has been extensively used as a model of MS for many decades. It has been shown that in chronic EAE in mice, inhibition of P2X4 receptors with TNP-ATP leads to exacerbation of the clinical signs, whereas their allosteric potentiation with ivermectin has the opposite effect [223]. Blockade of P2X7 receptors results in suppression of demyelination and axonal damage, reduction of gliosis, and amelioration of neurological symptoms in chronic EAE [224–226]. These data support the crucial role of P2X4 and P2X7 in MS pathology.
Psychiatric Disorders
There is growing evidence of a possible role of P2X receptors in various psychiatric disorders including major depressive disorder (MDD), bipolar disorder (BD), schizophrenia (SCZ), anxiety, and autism spectrum disorder (ASD) [99, 154, 227, 228]. For several psychiatric diseases, such as MDD, SCZ, and BD, it has been postulated that activation of the P2X7 receptor by ATP release mediates brain inflammation [229–231].
Both P2X2 and P2X7 receptors play an important role in depression disorders. P2X2 receptors in the medial prefrontal cortex mediate the antidepressant-like effects of ATP [232], whereas P2X7 receptors activated by synaptic ATP release stimulate the hyperactivation of microglia and interleukin-1β production reported in depressive-like behaviors [231]. In turn, their inhibition by Brilliant Blue G as well as knockout of the corresponding gene has antidepressant effects [233–235]. Only a few data are available on the role of P2X7 receptors in SCZ, One study showed that the tricyclic antipsychotics prochlorperazine and trifluoperazine induce allosteric inhibition of human P2X7 receptor activity [236]. Therefore, one may speculate that the antipsychotic-induced inhibition of the P2X7 receptor contributes to therapeutic efficacy [230, 237]. P2X receptors are also involved in BD: the animal model of mania induced by chronic administration of amphetamine, pharmacological blockade with A-438079 and genetic deletion of P2X7 receptor completely reverses the increased locomotor activity induced by amphetamine [229]. A-438079 also abolishes the release of the pro-inflammatory cytokines IL-1β and TNF-α as well as lipid peroxidation in the hippocampus [229]. P2X receptor actions have also been implicated in ASD. Genetic deletion of the P2X4 receptor results in a significant reduction in sensory and social task performance, vocalization responses, and plausible deficit in the ability to extract and filter relevant information from the external milieu, suggesting that P2X4 receptors regulate information processing [238] as well as perceptual and socio-communicative functions [239]. Downregulation of P2X7 and P2Y2 receptor expression has been described as a compensatory response in an animal model of ASD [240].
In summary, the role of P2X receptors in psychiatric disorders requires additional studies, especially regarding P2X2, P2X3, P2X4, and P2X7 subtypes that are strongly associated with neurodevelopment and psychiatric disorders [241, 242].
A Plausible Link between ASICs and P2X Receptors in Brain Disorders
Apart from the differences between ASICs and P2X receptors, as previously stated, there are also similarities, including trimeric structure, similarity in the pore architecture, and ionic permeability [6]. While emerging evidence indicates interactions of ASICs and P2X receptors in sensory neurons [36, 37], their possible roles in brain functions and pathologies are still virtually unexplored. The overlap of brain pathologies mediated (potentially) by ASICs/P2X receptors (Table 2, Fig. 3) suggests that such interactions may be relevant to brain pathologies as well.Table 2 Evidence supporting the involvement of brain ASICs or P2X receptors in pathologies
Pathology ASICs P2X receptors
Ischemic stroke Inhibition of ASIC1a protects the brain from ischemic injury [104, 112], ASIC1a is involved in ischemic-induced damage [105, 113] Acute ischemia up-regulates P2X4R in microglia or macrophages [163, 164]
P2X4R inhibition has a neuroprotective effect [165, 168]
Chronic pain Brain ASIC1a contributes to the processing of pain, its block may be analgesic [28] Brain P2X7R contributes to central post-stroke pain [158] whereas brain P2X3R in diabetes-induced neuropathic pain [157]
Epilepsy Inhibition of ASIC1a reduces the probability of paroxysmal discharges [107, 112]
(However) ASIC1a activation limits the duration and progression of seizures [118]
P2X4R is up-regulated, and its deletion decreases neuronal death [170]
P2X7R inactivation has the anticonvulsive effect [171–175]
Alzheimer's disease Use of amiloride is associated with reduced risk of AD [127, 128]
PcTx1 rescued the synaptic plasticity associated with genetic and non-genetic models of AD [129]
Memantine, one of the very few drugs approved for amelioration of AD, apart from its blocking action on NMDA-receptor also affects ASICS [131]
P2X7R mediates neuroinflammation, and its inhibition ameliorates neuronal damage [190–194] and diminishes spatial memory impairment and cognitive deficits [198]
Overexpression of P2X4R in neurons enhanced the Aβ toxic effect, while its silencing decreased Aβ associated cellular death [199]
P2X2R is up-regulated in AD [200]
Multiple sclerosis Upregulation of ASIC1a is associated with the disease, and ASIC1 inhibition has a protective role [133–135] P2X4R and P2X7R are overexpressed in human patients [219, 220]
P2X7R genetic variants may be neuroprotective [221] or a risk factor [222]
P2X7R inactivation reduces the symptoms in EAE [224–226]
P2X4 inhibition potentiates the symptoms in EAE, allosteric potentiation with ivermectin has the opposite effect [223]
Parkinson’s disease ASIC1a inhibition has a protective effect [137] P2X7R inhibition shows a neuroprotective effect [202–205]
Overexpression or activation of P2X4Rs is associated with increased dopaminergic degeneration [206, 207]
P2X1R mediates an increase in alpha-synuclein in response to ATP [208]
Huntington’s disease Amiloride derivative benzamil enhances the activity of the ubiquitin-proteasome system [139]
Blocking the expression of ASIC1a has a similar effect [139]
P2X7 expression is increased in HD, its inhibition decreases symptoms and neuronal loss [210, 211]
Amyotrophic lateral sclerosis P2X7R is up-regulated in mice [213] and human patients [212]
Its role in ALS genesis and progression is dual and complex [217]
P2X4 is increased in degenerating motoneurons [214]
P2X4 allosteric modulator ivermectin extends the life span in ALS mice [215]
Psychiatric disorders ASIC1a inhibition decrease depression-like behavior in mice [140, 141]
ASIC1/2 genetic variation is related to individual differences in panic response [143, 144]
P2X2Rs mediate the antidepressant-like effects of ATP [232]
P2X7Rs stimulate the hyperactivation of microglia associated with depressive-like behaviors [231]
P2X7 inhibition has antidepressant effects [233–235]
P2X7R inhibition normalizes locomotor activity in the bipolar disorder model [229]
P2X7R mediates the positive effects of antipurinergic therapy in the autism spectrum disorder model [240]
Fig. 3 ASICs and P2X receptors contribute to almost the same set of brain pathologies. Dashed lines indicate weak evidence or not much data available.
In this regard, it seems to be of interest to study the following questions: are ASICs and P2X co-localized in the brain? Are there interactions between ASICs and P2X in the brain? In particular, it may be of interest to study whether such interactions could contribute to the enigma noted above as to how rapidly inactivating ASICs can substantially contribute to the pathologies that are typically associated with slow acidification. Regarding these issues, it should be emphasized that although interactions of ASIC3 with P2X receptors have been a major focus of research in sensory neurons [36], nevertheless, “ASIC1 can also be sensitized through P2X receptors” [36].
Furthermore, given that novel technologies for investigating functional cell-surface ASIC1a channels have recently been developed and used to investigate the physiological roles of these channels in brain neurons [243], the application of such tools to investigating brain pathologies would be fascinating.
Conclusions
We felt the necessity of this review due to the quickly-expanding evidence on the physiological significance of the two receptor/channel families. This seems especially relevant for the ASICs.
Both families entered the “market of knowledge” substantially later than the more canonical families like cholino- or adrenoreceptors. In both cases, the path to recognition of ASICs or P2 receptors was quite difficult and long. Geoffrey Burnstock made prolonged efforts to demonstrate the importance of the dual metabo- and ionotropic roles of purinoreceptors. After being cloned [39], ASICs became a subject of widespread interest 14 years after their initial discovery [38].
The last decade demonstrated that this subject was a long shot: it became clear that every neuron in the mammalian nervous system expresses at least one member of the ASIC family. Correspondingly, we are bound to expect a great variety of functional roles and new possibilities for pharmacology. Thus, new pharmacological tools, not only ASIC-specific but ASIC subunit-specific, will allow the specification of new roles of these receptors in known functions or even reveal new functions.
Both receptor families examined here provide intriguing opportunities in the novel realm of research connected with the roles of receptors traditionally regarded as ion channels. Growing evidence indicates that these receptors play important roles connected with intimate intermolecular machinery in the brain rather than just allowing ions passage.
This review was supported by the National Research Foundation of Ukraine (https://nrfu.org.ua/en/, 2020.02/0263) and Die Deutsche Forschungsgemeinschaft (https://www.dfg.de/, 441694049).
Conflict of interest
The authors declare no competing interests.
Andrii Cherninskyi, Maksim Storozhuk, and Oleksandr Maximyuk have contributed equally to this review.
==== Refs
References
1. Zeng WZ Xu TL Proton production, regulation and pathophysiological roles in the mammalian brain Neurosci Bull 2012 28 1 13 10.1007/s12264-012-1068-2 22233885
2. Carattino MD Montalbetti N Acid-sensing ion channels in sensory signaling Am J Physiol Renal Physiol 2020 318 F531 F543 10.1152/ajprenal.00546.2019 31984789
3. Hill AS Ben-Shahar Y The synaptic action of Degenerin/Epithelial sodium channels Channels 2018 12 262 275 10.1080/19336950.2018.1495006 30001175
4. Vullo S Kellenberger S A molecular view of the function and pharmacology of acid-sensing ion channels Pharmacol Res 2020 154 104166 10.1016/j.phrs.2019.02.005 30731197
5. Burnstock G Kennedy C P2X receptors in health and disease Adv Pharmacol 2011 61 333 372 10.1016/B978-0-12-385526-8.00011-4 21586364
6. Gonzales EB Kawate T Gouaux E Pore architecture and ion sites in acid-sensing ion channels and P2X receptors Nature 2009 460 599 604 10.1038/nature08218 19641589
7. Baron A Lingueglia E Pharmacology of acid-sensing ion channels—Physiological and therapeutical perspectives Neuropharmacology 2015 94 19 35 10.1016/j.neuropharm.2015.01.005 25613302
8. Donkor ES Stroke in the 21st century: A snapshot of the burden, epidemiology, and quality of life Stroke Res Treat 2018 2018 3238165 30598741
9. Goldberg DS McGee SJ Pain as a global public health priority BMC Public Health 2011 11 770 10.1186/1471-2458-11-770 21978149
10. Walton C King R Rechtman L Kaye W Leray E Marrie RA Rising prevalence of multiple sclerosis worldwide: Insights from the Atlas of MS, third edition Mult Scler 2020 26 1816 1821 10.1177/1352458520970841 33174475
11. World Health Organization. Epilepsy, https://www.who.int/news-room/fact-sheets/detail/epilepsy. 2022.
12. Lee CH Chen CC Roles of ASICs in nociception and proprioception Adv Exp Med Biol 2018 1099 37 47 10.1007/978-981-13-1756-9_4 30306513
13. Rook ML Musgaard M MacLean DM Coupling structure with function in acid-sensing ion channels: Challenges in pursuit of proton sensors J Physiol 2021 599 417 430 10.1113/JP278707 32306405
14. Pattison LA Callejo G St John Smith E Evolution of acid nociception: Ion channels and receptors for detecting acid Philos Trans R Soc Lond B Biol Sci 2019 374 20190291 10.1098/rstb.2019.0291 31544616
15. Uchitel OD González Inchauspe C Weissmann C Synaptic signals mediated by protons and acid-sensing ion channels Synapse 2019 73 e22120 10.1002/syn.22120 31180161
16. Inoue K Tsuda M Nociceptive signaling mediated by P2X3, P2X4 and P2X7 receptors Biochem Pharmacol 2021 187 114309 10.1016/j.bcp.2020.114309 33130129
17. Martin-Aragon Baudel M Espinosa-Tanguma R Nieves-Cintron M Navedo MF Purinergic signaling during hyperglycemia in vascular smooth muscle cells Front Endocrinol (Lausanne) 2020 11 329 10.3389/fendo.2020.00329 32528416
18. Aslam M Gündüz D Troidl C Heger J Hamm CW Schulz R Purinergic regulation of endothelial barrier function Int J Mol Sci 2021 22 1207 10.3390/ijms22031207 33530557
19. Shokoples BG Paradis P Schiffrin EL P2X7 receptors: An untapped target for the management of cardiovascular disease Arterioscler Thromb Vasc Biol 2021 41 186 199 32998520
20. Agostinho P Madeira D Dias L Simões AP Cunha RA Canas PM Purinergic signaling orchestrating neuron-glia communication Pharmacol Res 2020 162 105253 10.1016/j.phrs.2020.105253 33080321
21. Zhang RJ Yin YF Xie XJ Gu HF Acid-sensing ion channels: Linking extracellular acidification with atherosclerosis Clin Chimica Acta 2020 502 183 190 10.1016/j.cca.2019.12.027
22. Gonzales EB Sumien N Acidity and acid-sensing ion channels in the normal and Alzheimer's disease brain J Alzheimers Dis 2017 57 1137 1144 10.3233/JAD-161131 28211811
23. Wang JJ Xu TL Acid-sensing ion channels as a target for neuroprotection: Acidotoxicity revisited Sheng Li Xue Bao 2016 68 403 413 27546501
24. Storozhuk M Cherninskyi A Maximyuk O Isaev D Krishtal O Acid-sensing ion channels: Focus on physiological and some pathological roles in the brain Curr Neuropharmacol 2021 19 1570 1589 10.2174/1570159X19666210125151824 33550975
25. Duveau A Bertin E Boué-Grabot E Implication of neuronal Versus microglial P2X4 receptors in central nervous system disorders Neurosci Bull 2020 36 1327 1343 10.1007/s12264-020-00570-y 32889635
26. Du J Reznikov LR Price MP Zha XM Lu Y Moninger TO Protons are a neurotransmitter that regulates synaptic plasticity in the lateral amygdala Proc Natl Acad Sci U S A 2014 111 8961 8966 10.1073/pnas.1407018111 24889629
27. Kreple CJ Lu Y Taugher RJ Schwager-Gutman AL Du J Stump M Acid-sensing ion channels contribute to synaptic transmission and inhibit cocaine-evoked plasticity Nat Neurosci 2014 17 1083 1091 10.1038/nn.3750 24952644
28. Li HS Su XY Song XL Qi X Li Y Wang RQ Protein kinase C lambda mediates acid-sensing ion channel 1a-dependent cortical synaptic plasticity and pain hypersensitivity J Neurosci 2019 39 5773 5793 10.1523/JNEUROSCI.0213-19.2019 31101759
29. González-Inchauspe C Urbano FJ di Guilmi MN Uchitel OD Acid-sensing ion channels activated by evoked released protons modulate synaptic transmission at the mouse Calyx of held synapse J Neurosci 2017 37 2589 2599 10.1523/JNEUROSCI.2566-16.2017 28159907
30. Mango D Nisticò R Acid-sensing ion channel 1a is involved in N-methyl D-aspartate receptor-dependent long-term depression in the Hippocampus Front Pharmacol 2019 10 555 10.3389/fphar.2019.00555 31178731
31. Pankratov Y Castro E Miras-Portugal MT Krishtal O A purinergic component of the excitatory postsynaptic current mediated by P2X receptors in the CA1 neurons of the rat hippocampus Eur J Neurosci 1998 10 3898 3902 10.1046/j.1460-9568.1998.00419.x 9875366
32. Pankratov Y Lalo U Krishtal O Verkhratsky A P2X receptor-mediated excitatory synaptic currents in somatosensory cortex Mol Cell Neurosci 2003 24 842 849 10.1016/S1044-7431(03)00233-1 14664830
33. Chiang PH Chien TC Chen CC Yanagawa Y Lien CC ASIC-dependent LTP at multiple glutamatergic synapses in amygdala network is required for fear memory Sci Rep 2015 5 10143 10.1038/srep10143 25988357
34. Li WG Liu MG Deng S Liu YM Shang L Ding J ASIC1a regulates insular long-term depression and is required for the extinction of conditioned taste aversion Nat Commun 2016 7 13770 10.1038/ncomms13770 27924869
35. Pankratov YV Lalo UV Krishtal OA Role for P2X receptors in long-term potentiation J Neurosci 2002 22 8363 8369 10.1523/JNEUROSCI.22-19-08363.2002 12351710
36. Birdsong WT Fierro L Williams FG Spelta V Naves LA Knowles M Sensing muscle ischemia: Coincident detection of acid and ATP via interplay of two ion channels Neuron 2010 68 739 749 10.1016/j.neuron.2010.09.029 21092862
37. Stephan G Huang L Tang Y Vilotti S Fabbretti E Yu Y The ASIC3/P2X3 cognate receptor is a pain-relevant and ligand-gated cationic channel Nat Commun 2018 9 1354 10.1038/s41467-018-03728-5 29636447
38. Krishtal OA Pidoplichko VI A receptor for protons in the nerve cell membrane Neuroscience 1980 5 2325 2327 10.1016/0306-4522(80)90149-9 6970348
39. Waldmann R Champigny G Bassilana F Heurteaux C Lazdunski M A proton-gated cation channel involved in acid-sensing Nature 1997 386 173 177 10.1038/386173a0 9062189
40. Gründer S Geissler HS Bässler EL Ruppersberg JP A new member of acid-sensing ion channels from pituitary gland Neuroreport 2000 11 1607 1611 10.1097/00001756-200006050-00003 10852210
41. Lingueglia E de Weille JR Bassilana F Heurteaux C Sakai H Waldmann R A modulatory subunit of acid sensing ion channels in brain and dorsal root ganglion cells J Biol Chem 1997 272 29778 29783 10.1074/jbc.272.47.29778 9368048
42. Price MP Snyder PM Welsh MJ Cloning and expression of a novel human brain Na+ channel J Biol Chem 1996 271 7879 7882 10.1074/jbc.271.14.7879 8626462
43. García-Añoveros J Derfler B Neville-Golden J Hyman BT Corey DP BNaC1 and BNaC2 constitute a new family of human neuronal sodium channels related to degenerins and epithelial sodium channels Proc Natl Acad Sci U S A 1997 94 1459 1464 10.1073/pnas.94.4.1459 9037075
44. Wemmie JA Askwith CC Lamani E Cassell MD Freeman JH Jr Welsh MJ Acid-sensing ion channel 1 is localized in brain regions with high synaptic density and contributes to fear conditioning J Neurosci 2003 23 5496 5502 10.1523/JNEUROSCI.23-13-05496.2003 12843249
45. Ugawa S Ueda T Takahashi E Hirabayashi Y Yoneda T Komai S Cloning and functional expression of ASIC-beta2, a splice variant of ASIC-beta Neuroreport 2001 12 2865 2869 10.1097/00001756-200109170-00022 11588592
46. Zha XM Acid-sensing ion channels: Trafficking and synaptic function Mol Brain 2013 6 1 10.1186/1756-6606-6-1 23281934
47. Baron A Waldmann R Lazdunski M ASIC-like, proton-activated currents in rat hippocampal neurons J Physiol 2002 539 485 494 10.1113/jphysiol.2001.014837 11882680
48. Wemmie JA Taugher RJ Kreple CJ Acid-sensing ion channels in pain and disease Nat Rev Neurosci 2013 14 461 471 10.1038/nrn3529 23783197
49. Coscoy S de Weille JR Lingueglia E Lazdunski M The pre-transmembrane 1 domain of acid-sensing ion channels participates in the ion pore J Biol Chem 1999 274 10129 10132 10.1074/jbc.274.15.10129 10187795
50. Akopian AN Chen CC Ding Y Cesare P Wood JN A new member of the acid-sensing ion channel family Neuroreport 2000 11 2217 2222 10.1097/00001756-200007140-00031 10923674
51. Cristofori-Armstrong B Saez NJ Chassagnon IR King GF Rash LD The modulation of acid-sensing ion channel 1 by PcTx1 is pH-, subtype- and species-dependent: Importance of interactions at the channel subunit interface and potential for engineering selective analogues Biochem Pharmacol 2019 163 381 390 10.1016/j.bcp.2019.03.004 30849303
52. Buta A Maximyuk O Kovalskyy D Sukach V Vovk M Ievglevskyi O Novel potent orthosteric antagonist of ASIC1a prevents NMDAR-dependent LTP induction J Med Chem 2015 58 4449 4461 10.1021/jm5017329 25974655
53. Joch M Ase AR Chen CXQ MacDonald PA Kontogiannea M Corera AT Parkin-mediated monoubiquitination of the PDZ protein PICK1 regulates the activity of acid-sensing ion channels Mol Biol Cell 2007 18 3105 3118 10.1091/mbc.e05-11-1027 17553932
54. Duan B Wang YZ Yang T Chu XP Yu Y Huang Y Extracellular spermine exacerbates ischemic neuronal injury through sensitization of ASIC1a channels to extracellular acidosis J Neurosci 2011 31 2101 2112 10.1523/JNEUROSCI.4351-10.2011 21307247
55. Gobetto MN González-Inchauspe C Uchitel OD Histamine and corticosterone modulate acid sensing ion channels (ASICs) dependent long-term potentiation at the mouse anterior cingulate cortex Neuroscience 2021 460 145 160 10.1016/j.neuroscience.2021.01.022 33493620
56. Krishtal OA Osipchuk YV Shelest TN Smirnoff SV Rapid extracellular pH transients related to synaptic transmission in rat hippocampal slices Brain Res 1987 436 352 356 10.1016/0006-8993(87)91678-7 2829992
57. Yu Z Wu YJ Wang YZ Liu DS Song XL Jiang Q The acid-sensing ion channel ASIC1a mediates striatal synapse remodeling and procedural motor learning Sci Signal 2018 11 eaar4481 10.1126/scisignal.aar4481 30087178
58. Wemmie JA Chen J Askwith CC Hruska-Hageman AM Price MP Nolan BC The acid-activated ion channel ASIC contributes to synaptic plasticity, learning, and memory Neuron 2002 34 463 477 10.1016/S0896-6273(02)00661-X 11988176
59. Cho JH Askwith CC Presynaptic release probability is increased in hippocampal neurons from ASIC1 knockout mice J Neurophysiol 2008 99 426 441 10.1152/jn.00940.2007 18094106
60. Liu MG Li HS Li WG Wu YJ Deng SN Huang C Acid-sensing ion channel 1a contributes to hippocampal LTP inducibility through multiple mechanisms Sci Rep 2016 6 23350 10.1038/srep23350 26996240
61. Mango D Braksator E Battaglia G Marcelli S Mercuri NB Feligioni M Acid-sensing ion channel 1a is required for mGlu receptor dependent long-term depression in the hippocampus Pharmacol Res 2017 119 12 19 10.1016/j.phrs.2017.01.028 28137639
62. Price MP Gong H Parsons MG Kundert JR Reznikov LR Bernardinelli L Localization and behaviors in null mice suggest that ASIC1 and ASIC2 modulate responses to aversive stimuli Genes Brain Behav 2014 13 179 194 10.1111/gbb.12108 24256442
63. Wang Q Wang Q Song XL Jiang Q Wu YJ Li Y Fear extinction requires ASIC1a-dependent regulation of hippocampal-prefrontal correlates Sci Adv 2018 4 eaau3075 10.1126/sciadv.aau3075 30417090
64. Ghobbeh A Taugher RJ Alam SM Fan R LaLumiere RT Wemmie JA A novel role for acid-sensing ion channels in Pavlovian reward conditioning Genes Brain Behav 2019 18 e12531 10.1111/gbb.12531 30375184
65. Jiang Q Wang CM Fibuch EE Wang JQ Chu XP Differential regulation of locomotor activity to acute and chronic cocaine administration by acid-sensing ion channel 1a and 2 in adult mice Neuroscience 2013 246 170 178 10.1016/j.neuroscience.2013.04.059 23644053
66. Gutman AL Cosme CV Noterman MF Worth WR Wemmie JA LaLumiere RT Overexpression of ASIC1A in the nucleus accumbens of rats potentiates cocaine-seeking behavior Addict Biol 2020 25 e12690 10.1111/adb.12690 30397978
67. Ziemann AE Allen JE Dahdaleh NS Drebot II Coryell MW Wunsch AM The amygdala is a chemosensor that detects carbon dioxide and acidosis to elicit fear behavior Cell 2009 139 1012 1021 10.1016/j.cell.2009.10.029 19945383
68. Lin B Alganem K O'Donovan SM Jin Z Naghavi F Miller OA Activation of acid-sensing ion channels by carbon dioxide regulates amygdala synaptic protein degradation in memory reconsolidation Mol Brain 2021 14 78 10.1186/s13041-021-00786-7 33962650
69. Mir FA Jha SK Locus coeruleus acid-sensing ion channels modulate sleep–wakefulness and state transition from NREM to REM sleep in the rat Neurosci Bull 2021 37 684 700 10.1007/s12264-020-00625-0 33638800
70. Khakh BS Burnstock G The double life of ATP Sci Am 2009 301 84–90 92 20058644
71. Holton FA Holton P The capillary dilator substances in dry powders of spinal roots; a possible role of adenosine triphosphate in chemical transmission from nerve endings J Physiol 1954 126 124 140 10.1113/jphysiol.1954.sp005198 13212735
72. Holton P The liberation of adenosine triphosphate on antidromic stimulation of sensory nerves J Physiol 1959 145 494 504 10.1113/jphysiol.1959.sp006157 13642316
73. Burnstock G Purinergic nerves Pharmacol Rev 1972 24 509 581 4404211
74. Khakh BS North RA Neuromodulation by extracellular ATP and P2X receptors in the CNS Neuron 2012 76 51 69 10.1016/j.neuron.2012.09.024 23040806
75. Krishtal OA Marchenko SM Pidoplichko VI Receptor for ATP in the membrane of mammalian sensory neurones Neurosci Lett 1983 35 41 45 10.1016/0304-3940(83)90524-4 6302606
76. Jahr CE Jessell TM ATP excites a subpopulation of rat dorsal horn neurones Nature 1983 304 730 733 10.1038/304730a0 6888539
77. Benham CD Tsien RW A novel receptor-operated Ca2+-permeable channel activated by ATP in smooth muscle Nature 1987 328 275 278 10.1038/328275a0 2439921
78. Kolb HA Wakelam MJO Transmitter-like action of ATP on patched membranes of cultured myoblasts and myotubes Nature 1983 303 621 623 10.1038/303621a0 6304532
79. Illes P Müller CE Jacobson KA Grutter T Nicke A Fountain SJ Update of P2X receptor properties and their pharmacology: IUPHAR Review 30 Br J Pharmacol 2021 178 489 514 10.1111/bph.15299 33125712
80. McCarthy AE Yoshioka C Mansoor SE Full-length P2X7 structures reveal how palmitoylation prevents channel desensitization Cell 2019 179 659 670.e13 10.1016/j.cell.2019.09.017 31587896
81. Burnstock G Introduction and perspective, historical note Front Cell Neurosci 2013 7 227 10.3389/fncel.2013.00227 24312014
82. Loesch A On P2X receptors in the brain: Microvessels. Dedicated to the memory of the late Professor Geoffrey Burnstock (1929–2020) Cell Tissue Res 2021 384 577 588 10.1007/s00441-021-03411-0 33755804
83. North RA Molecular physiology of P2X receptors Physiol Rev 2002 82 1013 1067 10.1152/physrev.00015.2002 12270951
84. Ruppelt A Ma W Borchardt K Silberberg SD Soto F Genomic structure, developmental distribution and functional properties of the chicken P2X(5) receptor J Neurochem 2001 77 1256 1265 10.1046/j.1471-4159.2001.00348.x 11389176
85. Hausmann R Kless A Schmalzing G Key sites for P2X receptor function and multimerization: Overview of mutagenesis studies on a structural basis Curr Med Chem 2015 22 799 818 10.2174/0929867322666141128163215 25439586
86. Mansoor SE Lü W Oosterheert W Shekhar M Tajkhorshid E Gouaux E X-ray structures define human P2X3 receptor gating cycle and antagonist action Nature 2016 538 66 71 10.1038/nature19367 27626375
87. Taruno A Nomura K Kusakizako T Ma Z Nureki O Foskett JK Taste transduction and channel synapses in taste buds Pflugers Arch Eur J Physiol 2021 473 3 13 10.1007/s00424-020-02464-4 32936320
88. Mittal R Chan B Grati M Mittal J Patel K Debs LH Molecular structure and regulation of P2X receptors with a special emphasis on the role of P2X2 in the auditory system J Cell Physiol 2016 231 1656 1670 10.1002/jcp.25274 26627116
89. Krajewski JL P2X3-containing receptors as targets for the treatment of chronic pain Neurotherapeutics 2020 17 826 838 10.1007/s13311-020-00934-2 33009633
90. Nieber K Poelchen W Illes P Role of ATP in fast excitatory synaptic potentials in locus coeruleus neurones of the rat Br J Pharmacol 1997 122 423 430 10.1038/sj.bjp.0701386 9351497
91. Silinsky EM Gerzanich V Vanner SM ATP mediates excitatory synaptic transmission in mammalian neurones Br J Pharmacol 1992 106 762 763 10.1111/j.1476-5381.1992.tb14408.x 1327385
92. Mori M Heuss C Gähwiler BH Gerber U Fast synaptic transmission mediated by P2X receptors in CA3 pyramidal cells of rat hippocampal slice cultures J Physiol 2001 535 115 123 10.1111/j.1469-7793.2001.t01-1-00115.x 11507162
93. Bardoni R Goldstein PA Lee CJ Gu JG MacDermott AB ATP P2X receptors mediate fast synaptic transmission in the dorsal horn of the rat spinal cord J Neurosci 1997 17 5297 5304 10.1523/JNEUROSCI.17-14-05297.1997 9204914
94. Jo YH Donier E Martinez A Garret M Toulmé E Boué-Grabot E Cross-talk between P2X4 and gamma-aminobutyric acid, type A receptors determines synaptic efficacy at a central synapse J Biol Chem 2011 286 19993 20004 10.1074/jbc.M111.231324 21482824
95. Jo YH Role LW Coordinate release of ATP and GABA at in vitro synapses of lateral hypothalamic neurons J Neurosci 2002 22 4794 4804 10.1523/JNEUROSCI.22-12-04794.2002 12077176
96. Pankratov Y Lalo U Verkhratsky A North RA Quantal release of ATP in mouse cortex J Gen Physiol 2007 129 257 265 10.1085/jgp.200609693 17325196
97. Gu JG MacDermott AB Activation of ATP P2X receptors elicits glutamate release from sensory neuron synapses Nature 1997 389 749 753 10.1038/39639 9338789
98. Lalo U Palygin O Verkhratsky A Grant SGN Pankratov Y ATP from synaptic terminals and astrocytes regulates NMDA receptors and synaptic plasticity through PSD-95 multi-protein complex Sci Rep 2016 6 33609 10.1038/srep33609 27640997
99. Montilla A Mata GP Matute C Domercq M Contribution of P2X4 receptors to CNS function and pathophysiology Int J Mol Sci 2020 21 5562 10.3390/ijms21155562 32756482
100. Ge WP Yang XJ Zhang Z Wang HK Shen W Deng QD Long-term potentiation of neuron-glia synapses mediated by Ca2+-permeable AMPA receptors Science 2006 312 1533 1537 10.1126/science.1124669 16763153
101. Duan S Anderson CM Keung EC Chen Y Chen Y Swanson RA P2X7 receptor-mediated release of excitatory amino acids from astrocytes J Neurosci 2003 23 1320 1328 10.1523/JNEUROSCI.23-04-01320.2003 12598620
102. Boscardin E Alijevic O Hummler E Frateschi S Kellenberger S The function and regulation of acid-sensing ion channels (ASICs) and the epithelial Na+ channel (ENaC): IUPHAR Review 19 Br J Pharmacol 2016 173 2671 2701 10.1111/bph.13533 27278329
103. Sivils A Yang F Wang JQ Chu XP Acid-sensing ion channel 2: Function and modulation Membranes 2022 12 113 10.3390/membranes12020113 35207035
104. Xiong ZG Zhu XM Chu XP Minami M Hey J Wei WL Neuroprotection in ischemia: Blocking calcium-permeable acid-sensing ion channels Cell 2004 118 687 698 10.1016/j.cell.2004.08.026 15369669
105. Wang YZ Wang JJ Huang Y Liu F Zeng WZ Li Y Correction: Tissue acidosis induces neuronal necroptosis via ASIC1a channel independent of its ionic conduction eLife 2016 5 e14128 10.7554/eLife.14128 26799507
106. Sluka KA Winter OC Wemmie JA Acid-sensing ion channels: A new target for pain and CNS diseases Curr Opin Drug Discov Devel 2009 12 693 704 19736627
107. Ievglevskyi O Isaev D Netsyk O Romanov A Fedoriuk M Maximyuk O Acid-sensing ion channels regulate spontaneous inhibitory activity in the hippocampus: Possible implications for epilepsy Philos Trans R Soc Lond B Biol Sci 2016 371 20150431 10.1098/rstb.2015.0431 27377725
108. Lai K Song XL Shi HS Qi X Li CY Fang J Bilirubin enhances the activity of ASIC channels to exacerbate neurotoxicity in neonatal hyperbilirubinemia in mice Sci Transl Med 2020 12 eaax1337 10.1126/scitranslmed.aax1337 32051225
109. Bolshakov KV Essin KV Buldakova SL Dorofeeva NA Skatchkov SN Eaton MJ Characterization of acid-sensitive ion channels in freshly isolated rat brain neurons Neuroscience 2002 110 723 730 10.1016/S0306-4522(01)00582-6 11934479
110. Alijevic O Bignucolo O Hichri E Peng Z Kucera JP Kellenberger S Slowing of the time course of acidification decreases the acid-sensing ion channel 1a current amplitude and modulates action potential firing in neurons Front Cell Neurosci 2020 14 41 10.3389/fncel.2020.00041 32180707
111. Xiong ZG Chu XP Simon RP Acid sensing ion channels—novel therapeutic targets for ischemic brain injury Front Biosci 2007 12 1376 1386 10.2741/2154 17127388
112. Xiong ZG Chu XP Simon RP Ca2+-permeable acid-sensing ion channels and ischemic brain injury J Membrane Biol 2006 209 59 68 10.1007/s00232-005-0840-x 16685601
113. Wang JJ Liu F Yang F Wang YZ Qi X Li Y Nat Commun 2020 11 475 10.1038/s41467-019-13873-0 31980622
114. Sirven JI Epilepsy: A spectrum disorder Cold Spring Harb Perspect Med 2015 5 a022848 10.1101/cshperspect.a022848 26328931
115. Simon RP Benowitz N Hedlund R Copeland J Influence of the blood-brain pH gradient on brain phenobarbital uptake during status epilepticus J Pharmacol Exp Ther 1985 234 830 835 4032292
116. Somjen GG Acidification of interstitial fluid in hippocampal formation caused by seizures and by spreading depression Brain Res 1984 311 186 188 10.1016/0006-8993(84)91416-1 6488041
117. Chesler M Kaila K Modulation of pH by neuronal activity Trends Neurosci 1992 15 396 402 10.1016/0166-2236(92)90191-A 1279865
118. Ziemann AE Schnizler MK Albert GW Severson MA Howard MA III Welsh MJ Seizure termination by acidosis depends on ASIC1a Nat Neurosci 2008 11 816 822 10.1038/nn.2132 18536711
119. Xiong ZG Pignataro G Li M Chang SY Simon RP Acid-sensing ion channels (ASICs) as pharmacological targets for neurodegenerative diseases Curr Opin Pharmacol 2008 8 25 32 10.1016/j.coph.2007.09.001 17945532
120. Mango D Nisticò R Neurodegenerative disease: What potential therapeutic role of acid-sensing ion channels? Front Cell Neurosci 2021 15 730641 10.3389/fncel.2021.730641 34690702
121. Ortega-Ramírez A Vega R Soto E Acid-sensing ion channels as potential therapeutic targets in neurodegeneration and neuroinflammation Mediators Inflamm 2017 2017 3728096 10.1155/2017/3728096 29056828
122. Salter MW Stevens B Microglia emerge as central players in brain disease Nat Med 2017 23 1018 1027 10.1038/nm.4397 28886007
123. Ma CL Sun H Yang L Wang XT Gao S Chen XW Acid-sensing ion channel 1a modulates NMDA receptor function through targeting NR1/NR2A/NR2B triheteromeric receptors Neuroscience 2019 406 389 404 10.1016/j.neuroscience.2019.03.044 30926548
124. Storozhuk M Kondratskaya E Nikolaenko L Krishtal O A modulatory role of ASICs on GABAergic synapses in rat hippocampal cell cultures Mol Brain 2016 9 90 10.1186/s13041-016-0269-4 27760555
125. Hynd MR Scott HL Dodd PR Glutamate-mediated excitotoxicity and neurodegeneration in Alzheimer's disease Neurochem Int 2004 45 583 595 10.1016/j.neuint.2004.03.007 15234100
126. Zholos AV Moroz OF Storozhuk MV Curcuminoids and novel opportunities for the treatment of Alzheimer's disease: Which molecules are actually effective? Curr Mol Pharmacol 2019 12 12 26 10.2174/1874467211666181012150847 30318014
127. Khachaturian AS Zandi PP Lyketsos CG Hayden KM Skoog I Norton MC Antihypertensive medication use and incident Alzheimer disease: The Cache County Study Arch Neurol 2006 63 686 692 10.1001/archneur.63.5.noc60013 16533956
128. Chuang YF Breitner JCS Chiu YL Khachaturian A Hayden K Corcoran C Use of diuretics is associated with reduced risk of Alzheimer's disease: The Cache County Study Neurobiol Aging 2014 35 2429 2435 10.1016/j.neurobiolaging.2014.05.002 24910391
129. Mango D Nisticò R Role of ASIC1a in aβ-induced synaptic alterations in the hippocampus Pharmacol Res 2018 131 61 65 10.1016/j.phrs.2018.03.016 29574226
130. Wu Y Qin D Yang H Fu H Evidence for the participation of acid-sensing ion channels (ASICs) in the antinociceptive effect of curcumin in a formalin-induced orofacial inflammatory model Cell Mol Neurobiol 2017 37 635 642 10.1007/s10571-016-0399-3 27383838
131. Shteinikov VY Tikhonova TB Korkosh VS Tikhonov DB Potentiation and block of ASIC1a by memantine Cell Mol Neurobiol 2018 38 869 881 10.1007/s10571-017-0561-6 29058095
132. Goldenberg MM Multiple sclerosis review P T 2012 37 175 184 22605909
133. Vergo S Craner MJ Etzensperger R Attfield K Friese MA Newcombe J Acid-sensing ion channel 1 is involved in both axonal injury and demyelination in multiple sclerosis and its animal model Brain 2011 134 571 584 10.1093/brain/awq337 21233144
134. Friese MA Craner MJ Etzensperger R Vergo S Wemmie JA Welsh MJ Acid-sensing ion channel-1 contributes to axonal degeneration in autoimmune inflammation of the central nervous system Nat Med 2007 13 1483 1489 10.1038/nm1668 17994101
135. Arun T Tomassini V Sbardella E de Ruiter MB Matthews L Leite MI Targeting ASIC1 in primary progressive multiple sclerosis: Evidence of neuroprotection with amiloride Brain 2013 136 106 115 10.1093/brain/aws325 23365093
136. Beitz JM Parkinson's disease: A review Front Biosci (Schol Ed) 2014 6 65 74 10.2741/S415 24389262
137. Arias RL Sung MLA Vasylyev D Zhang MY Albinson K Kubek K Amiloride is neuroprotective in an MPTP model of Parkinson's disease Neurobiol Dis 2008 31 334 341 10.1016/j.nbd.2008.05.008 18606547
138. Tabrizi SJ Flower MD Ross CA Wild EJ Huntington disease: New insights into molecular pathogenesis and therapeutic opportunities Nat Rev Neurol 2020 16 529 546 10.1038/s41582-020-0389-4 32796930
139. Wong HK Bauer PO Kurosawa M Goswami A Washizu C Machida Y Blocking acid-sensing ion channel 1 alleviates Huntington's disease pathology via an ubiquitin-proteasome system-dependent mechanism Hum Mol Genet 2008 17 3223 3235 10.1093/hmg/ddn218 18658163
140. Coryell MW Wunsch AM Haenfler JM Allen JE Schnizler M Ziemann AE Acid-sensing ion channel-1a in the amygdala, a novel therapeutic target in depression-related behavior J Neurosci 2009 29 5381 5388 10.1523/JNEUROSCI.0360-09.2009 19403806
141. Fedoriuk MP Cherninskyi AO Maximyuk OP Isaev DS Bogovyk RI Venhreniuk AV Inhibition of brain asics affects hippocampal theta-rhythm and openfield behavior in rats Fiziol Zh 2019 65 15 19 10.15407/fz65.01.015
142. Yellepeddi V Sayre C Burrows A Watt K Davies S Strauss J Stability of extemporaneously compounded amiloride nasal spray PLoS One 2020 15 e0232435 10.1371/journal.pone.0232435 32649677
143. Leibold NK van den Hove D Viechtbauer W Kenis G Goossens L Lange I Amiloride-sensitive cation channel 2 genotype affects the response to a carbon dioxide panic challenge J Psychopharmacol 2017 31 1294 1301 10.1177/0269881116686880 28121219
144. Gugliandolo A Gangemi C Caccamo D Currò M Pandolfo G Quattrone D The RS685012 polymorphism of ACCN2, the human ortholog of murine acid-sensing ion channel (ASIC1) gene, is highly represented in patients with panic disorder Neuromolecular Med 2016 18 91 98 10.1007/s12017-015-8380-8 26589317
145. Dietrich CJ Morad M Synaptic acidification enhances GABAA signaling J Neurosci 2010 30 16044 16052 10.1523/JNEUROSCI.6364-09.2010 21106843
146. Palmer MJ Hull C Vigh J von Gersdorff H Synaptic cleft acidification and modulation of short-term depression by exocytosed protons in retinal bipolar cells J Neurosci 2003 23 11332 11341 10.1523/JNEUROSCI.23-36-11332.2003 14672997
147. Stawarski M Hernandez RX Feghhi T Borycz JA Lu Z Agarwal AB Neuronal glutamatergic synaptic clefts alkalinize rather than acidify during neurotransmission J Neurosci 2020 40 1611 1624 10.1523/JNEUROSCI.1774-19.2020 31964719
148. Rash LD Acid-sensing ion channel pharmacology, past, present, and future … Adv Pharmacol 2017 79 35 66 10.1016/bs.apha.2017.02.001 28528673
149. Tikhonov DB Magazanik LG Nagaeva EI Ligands of acid-sensing ion channel 1a: Mechanisms of action and binding sites Acta Naturae 2019 11 4 13 10.32607/20758251-2019-11-1-4-13 31024743
150. Allen NJ Attwell D Modulation of ASIC channels in rat cerebellar Purkinje neurons by ischaemia-related signals J Physiol 2002 543 521 529 10.1113/jphysiol.2002.020297 12205186
151. Nagaeva EI Tikhonova TB Magazanik LG Tikhonov DB Histamine selectively potentiates acid-sensing ion channel 1a Neurosci Lett 2016 632 136 140 10.1016/j.neulet.2016.08.047 27574729
152. Park JH Kim YC P2X7 receptor antagonists: A patent review (2010–2015) Expert Opin Ther Pat 2017 27 257 267 10.1080/13543776.2017.1246538 27724045
153. Pevarello P Bovolenta S Tarroni P Za L Severi E Torino D P2X7 antagonists for CNS indications: Recent patent disclosures Pharm Pat Anal 2017 6 61 76 10.4155/ppa-2016-0044 28248151
154. Andrejew R Oliveira-Giacomelli Á Ribeiro DE Glaser T Arnaud-Sampaio VF Lameu C The P2X7 receptor: Central hub of brain diseases Front Mol Neurosci 2020 13 124 10.3389/fnmol.2020.00124 32848594
155. Pelegrin P P2X7 receptor and the NLRP3 inflammasome: Partners in crime Biochem Pharmacol 2021 187 114385 10.1016/j.bcp.2020.114385 33359010
156. Adinolfi E Giuliani AL de Marchi E Pegoraro A Orioli E di Virgilio F The P2X7 receptor: A main player in inflammation Biochem Pharmacol 2018 151 234 244 10.1016/j.bcp.2017.12.021 29288626
157. Guo J Fu X Cui X Fan M Contributions of purinergic P2X3 receptors within the midbrain periaqueductal gray to diabetes-induced neuropathic pain J Physiol Sci 2015 65 99 104 10.1007/s12576-014-0344-5 25367719
158. Kuan YH Shih HC Tang SC Jeng JS Shyu BC Targeting P2X7 receptor for the treatment of central post-stroke pain in a rodent model Neurobiol Dis 2015 78 134 145 10.1016/j.nbd.2015.02.028 25836422
159. Falzoni S Donvito G Di Virgilio F Detecting adenosine triphosphate in the pericellular space Interface Focus 2013 3 20120101 10.1098/rsfs.2012.0101 23853707
160. He Y Taylor N Fourgeaud L Bhattacharya A The role of microglial P2X7: Modulation of cell death and cytokine release J Neuroinflammation 2017 14 1 13 10.1186/s12974-017-0904-8 28086917
161. Lord B Ameriks MK Wang Q Fourgeaud L Vliegen M Verluyten W A novel radioligand for the ATP-gated ion channel P2X7: [3H]JNJ-54232334 Eur J Pharmacol 2015 765 551 559 10.1016/j.ejphar.2015.09.026 26386289
162. Bhattacharya A Lord B Grigoleit JS He Y Fraser I Campbell SN Neuropsychopharmacology of JNJ-55308942: Evaluation of a clinical candidate targeting P2X7 ion channels in animal models of neuroinflammation and anhedonia Neuropsychopharmacology 2018 43 2586 2596 10.1038/s41386-018-0141-6 30026598
163. Cheng RD Ren JJ Zhang YY Ye XM P2X4 receptors expressed on microglial cells in post-ischemic inflammation of brain ischemic injury Neurochem Int 2014 67 9 13 10.1016/j.neuint.2014.01.011 24486458
164. Li F Wang L Li JW Gong M He L Feng R Hypoxia induced amoeboid microglial cell activation in postnatal rat brain is mediated by ATP receptor P2X4 BMC Neurosci 2011 12 111 10.1186/1471-2202-12-111 22053919
165. Verma R Cronin CG Hudobenko J Venna VR McCullough LD Liang BT Deletion of the P2X4 receptor is neuroprotective acutely, but induces a depressive phenotype during recovery from ischemic stroke Brain Behav Immun 2017 66 302 312 10.1016/j.bbi.2017.07.155 28751018
166. Wixey JA Reinebrant HE Carty ML Buller KM Delayed P2X4R expression after hypoxia-ischemia is associated with microglia in the immature rat brain J Neuroimmunol 2009 212 35 43 10.1016/j.jneuroim.2009.04.016 19447505
167. Ozaki T Muramatsu R Sasai M Yamamoto M Kubota Y Fujinaka T The P2X4 receptor is required for neuroprotection via ischemic preconditioning Sci Rep 2016 6 25893 10.1038/srep25893 27173846
168. Xiao J Huang Y Li X Li L Yang T Huang L TNP-ATP is beneficial for treatment of neonatal hypoxia-induced hypomyelination and cognitive decline Neurosci Bull 2016 32 99 107 10.1007/s12264-015-0003-8 26769489
169. Beamer E Conte G Engel T ATP release during seizures - A critical evaluation of the evidence Brain Res Bull 2019 151 65 73 10.1016/j.brainresbull.2018.12.021 30660718
170. Ulmann L Levavasseur F Avignone E Peyroutou R Hirbec H Audinat E Involvement of P2X4 receptors in hippocampal microglial activation after status epilepticus Glia 2013 61 1306 1319 10.1002/glia.22516 23828736
171. Engel T Gomez-Villafuertes R Tanaka K Mesuret G Sanz-Rodriguez A Garcia-Huerta P Seizure suppression and neuroprotection by targeting the purinergic P2X7 receptor during status epilepticus in mice FASEB J 2012 26 1616 1628 10.1096/fj.11-196089 22198387
172. Huang C Chi XS Li R Hu X Xu HX Li JM Inhibition of P2X7 receptor ameliorates nuclear factor-kappa B mediated neuroinflammation induced by status epilepticus in rat hippocampus J Mol Neurosci 2017 63 173 184 10.1007/s12031-017-0968-z 28856625
173. Jimenez-Pacheco A Mesuret G Sanz-Rodriguez A Tanaka K Mooney C Conroy R Increased neocortical expression of the P2X7 receptor after status epilepticus and anticonvulsant effect of P2X7 receptor antagonist A-438079 Epilepsia 2013 54 1551 1561 10.1111/epi.12257 23808395
174. Mesuret G Engel T Hessel EV Sanz-Rodriguez A Jimenez-Pacheco A Miras-Portugal MT P2X7 receptor inhibition interrupts the progression of seizures in immature rats and reduces hippocampal damage CNS Neurosci Ther 2014 20 556 564 10.1111/cns.12272 24750893
175. Rodriguez-Alvarez N Jimenez-Mateos EM Engel T Quinlan S Reschke CR Conroy RM Effects of P2X7 receptor antagonists on hypoxia-induced neonatal seizures in mice Neuropharmacology 2017 116 351 363 10.1016/j.neuropharm.2017.01.005 28082183
176. Avignone E Ulmann L Levavasseur F Rassendren F Audinat E Status epilepticus induces a particular microglial activation state characterized by enhanced purinergic signaling J Neurosci 2008 28 9133 9144 10.1523/JNEUROSCI.1820-08.2008 18784294
177. Zhou X Ma LM Xiong Y Huang H Yuan JX Li RH Upregulated P2X3 receptor expression in patients with intractable temporal lobe epilepsy and in a rat model of epilepsy Neurochem Res 2016 41 1263 1273 10.1007/s11064-015-1820-x 26738991
178. Xia J Wang H Zhang Q Han Z Modulation of P2X purinoceptor 3 (P2X3) in pentylenetetrazole-induced kindling epilepsy in rats Med Sci Monit 2018 24 6165 6177 10.12659/MSM.910352 30181529
179. Edwards C Klekot O Halugan L Korchev Y Follow your nose: A key clue to understanding and treating COVID-19 Front Endocrinol (Lausanne) 2021 12 747744 10.3389/fendo.2021.747744 34867791
180. Ceban F Ling S Lui LMW Lee Y Gill H Teopiz KM Fatigue and cognitive impairment in Post-COVID-19 Syndrome: A systematic review and meta-analysis Brain Behav Immun 2022 101 93 135 10.1016/j.bbi.2021.12.020 34973396
181. Wan D Du T Hong W Chen L Que H Lu S Neurological complications and infection mechanism of SARS-CoV-2 Signal Transduct Target Ther 2021 6 406 10.1038/s41392-021-00818-7 34815399
182. Lukiw WJ Pogue A Hill JM SARS-CoV-2 infectivity and neurological targets in the brain Cell Mol Neurobiol 2022 42 217 224 10.1007/s10571-020-00947-7 32840758
183. Hamming I Timens W Bulthuis MLC Lely AT Navis GJ van Goor H Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis J Pathol 2004 203 631 637 10.1002/path.1570 15141377
184. Cui H Su S Cao Y Ma C Qiu W The altered anatomical distribution of ACE2 in the brain with Alzheimer's disease pathology Front Cell Dev Biol 2021 9 684874 10.3389/fcell.2021.684874 34249938
185. Di Virgilio F Tang Y Sarti AC Rossato M A rationale for targeting the P2X7 receptor in Coronavirus disease 19 Br J Pharmacol 2020 177 4990 4994 10.1111/bph.15138 32441783
186. Heppner FL Ransohoff RM Becher B Immune attack: The role of inflammation in Alzheimer disease Nat Rev Neurosci 2015 16 358 372 10.1038/nrn3880 25991443
187. McGeer PL Rogers J McGeer EG Inflammation, antiinflammatory agents, and Alzheimer's disease: The last 22 years J Alzheimers Dis 2016 54 853 857 10.3233/JAD-160488 27716676
188. Sáez-Orellana F Fuentes-Fuentes MC Godoy PA Silva-Grecchi T Panes JD Guzmán L P2X receptor overexpression induced by soluble oligomers of amyloid beta peptide potentiates synaptic failure and neuronal dyshomeostasis in cellular models of Alzheimer's disease Neuropharmacology 2018 128 366 378 10.1016/j.neuropharm.2017.10.027 29079292
189. Sáez-Orellana F Godoy PA Bastidas CY Silva-Grecchi T Guzmán L Aguayo LG ATP leakage induces P2XR activation and contributes to acute synaptic excitotoxicity induced by soluble oligomers of β-amyloid peptide in hippocampal neurons Neuropharmacology 2016 100 116 123 10.1016/j.neuropharm.2015.04.005 25896766
190. Rampe D Wang L Ringheim GE P2X7 receptor modulation of β-amyloid- and LPS-induced cytokine secretion from human macrophages and microglia J Neuroimmunol 2004 147 56 61 10.1016/j.jneuroim.2003.10.014 14741428
191. Sanz JM Chiozzi P Ferrari D Colaianna M Idzko M Falzoni S Activation of microglia by amyloid{beta}requires P2X7 receptor expression J Immunol 2009 182 4378 4385 10.4049/jimmunol.0803612 19299738
192. Skaper SD Facci L Culbert AA Evans NA Chessell I Davis JB P2X(7) receptors on microglial cells mediate injury to cortical neurons in vitro Glia 2006 54 234 242 10.1002/glia.20379 16817206
193. Kim SY Moon JH Lee HG Kim SU Lee YB ATP released from beta-amyloid-stimulated microglia induces reactive oxygen species production in an autocrine fashion Exp Mol Med 2007 39 820 827 10.1038/emm.2007.89 18160853
194. Lee HG Won SM Gwag BJ Lee YB Microglial P2X7 receptor expression is accompanied by neuronal damage in the cerebral cortex of the APPswe/PS1dE9 mouse model of Alzheimer's disease Exp Mol Med 2011 43 7 14 10.3858/emm.2011.43.1.001 21088470
195. Diaz-Hernandez JI Gomez-Villafuertes R León-Otegui M Hontecillas-Prieto L del Puerto A Trejo JL In vivo P2X7 inhibition reduces amyloid plaques in Alzheimer's disease through GSK3β and secretases Neurobiol Aging 2012 33 1816 1828 10.1016/j.neurobiolaging.2011.09.040 22048123
196. León-Otegui M Gómez-Villafuertes R Díaz-Hernández JI Díaz-Hernández M Miras-Portugal MT Gualix J Opposite effects of P2X7 and P2Y2 nucleotide receptors on α-secretase-dependent APP processing in Neuro-2a cells FEBS Lett 2011 585 2255 2262 10.1016/j.febslet.2011.05.048 21651910
197. Delarasse C Auger R Gonnord P Fontaine B Kanellopoulos JM The purinergic receptor P2X7 triggers alpha-secretase-dependent processing of the amyloid precursor protein J Biol Chem 2011 286 2596 2606 10.1074/jbc.M110.200618 21081501
198. Chen X Hu J Jiang L Xu S Zheng B Wang C Brilliant Blue G improves cognition in an animal model of Alzheimer's disease and inhibits amyloid-β-induced loss of filopodia and dendrite spines in hippocampal neurons Neuroscience 2014 279 94 101 10.1016/j.neuroscience.2014.08.036 25193238
199. Varma R Chai Y Troncoso J Gu J Xing H Stojilkovic SS Amyloid-β induces a caspase-mediated cleavage of P2X4 to promote purinotoxicity Neuromol Med 2009 11 63 75 10.1007/s12017-009-8073-2
200. Godoy PA Mennickent D Cuchillo-Ibáñez I Ramírez-Molina O Silva-Grecchi T Panes-Fernández J Increased P2X2 receptors induced by amyloid-β peptide participates in the neurotoxicity in Alzheimer's disease Biomed Pharmacother 2021 142 111968 10.1016/j.biopha.2021.111968 34343896
201. Carmo M Gonçalves FQ Canas PM Oses JP Fernandes FD Duarte FV Enhanced ATP release and CD73-mediated adenosine formation sustain adenosine A2A receptor over-activation in a rat model of Parkinson's disease Br J Pharmacol 2019 176 3666 3680 10.1111/bph.14771 31220343
202. Ferrazoli EG de Souza HDN Nascimento IC Oliveira-Giacomelli Á Schwindt TT Britto LR Brilliant blue G, but not fenofibrate, treatment reverts hemiparkinsonian behavior and restores dopamine levels in an animal model of Parkinson's disease Cell Transplant 2017 26 669 677 10.3727/096368917X695227 28403913
203. Jiang T Hoekstra J Heng X Kang W Ding J Liu J P2X7 receptor is critical in α-synuclein—mediated microglial NADPH oxidase activation Neurobiol Aging 2015 36 2304 2318 10.1016/j.neurobiolaging.2015.03.015 25983062
204. Kumar S Mishra A Krishnamurthy S Purinergic antagonism prevents mitochondrial dysfunction and behavioral deficits associated with dopaminergic toxicity induced by 6-OHDA in rats Neurochem Res 2017 42 3414 3430 10.1007/s11064-017-2383-9 28836128
205. Oliveira-Giacomelli Á Albino CM de Souza HDN Corrêa-Velloso J de Jesus Santos AP Baranova J P2Y6 and P2X7 receptor antagonism exerts neuroprotective/neuroregenerative effects in an animal model of Parkinson's disease Front Cell Neurosci 2019 13 476 10.3389/fncel.2019.00476 31787881
206. Ma J Gao J Niu M Zhang X Wang J Xie A P2X4R overexpression upregulates interleukin-6 and exacerbates 6-OHDA-induced dopaminergic degeneration in a rat model of PD Front Aging Neurosci 2020 12 580068 10.3389/fnagi.2020.580068 33328961
207. Zhang X Wang J Gao JZ Zhang XN Dou KX Shi WD P2X4 receptor participates in autophagy regulation in Parkinson's disease Neural Regen Res 2021 16 2505 2511 10.4103/1673-5374.313053 33907041
208. Gan M Moussaud S Jiang P McLean PJ Extracellular ATP induces intracellular alpha-synuclein accumulation via P2X1 receptor-mediated lysosomal dysfunction Neurobiol Aging 2015 36 1209 1220 10.1016/j.neurobiolaging.2014.10.037 25480524
209. Navarro G Borroto-Escuela DO Fuxe K Franco R Purinergic signaling in Parkinson's disease. Relevance for treatment Neuropharmacology 2016 104 161 168 10.1016/j.neuropharm.2015.07.024 26211977
210. Díaz-Hernández M Díez-Zaera M Sánchez-Nogueiro J Gómez-Villafuertes R Canals JM Alberch J Altered P2X7-receptor level and function in mouse models of Huntington's disease and therapeutic efficacy of antagonist administration FASEB J 2009 23 1893 1906 10.1096/fj.08-122275 19171786
211. Ollà I Santos-Galindo M Elorza A Lucas JJ P2X7 receptor upregulation in Huntington's disease brains Front Mol Neurosci 2020 13 567430 10.3389/fnmol.2020.567430 33122998
212. Yiangou Y Facer P Durrenberger P Chessell IP Naylor A Bountra C COX-2, CB2 and P2X7-immunoreactivities are increased in activated microglial cells/macrophages of multiple sclerosis and amyotrophic lateral sclerosis spinal cord BMC Neurol 2006 6 12 10.1186/1471-2377-6-12 16512913
213. D'Ambrosi N Finocchi P Apolloni S Cozzolino M Ferri A Padovano V The proinflammatory action of microglial P2 receptors is enhanced in SOD1 models for amyotrophic lateral sclerosis J Immunol 2009 183 4648 4656 10.4049/jimmunol.0901212 19734218
214. Casanovas A Hernández S Tarabal O Rosselló J Esquerda JE Strong P2X4 purinergic receptor-like immunoreactivity is selectively associated with degenerating neurons in transgenic rodent models of amyotrophic lateral sclerosis J Comp Neurol 2008 506 75 92 10.1002/cne.21527 17990272
215. Andries M van Damme P Robberecht W van den Bosch L Ivermectin inhibits AMPA receptor-mediated excitotoxicity in cultured motor neurons and extends the life span of a transgenic mouse model of amyotrophic lateral sclerosis Neurobiol Dis 2007 25 8 16 10.1016/j.nbd.2006.08.018 17045808
216. Volonté C Amadio S Liguori F Fabbrizio P Duality of P2X7 receptor in amyotrophic lateral sclerosis Front Pharmacol 2020 11 1148 10.3389/fphar.2020.01148 32792962
217. Apolloni S Amadio S Montilli C Volonté C D'Ambrosi N Ablation of P2X7 receptor exacerbates gliosis and motoneuron death in the SOD1-G93A mouse model of amyotrophic lateral sclerosis Hum Mol Genet 2013 22 4102 4116 10.1093/hmg/ddt259 23736299
218. Fabbrizio P Amadio S Apolloni S Volonté C P2X7 receptor activation modulates autophagy in SOD1-G93A mouse microglia Front Cell Neurosci 2017 11 249 10.3389/fncel.2017.00249 28871219
219. Vázquez-Villoldo N Domercq M Martín A Llop J Gómez-Vallejo V Matute C P2X4 receptors control the fate and survival of activated microglia Glia 2014 62 171 184 10.1002/glia.22596 24254916
220. Amadio S Parisi C Piras E Fabbrizio P Apolloni S Montilli C Modulation of P2X7 receptor during inflammation in multiple sclerosis Front Immunol 2017 8 1529 10.3389/fimmu.2017.01529 29187851
221. Gu BJ Field J Dutertre S Ou A Kilpatrick TJ Lechner-Scott J A rare P2X7 variant Arg307Gln with absent pore formation function protects against neuroinflammation in multiple sclerosis Hum Mol Genet 2015 24 5644 5654 10.1093/hmg/ddv278 26188005
222. Sadovnick AD Gu BJ Traboulsee AL Bernales CQ Encarnacion M Yee IM Purinergic receptors P2RX4 and P2RX7 in familial multiple sclerosis Hum Mutat 2017 38 736 744 10.1002/humu.23218 28326637
223. Zabala A Vazquez-Villoldo N Rissiek B Gejo J Martin A Palomino A P2X4 receptor controls microglia activation and favors remyelination in autoimmune encephalitis EMBO Mol Med 2018 10 e8743 10.15252/emmm.201708743 29973381
224. Lang PA Merkler D Funkner P Shaabani N Meryk A Krings C Oxidized ATP inhibits T-cell-mediated autoimmunity Eur J Immunol 2010 40 2401 2408 10.1002/eji.200939838 20683833
225. Matute C Torre I Pérez-Cerdá F Pérez-Samartín A Alberdi E Etxebarria E P2X(7) receptor blockade prevents ATP excitotoxicity in oligodendrocytes and ameliorates experimental autoimmune encephalomyelitis J Neurosci 2007 27 9525 9533 10.1523/JNEUROSCI.0579-07.2007 17728465
226. Sharp AJ Polak PE Simonini V Lin SX Richardson JC Bongarzone ER P2x7 deficiency suppresses development of experimental autoimmune encephalomyelitis J Neuroinflammation 2008 5 33 10.1186/1742-2094-5-33 18691411
227. Cheffer A Castillo ARG Corrêa-Velloso J Gonçalves MCB Naaldijk Y Nascimento IC Purinergic system in psychiatric diseases Mol Psychiatry 2018 23 94 106 10.1038/mp.2017.188 28948971
228. Krügel U Purinergic receptors in psychiatric disorders Neuropharmacology 2016 104 212 225 10.1016/j.neuropharm.2015.10.032 26518371
229. Gubert C Fries GR Pfaffenseller B Ferrari P Coutinho-Silva R Morrone FB Role of P2X7 receptor in an animal model of Mania induced by D-amphetamine Mol Neurobiol 2016 53 611 620 10.1007/s12035-014-9031-z 25502294
230. Mingam R Smedt VD Amédée T Bluthé RM Kelley KW Dantzer R In vitro and in vivo evidence for a role of the P2X7 receptor in the release of IL-1β in the murine brain Brain Behav Immun 2008 22 234 244 10.1016/j.bbi.2007.08.007 17905568
231. Stokes L Spencer SJ Jenkins TA Understanding the role of P2X7 in affective disorders-are glial cells the major players? Front Cell Neurosci 2015 9 258 10.3389/fncel.2015.00258 26217184
232. Cao X Li LP Wang Q Wu Q Hu HH Zhang M Astrocyte-derived ATP modulates depressive-like behaviors Nat Med 2013 19 773 777 10.1038/nm.3162 23644515
233. Basso AM Bratcher NA Harris RR Jarvis MF Decker MW Rueter LE Behavioral profile of P2X7 receptor knockout mice in animal models of depression and anxiety: Relevance for neuropsychiatric disorders Behav Brain Res 2009 198 83 90 10.1016/j.bbr.2008.10.018 18996151
234. Csölle C Andó RD Kittel Á Gölöncsér F Baranyi M Soproni K The absence of P2X7 receptors (P2rx7) on non-haematopoietic cells leads to selective alteration in mood-related behaviour with dysregulated gene expression and stress reactivity in mice Int J Neuropsychopharmacol 2013 16 213 233 10.1017/S1461145711001933 22243662
235. Csölle C Baranyi M Zsilla G Kittel A Gölöncsér F Illes P Neurochemical changes in the mouse Hippocampus underlying the antidepressant effect of genetic deletion of P2X7 receptors PLoS One 2013 8 e66547 10.1371/journal.pone.0066547 23805233
236. Hempel C Nörenberg W Sobottka H Urban N Nicke A Fischer W The phenothiazine-class antipsychotic drugs prochlorperazine and trifluoperazine are potent allosteric modulators of the human P2X7 receptor Neuropharmacology 2013 75 365 379 10.1016/j.neuropharm.2013.07.027 23954492
237. Söderlund J Schröder J Nordin C Samuelsson M Walther-Jallow L Karlsson H Activation of brain interleukin-1β in schizophrenia Mol Psychiatry 2009 14 1069 1071 10.1038/mp.2009.52 19920835
238. Zhang PA Sun Q Li YC Weng RX Wu R Zhang HH Overexpression of purinergic P2X4 receptors in Hippocampus rescues memory impairment in rats with type 2 diabetes Neurosci Bull 2020 36 719 732 10.1007/s12264-020-00478-7 32198702
239. Wyatt LR Godar SC Khoja S Jakowec MW Alkana RL Bortolato M Sociocommunicative and sensorimotor impairments in male P2X4-deficient mice Neuropsychopharmacology 2013 38 1993 2002 10.1038/npp.2013.98 23604007
240. Naviaux RK Zolkipli Z Wang L Nakayama T Naviaux JC Le TP Antipurinergic therapy corrects the autism-like features in the poly(IC) mouse model PLoS One 2013 8 e57380 10.1371/journal.pone.0057380 23516405
241. Cheung KK Chan WY Burnstock G Expression of P2X purinoceptors during rat brain development and their inhibitory role on motor axon outgrowth in neural tube explant cultures Neuroscience 2005 133 937 945 10.1016/j.neuroscience.2005.03.032 15964486
242. Huang L Otrokocsi L Sperlágh B Role of P2 receptors in normal brain development and in neurodevelopmental psychiatric disorders Brain Res Bull 2019 151 55 64 10.1016/j.brainresbull.2019.01.030 30721770
243. Song XL Liu DS Qiang M Li Q Liu MG Li WG Postsynaptic targeting and mobility of membrane surface-localized hASIC1a Neurosci Bull 2021 37 145 165 10.1007/s12264-020-00581-9 32996060
| 36445556 | PMC9707125 | NO-CC CODE | 2022-12-01 23:20:20 | no | Neurosci Bull. 2022 Nov 29;:1-18 | utf-8 | Neurosci Bull | 2,022 | 10.1007/s12264-022-00986-8 | oa_other |
==== Front
Learn Environ Res
Learn Environ Res
Learning Environments Research
1387-1579
1573-1855
Springer Netherlands Dordrecht
9440
10.1007/s10984-022-09440-y
Article
Exemplification process in online education: a longitudinal study of mathematics teachers
http://orcid.org/0000-0002-2083-688X
Sevimli Eyüp [email protected]
grid.411776.2 0000 0004 0454 921X Faculty of Educational Science, İstanbul Medeniyet University, Istanbul, Turkey
28 11 2022
124
29 10 2021
3 11 2022
© The Author(s), under exclusive licence to Springer Nature B.V. 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The examples used in the teaching–learning process of mathematics have a crucial role in fostering conceptual understanding, and some variables can affect instructors' qualified example usage. This longterm study focused on mathematics teachers’ exemplification process in face-to-face and online learning environments. In this regard, the change in examples used by mathematics teachers were evaluated in terms of content preparation and presentation during the shift from face-to-face lectures to online classes. A longitudinal design was used in the study and the teaching processes of 14 middle-school mathematics teachers were observed over two semesters. Observation notes, course documents, and semi-structured interview data were analyzed, and content analysis findings were presented through descriptive statistics in order to compare content preferences in two different learning environments. The use of worked examples decreased, while the use of conceptual examples increased with the shift from face-to-face lectures to online classes. Also the length of time devoted to examples in online classes decreased, and examples were more teacher-centered. The interview revealed that mathematics teachers need support in terms of example preparation and presentation aspects in online learning environments. The other technological-pedagogical competencies that teachers might need to choose qualified examples in different teaching–learning environments are discussed in the light of relevant literature.
Keywords
Exemplification
Mathematics teachers
Online education
Teaching practices
==== Body
pmcIntroduction
One of the indicators explaining the effectiveness of the learning–teaching process is the quality of the interaction between the teacher, the student and the knowledge components. The pedagogical content knowledge of teachers, interactions with students, and presentation tools of knowledge in the different learning environments (e.g. face-to-face, hybrid, or online) can affect teaching practices (Borba et al., 2018; Francis & Jacobsen, 2013; Trenholm et al., 2016). Since the day Covid-19 started, many countries decided to switch from traditional to online education for different classroom levels (from K–12 to postgraduate level). After the outbreak of the pandemic, more emphasis has been put on how online learning environments can be used more effectively in the transition process between textbook, teacher and student knowledge (Demir et al., 2021; Rapanta et al., 2020; Sevimli et al., 2022). Because the discipline of mathematics has an abstract nature and accumulated knowledge structure, as well as requiring problem-solving processes, the content preferences of teachers in the Online Learning Environments (herein after OLE) can differ according to the Face-to-Face Learning Environments (herein after FLE).
Examples are one of the most important teaching contents of mathematics, because they have the potential to combine the abstract nature of mathematics and the problem-solving process with the visualization, proof and argumentation process in different learning environments (Mason & Pim, 1984; Rø & Arnesen, 2020; Watson & Chick, 2011). In this sense, determining which types of examples are used by teachers in the online education process and addressing the change in the time allocated to the examples will enable examination of the technological-pedagogical preferences of instructors. A growing body of literature has examined the role of examples (student-generated or teacher-centered) in mathematical understanding in face-to-face education (Bentley & Stylianides, 2017; Peled & Zaslavsky, 1997; Zaslavsky, 2019). Until recently, there was no answer for the question about how mathematics teachers choose mathematical examples and present that content in changing learning environments. Whereas it is well-known that the type of examples chosen in the teaching can differ depending on the type of knowledge to be taught, the teacher's experience, and the teaching environment (Atkinson et al., 2000; Rowland, 2008; Sevimli, 2016). Also, the advantages offered or the limitations caused by the changing learning environments can affect the presentation methods of the examples, which are the important content of mathematics, in the teaching process. Especially after the Covid-19 pandemic, there is the need for research in this area to determine the time allocated to examples and how mathematics teachers' choices of examples have changed in online classes.
The main purpose of this research was to examine how the examples used by middle-school mathematics teachers have changed in terms of content preparation and presentation during the Covid-19 pandemic. The research questions were formulated as follows: (1) how did the presentation methods of examples used by mathematics teachers change during the shift from face-to-face lectures to online classes? and (2) how did the types of examples used by mathematics teachers change during the shift from face-to-face lectures to online classes? By answering these research questions, knowledge about the characteristics of the examples prepared by mathematics teachers for use in classroom practices would be obtained. Besides, determining the time allocated for examples in online learning environments compared with face-to-face classes would provide an opportunity to understand teachers’ presentation methods of the examples. In this way, the quality of the teaching content used by teachers during the exemplification process could be evaluated from the perspective of the advantages and limitations of the OLE.
Conceptual framework
Exemplification in mathematics education
In a dictionary entry, an example is a noun that means “a little piece taken from the whole to illustrate the characteristics of it”. Examples are used to describe a concept or phenomenon, as well as to show the nature of a general rule (Mason & Pim, 1984). In educational literature, a glossary of example terms includes teaching content, both in the context of a textbook and in the context of teacher narration and student explanations. The key role that examples play in the process of developing and teaching mathematics as a discipline has required greater emphasis on the exemplification process of research on mathematics education compared with other fields (Watson & Mason, 2005). In parallel with researchers’ increasing interest in this issue, there have also been different variations in the definition of the concept. For instance, the mathematics education literature includes a special representation chosen to explore or explain a general principle (Watson & Chick, 2011), a tool used to explain mathematical knowledge (Bills et al., 2006), a kind of practice technique that allows conceptual communication between the student and the teacher (Peled & Zaslavsky, 1997), and content that allows the general to be seen in the particular (Mason & Pimm, 1984). Upon synthesizing the characteristics that all these definitions reveal for examples, it is observed that examples can have different roles such as summarization, explanation, interpretation, and confirmation.
Several classifications for examples have been developed to understand the characteristics of examples in mathematics education. In early stages, examples were divided into two categories such as generic (generalizable examples aimed at clarifying the subject) and specific (limited examples used to explain more specific cases) (Mason & Pim, 1984). With a similar perspective, Watson and Mason (2005) pointed to contents that emphasize the critical features of a concept while defining the boundary examples. The classification that is most used in the mathematics education research is that formed by Bills et al. (2006) who divide examples into generic examples, counter-examples, and non-examples. Counter examples can be used to counter a hypothesis or assertion, while non-examples can serve to clarify the boundaries of a concept. Rowland (2008) considered the use of examples by teachers under the categories of variables, sequencing, representations and learning objective, suggesting that novice teachers need specific guidance and assistance in appreciating the different roles of examples in mathematics in teaching. Across these categories of examples, there are also three other special labels of example types that researchers independently observe during the teaching process: prototype example, start-up example, proof example (Peled & Zaslavsky, 1997; Sevimli, 2016; Watson & Mason, 2005; Zaslavsky, 2019). While studies of the use of examples as a proof or argumentation tool have increased in the last decade (Rø & Arnesen, 2020), the classification was built on the use of examples during generalizing and disproving processes (Yopp & Ely, 2016).
The presentation method of the examples also can be an indicator factor for categorization. For instance, an example can be present as the teacher-centered methods (teacher is dominant in the process of sharing content) or learner-generated (students are actively involved in the process of generating, solving, or interpreting examples) methods. The related literature has demonstrated that it is important for learners to generate and solve examples and also for the types of examples presented in the classroom (Bentley & Stylianides, 2017; Watson & Chick, 2011). Much of the current literature on the exemplification process in mathematics education pays particular attention to the following topics: (1) the determination of the characteristics of the examples used in the teaching–learning (Bills et al., 2006; Sinclair et al., 2011; Watson & Chick, 2011), and (2) evaluation of the role of exemplification processes on mathematical reasoning or proving (Bentley & Stylianides, 2017; Peled & Zaslavsky, 1997). However, there are no previous reports in the literature on what kind of changes have occurred in the choice of examples in technology-supported teaching environments or how teachers' choice of examples is shaped in the process of reflecting on knowledge to be taught (textbook knowledge) to knowledge actually taught (teacher practice). Examples are crucial reference sources for comprehending the knowledge provided by teachers in the classroom and for knowing which sources are used for what purpose in the process of didactic transposition of scholarly knowledge (Chevallard & Bosch, 2013). In the next section, the didactic characteristics of these environments are discussed in order to evaluate reflections of the exemplification process in online learning environments.
Teaching mathematics in online learning environments
With the frequent use of Internet technology and interactive digital tools in the teaching process, classes have started to involve virtual, online, or digital adjectives next to their names. These terms actually mean to conduct education on web-based platforms. With the shift from traditional (physical) classes to virtual classes, new pedagogical information began to be needed. Because it is not only enough to integrate digital tools into the classroom in the OLE, but it is also necessary for teachers to be able to prepare suitable teaching content and present this content in accordance with the constructivist approach in these classes (Engelbrecht et al., 2020). Digital pedagogy stands out as a skill that teachers should have for planning, implementing, and evaluating teaching in the OLE, especially in online or blended learning approaches. Digital pedagogy is a domain of competence based on a constructivist approach aimed at effective use of modern technologies in the teaching–learning environment. Digital pedagogy includes dimensions such as digital literacy, system management skills, knowledge of appropriate digital teaching tools and creating effective teaching content, while the present study focuses specifically on the dimension of preparing effective teaching content. To ensure effective learning environments, teachers need to prefer and use more pragmatically digital resources when teaching content in live online classes (Bennison et al., 2020). Criteria that can be considered to assess whether effective teaching content is presented in accordance with appropriate pedagogical approaches when teaching mathematics are whether pedagogically-high cognitive content is used, and how it is presented and scheduled (Trenholm & Peschke, 2020).
A series of recent studies have indicated that mathematics teaching practices in the OLE differ from the traditional classroom (face-to-face lectures) and that technological-pedagogical competencies shape applications in mathematics classes (Arcavi, 2020; Bennison et al., 2020; Borba et al., 2018). With the help of tablets with flexible writing tools that allow online sharing (Karal et al., 2015), and with the support of web 2.0 tools designed particularly for mathematical learning areas (such as dynamic geometry software and online data analysis platforms), mathematical concepts, symbols and solution process steps can now be easily presented to students interactively in the OLE. Arcavi (2020) also pointed out that digital tools used in mathematics education increase the capabilities of teachers to access and manage teaching content. Moreover, Moodley (2019) claims that effective use of online social media networks can support the professional development of teachers in virtual classrooms. Apart from advantages, the difficulties encountered during the mathematics teaching process in the OLE also are discussed in the literature. For instance, when Demir et al. (2021) examined the attitudes and opinions of the teachers in mathematics lessons conducted with online education during the first year of Covid-19 pandemic, they determined that teachers had difficulties, especially with technical problems and lack of interaction. It was also reported in the literature that interaction is limited when teaching mathematics in the OLE, communication tools are insufficient and gestures/facial expressions cannot be observed (Trenholm & Peschke, 2020). Mathematics, which has more symbolic language, can cause communication limitations because of the lack of software or hardware in the process of teaching online classes. In this context, instructors cannot use drawing tools flexibly when presenting content (text adherence or limited typing flexibility with the Mouse), and the learning management system (LMS) does not have a discussion web-log or cooperative learning activities. When the experiences of instructors in the compulsory transition to online education in undergraduate level mathematics were examined, Sevimli (2022) found that content that requires the proof of integral theorems and the use of integration techniques in virtual classrooms was reduced by lecturers because of the limitations of menu tools and more dependence on the calculus textbook in the OLE.
Upon analyzing the relevant literature, no previous research was found to have evaluated the process of preparing effective teaching content, which is one of the technological-pedagogical competencies, through online classroom practices of teachers. After the Covid-19 pandemic, the impact of the transition from the FLE to the OLE on teacher practice is still important. During this urgent and unprepared transition, it is unclear whether teachers have changed their teaching practices in the FLE and prepared content appropriate to the OLE, and how teachers have prepared these contents. This study was conducted to determine the types of examples used in the teaching of mathematics in the online education process, and thus to understand the technological-pedagogical competencies needed for effective teaching of content.
Method
Research design and participants
The examples preferred by mathematics teachers in two different learning environments were examined in this study, with a longitudinal study being used in the design of the method. The longitudinal study is a pattern of developmental research that requires working with the same participants over an extended period of time to reveal the changes in their specific behavior in different environments. Participants in the study were mathematics teachers and their choices of examples were obtained via a longitudinal design by accessing the same group of individuals (minimizing individual differences). Because the teachers' previous experiences and the type of schools in which they work also could affect the exemplification process with the learning environments, the study participants were selected according to purposeful sampling. The participants comprised 14 mathematics teachers working in five different middle schools in the Middle Black Sea Region of Turkey. Nine participants were female and five were male, and all participants had professional experience between 11 and 20 years. All participants declared that they participated in a course (2 h) with web 2.0 tools within the scope of inservice training before the pandemic.
Factors affecting school choice are socioeconomic status and cooperation protocol. Before participant teachers were selected, a pool was created and preliminary interviews were held with the Psychological Counseling and Guidance Services in these schools. Within the scope of the interview, the data that can be accepted as an indicator for socioeconomic levels of the school students were collected. According to this, the students in the schools where the participants work are similar according to the following characteristics: the rate of having a personal computer (62–70%), the rate of having an internet connection (80–93%); and the average years of education of the members living in the family (8–10 years). The student profile in the schools included in the sample represents the national average in terms of family income. The use of a smart board in the FLE, and access to the internet in the OLE, can affect teacher preparation and student motivation. For this reason, the schools where participants work are located in the city center, and these schools use a smart board during face-to-face education. Within the scope of the cooperation protocol, preservice teachers were given the opportunity to practise in the relevant school and the academics made observations at the school.
As part of the study, the examples used by mathematics teachers in two different learning environments were examined and the fall semesters of the 2019–2020 and 2020–2021 academic years were included in the research. In the spring of 2020, the transition from face-to-face training to online education occurred because of the Covid-19 pandemic but, in the first place, training was carried out with television broadcasts instead of the OLE in Turkey. The spring semester was not considered in the research process in order to prevent data loss in the process of comparing learning environments and also because of to the lack of sufficient motivation and time devoted for teachers to prepare content. In addition, it was possible to study again with the same participants during two semesters (fall semesters of the 2019–2020 and 2020–2021 academic years), by virtue of the fact that there is a cooperation protocol between the middle schools. Presenting the characteristics of learning environments contributes to a better understanding of the findings obtained within the scope of the research. For this reason, general features related to face-to-face lectures and online classrooms are included in the next section.
Settings
In this study, we presented a general picture of the experiences of mathematics teachers regarding content preference during the sudden transition to the OLE from the FLE. For this reason, it is important to understand how the change in the learning environment reflects on the content preference of participating teachers under the conditions that the professional competencies of the teachers and the socioeconomic levels of the students have similar characteristics. These characteristics were obtained through the researchers' observation notes and video recordings of the teaching process. The common characteristics of both the FLE and OLE can be listed as follows. Observation data supporting these characteristics are also given, respectively: (1) direct instruction method (FLE: 47%; OLE: 51%) and question–answer technique (FLE: 39%; OLE: 45%) are often implemented more by the participants; (2) the order of the units of study and the length of time devoted to the unit are similar in both learning environments (with the time devoted to natural numbers, sets and fractions being common in the curriculum and the participants adhering to it throughout the teaching period); (3) the introduction of units or topics begins with activities, and then examples and definitions are presented (In all units in the 6th grade mathematics textbook, the introduction of the concept begins with an activity); and (4) middle school mathematics textbooks printed by the Ministry of Education were sent to students free of charge.
Characteristics of learning environments in which research data were collected were also examined according to their differences. The characteristics that distinguish FLE and OLE can be listed as follows: (1) classroom attendance was higher in the FLE than in the OLE (According to the course attendance chart at FLE, the average attendance rate was 94%. The live courses attendance rate was 80% in the statistics of OLE); (2) participating teachers chose supplementary sources in addition to the textbook in the FLE (all participants) while, in the OLE, 11 out of 14 participants used web-based educational portals in addition to individual narration; (3) compared to the FLE, students were given more assignments in the OLE; and (4) students’ generated examples decreased during the transition from FLE (29%) to OLE (12%) and there was an increase in the teacher-centered example presentations.
One participant in the OLE shared a document camera, while five participants shared a graphic tablet, and eight participants shared pdf annotator tools (write notes, edit drawings, solve problems, etc.) as teaching content. Teachers accessed online classes through the e-content platform, for which the Turkish acronym is EBA (The Educational and Informatics Network). EBA includes live video conference applications with supportive teaching materials such as video, games, z-books, TV, radio, discussion platforms and also teachers can be streamed live course on online cloud-based platforms such as ZOOM Cloud Meeting (Fig. 1). Within the context of this study, live lecture content in the OLE and teaching processes in the classroom in the FLE were considered.Fig. 1 An image from the EBA platform
In order to understand the research framework, it is important to understand the level of the content presented in teaching environments and the description of the learning area studied. Middle-school education lasts for 4 years (Grade 5–8) in Turkey, and there are five mathematical learning areas at this level. The study was carried out through the ‘numbers and operations’ learning area of 6th-grade mathematics. Grade 6 was selected for the study to reduce the effect of the familiarization phase of students on the teaching content preference during the transition from the classroom teacher to the branch teacher (from primary to the middle-school level). The content presented in Table 1 illustrates the average length of follow-up time of the lectures observed in the learning area of numbers and operations. While determining the distribution of lecture hours followed in learning environments, the time allocated to the sub-learning areas in the annual plans was taken into account. These data show that lectures given in both FLE and OLE are significantly represented and that a similar number of lectures have been selected in such a way as to ensure the validity of the content. In the scope of the study, an average of 33 lecture hours were provided to FLE participants and an average of 35 lecture hours were provided in the OLE, so that a total of 462 lectures were observed in the FLE and 490 lectures were observed in the OLE.Table 1 Distribution of lecture hours in learning environments
Learning area Sub-learning area Lecture hours
FLE OLE
Numbers and operations Operations with natural numbers 7 6
Factors and multiples 8 11
Sets 5 5
Integers 5 4
Operations with fractions 8 9
Total 33 35
Data-collection process
In the data-collection process, it is necessary to determine the types of examples that middle-school mathematics teachers prefer to use in their lectures and to determine the issues affecting the choice of examples. From this viewpoint, observing multiple teachers only at a specific lecture time, or observing a limited number of teachers for a large number of lectures, could limit the trustworthiness and transferability of the study. Observing the lessons of different teachers in different schools over the lecture of a teaching period is quite time-consuming and a process that few researchers can handle alone. It was decided to collect data over on the grounds that it would be incomplete to conduct analysis through the participants' lecture plan or lecture notes (the teacher does not reflect some content in the lecture notes, uses technology support, etc.). The data in this research were obtained over multiple observation sources for the periods of two years and within the scope of the School Experience course, which is given in the last year of teacher training programs in Turkey and includes observations and activities made by preservice mathematics teachers over a period of 12 weeks under the guidance of a master teacher. The theoretical part of this course (for one hour each week) was conducted by the researcher and the master teachers were observed during the practical part of the course (for four hours each week). This course has an ideal design for the research data collection strategy because it allows a large number of live lecture observations and interviews with teachers. In this study, 16 preservice mathematics teachers enrolled in the School Experience course participated in the workshop on ‘characteristics and taxonomy of examples used in mathematics teaching’ in the theoretical part of the lecture, and the examples used in the mathematics textbook were classified together with the preservice mathematics teachers to enable them to gain experience. Preservice mathematics teachers who had received necessary training to evaluate the different example types used in the teaching of mathematics observed teachers participating in the practical part of the School Experience course, as well as accompanying the researcher in the data-collection process of this research. Thus, the examples used by participating teachers in classroom practices were obtained as a result of synthesizing observation reports of multiple preservice mathematics teachers.
Data-collection tools
In the present study, the exemplification processes of mathematics teachers in teaching practices were evaluated under the themes of content presentation and preparation processes. For this purpose, data related to the length of time devoted to the use of examples and types of examples used were achieved through observations and interview techniques.
Lesson observation form
The Lesson Observation Form (LOF) was needed to access standardized data in lectures, observed by researchers and preservice mathematics teachers. The LOF was designed to include observation notes for two criteria (types of examples used and duration), which were considered in this study. The exemplification framework in the mathematics education literature, field observations, and expert opinions were considered when developing the LOF. First, because it is necessary to determine the types of examples used by teachers in the teaching process, the classification of examples by Bills et al. (2006) and Smith and Stein's (1998) taxonomy were blended together, building a classification that also refers to the epistemological characteristics of examples as well as their pedagogical qualities. This classification schema synthesizes the purpose of using examples by teachers in classroom practice and the reflection of scholarly knowledge in the textbook through examples. According to this classification schema, the four main types of examples are start-up, worked, conceptual, and proof examples. The length of time devoted to the use of examples in the class is one of the criteria, which is recorded in the LOF. In field observations, it was determined that the time spent for presentation and solution of examples was at least 1–9 min. In that sense, the “duration” criterion was also added to the LOF to determine the average time devoted to each example using an equal interval classification criterion of two minutes.
With the usage of LOF, the teaching content of each participant was coded for 12 weeks (three hours for each week). In addition, the researcher directly followed the lessons of the participating teachers along with the preservice mathematics teachers for some weeks, with the validity and reliability of the observations being confirmed again by comparing them with the student notebooks. In this way, the vast majority of the examples in the dataset included the content shared by multiple sources (the researcher's note along with the preservice mathematics teachers’ observations and the students' notebooks). In the selection of the student notebook, the students who took the most-comprehensive notes and largely matched the observers' previous observation notes were included in the data of the study. A total of 90 examples in textbooks, 2436 used in the FLE and 3290 used in the OLE, were evaluated within the scope of the study.
Semi-structured interviews
Semi-structured interviews were carried out after the LOF data were collected and used to support these data. The interview form was used for all participants to understand their views on the advantages and limitations of the OLE during exemplification process. During the preparation of the interview form, the findings of the LOF were used in the context of duration and types of used examples. In this sense, the experiences of the participants in the process of preparing and presenting the examples used in online learning environments were questioned. The interviews were conducted through an online conference program and recorded with permission. In the interview protocol, it was stated that the participants could leave the interview at any time, some examples of the content that they used during the OLE were presented, and their consent was obtained by sharing the personalized LOF findings. In the interview, which lasted between 20 and 30 min for each interviewee, participants were asked what are the advantages and limitations of the OLE in (1) the presentation of mathematical examples; and (2) the preparation of mathematical examples? After the interview questions were prepared, the opinion of two experts who completed their PhD in mathematics education was obtained. The suggestions given by experts about the participants’ reactions to their the LOF findings were taken into consideration. Interview transcripts were also sent to the participants and, after their approval, the analysis phase was started.
Data analysis
Two analysis methods were used for two types of data (observation notes and interviews). The categorical data obtained by the LOF in classroom observations were analyzed through descriptive statistics and presented via percentage-frequency distributions. The LOF, which was the primary data-collection tool of the research, was used by the researcher and 16 preservice mathematics teachers synchronously to observe the lessons of participating teachers. As it can be seen in Table 2, each example used by the teacher in the classroom was first numbered before coding was carried out according to the criteria of sub-learning area, types of example used, duration, and sources used. Types of examples used by participating teachers in both environments (FLE and OLE) were encoded under one of the example types (start-up, worked, conceptual, or proof) and represented by descriptive statistics. Start-up examples were employed to clarify a concept and present the characteristics of the concept. For instance, an example presented to show that the value of the distance will not be negative when introducing the concept of absolute value can have a start-up type. Worked examples are used to practise and calculate to reinforce the mathematical concept being taught. For instance, after the concept of absolute value is given, the calculation of absolute values of integers between -5 and 0 can be evaluated within the context of the worked example. Conceptual examples contribute to establishing a relationship between different knowledge and concepts. For instance, calculating the total length of the path taken by two vehicles moving away from each other in the opposite direction and at different speeds, with a starting point of 0, and proportioning the paths taken connects the concepts of absolute value and fraction. Proof examples are used to generate arguments to convince other people that something is true, and they include tasks such as generalization, inference, and inverse exemplification. It can be intuitively felt by students with proof examples that the sum of the absolute values of any two integers must be less than the absolute value of the sums of these numbers (triangle inequality). This classification schema, which categorizes the examples that teachers use in classroom practice, is important because it both takes previous classifications as a reference and is structured according to the teachers' reasons for the choice.Table 2 The example classification criteria in the LOF
The presentation methods of examples as the teacher-centered (teacher is dominant in the process of sharing content) or student-centered (students are actively involved in the process of generating, solving, or interpreting examples) were also shared with a percentage rate. The place of the examples in the teaching contents was also examined over the duration (in minutes) allocated to each example. While analyzing the interview data, the responses of each participant to the interview questions were coded and the distribution of the participants’ responses in the framework of the common categories corresponding to these codes were described via the table. The opinions upon which at least two participants agreed were taken as a reference in categorization processes. The interview data gathered from the participants were evaluated under the advantages and limitations categories in line with the questions in the interview form. Under these categories, participants' views of content preparation and presentation processes in online classes were coded together with sub-categories. Because some participants had views about both advantages and limitations, participants were coded under more than one category. The examples in the figures were translated from Turkish to English by the researcher to describe the teaching process in the learning environments.
Validity and reliability
In this study that was structured according to the qualitative approach, credibility, transferability, and consistency processes were evaluated respectively for validity and reliability. The method used for the credibility of the research is as follows. First, the longitudinal research design was used in the study and the data collection-analysis process took two years, enabling us to establish a long-term interaction with the data. Second, multiple observations were carried out with the same participants through purposeful sampling to ensure that the data could be transferred to other situations. The physical characteristics of learning environments and the participants' teaching experiences in these environments were described in detail via observation (under the Setting heading), thus determining the transferability framework of the findings. During the observation process, because the lectures considered the introduction to units and/or concepts, the study focused on the ways of using examples regarding concepts used in the teaching process. In this way, although the lectures followed through observation notes correspond to a third of the semester, the fact that the observed lectures were introductory to the concept/unit increased the generalizability of findings. In addition, even though a limited number of teachers (14 mathematics teachers) took part in the study as participants, it was thought that comprehensive findings would be achieved in the selection of these teachers because of the requirement set by the Ministry of National Education (in Turkey) to be master teachers.
In order to ensure the consistency of the research, methods of diversification in the data source, multiple observation notes in the data-collection process, and inter-rater reliability in the data analysis process were used. Therefore, student notebooks and textbooks were considered along with observations when selecting the examples participants used in the classroom practice. In addition, the diversification of data sources was ensured by checking the consistency of observation findings with the interview process. The researcher's biases were reduced by considering the notes of more than one observer in the lecture follow-up. The coding consistency of different evaluators, who were trained in the classification of example types and processed observation notes into standardized forms, were checked. The reliability coefficient was calculated by the Miles and Huberman’s (1994) inter-rater reliability formula [number\, of\, subjects \,on\, which\, consensus\, is\, reachednumber \,of\, all\, subjects×100]. Consequently, 278 randomly-selected examples were encoded by three different raters and, as a result of cross-comparisons, the inter-rater reliability coefficient ranged from 0.81 to 0.88. These results indicate that the coding for characteristics of examples was highly compatible. In order to check whether there was any data loss in the observation notes of the prospective teachers, student notebooks in the FLE and lecture videos in the OLE were examined; both sources showed that observation notes had high comprehensiveness.
Findings
While presenting the descriptive findings, the examples used by participants in the FLE and OLE were primarily considered in terms of time devoted to example presentation. Later, the results of the example types used in different learning environments were evaluated by comparing them with the textbook. The interview findings were presented categorically to facilitate understanding of the advantages and limitations of the OLE compared with the FLE during the exemplification process.
Duration and presentation methods of examples
One of the most-important objectives of the research was to evaluate the examples that mathematics teachers use when teaching during their lectures in the FLE and OLE. We determined the average number of examples used per lecture and the total number of examples in each learning environment. The findings related to the analyzed contents showed that the average of the total number of examples used by per participant was 174 in the FLE, whereas this number was 234 in the OLE (Table 3).Table 3 Frequency of characteristics of the analyzed contents
Characteristics Frequency
FLE OLE
Total observed lesson hours 462 490
Average observation hours for per participant 33 35
Total number of examples in lecture notes 2436 3290
Average number of examples used by per participant 174 234
Average number of examples used in per lesson 5.2 6.7
Average minute devoted to an example 4.5 3.3
It was also found that the average number of examples used each lesson was 5.2 in the FLE and 6.7 in the OLE. This finding suggests that participants used up to a third more examples in the OLE than in face-to-face lectures. In order to scrutinize this finding more thoroughly, the time devoted to examples in teaching environments was also examined. The average time participants devoted to an example for presentation was 4.5 min in the FLE and 3.3 min in the OLE. The data available in Fig. 2 include the durations (in minutes) that participants used for presenting examples in the FLE and OLE. Participants devoted between 4 and 6 min to 34% of the examples that they used in the FLE. In the FLE, examples lasting longer than 6 min include 29% of all examples, while examples lasting less than 2 min constitute 10% of all examples. It was found that participants used examples requiring 2–4 min in the OLE in the teaching process (40%). Examples lasting less than 2 min are also often available in the OLE (27%). When comparing the two learning environments, it was observed that participants frequently used examples taking less than two minutes in the OLE, while they rarely tend to use examples taking longer than 6 min. Besides, it was found that the use of examples requiring more than 4 min in the OLE showed a trend of decreaseing compared with the FLE.Fig. 2 Distribution of the time devoted to the examples
The participants’ teaching content preferences can be affected by the way in which the content is used in classroom practice, as well as the length of time devoted to the content. To determine the presentation method, the use of each example in the FLE and OLE is classified as teacher-centered or student-centered. When the two learning environments are compared in terms of presentation methods, student-centered presentations take more time in the FLE compared to the OLE (Fig. 3). In this sense, 21% examples were completed in 4–6 min and 17% of examples last 6–8 min under the student-centered presentations in the FLE. Also student-centered presentations usually were completed in a short span of time (between 2 and 4 min) in the OLE. When the time allocated to teacher-centered presentations were evaluated, it was found that the exemplification process was completed in a shorter time in the OLE, but no significant difference was encountered in terms of the time allocated to teacher-centered presentations in the FLE. In this regard, 22% of the examples were completed in between 2–4 min and 17% of examples lasted between 0–2 min under the teacher-centered presentations in the OLE. When the findings in Fig. 3 were evaluated as a whole in terms of the learning environment versus presentation methods, it can be seen that 53% of the examples were student-centered while 47% were teacher-centered in the FLE. In the OLE, teacher-centered examples were more frequent than student-centered examples.Fig. 3 Distribution of the duration and presentation method of examples
To compare the exemplification process of the same participant in different learning environments in terms of the presentation-duration relationship, two examples that P7 used while explaining the fraction topic were shared. For instance, the example of fraction in Fig. 4 used by P7 in the FLE was first presented to students with a worksheet and then it took about 4 min with a student explaining his solution on the board. While the student solved the example on the board, the other students wrote in their notebooks.Fig. 4 A student-centered example used by P7 in the FLE
The example used for the same topic (fractions) in the OLE was completed in less than 3 min, although it requires more steps and solutions (Fig. 5). P7 used a learning toolkit to simulate the process of adding denominators in unequal fractions through the Geogebra software and she shared a link with the students for simultaneous participating in the solution of the example.Fig. 5 A student-centered example used by P7 in the OLE
When participants were asked about the advantages and limitations of online environments in terms of the time devoted to the examples within the scope of the interview, the sub-categories in Table 4 were obtained. The sub-category to which the participants referred more frequently under the advantages of online teaching environments category was the availability of Synchronized learning tools (6 out of 14 participants). For instance, P9 stated that the availability of web 2.0 tools in online environments offers the opportunity for students to work together (see Table 4). In addition, four participants considered as an advantage that Recorded lectures can be presented to students without wasting time, referring to the fact that the content can be re-watched according to students' own learning speed in the OLE. For instance, P14 considers the opportunity to return to the recorded content as an advantage in terms of using more examples and being able to share video content in and out of class. Consequently, when the participants' views of the advantages and limitations of the OLE were compared in terms of content presentation, it was determined that the participants referred more to the technological advantages offered by online classes (Table 4).Table 4 Participant views on the advantages and limitations of the OLE in terms of content presentation
Categories Sub-categories Participants Excerpt from participants
Advantages Synchronized learning tools P2, P7, P5, P6, P9, and P13 P9: I think it is an advantage that math-based applications such as Geogebra or Mathingon, where students can be engaged simultaneously on the same example, are easily available in online classes
Recorded lectures P4, P11, P12, and P14 P14: It can take time to wait for the examples to be written by the students. Since the lessons are recorded in live online classes, the students can return to the course content they want, and I was able to use the increased time by giving more examples in or out of class
Limitations Classroom management P3, P7, P10, and P12 P10: I had to give some content, which I solved with the collaborative approach in face-to face classes, with direct instruction in online education. Because it is difficult to provide classroom management in online classes and to determine who participates in the example solution process in detail
Motivation P1 and P8 P1: It is not easy to keep students' attention in front of the computer screen for a long time. When I asked students a question, very few are motivated to answer. I shared most of examples in the worksheets after the course, so students can take a deeper look at the teaching content
When participants were also asked why they prefer more teacher-centered presentations, although student-centered presentations require less time in the OLE, answers generally focused on the limitations of OLE. Under the limitations category, the sub-categories of classroom management and student motivation were identified. Four participants (P3, P7, P10, and P12) pointed out that they preferred teacher-centered presentations to student-centered ones because of difficulties in classroom management. Two participants, on the other hand, stated that they use examples that do not require a long time to be presented in the OLE, citing the decrease in motivation over time. For instance, P1 pointed out that student motivation is low over the long term in the online course and therefore he shared the worksheets which contained examples after live online classes.
Types of examples used
Another purpose of this research was to examine the types of examples used by teachers in different learning environments. Upon examining the descriptive results of the distribution of examples in Table 5, the most-common type of example involved in the 6th-grade mathematics textbook, which is commonly used by participants, is the worked example (58%). It was found that 28% of examples in the textbook were conceptual examples, but that a limited number of start-up and proof examples were available. Upon examining the examples chosen by the participants in the teaching process in the FLE, it was observed that about two out of every three examples used in the classroom were worked-type examples, followed by conceptual examples. While more than half of the examples chosen by participants in the OLE were still of the worked type, the proof example was rarely used in these classes.Table 5 Example types used in learning environments
Learning environment Example types Agreement rate of example
Start-up Worked Conceptual Proof
n % n % n % n % n %
Textbook 10 11 52 58 25 28 3 3
FLE 26 15 113 65 30 17 5 3 42 47
OLE 51 22 119 51 61 26 3 1 68 76
*n: the average number of examples used per participant
Comparing the learning environments, it was found that the types of examples which participants chose in both teaching–learning processes were similar to each other, but there were differences between the frequency of choices of examples according to the learning environment. Even though the most-commonly chosen example type in both the FLE and OLE was the worked type, there was a decrease in the percentage of preferences of worked types, while there was an increase in the choice of conceptual types, when switching from face-to-face lectures to online classes. While the utilization rate of the start-up example in the OLE was higher than for the FLE, the least-chosen type of example in both learning environments was the proof example. When the examples that are used in learning environments are compared to the textbook, the most obvious change in the FLE is the decrease in the conceptual type of examples whereas there is a remarkable change in the OLE in terms of the start-up type of examples. To examine the reflection of the examples that are contained in textbooks on the teaching practice, how many of these examples were used in the FLE and in the OLE also was investigated. Textbook examples were used more often in the OLE when compared with the FLE. It was determined that 47% of the examples in the textbook were used in the FLE while 76% of examples were used in the OLE by direct quotation.
It was also ascertained through interviews how online learning environments affected the participants’ content (example) preparation processes. Five participants stated that online learning environments contributed to the process of preparing examples in terms of resources diversity (Table 6). Participants, who were evaluated their opinions in this sub-category, pointed out that qualified examples can be presented to students without wasting time via e-textbooks that can be easily integrated into the OLE. For instance, P3 stated that video-supported examples in the OLE were chosen from the EBA platform so that a wide variety of examples could be delivered to students faster. When asked about the effect of the shift from face-to-face lectures to online classes on the chosen sources during access to examples, P2 and P7 stated that they benefited from the online platforms (e.g. CODAP, Learning apps, Pool Everywhere, etc.) in order to use different type of representations in the OLE during the exemplification process. P5, P8 and P11 considered it as an advantage that more-personalized content could be generated in the OLE through interactive group work. Thus, the teacher can organize suitable content for different levels of students, and students can choose appropriate examples from among these content in the OLE. When the participants' views on the advantages and limitations of the OLE were compared in terms of content preparation, participants referred more to the limitations caused by the changing learning environments (Table 6; Fig. 6).Table 6 Participant views on the advantages and limitations of the OLE in terms of content preparation
Categories Sub-categories Codes/participants* Excerpt from participants
Advantages Resources diversity P2, P3, P7, P12, and P13 P3: I was able to easily share different types of examples on digital platforms or ready-made electronic resources with the students during the online lesson. For instance, I was able to support the topics in the "numbers and operations" learning area with available examples, which were supported with video on the EBA platform
Personalized content P5, P8, and P11 P11: I give students the opportunity to generate their own examples in the breakout rooms. Thus, we were able to generate personalized contents and share them in the classroom in a short time
Limitations Content-oriented platforms P1, P3, P4, P6, P9, P10 and P14 P6: Online learning platforms or virtual manipulatives generally include start-up examples. It can be difficult to find content suitable for each outcome in the curriculum. For this reason, I would like to use the content that has been developed specifically for the learning outcomes
Synchronized writing tools P2, P5 and P12 P2: There are some examples in my lecture notes that I used as prototypes in the lessons, but I could not use them in online classes because we suddenly switched to online education. Since the writing tools only allowed me to write, I could not solve the worked type examples with the participation of students
Fig. 6 Example types used in the classrooms
Under the limitations category, the sub-categories of content-oriented platforms and synchronized writing tools were identified. Half of the participants stated that content-oriented platforms that are compatible with the curriculum are limited or that they are not aware of the qualified resources available, if any. Three participants, on the other hand, stated that they could not use the examples in their lecture notes because of the lack of synchronized writing tools. For instance, P6 stated that she referred to web sources that can be used more flexibly in the OLE, but that she does not have clear knowledge of their compatibility with the curriculum because she has reached these contents via her own experience. P2, on the other hand, stated that he benefited from his personal example space in the FLE, while he stays more committed to the textbook in the OLE. Figure 5 includes two different types of examples about fractions that P2 uses in face-to-face lectures and online classes. It seems that P2 reflects a proof-type example in a face-to-face classroom using the whiteboard, but presents a start-up example in an online classroom via web 2.0 tools. When it was asked why he (P2) gave limited places in both classes to proof examples, P2 stated that he carried out the introduction to the ‘numbers and operations’ in the FLE via tasks and thus used examples after the definition of the concept by practising instead of providing knowledge about the concept. According to him, it is more difficult to present tasks in the OLE. For this reason, P2 stated that there is more need for the use of start-up examples that include the definition of the concept in the OLE.
Discussion
The research findings were discussed under the themes of content presentation and preparation processes. While the findings regarding the presentation methods of the examples were discussed, the time allocated to the examples and the number of used examples were emphasized. The main findings regarding the presentation of the examples showed that more examples were used in the OLE compared with the FLE, and that the average length of time devoted to the presentation of an example in the OLE decreased. Taking into account the average number of examples (5.2 for FLE and 6.7 for OLE) given by participants per lesson and the average time devoted to an example (about 4.5 min for FLE and 3.3 min for OLE) in the classroom practices by participants, it seems that example content is used in more than half of the lessons (lesson duration is 40 min) in both learning environments. Based on this finding, it can be concluded that examples are an important component of online mathematics teaching as well as face-to-face instruction (Bills et al., 2006). One of the reasons why the participants used more examples in the OLE might be that they needed more examples when presenting the mathematical definition or that they included less of other content (tasks, exercises, proofs, etc.) in the teaching process. The findings supporting this inference were actually encountered in the interviews. One of the participants (P10) stated that, instead of introducing the concept with teaching contents (tasks, problem situations, etc.) that require collaborative work, lectures are organized through giving definitions and presenting examples. Nonetheless, in accordance with the constructivist approach, the student should be able to reach an understanding with their previous knowledge instead of presenting the definition of the concept directly. The teaching content in the online classrooms should be organized in collaborative participation to foster conceptual understanding and promote mathematical communication (Borba et al., 2018).
The finding that participants use more examples in the OLE compared with the FLE might provide the basis for two inferences. First, with the technological advantages offered by online classes, teaching content (examples) can be brought together with the students in a shorter period of time (Arcavi, 2020; Trenholm et al., 2016). In the interview findings, participants frequently stated that they benefited from the advantages of the OLE by using digital learning tools and recording the content. The second inference is that there is a change in the way in which the examples are presented. In this study, how examples are presented in classroom practice was also examined. The example presentation, which is student-centered and more suitable for the constructivist approach, decreased with the transition from face-to-face lectures to online classes and a significant rate of the examples were used in the dominance of teachers. From this perspective, in an environment where teacher–student interaction is more limited (OLE), teachers can be considered to have a more-dominant role in managing the teaching process. The greatest limitation of online classes compared with face-to-face classes is the lack of interaction (Borba et al., 2018; Demir et al., 2021; Trenholm & Peschke, 2020). In fact, it is possible for the student to take the floor and generate examples to carry out solutions on the virtual board in the OLE, which in short means that the lectures are interactive. To increase interaction, Borba et al. (2018) suggest that interactive digital teaching materials should be prepared in the OLE to foreground student-centered teaching in accordance with the constructivist understanding. From the beginning of the Covid-19 pandemic, teachers in Turkey accessed online classes through the e-content platform, which is known as EBA and is freely available. But it was observed that participants do not benefit from the advantages offered by this platform. This finding also indicates that participants need support in choosing and using effective teaching content, which is one of the technological-pedagogical competencies.
The other theme that represents the focus of the discussion is the type of examples that teachers use when preparing the teaching content. Participating teachers used worked types of examples in the FLE, while they used conceptual and start-up examples more frequently in the OLE. It is more important to use different types of examples with similar rates in the class than to use a single type of examples (Sevimli, 2016; Watson & Chick, 2011). From this point of view, it can be claimed that the example diversity in the OLE is greater and that therefore more rich contents are preferred. The finding distinguishing this study from previous research is that, with the transition to online classes, the use of conceptual examples that support relational understanding increases, while the use of the worked type of examples decreases. Because of the preference for textbook-compatible content in the OLE, more varied types of examples might have been used in these classes. As a matter of fact, when the sources from which the examples were selected were examined, it was found out that the examples in the textbook were more directly used in the OLE. Similarly, Sevimli et al. (2022) found that the frequency of use of textbooks by middle-school mathematics teachers during the transition from face-to-face education to distance education changed in that teachers used the textbook more often in distance education. Another reason for more-diverse example types being used in the OLE is that digital sources can be used flexibly in these classes. Although teachers have been caught unprepared for the transition to online classes, materials supported by internet technology can be used in these classrooms, which contain various types of examples. Provided that these teaching materials are defined specifically to textbooks, it is important that the interactive-dynamic teaching contents that can be used in online environments are integrated with internet technology (video-supported, QR-coded, web 2.0-integrated contents).
Just as online education offers an advantage in terms of content preparation, inability to use teaching habits in online classes might also have affected teachers’ content preferences. For instance, the decline in participants' digital writing ability, their lack of sufficient knowledge, and lack of time to prepare content in the OLE also might have forced them to adhere more to the textbook. Taking into account that participants in this research frequently used the pdf version of the textbook in the OLE, it can be confirmed that they more often refer to teacher-centered approaches because of their inability to use digital writing tools flexibly. The interview data also revealed that some participants had difficulties because of a lack of writing tools while sharing the examples of their previous lecture notes in online classes, and therefore criticized the students' not taking notes. The limited teacher–student interaction and lack of teachers' knowledge of choosing the appropriate digital tool can be an obstacle to the use of effective content teaching in the OLE. Similar to the findings of Francis and Jacobsen’s (2013), participants in the current study requested professional development on web 2.0 tools that can be used in mathematics classes in order to reflect the various sources offered by online classes on effective teaching practice. This finding indicates that teachers are aware of their shortcomings in terms of the technological-pedagogical aspects and are motivated to participate in professional development seminars in order to improve the quality of teaching in the OLE. On the other hand, Li et al. (2022) found that teachers involved in the online inservice teacher training course felt less satisfied with their experience of motivation and interaction during the training. Before the Covid-19 pandemic, the mathematics teachers in this study also participated in an online inservice teacher training course about the use of web 2.0 tools. Based on the findings of the present research and the results of previous studies, it is better to engage teachers with interactive digital learning tools that can enable with the mathematical examples, instead of explaining the general use of web 2.0 tools in inservice training.
Conclusion and suggestions
The study suggests that the length of time devoted to examples in the OLE was shorter than in the FLE, but also that examples were used more frequently in these classes. Besides, students-centered methods are often preferred in the FLE while teacher-centered methods are preferred in the OLE during the content presentations. Differentiation in the learning environment has also affected teachers’ content preparation process. It was determined that worked examples were used frequently in the FLE, whereas start-up and conceptual examples were preferred more often in the OLE. Use of proof examples, which require advanced mathematical thinking processes, was limitedly in both learning environments. Compared with face-to-face classes, types of example preferred by teachers in online education showed more similarity with the textbook examples. Interviews revealed that teachers needed the knowledge to use content-oriented digital learning platforms to provide variety in mathematical example types in technological aspects. It has also been concluded that teachers should be supported in terms of online classroom management to organize student-centered presentation methods interactively in pedagogical aspect.
Based on the results, some recommendations can be formulated for both researchers and educational practitioners. It is recommended that the content in both the textbook and teaching practice be further equipped with example types that require advanced mathematical thinking. Through inservice training, mathematics teachers can be aware of technologies that they can use more effectively in the OLE, and teachers can be supported to develop contents to enable interaction in the OLE. But it is not enough to form a framework for inservice training based only on the results of this research. Because in this research, limited learning areas, subject content, and examples were studied to make a holistic analysis. With other research in this area, technological content knowledge that teachers need in the OLE and their pedagogical beliefs can be examined in more detail. If the time devoted to teaching content had been studied separately in this study, the way in which examples are presented and the types of examples could have been examined together (accompanied by cross matches in future research), new contributions to the process of developing qualified teaching content might be provided.
The present study had certain limitations in terms of achieving more generalizable results. The first limitation is that only 14 mathematics teachers participated in the study and their lectures the Numbers and operations learning area were evaluated in terms of exemplification processes. It is believed that future research that involves evaluating different learning areas (such as algebra, geometry, statistics) and different teaching contents (such as tasks, definition, and/or exercises) over a broader sampling is likely to reveal generalizable results. Another limitation is that the online education approach was carried out with a fully-online model during the Covid-19 pandemic. Future studies could compare the types of examples used in different online learning environments (e.g. hybrid, open, flipped-classroom, etc.) after the pandemic. In this way, the technological and pedagogical aspects that affect the content preferences of teachers in different learning environments could be investigated comparatively.
Author contributions
One hundred percent contribution by the corresponding author (ES). The author read and approved the final manuscript.
Funding
The author confirms that he received no funding.
Data availability
The author has the data available in video recording form for online learning environment and in paper form for the face-to-face classrooms.
Declarations
Competing interests
The author declares that he has no competing interests.
Ethics approval
Granted by my university’s Scientific Research and Publication Ethics Board.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Arcavi A Llinares S Chapman O From tools to resources in the professional development of mathematics teachers International Handbook of Mathematics Teacher Education 2020 Brill Leiden
Atkinson RK Derry SJ Renkl A Wortham D Learning from examples: Instructional principles from the worked examples research Review of Educational Research 2000 70 2 181 214 10.3102/00346543070002181
Bennison A Goos M Geiger V Utilising a research informed instructional design approach to develop an online resource to support teacher professional learning on embedding numeracy across the curriculum ZDM-Mathematics Education 2020 52 1017 1031 10.1007/s11858-020-01140-2
Bentley J Stylianides GJ Drawing inferences from learners’ examples and questions to inform task design and develop learners’ spatial knowledge Journal of Mathematical Behavior 2017 47 35 53 10.1016/j.jmathb.2017.06.001
Bills, E., Dreyfus, T., Mason, J., Tsamir, P., Watson, A. & Zaslavsky, O. (2006). Exemplification in mathematics education. In J. Novotna (Ed.), Proceedings of the 30th PME Conference (pp. 126–154).
Borba MC Chiari AS de Almeida HR Interactions in virtual learning environments: New roles for digital technology Eductional Study in Mathematics 2018 98 269 286 10.1007/s10649-018-9812-9
Chevallard Y Bosch M Lerman S Didactic transposition in mathematics education Encyclopedia of Mathematics Education 2013 Springer
Demir B Kaleli-Yılmaz G Sert-Celik H Teachers’ attitudes and opinions on mathematics lessons conducted with distance education due to Covid-19 pandemic Turkish Online Journal of Distance Education 2021 22 4 147 163 10.17718/tojde.1002812
Engelbrecht J Llinares S Borba MC Transformation of the mathematics classroom with the internet ZDM-Mathematics Education 2020 52 825 841 10.1007/s11858-020-01176-4 32837579
Francis K Jacobsen M Synchronous online collaborative professional development for elementary mathematics teachers The International Review of Research in Open and Distributed Learning 2013 14 3 319 343 10.19173/irrodl.v14i3.1460
Karal H Kokoc M Colak C Yalcin Y A case study on online mathematics teaching with pen-based technology: Experiences of two instructors Contemporary Educational Technology 2015 6 4 319 337 10.30935/cedtech/6157
Li N Taconis R den Brok P Chinese teachers’ perceptions of an online teacher course and its results Learning Environments Research 2022 25 115 139 10.1007/s10984-021-09353-2
Mason J Pimm D Generic examples: Seeing the general in the particular Educational Studies in Mathematics 1984 15 3 277 290 10.1007/BF00312078
Miles MB Huberman AM Qualitative Data Analysis: An Expanded Sourcebook 1994 2 Sage
Moodley M WhatsApp: Creating a virtual teacher community for supporting and monitoring after a professional development programme South African Journal of Education 2019 39 2 1 10 10.15700/saje.v39n2a1323
Peled I Zaslavsky O Counter-examples that (only) prove and counter-examples that (also) explain FOCUS on Learning Problems in Mathematics 1997 19 3 49 61
Rapanta C Botturi L Goodyear P Online university teaching during and after the Covid-19 crisis: Refocusing teacher presence and learning activity Postdigital Science Education 2020 2 923 945 10.1007/s42438-020-00155-y
Rø K Arnesen KK The opaque nature of generic examples: The structure of student teachers’ arguments in multiplicative reasoning The Journal of Mathematical Behavior 2020 58 100755 10.1016/j.jmathb.2019.100755
Rowland T The purpose, design and use of examples in the teaching of elementary mathematics Educational Studies in Mathematics 2008 69 2 149 163 10.1007/s10649-008-9148-y
Sevimli E Investigation of structural features of examples which textbook located and lecturers’ preferred in Calculus Education and Science 2016 41 183 319 339
Sevimli E Evaluation of the didactic transposition process in teaching integral: Face-to-face versus online education International Journal for Technology in Mathematics Education 2022 29 1 37 48
Sevimli E Çelik S Kul Ü Comparison of the frequency of mathematics teachers’ use of mathematics textbooks in face-to-face and distance education Turkish Journal of Education 2022 11 1 16 35 10.19128/turje.896355
Sinclair N Watson A Zazkis R Mason J The structuring of personal example spaces Journal of Mathematical Behavior 2011 30 4 291 303 10.1016/j.jmathb.2011.04.001
Smith MS Stein MK Selecting and creating mathematical tasks: From research to practice Mathematics Teaching in the Middle School 1998 3 5 344 350 10.5951/MTMS.3.5.0344
Trenholm S Alcock L Robinson C The instructor experience of fully online tertiary mathematics: A challenge and an opportunity Journal for Research in Mathematics Education 2016 47 2 147 161 10.5951/jresematheduc.47.2.0147
Trenholm S Peschke J Teaching undergraduate mathematics fully online: A review from the perspective of communities of practice International Journal of Educational Technology in Higher Education 2020 17 17 37 10.1007/s11858-010-0301-6
Watson A Chick H Qualities of examples in learning and teaching ZDM-Mathematics Education 2011 43 2 283 294 10.1007/s11858-010-0301-6
Watson A Mason J Mathematics as a Constructive Activity: Learners Generating Examples 2005 Lawrence Erlbaum
Yopp DA Ely R When does an argument use a generic example? Educational Study in Mathematics 2016 91 37 53 10.1007/s10649-015-9633-z
Zaslavsky O There is more to examples than meets the eye: Thinking with and through mathematical examples Journal of Mathematical Behavior 2019 53 245 255 10.1016/j.jmathb.2017.10.001
| 36467116 | PMC9707127 | NO-CC CODE | 2022-12-01 23:20:21 | no | Learn Environ Res. 2022 Nov 28;:1-24 | utf-8 | Learn Environ Res | 2,022 | 10.1007/s10984-022-09440-y | oa_other |
==== Front
Eur J Orthop Surg Traumatol
Eur J Orthop Surg Traumatol
European Journal of Orthopaedic Surgery & Traumatology
1633-8065
1432-1068
Springer Paris Paris
36445460
3431
10.1007/s00590-022-03431-8
Original Article
Home-based rehabilitation following anterior cruciate ligament reconstruction in the Kurdistan region of Iraq: epidemiology and outcomes
http://orcid.org/0000-0001-8665-1231
Kader Nardeen [email protected]
1
http://orcid.org/0000-0001-6393-5274
Jones Samantha [email protected]
2
Serdar Ziyad [email protected]
3
http://orcid.org/0000-0002-3822-6203
Banaszkiewicz Paul [email protected]
45
http://orcid.org/0000-0002-8332-2296
Kader Deiary [email protected]
67
1 grid.451349.e St George’s University Hospitals NHS Foundation Trust, London, UK
2 South West London Elective Orthopaedic Centre, Epsom, UK
3 Shar Teaching Hospital, Sulaymaniyah, Kurdistan Iraq
4 grid.42629.3b 0000000121965555 Faculty of Health and Life Sciences, Northumbria University, Newcastle Upon Tyne, Tyne and Wear UK
5 grid.415506.3 0000 0004 0400 3364 Queen Elizabeth Hospital Gateshead, Gateshead, UK
6 South West London Elective Orthopaedic Centre, Epsom, UK
7 grid.449828.b 0000 0004 0404 9231 University of Kurdistan Hewlêr, Erbil, Iraq
29 11 2022
18
18 7 2022
9 11 2022
© The Author(s), under exclusive licence to Springer-Verlag France SAS, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose
This study, set in in the Kurdistan region of Iraq, describes the epidemiology and outcomes of anterior cruciate ligament reconstruction (ACLR) followed by home-based rehabilitation alone.
Methods
A cohort observational study of patients aged ≥ 16 years with an ACL rupture who underwent an ACLR under a single surgeon. Followed by a home-based rehabilitation programme of appropriate simplicity for completion in the home setting; consisting of stretching, range of motion and strengthening exercises. Demographics, mechanism of injury, operative findings, and outcome data (Lysholm, Tegner Activity Scale (TAS), and revision rates) were collected from 2016 to 2021. Data were analysed using descriptive statistics.
Results
The cohort consisted of 545 patients (547 knees), 99.6% were male with a mean age of 27.8 years (SD 6.18 years). The mean time from diagnosis to surgery was 40.6 months (SD 40.3). Despite data attrition Lysholm scores improved over the 15-month follow-up period, matched data showed the most improvement occurred within the first 2 months post-operatively. Post-operative TAS results showed an improvement in level of function, but did not reach pre-injury levels by final follow-up. At final follow-up, six (1.1%) patients required an ACLR revision.
Conclusion
Patients who completed a home-based rehabilitation programme in Kurdistan had low revision rates and improved Lysholm scores 15 months post-operatively. To optimise resources, further research should investigate the efficacy of home-based rehabilitation for trauma and elective surgery in low- to middle-income countries and the developed world.
Keywords
Anterior cruciate ligament reconstruction
Anterior cruciate ligament reconstruction rehabilitation
Orthopaedic surgery
Home-based rehabilitation
Home-based physiotherapy
Self-directed rehabilitation
==== Body
pmcIntroduction
The annual incidence of anterior cruciate ligament (ACL) tears is estimated to be 68.6 per 100,000 person-years in an American-based study [1]. In Iraq, equivalent statistics are unavailable because credible data are not systematically collected and/or paper-based systems are utilised which are inappropriate for data analysis [2, 3]. Consequently, there is limited understanding of the epidemiology and financial burden of musculoskeletal problems in both Iraq and other low- to middle-income countries (LMIC) [4]. ACL ruptures are likely to represent a common injury in Iraq due to its relatively young population (median age 21.7 years) [5, 6] and the popularity of football [5].
Iraq is an economically developing country [6] in which the quality of health care has declined over the past 30 years as a result of multiple wars and embargoes [3]. Kurdistan is an autonomous region of Iraq which has become a safe place for many refugees and internally displaced people resulting in increasing pressure on local health care [3]. It is envisaged that improved information management systems, a trained workforce and specialised equipment are needed to improve health care provision in this region [2].
The generic aims of anterior cruciate ligament reconstruction (ACLR) are to improve function, quality of life, prevent further injury, osteoarthritis, and enable patients to resume sport and/or work [7]. In developed countries physiotherapists or rehabilitation specialists play a critical role in facilitating the rehabilitation required to return patients to pre-injury function [7]. However, in Iraq this approach is problematic because of the limited number of physiotherapists [8, 9]. Furthermore, patients often forego physiotherapy to afford the costs of an ACLR. To address such challenges the surgeon in Kurdistan provides patients with instruction on a home-based ACL rehabilitation programme.
This study is reporting on a unique experience of a specialist knee surgeon in Kurdistan who has established and maintained his own ACL database to monitor, inform and improve the outcomes of his ACL reconstruction (ACLR) surgery.
The aims of this study are to describe the epidemiology and outcomes for patients in Kurdistan who underwent an ACL reconstruction and completed a home-based rehabilitation programme over a 5-year period.
Method
This is an observational cohort study of patients who underwent an ACLR and completed a home exercise programme in the Kurdistan region of Iraq. Data were collected as part of the surgeon’s routine practice and patient follow-up.
The study included all patients aged ≥ 16 years who had an MRI confirmed ACL rupture and underwent ACLR during the study period (November 2016–November 2021). Patients were excluded from this study if they had concomitant posterior cruciate ligament injuries, multi-ligament injuries or fractures.
Surgery and home-based rehabilitation programme
All ACLRs were performed arthroscopically by the same surgeon using a hamstring autograft and the following method: The knee is examined under anaesthesia to confirm isolated ACL rupture. Intravenous ceftriaxone 1 g is given at induction, followed by 160 mg Gentamicin at tourniquet deflation. An anteromedial longitudinal incision over the pes anserinus insertion is utilised. The sartorius fascia is incised at the superior border of the gracilis tendon. Both gracilis and semitendinosus tendons are identified and soft tissue accessory attachments released. Both tendons are harvested using a closed tendon stripper and detached from the tibia. The tendons are then prepared as a four to six strand graft with "whip stitching" depending on thickness, achieving a minimum of 8 mm in graft diameter. The graft is then stored in moist saline soaked gauze.
Systematic examination of the knee is carried out with the arthroscope after creating an accessory anteromedial working portal under direct vision. The arthroscopic shaver and curette are used to debride residual ACL tissue from the lateral wall and posterior aspect of the notch. An entry point is made on the medial wall of the lateral femoral condyle using a microfracture awl. The knee is flexed to more than 110°. Long 2.4 mm guide wires with suture eyes are passed, exiting through the anterolateral skin of the thigh. The femoral tunnel is over-drilled with a 4.5 mm drill and then drilled again with a larger cannulated drill matching the graft dimeter. A minimum of 20–30 mm length of graft in the tunnel was deemed acceptable. The Acufex (Smith + Nephew) drill guide is used, set at 45–55° depending on the graft length. The tibial guide is placed on the ACL footprint, under arthroscopic vision. The same sequence of drilling is applied. Residual ACL tissue is removed from the tunnel margins. The graft is then passed and inspected for lateral wall or roof impingement. An RCI titanium (Smith & Nephew) interference screw (7 × 25 mm standard or reverse thread) or an Endobutton device is used for the femoral side. Either an RCI titanium interference screw or (Bio) RCI screw is used to fix the tibial side in full extension. Additional tibial fixation with suture disc or staple (10 mm) was used if there were any concerns about tibial bone quality. The knee is inspected and graft tension assessed. Range of motion and pivot shift are checked. Copious washout is performed followed by closure using 3:0 undyed Vicryl.
Following ACLR patients completed a home exercise programme. The surgeon provided each patient with a written (booklet) and video instruction (DVD, YouTube, Facebook) of the home-based rehabilitation programme available in Arabic, English and Kurdish. All exercises on the video instruction were demonstrated by the NGMV medical volunteers rehabilitation team (a UK registered charity) [10]. Table 1 provides an outline of the home exercise programme which an appropriate level of simplicity for completion in the home setting. The home-based programme consists of stretching, range of motion and strengthening exercises which are proven to be the most important exercises in improving outcomes [11, 12]. Exercises are divided into four stages: 0–3 weeks, 3–6 weeks, 6–12 weeks and 12–24 weeks. Although timescales are indicated throughout the programme, stages 3 and 4 rely on criteria-based progression (specific criteria that need to be achieved before patients can progress to the next stage of rehabilitation) [13].Table 1 Home-based rehabilitation programme
Exercises Duration (seconds) Repetitions (times) Frequency (hourly)
Stage 1 (Weeks 0–3): advised maximum rest, icing every 2 h, elevation and analgesia to allow swelling to reduce and incision sites to heal with the following exercises:
Knee range of movement (ROM) N/A 10–20 2
Static gluteal contractions 10 10 2
Static quadriceps contractions 10 10 2
Calf stretch (standing) 30 2 2
Hamstring stretch (standing, starting at week 2) 30 2 2
Stage 2 (Weeks 3–6):
Exercises Duration (seconds) Repetitions (times) Frequency (times daily)
Calf raises N/A 10–15 3–5
Isometric hamstring contraction (supine) 5 10 3–5
Hip abduction (side lying) N/A 10–15 3–5
Sit to stand (without hands) N/A 10–15 3–5
One legged sit to stand (starting at 5 weeks) N/A 10–15 3–5
Step ups N/A 10–15 3–5
Prone hangs (seated) (starting at 5 weeks) 300 N/A 3–5
Stage 3 (Weeks 6–12): Progression to the following exercises if swelling is controlled and patient has full knee extension with normal gait
Exercises Duration (seconds) Repetitions (times) Frequency (times daily)
Squats (double leg) N/A 10–15 3
Squats (operated leg only starting at 8 weeks) N/A 10–15 3
Hamstring curls (standing) N/A 10–15 3
Hamstring curls (prone, using elastic band) N/A 10–15 3
Bridging (double leg) 5 10–15 3
Stage 4 (Weeks 12–24): Progression to the following exercises if approximately 70% of quadriceps and hamstrings strength has returned
Exercises Duration (minutes) Repetitions (times) Frequency (times daily)
Lunges N/A 10–15 3
Hopping (one leg) N/A 10 3
Shuttle walking and running (10 m, starting at month 4 and 5, respectively) 5–10 1 3
Crossover steps (starting at month 5) 2–5 1 3
Data collection
Demographic data collection included the patient’s age, gender, height, weight and occupation. Details were also recorded about their injury (dominant limb, mechanism of injury, and date of injury) and the operation (tourniquet time, time to surgery and the surgical procedure). See Table 2 for all details.Table 2 Demographics, injury and operative details of 545 patients
Descriptive statistics
Patients (n = 545) 545 patients, 547 ACL reconstructions performed
November 2016 to November 2017 (Year 1), n = 44
December 2017 to November 2018 (Year 2), n = 80
December 2018 to November 2019 (Year 3), n = 101
December 2019 to November 2020 (Year 4), n = 123
December 2020 to November 2021 (Year 5), n = 199
Gender (n = 545) Male n = 543, 99.63%, Female n = 2, 0.37%
Age (years) (n = 544) Mean 27.87, SD 6.18, Range 17 to 46
Height (cm) (n = 498) Mean 175.67, SD 5.98, Range 155 to 193
Weight (kg) (n = 508) Mean 77.53, SD 11.82, Range 50 to 120
Occupation (in order of frequency) Education n = 103, 24.3%, Manual labour n = 76, 17.9%, Retail industry /services n = 48, 11.3%, Administration n = 45, 10.6%, Military n = 33, 7.8%, Skill/craft n = 33, 7.8%, Police and security n = 25, 5.9%, Hospitality services n = 25, 5.2%, Professionals n = 17, 4.0%, Medical professional/Industry n = 10, 2.4%, Sport and leisure industry n = 7, 1.7%, Managers n = 3, 0.7%, Housewife n = 2, 0.5%
Mechanism of injury (n = 508) Team Sport n = 467, 91.9%:
Soccer n = 462, 90.9%,
Handball n = 2, 0.4%
Volleyball n = 2, 0.4%
Basketball n = 1, 0.2%
Falls n = 13, 2.6%, Road traffic accidents n = 10, 2.0%, Contact sports n = 8, 1.6%, Jumping n = 6, 1.2%, Other n = 4, 0.8%
Injury to dominant or non-dominant limb (n = 505) Dominant limb n = 294, 58.2%
Non-dominant limb n = 209, 41.4%
Presenting with both limbs (injured separately) n = 2, 0.4%
Side of ACLR Left sided n = 223, 40.8% and Right sided n = 324, 59.2%
Date of injury to operation (n = 505) Mean 40.6 months, SD 40.3 months, (Range 0–212 months)
Outcome measures
When assessing the outcome of surgery the revision rate was regarded as the primary determining factor in addition to Lysholm and Tegner scores. The Lysholm is a rating scale (0–100), with higher scores indicating less symptoms and better levels of function [14]. The Tegner activity scale (TAS) (0–10) quantifies the highest current level of activity achievable by the patient with higher scores representing function at a demanding level of competitive sport (e.g. national elite football) [14]. Lysholm scores, mid-thigh girth and the Lachman’s tests were completed by the operating surgeon pre-operatively (at the time of initial clinic assessment) and repeated at 2, 6, 9 and 15-months post-operatively. TAS scores were recorded by the operating surgeon to include pre-injury, post-injury (at time of clinic assessment) and final follow-up, which varied for each patient. The Lysholm questionnaire and TAS are not validated for use in the native Kurdish language therefore an English version was used and some patients required help from the surgeon to interpret the questions.
ACLR failure was determined by patient reported instability and clinical findings of laxity on examination. All patients who had recurrence of instability had a repeat MRI scan to confirm diagnosis of graft failure and the need for revision. Due to limited number of specialist knee surgeons in Kurdistan, it is likely that all patients would return to the operating surgeon in this study if they had further problems following the ACLR. To confirm this, in June 2022 a telephone review was conducted to determine whether any patients had undergone a repeat arthroscopy, ACL revision or experienced pain and/or functional difficulties that were not reported at follow-up. Due to the different operation dates within the cohort the minimum telephone follow-up was 6 months and the maximum was for 5.5 years.
Data analysis
Data were cleaned and anonymised. Descriptive statistics (number, percentage, mean, and standard deviation (SD)) were calculated using Excel Version 16.61.1. Occupation was categorised into the following groups: hospitality services, skill/craft work, medical professional/Industry, military, sport and leisure industry, manual labour, retail industry/services, police and security, education, administration, professionals, housewife, and managers. Mechanism of injury was categorised as: falls, team sports (handball, volleyball, and basketball), and contact sport, road traffic accidents and other. Surgical findings were categorised according to those identified in the patellofemoral, medial, and lateral compartments.
Results
From November 2016 to November 2021, 547 ACLRs were performed in 545 patients (two patients had bilateral ACL injuries, operated on at separate times). The demographics, injury and operative details are described in Table 2. All procedures were performed by a single senior surgeon. ACL injuries occurred almost exclusively in males (n = 543, 99.63%), who injured their dominant limb playing soccer (90.9%). The average age was 27.87 (SD 6.18) and the average time to operation from date of injury was 40.6 months (SD 40.3).
Table 3 shows categorisation of the intra-operative findings for 547 knees. Only one patient had a partial ACL tear, and the remainder had full ACL tears. Just over half the cohort presented with medial meniscal injuries and approximately a third with lateral meniscus injuries.Table 3 Intra-operative findings
Anterior cruciate Patellofemoral joint Medial compartment Lateral compartment
Full tear 546 (99.8%)
Partial tear 1 (0.2%)
Chondral lesion/OA n = 5, 0.9% Meniscal injury n = 287, 52.3%
Chondral lesion/OA n = 114 20.8%
Meniscal injury n = 191, 34.9%
Chondral lesions/OA n = 16, 2.9%
Table 4 presents mean Lysholm scores and mid-thigh girth pre-operatively and post-operatively at the different follow-up intervals (2, 6, 9 and 15 months). Outcome data are presented for the full cohort at each follow-up, with data attrition (not all patients attended at every follow-up interval). Despite data attrition over the 15 month follow-up period average Lysholm scores showed overall improvement throughout the follow-up intervals from 72.11 (SD 12.66) at baseline to 91.9 (SD 8.58) at 15 months. Matched data were available for patients who attended both pre- and post-operative follow-up. Matched data showed the greatest improvement occurred within the first 2 months post-operatively increasing by 6.33 points a month from a baseline of 72.28 (n = 240, SD 12.42) to 84.94 (n = 240, SD 6.89).Table 4 Pre- and post-operative Lysholm scores, injured thigh girth and thigh girth difference
Follow-up interval Pre-operation 2 months 6 months 9 months 15 months
Mean Lysholm scores (0–100)
(with attrition)
(n = 507)
72.11 (SD 12.66)
(n = 420)
84.94 (SD 6.89)
(n = 349)
90.31 (SD 7.05)
(n = 250)
91.88 (SD 7.22)
(n = 130)
91.9 (SD8.58)
Mean Thigh Girth of injured limb (cm)
(with attrition)
(n = 455)
55.89 (SD 5.24
(n = 442)
54.35 (SD 5.75)
(n = 342)
55.37 (SD 5.16)
(n = 240)
56.20 (SD 4.50)
(n = 128)
56.60 (SD 4.25)
Mean Thigh Girth difference (cm)
(with attrition)
(n = 455)
− 0.88 (SD 1.39)
(n = 406)
− 1.04 (SD 1.51)
(n = 322)
− 0.75 (SD 1.33)
(n = 212)
− 0.23 (SD 1.35)
(n = 92)
− 0.31 (SD 1.45)
For matched patients at 6 months follow-up: Lysholm scores increased from 72.76 (n = 349, SD 12.15) to 90.31 (n = 349, SD 7.05)); at 9 months from 72.96 (n = 248, SD 12.03) to 91.85 (n = 248, SD 7.39); and at 15 months from 72.10 (n = 129, SD 12.62) to 91.98 (n = 129, SD 8.57). The highest Lysholm scores were attained at 9 and 15 months post-operatively (after the home programme was completed).
Table 4 presents mean mid-thigh girth of the injured limb pre-operatively and post-operatively at the same follow-up intervals, with data attrition. This reduced by over a centimetre and half at 2 months post-operatively, however by 9 months pre-operative girth was exceeded. Matched data were also available for patients who attended both pre- and post-operative follow-up. Matched data showed a decrease in mid-thigh girth at 2 months from 56.03 (n = 406, SD 5.15) to 54.91 cm (n = 406, SD 5.51); at 6 months follow-up from 55.79 (n = 325, SD 5.08) to 55.35 cm (n = 325, SD 5.17); and at 9 months from 56.21 (n = 213, SD 5.08) to 56.08 cm (n = 213, SD 4.58). In contrast to unmatched data, matched thigh girth was not regained until 15 months, increasing from 56.36 (n = 92, SD 4.96) to 56.65 cm (n = 92, SD 4.37).
Table 4 also presents mean thigh girth difference measured at baseline and at follow-up, with data attrition. Mean thigh girth difference was calculated from injured side mid-thigh girth subtracted by contralateral side mid-thigh girth, in centimetres. Findings closely mirror those reported above, with an initial increase in thigh girth difference, followed by a significant improvement with the injured side being within 0.23 cm (SD 1.35) of the contralateral side by 9 months.
Table 5 presents pre-injury, post-injury, and post-operative TAS scores. The post-operative final follow-up Tegner scores improved compared to the post-injury scores, but patients did not reach their pre-injury levels of function.Table 5 Pre-injury, post-injury, and post-operative Tegner scores
Mean Tegner Scores (0–10) Pre-injury scores, assessed at first clinic appointment (n = 496) Post-injury scores, assessed at first clinic appointment (n = 496) Post-operative scores, recorded at final follow-up (final follow-up variable) (n = 76)
7.13 (SD 1.01) 3.45 (SD 1.71) 4.96 (SD 2.14)
Table 6 outlines the outcome of the telephone review to determine re-rupture rates. 484 (88.5%) knees required no further surgery with the patient satisfied with the outcome. Six males with a mean age of 22.2 years (SD 3.9) required an ACL revision and one had the revision completed within the timeframe of the study. In four cases ACL graft failure was caused by return to sport or strenuous physical activities prior the time frame recommended by the surgeon, in the other two cases the cause for failure was unknown.Table 6 Telephone review to determine revision rate
Responses from telephone follow-up N %
No further surgery and satisfied with outcome 484 88.50
Pain and functional difficulties (requiring diagnostic arthroscopy n = 4, 0.7%) 9 1.60
Requires ACL revision 5 0.90
Revision ACL completed 1 0.20
Unable to contact 48 8.80
Total 547 100.00
Discussion
To our knowledge this is one of the first studies to describe the epidemiology and outcomes for patients who underwent ACLR in Kurdistan, Iraq. In this region ACL injuries occur almost exclusively in males, with a mean age of 27.8 years and were sustained playing soccer. Following the completion of a home-based exercise programme re-rupture rates were low (1.1%), and Lysholm scores improved over the 15 months follow-up period, with the most marked improvement in the first 2 months after ACLR. At the final follow-up, TAS scores had not returned to pre-operative activity levels.
The mean age of injury (27.8 years) aligns with that reported in previous studies [1]. However, the occurrence of ACL injuries almost exclusively in males is a novel finding, contrasting with 41% females reported in a large cohort study [1]. Despite the positive steps to increase women’s participation in sport in Iraq this has declined following the war [15]. The reasons for this are multi-factorial and relate to radical political, cultural and religious changes, in addition to concerns about the safety of women [15].
Findings from this study highlight key distinctions between health care delivery in Kurdistan and that of the developed world. It is unsurprising that the timeframe from date of injury to ACL surgery was much longer (mean of 40 months), compared to 6–12 months reported in developed countries [1, 16]. This may have contributed to the high percentage of meniscal tears and chondral lesions among this cohort and likely reflects the inaccessibility and inadequacies of health care in Iraq, particularly the limited number of specialist surgeons [3, 8, 8]. Limited research capacity [8], accredited courses and qualified health care educators hinder the development of specialist skills across all health care professions [17]. Patients in this study were reviewed by the surgeon because of ‘word-of-mouth referrals’, which may explain the increasing number of cases performed each year, even during the COVID-19 pandemic. However, knee surgeons in Kurdistan should raise their profiles amongst local soccer teams to help capture all ACL injuries.
The lack of physiotherapists necessitated the adoption of a simple home-based programme. In a small prospective randomised study, Hohmann et al. (2011) found no additional benefit of ACL rehabilitation supervised by a physiotherapist compared to an unsupervised approach [18]. The low revision rates and improved Lysholm scores may indicate that the home-based programme was appropriate for the needs of patients in this study, but a larger randomised controlled trial would be needed to draw any conclusions about its efficacy in this setting. In their 2012 systematic review, Kruse et al. found six studies comparing home-based therapies alone versus physiotherapy led rehabilitation following ACLR [12]. Most studies included patient reported outcome measures such as Lysholm, Tegner and anterior cruciate ligament quality of life (ACL-QOL) scores as well as post-operative measurements of knee range of motion (ROM). The studies each found either no difference in functional outcomes or differences in favour of the home-based therapy group (such as a higher ACL-QOL scores in one study and better knee ROM in another). A more recent study by Lim et al. found that although patients treated with home-based therapy alone regained their knee strength, those who received physiotherapy led rehabilitation had better recovery of proprioception and functional knee movement [19].
The TAS scores indicate that patients in this study did not return to their pre-injury level of activity on average, yet the main reason for ACL failure was the resumption of sport or strenuous activity earlier than advised. Return to sport or high-level activity is based on a range of factors which include: physical examination findings, functional assessment, return to sport tests and psychological readiness [20]. In light of these considerations, our patients may have needed more support or closer supervision with this challenging aspect of ACL rehabilitation. Nevertheless, the lack of resources and health care professionals provides a strong case for further research to investigate the efficacy of home-based rehabilitation programmes following elective and trauma surgery in Iraq and other LMIC. Although the NHS represents a distinctly different health care structure, the COVID-19 pandemic has placed increasing strains on the workforce due to the introduction of the elective recovery plan [21]. Furthermore, supported self-management (democratising self-care) is part of the NHS Long-Term Plan which aims to give patients choice and control over how their care is delivered. The option of ACL home-based rehabilitation, for example, may enable patients to tailor what can be long and intensive treatment around their individual needs [22].
Our findings also prompt consideration of conservative management as a treatment option in place of surgical reconstruction of the ACL. A 2021 umbrella study of systematic reviews and meta-analyses found only two randomised controlled trials comparing conservative management with surgical treatment, both of which found no difference between the two interventions [23], unfortunately these were the only high-quality studies found on this subject, and heavily rely on physiotherapy led rehabilitation which is largely unavailable to the population studied in this report making this infeasible option.
Strengths
Due to the lack of robust health information systems and specialist knee surgeons [9], this is likely to be one of the only studies of ACL injuries in Kurdistan. Real-time data entry by the performing surgeon is a potential source of bias, but it may have also enhanced data accuracy.
Limitations
The limitations of this study should be considered in the context of the poor resource setting in which it was conducted, with no national surveillance on orthopaedic elective procedures. We do not have data about patient adherence, rate of progression through the criteria-based stages or patients’ experiences of managing the programme at home.
Unsurprisingly attrition in the outcome data increased with time post-surgery. In Iraq, patients may have to travel long distances from neighbouring cities or regions to access specialist surgeons, which impact follow-up attendance along with the lack of systems to track patients after discharge. Alternatively, patient’s may not have been motivated to attend follow-up because of their improvement in symptoms and function, as indicated by the Lysholm scores.
The requirement for revision was assessed throughout follow-up and was confirmed by telephone contact in June 2022, thus patients were at different stages of their post-operative recovery. However, the majority of patients were beyond one-year post-injury which can be considered as adequate time for graft integration and muscular function required for return to the majority of activities [24].
Conclusion
This study provides insights into the management and rehabilitation of ACL injuries in the Kurdistan region of Iraq, highlighting key health care challenges in this country. Patients who completed a home-based rehabilitation programme after ACLR had low revision rates and improved Lysholm scores over a 15-month period. Further research is required to investigate the efficacy and feasibility of home-based rehabilitation programmes following trauma and elective surgery in both LMIC and the developed world.
Author contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by ZS and NK. The first draft of the manuscript was written by NK and SJ and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Declarations
Conflict of interest
The authors did not receive support from any organization for the submitted work. The authors have no relevant financial or non-financial interests to disclose and have no competing interests to declare that are relevant to the content of this article.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Sanders TL MaraditKremers H Bryan AJ Larson DR Dahm DL Levy BA Stuart MJ Krych AJ Incidence of anterior cruciate ligament tears and reconstruction: A 21-year population-based study Am J Sports Med 2016 44 1502 1507 10.1177/0363546516629944 26920430
2. Anthony CR, Moore M, Hilborne LH, Rooney A, Hickey S, Ryu Y, Botwinick L (2010) Strengthening health care in the Kurdistan Region of Iraq
3. Moramarco S Basa FB Alsilefanee HH Qadir SA EmbertiGialloreti L Developing a public health monitoring system in a War-torn Region: a field report from Iraqi Kurdistan Disaster Med Public Health Prep 2020 14 620 622 10.1017/dmp.2019.116 31685062
4. Joshipura M Gosselin RA Surgical burden of musculoskeletal conditions in low- and middle-income countries World J Surg 2020 44 1026 1032 10.1007/s00268-018-4790-8 30238386
5. Mousa S Building social cohesion between Christians and Muslims through soccer in post-ISIS Iraq Science 2020 369 866 870 10.1126/SCIENCE.ABB3153 32792403
6. International monetary fund (2019) World economic outlook database October 2019–WEO groups and aggregates information. https://www.imf.org/external/pubs/ft/weo/2019/02/weodata/groups.htm#oem. Accessed 14 Jun 2020
7. Filbay SR Grindem H Evidence-based recommendations for the management of anterior cruciate ligament (ACL) rupture Best Pract Res Clin Rheumatol 2019 33 33 47 10.1016/j.berh.2019.01.018 31431274
8. Hilfi TK AL Lafta R Burnham G Health services in Iraq The Lancet 2013 381 939 948 10.1016/S0140-6736(13)60320-7
9. World Confederation for Physical Therapy (WCPT) (2018) Survey reveals global state of the physical therapy profession | World Physiotherapy. https://world.physio/news/surveys-reveal-global-state-of-the-physical-therapy-profession. Accessed 22 May 2022
10. NGMV Charity. https://ngmvcharity.co.uk/. Accessed 19 Jun 2022
11. Wright RW Preston E Dunn WR Kuhn JE Spindler KP Fleming BC Amendola A Wolf BR Williams GN Andrish JT Bergfeld JA Parker RC Kaeding C Marx RG Mc Carty EC Wolcott M ACL reconstruction rehabilitation: a systematic review part I J Knee Surg 2008 21 217 10.1055/S-0030-1247822 18686484
12. Kruse LM Gray B Wright RW Rehabilitation after anterior cruciate ligament reconstruction: a systematic review J Bone Joint Surg–Ser A 2012 94 1737 1748 10.2106/JBJS.K.01246
13. Cavanaugh JT Powers M ACL rehabilitation progression: where are we now? Curr Rev Musculoskelet Med 2017 10 289 296 10.1007/s12178-017-9426-3 28791612
14. Briggs KK Lysholm J Tegner Y Rodkey WG Kocher MS Steadman JR The reliability, validity, and responsiveness of the Lysholm score and Tegner activity scale for anterior cruciate ligament injuries of the knee: 25 years later Am J Sports Med 2009 37 890 897 10.1177/0363546508330143 19261899
15. Al-Wattar NSY Hussein F Hussein AA Women’s narratives of sport and war in Iraq Muslim Women and Sport 2010 10.4324/9780203880630-28
16. Collins JE Katz JN Donnell-Fink LA Martin SD Losina E Cumulative incidence of ACL reconstruction after ACL injury in adults: role of age, sex, and race Am J Sports Med 2013 41 544 549 10.1177/0363546512472042 23302260
17. Jalal Al-Mosawi A Iraq healthcare system before covid-19 pandemic Int J Res Studies Med Health Sci 2020 5 2456 6373
18. Hohmann E Tetsworth K Bryant A Physiotherapy-guided versus home-based, unsupervised rehabilitation in isolated anterior cruciate injuries following surgical reconstruction Knee Surg Sports Traumatol Arthrosc 2011 19 1158 1167 10.1007/s00167-010-1386-8 21267543
19. Lim JM Cho JJ Kim TY Yoon BC Isokinetic knee strength and proprioception before and after anterior cruciate ligament reconstruction: A comparison between home-based and supervised rehabilitation J Back Musculoskelet Rehabil 2019 32 421 429 10.3233/BMR-181237 30507563
20. Meredith SJ Rauer T Chmielewski TL Fink C Diermeier T Rothrauff BB Svantesson E Hamrin-Senorski E Hewett TE Sherman SL Lesniak BP Bizzini M Chen S Cohen M Villa-Della S Engebretsen L Feng H Ferretti M Fu FH Imhoff AB Kaeding CC Karlsson J Kuroda R Lynch AD Menetrey J Musahl V Navarro RA Rabuck SJ Siebold R Snyder-Mackler L Spalding T van Eck C Vyas D Webster K Wilk K Return to sport after anterior cruciate ligament injury: panther symposium ACL injury return to sport consensus group Orthop J Sports Med 2020 10.1177/2325967120930829 32754624
21. NHS England (2022) Delivery plan for tackling the Covid-19 backlog of elective care
22. NHS England and NHS Improvement (2020) Supported self-management summary guide
23. Blom AW Donovan RL Beswick AD Whitehouse MR Kunutsor SK Common elective orthopaedic procedures and their clinical effectiveness: umbrella review of level 1 evidence The BMJ 2021 10.1136/bmj.n1511 34233885
24. Salmon L Russell V Musgrove T Pinczewski L Refshauge K Incidence and risk factors for graft rupture and contralateral rupture after anterior cruciate ligament reconstruction Arthroscopy 2005 21 948 957 10.1016/J.ARTHRO.2005.04.110 16084292
| 36445460 | PMC9707130 | NO-CC CODE | 2022-12-01 23:21:38 | no | Eur J Orthop Surg Traumatol. 2022 Nov 29;:1-8 | utf-8 | Eur J Orthop Surg Traumatol | 2,022 | 10.1007/s00590-022-03431-8 | oa_other |
==== Front
Am J Cult Sociol
Am J Cult Sociol
American Journal of Cultural Sociology
2049-7113
2049-7121
Palgrave Macmillan UK London
36466044
178
10.1057/s41290-022-00178-x
Original Article
Gliding on the edge of the iron cage: performing rationality and artistry in the sport of figure skating
Ji Zaoying [email protected]
Zaoying Ji
is a Ph.D. candidate in the Department of Sociology at the University of California, Irvine.
grid.266093.8 0000 0001 0668 7243 Sociology Department, University of California, Irvine, 3151 Social Science Plaza, Irvine, CA 92697 USA
29 11 2022
2022
10 4 657675
3 10 2022
© The Author(s), under exclusive licence to Springer Nature Limited 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Prior studies on growing formal rationalization in evaluation systems have overwhelmingly shown that they operate as “iron cages,” which redefine standards of excellence around quantifiable metrics. However, existing literature may have overestimated the extent of isomorphism in individual or organizational practices under highly rationalized systems of assessment. The judging system in figure skating both rewards quantifiable technical merit and valorizes a circumscribed notion of artistry characterized by maturity and authenticity. It constitutes a combination of formal rationalization and culturally specific productions of artistry, offering participants a distinct set of options in constructing their competition routines. Drawing on exclusively obtained interview data with 40 Olympic-level figure skaters from the U.S. and 5 other countries, and following Alexander’s (Sociol Theory 22(4):527–573, 2004) social performance theory, I explore the contingent process by which figure skaters maneuver between conformity to the formally rationalized rulebook and the non-isomorphic yet culture-structured performances of artistry. In doing so, I identify three types of skaters based on their embodiment of artistry: the emerging phenoms, the athletic performers and the “authentic” artists. My article reveals that despite rationalizing tendencies of the scoring system, skaters strive for social performances of artistry based on shared cultural scripts of greatness in the sport.
Keywords
Formal rationalization
Culture
Authenticity
Embodiment
Agency
Aesthetics
issue-copyright-statement© Springer Nature Limited 2022
==== Body
pmcIntroduction
At the 2022 Beijing Olympic Winter Games, 17-year-old Russian figure skater Alexandra Trusova punched her fist at the conclusion of her free skate, barely able to hold her emotions in after what appeared to be a triumphant skate. Performing to the soundtrack of Disney movie Cruella, Trusova achieved an unprecedented technical feat by landing five quadruple jumps in one program, a monumental accomplishment only rivaled by the most skilled male skaters. Sitting in the Kiss and Cry awaiting her scores, Trusova seemed reassured and rejoiced, almost anticipating her crowning moment with three competitors remaining. However, in a somewhat surprising turn of events approximately 15 minutes later, Trusova’s training mate and reigning world champion Anna Shcherbakova overtook her in the final standings and eventually claimed the Olympic title. Despite only executing two quadruple jumps compared to Trusova’s five, Shcherbakova dazzled the audience with a delicate balancing act between complicated technical maneuvers and unmanufactured artistry. Her free skate featured varied facial expressions, complex transitional steps and balletic extensions, while Trusova’s performance placed an overwhelming emphasis on her jumping elements, which left her sporadic attempts of performing to the music feeling contrived and her routine disjointed. Shcherbakova, in an elegant plum dress, utilized multiple means of symbolic production as she seamlessly weaved her technical elements in and out of her choreography. Not only did she transition effortlessly between one jump and the next, but she also embodied a versatile artist within a four-minute span: she was melancholy skating to Ruska by Apocalyptica, slightly flirtatious as the music transitioned to The Master and Margarita. Finally, as the program culminated in Mozart’s Lacrimosa, she displayed a level of solemnity and gravitas uncharacteristic of her tender age of 17. As the scores for Shcherbakova were posted, Trusova, having championed a movement to push the sport forward solely on technical grounds over the past four years and broken various Guinness World Records in the process, was inconsolable. Her strategy of stacking up points from higher levels of technical difficulty and displaying remarkable athleticism at the expense of artistry has proven less than fruitful.
The epic battle between Shcherbakova and Trusova sheds light on the different ways in which figure skaters draw on, mediate and perform narratives of rationality and artistry under the current judging system in figure skating. Known as the International Judging System (IJS), it was adopted by the International Skating Union (ISU) following the judging scandal at the 2002 Olympic Winter Games in order to minimize geopolitical biases and introduce greater transparency to the evaluation process. During the pairs skating competition at the 2002 Olympic Winter Games in Salt Lake City, Russia’s Elena Berezhnaya/Anton Sikharulidze and Canada’s Jamie Salé/David Pelletier both produced gold-medal worthy performances. The 9-person judging panel, split in half by the former Communist bloc (Russia, China, Poland, Ukraine) and the Western liberal bloc (U.S., Canada, Germany, Japan), reached a stalemate with the French judge holding the swing vote. The French judge ruled in support of the Russian pair in exchange for favors from a Russian judge for a French couple in the ice dance competition. When such allegations surfaced, the ISU and the International Olympic Committee ultimately invalidated the French vote in the pairs competition and awarded both the Russian and the Canadian pairs a gold medal, thereby producing the sport’s only dual Olympic champions to date. The 6.0 system used at the 2002 Olympic Games, under which judges used ordinal rankings and easily conspired with one another based on geopolitical interests, came under microscopic scrutiny after the disreputable incident. After months of discussions, the ISU came up with an alternative judging system, the International Judging System, which relies on its quantitative metrics to undermine collusion amongst judges. Unlike the previous scoring system that capped the perfect score at 6.0, the IJS quantifies both technical merit and artistic impression with much greater precision. On the technical side, it designates a base value to each trick and an execution score between −3 and +3 (later changed to −5 to +5). On the artistic side, the judges evaluate skaters based on five categories (Skating Skills, Transitions, Performance, Composition and Interpretation of the Music) on a scale from 0.25 to 10. To counteract judges’ predilection to score their compatriots favorably and deflate the scores of their chief competitors, the IJS drops the highest and lowest scores received by each skater among a panel of 9 judges. Competitors’ total scores, accurate up to two decimal places, also make it much less likely for athletes to tie in the final standings.
Figure skating’s International Judging System is emblematic of our increasing reliance on formally rational means to achieve unquantifiable and ineffable ends in the modern world. In higher education, for example, institutional prestige and quality of instruction have been largely gauged by the proliferation of college rankings published each year (Espeland and Sauder 2016). Similarly, UNESCO has relied heavily on rationalized metrics to demonstrate the transnational relevance of cultural heritage (DeSoucey et al. 2019). In the context of skating, concerns for transparency and fairness have led to the exclusive use of numbers to present seemingly disinterested evidence of competence and relative achievement: not only has the new judging system established a clear pecking order among technical elements by their levels of difficulty and quality of execution, it has also assigned numeric values to skaters’ artistic expressions.
However, in a sport that valorizes something as intangible and ineffable as artistry, how do skaters as embodied performers make sense of the drive to submit their performances to formally rationalized evaluations and enact shared cultural notions of greatness in figure skating, particularly the social meaning of artistry? To answer this question, I gained exclusive access to some of the most acclaimed athletes in the sport and conducted 40 semi-structured interviews on skaters’ lived experiences navigating growing rationalization in the judging system and conceiving of achieving authentic social performances on the ice. Contrary to institutional theorists’ prediction of tight coupling and isomorphic practices, I observe a range of skaters’ responses as they work around the constraints imposed by the judging system, enact shared cultural beliefs about the social meaning of excellence in the sport and attempt to touch the audience at both a cognitive level and an emotional level. I argue that the manner in which athletes adopt cultural scripts from the broader figure skating community and strive to create psychological identification with the judges is an ongoing contingent process mediated by individual agency. While neoinstitutionalism has focused exclusively on instrumental reasons in terms of survival and legitimacy in organizational fields, it has overlooked non-instrumental ones (Larsen 2016), in this case, figure skaters’ passion for and belief in their craft. Excellence in the sport of figure skating, while subject to formally rationalized metrics of evaluation, has not been stripped of cultural meanings or of performance artistry. Cultural sociologists view rationality itself not as an empirical fact but as a cultural construct bound by actors’ sense-making capacities (Broch 2020). Formal rationalization has rarely overtaken any organizational field, nor has it eliminated the space for agentic actors to actively interpret and enact cultural pragmatics (Alexander 2004).
According to Alexander (2004), audiences in modern society are fragmented, as they hold divergent values and may experience different degrees of psychological identification with any given performance. In figure skating, the social meaning of excellence has long been contested, as evidenced by the epic battle between Russian sensations Trusova and Shcherbakova. For some, the most glorious and exciting moments of the sport stem from extraordinary athletic feats, especially when accomplished by young skaters.
For many others, however, the presentation of an artistic persona, as skaters put out social performances that publicly convey recognizable embodiment of artistry, has remained a pillar in defining standards of excellence in the sport, as judges and audiences alike respond emotionally to the "goosebump factor" and "bigger than life charisma" (McCormick 2015, p. 171) that skaters cultivate through their performances. Artistry takes on a contextualized meaning of authenticity and conveys distinct social expectations of what a figure skater should embody on the ice. Importantly, authenticity in artistic expression is not so much a default state as it is the culmination of elaborate training and a result of a social performance. It represents a state of mastery that judges have incorporated into the evaluation process and tokenized by the presentation component (i.e. artistic) scores that skaters receive. In this paper, I explore the various ways in which elite figure skaters interpret the social meaning of excellence in the sport and how they describe their efforts to create social performances for the judges and the audience.
From ritual to record: growing formal rationalization of sports evaluations
The drastic changes that have taken place in the sport of figure skating need to be understood within the broader context of growing formal rationality and escalating demands for accountability, transparency and legitimacy in sports and beyond. Scholars have used terms such as “audit society” (Power 1997) and “audit culture” (Strathern 2000) to describe increasing pressure to monitor individual and organizational performances. Quantitative measures, which appear to be stripped of emotions and personal biases, are often considered the most reliable form of evidence for performance and competency (Espeland and Sauder 2016). Consequently, in recent decades, we have seen the rise of formally rationalized evaluation regimes in many industries, such as rankings in higher education, ratings for hospitals and international indices for wellbeing and affluence. Rising neoliberal ideologies and their market-based principles have reinforced the emphasis on measurements (Bromley and Powell 2012).
In modern sports, Allen Guttman (2004) has famously observed a movement from “ritual” to “record,” where principles of equality and fairness have contributed to the rationalization of almost all athletic activities. Modern sports, from America’s national games of football and baseball to artistic ones like gymnastics and figure skating are rationalized in “Max Weber’s sense of Zweckrationalitaet, i.e., there is a logical relationship between means and ends” (Guttman 2004, p. 40). Competitors enter local, national or international contests under the assumption that they will be rule-governed events and judged transparently and rationally, often with quantitative metrics (McCormick 2015). The rules are ever-changing cultural artifacts instead of divine instructions, but they do inform players’ training regimens and competition strategies. The drive to quantification and rationalization has also led to a newfound fascination with quantitative records, which, at a high level of abstraction, make the most impressive performances today comparable to those in the past. These records and comparisons fascinate sports followers and garner considerable attention in mass media. When it comes to aesthetic-based sports, quantification creates a perpetual quest for perfection. People to this day marvel at the first perfect ten in gymnastics achieved by Romanian gymnast Nadia Comaneci, with athletes continuing to chase that level of perfection even after the scoring system has been adapted (Guttman 2004).
The drive to submit all spheres of public life to the scrutiny of quantifiable metrics has fundamentally shifted how institutions operate. Organizations have felt growing pressure to conform to the external environment and have altered the way they operate on a daily basis. They have redistributed attention, strategy and efforts, focusing specifically on areas explicitly assessed by these rationalizing regimes (Bromley and Powell 2012). Unfortunately, efforts to align one’s practices based on quantifiable standards often come with unexpected, negative consequences. In K-12 public education, for example, middle school teachers in Atlanta who faced sanctions due to poor test results admitted to routinely changing their students’ answers on tests that determine if their school is meeting federal standards outlined by the No Child Left Behind (NCLB) initiative (Espeland and Sauder 2016). NCLB relied on formally rational methods of comparing standardized test scores to achieve its goal of offering better and fairer education. Likewise, in classical music competitions judged by numeric scores, efforts to ensure objectivity and fairness work against the musical way of being and “interferes with the ritual dynamics of transcendent aesthetic experience” (McCormick 2015, pp. 6–7). Musicians are acutely aware that the formally rationalized criteria used to judge their performances undermine their musical values, but they feel compelled to prioritize technical accuracy in order to maximize their chances at becoming a laureate.
How do elite figure skaters navigate the submission of their performances to formally rationalized evaluations, which in turn shapes their training regimens, competition routines and standings? I draw on insights from Jeffrey Alexander (2004)’s social performance theory to understand athletes’ divergent attempts to create psychological identification with the judges and the audience.
Sports as social performances
While institutionalists’ conceptualization of the iron cage highlights how culture constrains and homogenizes action, cultural sociologists (Alexander 2004; Broch 2020; Larsen 2016) posit that culture possesses relative autonomy in shaping actions and institutions and that it is as consequential as material or instrumental forces (Alexander and Smith 2004). Actions are never simply coerced but invariably embedded in shared cultural meanings, which provide for both routine and creativity, “allowing for the reproduction and transformation of structure” (Alexander and Smith 2004). Contrary to the iron cage perspective that paints individuals as ideological dopes, cultural sociologists perceive actors as engaging in cultural meaning-making processes that shape their behaviors in a contingent manner. While actors do think strategically about gaining legitimacy in their respective fields and being perceived as relevant and excellent in what they do, what the neoinstitutional perspective has totally missed is that these actors also have a genuine passion for and belief in what they do that stretches beyond the need for survival (Larsen 2016). A successful performance of legitimacy relies upon both instrumental and non-instrumental reasons, especially in social contexts that value authenticity (Alexander 2004).
In this article, I draw on Alexander (2004) to investigate how elite figure skaters enact different cultural meaning systems and present their “authentic” artistic selves while maneuvering technicalities imposed by the quantitative metrics in the judging system. Alexander’s theory of social performance is useful for making sense of elite figure skating because it illustrates how the sport is not a purely mechanical behavior but rather one that valorizes culturally distinct notions of artistry characterized by maturity and authenticity. It brings culture together with situational pragmatics and outlines an agentic process by which individuals or groups put culture narratives, myths and codes into action. It is an iconic process where cultural meaning intersects with materiality and corporal acts (Broch 2020). Actors, in this case competitive figure skaters, project cultural meanings and enact patterned representations that are encoded in their cultural scripts, which are systems of collective representation that invoke cognitive resonance with the audience (McCormick 2015). During our interviews, skaters reveal how the figure skating community lacks consensus when it comes to cultural scripts on the competition circuit. Athletes, coaches, judges and the audience vary in how they interpret standards of excellence in the sport, particularly when it comes to the embodiment of artistry on the ice. Like in many other sports, collective representations in figure skating are mediated by loosely coupled institutionalized logics (Broch 2020). Given the lack of a set script, we observe a range of plausible interpretations, which are in turn embodied through diverging training regimens in the “back stage” (Goffman 1959) as well as varied presentation of one’s self on the front stage, where skaters glide around the edges of the iron cage and present their artistic, musical and gendered selves.
According to Alexander (2004, p. 529), the key to a successful cultural performance lies in authenticity through cultural pragmatics, or the “social process by which actors, individually or in concert, display for others the meaning of their social situation.” The primary task for actors is not to embody social meanings that they personally subscribe to. Rather, it is to signal what they wish to have others believe and to utilize actions and gestures to convince the audience of the authenticity of their performances. Alexander argues that as societies become more complex and fragmented, elements of social performance become increasingly de-fused. Therefore, actors are tasked with re-fusing multiple multiple codes, myth and narratives back together into an indivisible whole and maintaining a seamless flow throughout their performances. These means of symbolic production also “serve as iconic representations to help them dramatize and make vivid the invisible motives and morals they are trying to represent” (Alexander 2004, p. 532). They range from performers’ outfits to many other ‘‘standardized expressive equipment’’ that is “intentionally or unwittingly employed” (Goffman 1959, pp. 22–51). In figure skating, athletes’ costumes, hairstyles, makeup and accessories play an instrumental role in materializing skaters’ cultural narratives and artistic visions. Competition outfits are designed to flatter skaters’ bodies and align them with valorized aesthetics in the sport: slender, balletic long lines for female skaters and toned muscular build for male skaters. Makeup, hairstyles and jewelries help enhance the narratives and emotions but are not overly elaborate so as to appear staged. The ultimate goal of figure skaters transcends technical mastery. It establishes an emotional connection and “create[s] the conditions for projecting cultural meaning from performance to audience” via psychological identification (Alexander 2004, p. 547). When putting together elaborate choreography for their competition routines, figure skaters engage in emotion management and dramaturgy through the practice of mis-en-scèn: they put culture “into the scene” (Broch 2020) through communicative gestures beyond the execution of technical elements. They painstakingly choose the styles of music that they are skating to and diligently control the way they emote throughout their performances, hoping to present an artistic persona that is believable to the audience. In competition, as figure skaters take center ice, hit their starting pose and wait for the music to start playing, they participate in the ritual of elite organized sport, where each contestant is allotted approximately 7 minutes (short and long programs combined) to showcase their craft under the spotlight, with a panel of 9 judges sitting in the front row, carefully examining each minute detail of one’s performance within the confines of the scoring system while the audience sits in the bleachers, applauding, waving flags and banners in support of their favorite athletes.
In trying to strike a balance between being in control and overtaken by the plausible limits of the iron cage of organized play, actors signal mastery by marshaling out an appearance of naturalness and effortlessly embodying authentic performers without shedding light on the constructed nature of performance. That said, authenticity is not so much a default state as it is achieved through a highly elaborate and rehearsed process. Building a reputation as an authentic performer requires social, cultural and economic capital to access various means of symbolic production, ranging from custom-made skates to designer costumes and elaborate makeup. The unequal distribution of social power (Alexander 2004) among national governing bodies and training sites also means that some athletes are more able to rely on the support of their federations than others for guidance and feedback. U.S. Figure Skating and other resourceful nations invite skaters on their national team to an annual training camp to assess the materials that skaters have come up with for the upcoming competition season and send expert judges to their training bases throughout the season to monitor them and give them feedback. It is not uncommon for skaters to be advised to change their programs in the middle of a competition season if officials deemed that these routines deviate from the desirable forms of artistic expressions in the sport. Such was the case for one Team USA skater I interviewed, who debuted their short program during the off season in front of a panel of international judges. “For a lack of a better word, they tore that program apart,” the skater (skater 3) recalled. The judges vetoed that program on the ground that it failed to build up to that exciting Olympic moment and felt lukewarm.
Moreover, status hierarchies in the sport of figure skating and society at large endow nearly monopolistic power in the judging panel, who assigns symbolic values to skaters’ display of naturalness and artistry. Intersecting cultural meanings with materiality and corporal acts (Broch 2020), judges and officials subject athletes to a regulated social order in the sport and construct a range of options when it comes to presenting artistry and an image of a docile body (Foucault 1995) that has been improved and transformed through prolonged rigorous training. They valorize gendered and heteronormative aesthetics and bodily practices that are commonly associated with upper-class, European (American) bodily practices, drawing inspiration from classical ballet and musical accompaniment (Kestnbaum 2003).
Following Alexander (2004)’s social performance theory, I investigate how cultural codes inform and shape athletes’ performances in general and their embodiment of artistry specifically. I argue that elite figure skaters’ performances are balancing acts between conformity to the iron cage of rules and the non-isomorphic yet culturally structured cultivation of authentic performances. In the next section, I show how athletes’ diverse approaches to competitive figure skating are contingent on different cultural narratives that they subscribe to regarding being an artist on the ice. Based on the ways in which they navigate and negotiate within or on the edges of the iron-cage judging system, I developed a three-way typology of elite skaters, namely the emerging phenoms, the athletic performers and the “authentic” artists.
Data and method
To examine how elite figure skaters as embodied performers enact shared cultural notions of greatness in the face of growing rationalization in the scoring system, I conducted semi-structured, in-depth interviews with current elite figure skaters who bear the consequences of judgments made under the IJS, which dictate their standings and reputation in the sport. In particular, I highlight how athletes’ cultural understanding of formally rationalized stipulations in the rulebook shapes the ways they construct their competition routines. By focusing on the embodiment of culture in actual practices, my interviews illuminate how athletes’ actions are both enabled and constrained by background representations in the skating community (Alexander and Smith 2004; Sewell 1992). In focusing on one case, I have sacrificed breadth for depth, but the merit of in-depth, interview-based projects lies in their unique ability to account for narratives that large-scale quantitative research fails to capture. Other researchers who study how rules, numbers and formally rationalized systems affect behaviors and outcomes have used similar types of data and methodological tools (e.g. Lom 2016; Espeland and Sauder 2007; Christin 2018).
As a former competitive skater and one of the select few individuals who are granted media credentials at major international figure skating competitions, I gained exclusive access to the most accomplished athletes who are currently active in the sport. From October, 2020 to January, 2022, I collected interview data at all international events held on U.S. soil as well as two U.S. Figure Skating Championships. I spoke to athletes in the mixed zone and press conferences immediately following their competitions. The bulk of my interview data, however, has come from in-depth interviews, each ranging from 30 minutes to an hour, that I conducted the day after the conclusion of each competition. During the 2020/2021 season, per COVID-19 protocols, all of my interviews took place either on Zoom or over the phone. Beginning in July, 2021, I was able to conduct interviews on site and broaden my sample beyond Team USA skaters as travel restrictions eased.
In total, I conducted 40 one-on-one, in-depth interviews with figure skaters from the U.S. and 5 other countries. My sample of skaters includes both novice athletes who recently became age-eligible at the Olympic level and veteran skaters holding National and/or World Championships titles and medals. Respondents are roughly evenly distributed among the four subfields of figure skating, namely men and women singles, pair skating and ice dance. Given the small number of athletes who are able to make the national team in figure skating (for reference, Team USA has 55 skaters across four disciplines this season on its Olympic-level roster), my sample is representative of elite figure skaters in the U.S. while also attending to international perspectives. All interviewees gave me permission to record our conversations and I have transcribed them. National governing bodies of the sport, who give permission for accredited journalists to conduct exclusive interviews following each event, have fulfilled all my interview requests, so I have a response rate of 100%. That said, with respect to limited capacity in processing interview requests, I did not put in an excessive number at any given competition. Instead, I accumulated a cross-national sample of elite athletes over time.
Data analysis involves thematic coding and content analysis of interview transcripts. Revolving my research questions, I focused on skaters’ lived experience of striving for psychological identification with the judges and the audience based on shared cultural beliefs of excellence in the sport. I also made notes around each discussion that invokes the 6.0 system or the IJS as well as the way that elite skaters balance their time between technique and artistry in training to zoom in on the impact of the judging system on skaters’ program construction.
The iron-cage judging system in the sport of figure skating
The growing formal rationalization in figure skating’s judging system has restructured the institutional environment that gives top-level skaters a circumscribed set of options when it comes to constructing their competition routines and crafting their day-to-day training regimens. During our interviews, most skaters described the constant tug of war that occurs between the two major components of their performance: on one hand, skaters perform up to a dozen technical elements in a competition program, all of which are closely scrutinized by the judging panel who numerically assesses their levels of difficulty and quality of execution. On the other hand, the artistic side of the sport calls for musicality, character portrayal and emotional execution. However, the technical and artistic aspects of their skating frequently work against one another. Commitments to the choreography and transitional steps, even something as minuscule as a hand gesture and head movement can easily throw off a jump completely. Therefore, striving to put together a fused social performance is a tremendously challenging feat.
In addition, as the new judging system spells out the exact features required and their respective numeric values, technical elements have become much more time-consuming than they were before, which leave skaters with limited opportunities to showcase their artistry. My interview data explores the ways in which skaters reconcile the exacting technical requirements imposed by the scoring system with their artistic visions. A pair skater (skater 2) shared that in their four-minute long program that consisted of 11 elements, just a spin alone can take up to 20 seconds. Sometimes, they go to extra lengths to ensure that they do not lose points due to human errors in operating this incredibly elaborate scoring machine: “you want to hold your death spiral for an additional half rev just in case the technical panel blinked and missed half a rev (laugh).” The time-consuming nature of technical elements under IJS has undermined their ability to dedicate considerable time and effort to performance and interpretation.
Stacy Lom (2016), who studies the effect of IJS on the sport of figure skating, contends that the increasingly quantified nature of the judging system has led to a disproportionate emphasis on technical merit and a devaluation of artistry, which is harder to be accounted for through formally rationalized means. Skaters play the “numbers game” and produce performances that are highly homogeneous to their competitors’. However, I find that skaters’ responses to the judging system have been far from unanimous: while some treat each aspect of their performance, technical or artistic, as highly intentional athletic endeavors, others strive to preserve their artistry from being mastered by constraints imposed by the numeric stipulations of the rulebook. As skaters navigate their institutional environment governed by the IJS, they invoke shared cultural codes regarding embodiment of artistry and standards of excellence in the sport, which in turn shape the way they present themselves in competition. A highly circumscribed notion of artistry that valorizes maturity and authenticity creates an additional axis of stratification within a highly rationalized system. Skaters who intend to present themselves as authentic artists work towards enacting what cultural sociologists understand to be “cultural pragmatics:” they dedicate considerable efforts to persuasively integrate various elements of their routines, hoping to present them as a fused, convincing whole.
The emerging phenoms
According to Goffman (1959), actors intentionally or unwittingly employ “expressive equipment,” hoping to present to the audience an idealized version of themselves aligned with distinct cultural expectations. In recent years, mass media have hyped up novice skaters who upset their more experienced counterparts by executing extraordinarily complex technical elements. For example, when 14-year-old American phenom Alysa Liu defended her U.S. National title in a stacked field of seasoned competitors, major media outlets like NBC Sports (2020) applauded her for being one of the only female skaters to have ever landed the historically difficult Triple Axel, completing 3.5 revolutions in the air, but made no remarks on her artistry. Indeed, in the elite figure skating world, young skaters who burst onto the scene are usually expected to be some of the most technically advanced. However, when it comes to the artistic aspects of their skating, they are often perceived as lacking maturity and life experience to authenticate the complex emotions that they hope to convey. Therefore, just like novice rock climbers (Beames and Pike 2008) who deliberately try to misrepresent their abilities through overly elaborate, pro-level clothing, newcomers to the senior level often wear custom-made outfits bedazzled with rhinestones and makeup geared for upscale elegance to make them appear older (Rand 2012). Some even go so far as to choose pieces of music portraying themes and life experiences beyond their age in a primitive attempt to signal maturity and sophistication in style and expressions.
In spite of their surface-level efforts to embody the mature, artistic skater, the emerging phenoms in the sport are limited by a lower ceiling on artistry given their small body of work at the senior level. Judges routinely reward veteran skaters who present themselves as mature, sophisticated performers and penalize younger skaters for their contrived and disingenuous attempts at enacting artistry. Therefore, in alignment with the cultural terms of reference in the sport, the novice skaters in my study unanimously claimed that when they are perceived as newcomers to the senior scene, they strive to create exhilarating moments on competition ice by attempting spectacular technical elements, such as jumps with more than three revolutions or hyper-extended posture on spins. A skater in my study who made the Olympic team in their first senior season (skater 4) recalled that they had to upgrade their technical arsenal in order to boost their chances. “Because it was my first year senior, my [artistic] components wouldn’t just magically go up,” they explained, “I knew that I needed those quadruple jumps so that I could undeniably raise my scores to where I could qualify for the Olympic team.” And they did make it to the Olympic Games that year.
However, even though spectacular technical elements give newcomers fleeting moments of glory, as judges and audiences alike are amazed by their mind-blowing jumping abilities, it only goes so far in the sport. Prioritizing the easily quantifiable elements may have become a desirable approach in other fields, but in a sport like figure skating that valorizes something as ineffable as artistry, immense technical prowess will not be sufficient to fully capture the essence of its excellence. Therefore, primarily highlighting technical merit is only common practice amongst newcomers to the scene. Seasoned competitors who strive to claim top accolades acknowledge that the only viable way to the pinnacle of their sport is to excel both technically and artistically. That said, they find different ways to enact artistry within the constraints of their highly restrictive institutional environment.
The athletic performers
In the sport of figure skating, one’s perceived ability to enact authentic artistry depends heavily upon the most fundamental means of symbolic production an actor can access-their body. For female skaters in particular, perceptions of authenticity and artistry are profoundly governed by culturally distinct standards of femininity that revolve around the “artistic” and “athletic” dichotomy, which frequently function as codes akin to butch and femme (Rand 2012). “Athletic” skaters, especially when described in opposition to “artistic” skaters, are often known for technical prowess that are unattainable for most female skaters. Tonya Harding and Midori Ito, two renowned “athletic” skaters of the past generation, are not perceived artistic because of their mastery of the once elusive jump: the triple Axel, an accomplishment that was gendered masculine at the time for its rarity in women’s skating. “Athletic” also implies that such technical feats are achieved at the expense of artistry.
Today, as more female skaters have entered the previously male-exclusive territories of triple Axels and quadruple jumps, gender coding in the sport becomes increasingly nuanced. It is not just about whether women singles competitors can accomplish such remarkable technical feats, as triple Axels and quadruple jumps are usually required to succeed at the highest level of the sport. Instead, it is how one accomplishes these “ultra-C” elements that dictates the way they are gender coded. On the one hand, if a female skater is able to complete four revolutions in the air using the “spinny” technique, which requires the athlete to be extraordinarily slim so as to speed up rotations, they are usually still coded as “female” for their delicate, ballerina-like style of movement. On the other hand, when female skaters acquire the necessary air time to rotate four times through the height and distance of their jumps as well as speed and power, just like their predecessors Harding and Ito had done, they are usually perceived as “athletic.” Their physical stature tends to be more muscular and bulkier, which also contributes to their perceived masculinity.
As such, for skaters in my study who are perceived as athletic and lacking femininity, curating an artistic persona on the ice, especially one that aligns with the valorized aesthetics in the sport, has been an ongoing struggle. Being viewed as lacking the culturally valorized physical attributes and lyrical styles of performance, they adopt a heavily intentional approach to crafting their competition programs. Instead of viewing artistry as an outlet of self-expression, they treat it as athletic endeavors and work closely with the judging system to carefully check off each box on the rubric. For example, when a single skater (skater 5) holds their leg up (the skating term for it is a “spiral sequence”) to highlight their flexibility and strength in between their jumps, they calculate in their head how the transitional move would elevate their scores. “When I’m doing the spiral,” they explained, “in my head I’m thinking interpretation and transitions [marks] may go up, especially if it’s put later on in the program and it’s that exciting moment.” Skaters who view artistic expression as athletic labor consider it to be a highly strenuous activity that needs to be balanced against the execution of technical elements. A pair skater in my study carefully anticipates the energy required to perform their choreography and match the intensity of the music at the end of their long program. To these athletes, artistic expressions are not categorically different from technical elements. They are merely another task assigned by the judging system and may be attained through predominantly athletic endeavors.
The “Authentic” artists
Although some skaters in my study treat the artistic elements of their skating as highly intentional and closely align the construction of their programs with the highly quantified judging system, others approach their choreographic process to showcase artistry in a more seemingly natural and unmanufactured manner. Granted, the artistic persona in figure skating is not innate. Rather, it is a combination of desirable physical attributes and years of practice and curation. One needs to understand the distinct cultural expectations in the world of figure skating and go through a prolonged apprenticeship period in order to master and naturalize their embodiments of such desired qualities. A veteran on Team USA (skater 7) shared that skaters have to navigate various “hidden curricula” in the sport when it comes to how artistry is assessed, as there could be discrepancies in preferences of music choice and choreographic packaging between international and domestic judges. In recent years, while the widely acclaimed French Olympic champions Gabriella Papadakis/Guillaume Cizeron have popularized a minimalist style accompanied by modern piano pieces on the international competition scene, American judges continue to reward crowd-pleasing programs set to Broadway and other Americana music that is more accessible to the audience at U.S. National Championships. Therefore, skaters learn to weigh their battles against one another and come up with competition routines that are eclectic enough to cater towards both sets of judges and crowds.
Skaters who strive to become “authentic” artists engage in cultural pragmatics (Alexander 2004), trying to make their social performance appear seamless and convincing as a whole. They convey their genuine passion for the performance aspect of the sport and prioritize the artistic aspect of their skating, even if it potentially jeopardizes the execution of their technical elements. One single skater (skater 6) in my study, for example, views creating a skating routine as analogous to painting a picture with a brush. Instead of thinking how individual movements fulfill the IJS requirements, they draw the picture as one piece of art. “Each step, each jump and weaving in and out is like a brush stroke,” the skater said. “At the end of the program you have this painting. That’s how I see it. It’s very holistic. You can’t see the whole picture until it’s done and appreciate it as a whole.”
They acknowledge that their artistic vision can potentially undermine the consistency of their jumps, but they have refused to let their integrity falter when it comes to artistry, which is what made them fall in love with skating in the first place. “I try my best to push the artistic side and make the transitions as difficult as possible,” the skater claimed. “It obviously makes the elements harder but I definitely want to push that way because that’s what drew me in when I was 5 years old and my love for performing and for the artistic side of the sport has only grown over the years. I don’t want to hold back artistically for the technical.”
Similarly, a pairs team (skater 7 and 8) who values musicality places an incredibly tricky choreographic pattern right before their most intricate lift on the crescendo of the music. While their choreographic pattern will unlikely make a significant contribution to their overall score, it drastically enhances their musicality and quality of their artistic expressions. They specifically hope to counteract the homogenizing nature of the IJS and avoid the tendency to “go from one [technical] element to another” in their competition programs merely to satisfy the rigid and rationalized requirements laid out by the judging system.
Skaters hoping to be perceived as authentic artists engage in elaborate training regimens focused on the way they present their artistic personas in front of the judges and the audience. When asked about how they cultivate a sense of naturalness when it comes to projecting artistry on the ice, skater 6 rejected the idea of “innate talent” and pointed to an elaborate, prolonged rehearsal process that includes a myriad of dance practices as a supplement to their skating training. As they (skater 6) put it:Did artistry come naturally to me? No, Naturally I love performing but to be honest, throughout the years, I have taken so many dance classes, from ballet to modern to hip hop. I have worked really hard with my choreographers each year to cultivate a chosen artistic genre. Every single year I learn a different type of style and a different range of performances, whether it’s Latin, classical, Broadway, jazz and it’s always changing and growing my repertoire.
In the face of increasingly complex technical requirements mandated by the judging system, skaters acknowledge the inevitable tradeoff involved in the process of cultivating an authentic artistic persona on the ice. They recognize that their artistic vision can potentially undermine the consistency of their jumps, as even a tiny gesture intended to enhance their artistic vision could completely throw their balance off and interfere with their ability to execute technical elements. However, because of their genuine passion for the artistic aspects of the sport, they have non-instrumental reasons (Larsen 2016) for not watering down their choreography for the sake of technicality.
In trying to master transitional steps that could enhance the seamlessness of their routines, figure skaters who strive to enact authentic artistry dedicate a considerable amount of time in their daily training regimens to refine their toolkits used to fuse elements of their performances back together. One such important tool is skating skills, which refer to an athlete’s ability to execute complicated transitional steps that integrate their technical elements seamlessly and with ample flow. Skaters who valorize and actively maintain the image of authentic artistry through their performances hone their skating skills daily. A singles skater (skater 9) told me that their training group, one of the most prominent in their country, begins each morning with a skating skills class, which helps athletes learn to control their edges with the utmost precision and convey their artistic visions with clear intentions through each movement. They also frequently access their choreographers throughout the season and dedicate entire training sessions to polishing nuanced in their choreographies only.
Furthermore, skaters who strive to become “authentic” artists constantly work on their ability to emote, hoping to forge psychological identification with the audience (Alexander 2004). To hone their ability to resonate with the audience at an emotional level and “make them feel something when [we] perform,” multiple skaters (e.g. skaters 3, 6–9) in my study make use of the mirrors in ballet studios to practice engaging their facial expressions during their performances in accordance with their music, characters and choreography. A national champion single skater (skater 9) explained that efforts to naturalize these emotions in competition is a time-consuming one, but “if you do something like that [engaging with your emotions and sharing a story] in your practice and you’re always cognizant of it, when you go and compete, it becomes little bit easier to share.”
That said, processes of enacting authentic artistry are much smoother for those who possess the physical attributes and personality traits that are coded as artistic in figure skating. The figure skating world is an archetypical field where background representations intersect with materiality and corporal acts (Broch 2020). Judges and officials in the sport actively shape the notion of artistry and its proper forms of embodiment on the ice. While it is true that athletes are encouraged to project their authentic selves on the ice, specific cultural expectations in the sport dictate that some athletes’ artistic personas are given more currency than others’. Figure skating, like other aesthetics-based sports and art forms, are built upon an elitist culture that espouses a distinct, Eurocentric vision: when it comes to female skaters, for instance, long, toned lines are coded as artistic whereas stockier bodies are coded as athletic and powerful, which is a less desirable look on the ice. Many Olympic skaters that I spoke to predict that the refined ballerina look will continue to be the enduring image of the sport for years to come. In pairs and ice dance, while teams can portray a wide range of relationships with one another, heterosexual, romantic narratives have been disproportionately rewarded. “I do think our chemistry plays to our advantage when it comes to our interpretation,” a pair skater (skater 3) confessed. “We are the classic cliche of pair teams who are a couple off the ice. Our program is about a girl who is a free spirit and this man falls in love with her and vice versa.” Pair skaters who can cultivate romantic chemistry between one another are often perceived as more artistic, whereas sibling teams have been known to be held down in presentation component scores for their inability to do so.
Thus, cultivating an authentic artistry persona is a concerted effort that centers around the skater’s balancing act between the iron cage of technicality and the social performance of authentic artistry. It hinges upon different cultural narratives that skaters subscribe to regarding being an artist on the ice as well as judges’ perceptions of artistry and gender coding dynamics in the sport. One skater (skater 7) in my study illustrated such contingent processes of cultivating authenticity and performing artistry as follows: “[artistry] can be enhanced through training as one gets older, but I haven’t witnessed too many skaters have a rebirth of their ability to interpret. I’ve seen some get better or learn how to move in a more pleasing way to the eye, but if they don’t have [the desired physical attributes] from the start, they will likely shine less than the skater who breathes artistry.” The culturally specific perception of artistry based on one’s physicality and styles of performance valorizes some skaters’ presentation of an artistic persona while diminishing others’. That said, skaters who strive for authenticity and artistic integrity exert considerable agency in their training regimens and engage in a prolonged rehearsal process to hone various means of symbolic production that are employed to invoke cultural scripts that signify artistry and resonate with the audience.
Discussion and conclusion
This article contributes broadly to a growing body of literature that examines the consequences of formally rationalized instruments of evaluation (Espeland and Sauder 2016; Christin 2018; Rivera and Tilcsik 2019). Similar to Lisa McCormick (2015)’s research on classical musicians, who compromise their musical values to prioritize technical accuracy in response to the formally rationalized criteria in competition, I find that there is tension between striving for technical prowess and artistic expressions in the sport of figure skating. On one hand, athletes struggle with the increased time constraints imposed by the exacting requirements on the technical side under IJS, which compete with their time and energy spent on the artistic side of their skating. On the other hand, commitment to choreography and emotional execution is draining, which can impede the execution of technical elements.
Drawing on Alexander (2004), this article explores the various ways in which skaters maneuver between the iron cage of rationalized technical requirements and their presentation of an artistic persona. I illustrate the contingent process by which elite figure skaters enact cultural structures and their varied success at persuading the audience of the authenticity of their performances. Newcomers to the Olympic level make contrived attempts to signify their maturity and mastery of the sport through overly elaborate costumes and makeup as well as age-inappropriate music selections. Veterans in the sport, on the other hand, present bifurcated strategies when it comes to performing such balancing acts. Some adopt a heavily intentional approach to crafting their competition programs, whereas others make a concerted effort to seamlessly re-fuse multiple elements of their performances back into a convincing whole, striving to achieve an appearance of authentic artistry. As I have shown in my study, the notion of artistry comes with specific, socially constructed expectations in the sport, while “authenticity,” as paradoxically as it sounds, is a highly elaborate and rehearsed state that takes years of practice (McCormick 2015). Skaters who work towards presenting an authentic artist persona constantly hone their transitional steps, skating skills among other means of symbolic production that facilitate their dramatization of “invisible motives and morals they are trying to represent” (Alexander 2004, p. 532).
While my work accounts for the current desired aesthetics in the sport, it does not adequately explain how those who hold social power have actively defined and perpetuated their aesthetic judgements and whether or not these standards have or could evolve over time. In figure skating, judges and the International Skating Union hold considerable social power (Alexander 2004) in defining and assessing proper embodiment of artistry on the ice. Scholars have found that aesthetic judgments are influenced by the judgments of others in the community (e.g. Childress and Friedkin 2012; Lamont 2009). Although privately held views may or may not change through social interactions, one’s expressed judgments tend to conform to community standards (Wohl 2015). Therefore, it would be worthwhile to study conferences hosted by the International Skating Union to examine discussions regarding the proper embodiment of artistry in the sport of figure skating to gain a better understanding of how standards of excellence are defined and negotiated among judges and officials.
Due to inherent challenges associated with accessing an elite and exclusive population, the majority of my respondents represent Team USA with a few others competing on behalf of other countries in Asia, Europe and North America. As Angèle Christin (2018) has observed in her study on how media outlets in the U.S. and France respond to quantifiable metrics in their external environments, actors’ responses to constraints and pressures imposed by formally rationalized forms of assessments depend on their national and institutional contexts. Therefore, while I have tried to recruit participants from a range of countries, this study does not capture the impact of IJS on some key actors in the field, most noticeably athletes from Russia. I have anecdotally observed that Russian skaters disproportionately attempt to display artistry in a more mechanical manner, but I am unable to test my hypothesis given the language barrier. Future studies may investigate how processes of meaning-making and legitimacy-building diverge across national contexts despite a nearly universal drive to submit to quantifiable metrics.
In conclusion, the scoring system in figure skating represents a hybrid evaluation system that constitutes a combination of formal rationalization and culturally specific production of authentic artistry. The formally rationalized scoring system, coupled with a multifaceted and globalized definition of artistry, elicits a distinctive range of responses and strategies from skaters who strive to offer authentic social performances. Skaters work towards embodying shared cultural notions of greatness on the ice and engage in agentic work to maneuver between the iron cage of technical requirements and the symbolic production of artistry. Despite risking the consistency of their technical elements, athletes who strive to convey authentic artistry through their social performances on the ice constantly work on integrating elements of their performance into a fused whole and honing their ability to create psychological identification with the audience through their movements and emotions. Findings in my study suggest that in spite of coercion in the external environment, in this case the growing quantification of the scoring systems in figure skating, actors never behave purely due to instrumental reasons. In fact, instrumental reasons themselves are codes and narratives that shape figure skaters’ performances. Therefore, the sport of figure skating is embedded in a wealth of shared cultural scripts, as athletes actively display and shape meanings of their social situations through performances on the ice (Alexander 2004).
Acknowledgements
The author wishes to thank Dr. Trygve Broch for his editorial wisdom and anonymous reviewers for their insightful comments. This paper benefited from perceptive feedback from Nina Bandelj, Francesca Polletta, Evan Schofer, Julia Lerch and Alexa Knierim. Participants at the American Sociological Association’s Annual Meeting also provided useful suggestions on earlier drafts.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Alexander J Cultural Pragmatics: Social Performance Between Ritual and Strategy Sociological Theory 2004 22 4 527 573 10.1111/j.0735-2751.2004.00233.x
Alexander J Performance and Power 2011 Cambridge, UK Polity Press
Alexander J Smith P The Strong Program in Cultural Sociology 2004 New Haven Yale University Center for Cultural Sociology
Beames, S., and E. Pike. 2008. Goffman goes rock climbing: Using creative fiction to explore the presentation of self in outdoor education. Australian Journal of Outdoor Education 12 (2): 3–11.
Broch T A Performative Feel for the Game: How Meaningful Sports Shape Gender, Bodies, and Social Life (Cultural Sociology) 2020 London Palgrave Macmillan
Bromley P Powell W From Smoke and Mirrors to Walking the Talk: Decoupling in the Contemporary World The Academy of Management Annals. 2012 6 1 483 530 10.5465/19416520.2012.684462
Childress C Friedkin N Cultural Reception and Production: The Social Construction of Meaning in Book Clubs American Sociological Review 2012 77 1 45 68 10.1177/0003122411428153
Christin A Counting Clicks: Quantification and Variation in Web Journalism in the United States and France American Journal of Sociology. 2018 123 5 1382 1415 10.1086/696137
Coburn C Beyond Decoupling: Rethinking the Relationship Between the Institutional Environment and the Classroom Sociology of Education 2004 77 211 244 10.1177/003804070407700302
DeSoucey M Elliott M Schmutz V Rationalized Authenticity and the Transnational Spread of Intangible Cultural Heritage Poetics 2019 75 2019 101332 10.1016/j.poetic.2018.11.001
DiMaggio P Powell W The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields American Sociological Review 1983 48 2 147 160 10.2307/2095101
Espeland W Sauder M Rankings and Reactivity: How Public Measures Recreate Social Worlds American Journal of Sociology 2007 113 1 1 40 10.1086/517897
Espeland W Sauder M Engines of Anxiety: Academic Rankings, Reputations and Accountability 2016 New York Russell Sage Foundation
Foucault M Discipline and Punish: The Birth of the Prison 1995 New York Vintage Books
Goffman E The Presentation of Self in Everyday Life 1959 New York Anchor Books
Guttman A From Ritual to Record: The Nature of Modern Sports 2004 New York Columbia University Press
Hallett T The Myth Incarnate: Recoupling Processes, Turmoil, and Inhabited Institutions in an Urban Elementary School American Sociological Review 2010 75 1 52 74 10.1177/0003122409357044
Kellogg K Operating room: Relational spaces and microinstitutional change in surgery American Journal of Sociology. 2009 115 3 657 711 10.1086/603535
Kestnbaum E Culture On Ice: Figure Skating and Cultural Meaning 2003 Middletown Wesleyan University Press
Lamont M How Professors Think: Inside the Curious World of Academic Judgment 2009 Cambridge Harvard University Press
Larsen H Performing Legitimacy: Studies in High Culture and the Public Sphere 2016 London Palgrave Macmillan
Lom S Changing Rules, Changing Practices: The Direct and Indirect Effects of Tight Coupling in Figure Skating Organization Science 2016 27 1 36 52 10.1287/orsc.2015.1018
McCormick L Performing Civility: International Competitions in Classical Music 2015 Cambridge Cambridge University Press
Power M The Audit Society 1997 Oxford Oxford University Press
Rand E Red Nails, Black Skates: Gender, Cash, and Pleasure On and Off the Ice 2012 Durham Duke University Press
Rivera L Tilcsik A Scaling Down Inequality: Rating Scales, Gender Bias, and the Architecture of Evaluation American Sociological Review 2019 84 2 248 274 10.1177/0003122419833601
Sewell W A Theory of Structure: Duality, Agency and Transformation American Journal of Sociology 1992 98 1 1 30 10.1086/229967
Strathern M Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy 2000 New York Routledge
Swidler A Talk of Love: How Culture Matters 2001 Chicago The University of Chicago Press
Witz A Warhurst C Nickson D The Labor of Aesthetics and the Aesthetics of Organization Organization 2003 10 1 33 54 10.1177/1350508403010001375
Wohl H Community Sense: The Cohesive Power of Aesthetic Judgment Sociological Theory 2015 33 4 299 326 10.1177/0735275115617800
| 36466044 | PMC9707132 | NO-CC CODE | 2022-12-15 23:15:18 | no | Am J Cult Sociol. 2022 Nov 29; 10(4):657-675 | utf-8 | Am J Cult Sociol | 2,022 | 10.1057/s41290-022-00178-x | oa_other |
==== Front
Univers Access Inf Soc
Univers Access Inf Soc
Universal Access in the Information Society
1615-5289
1615-5297
Springer Berlin Heidelberg Berlin/Heidelberg
952
10.1007/s10209-022-00952-1
Long Paper
Investigating the impact of virtual tourism on travel intention during the post-COVID-19 era: evidence from China
Ye Dingyu [email protected]
1
http://orcid.org/0000-0003-2328-9233
Cho Dongmin [email protected]
2
Liu Fufan 3
Xu Yu 4
Jia Zhengzhi 5
Chen Jianyu 6
1 grid.411545.0 0000 0004 0470 4320 Department of Design and Manufacturing Engineering, Jeonbuk National University, Jeollabuk-do Jeonju, Republic of Korea
2 grid.411545.0 0000 0004 0470 4320 Department of Industrial Design, Jeonbuk National University, Jeollabuk-do Jeonju, Republic of Korea
3 grid.9681.6 0000 0001 1013 7965 Faculty of Information Technology, University of Jyväskylä, Jyvaskyla, Finland
4 Department of Business Economics, Chent University, Ghent, Belgium
5 grid.19822.30 0000 0001 2180 2449 Department of International Communication, Birmingham City University, Birmingham, UK
6 grid.495269.5 Department of Economics and Management, Sichuan Tourism University, Chengdu, China
29 11 2022
117
14 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
This study explores the mechanism that contributes to travel intention in the field of virtual tourism. The overall research method is based on the “Stimulus-Organism-Response” theory. In the research model, the effects of content quality, system quality, and interaction quality in virtual tourism on tourism experience and travel intention are explored, as well as the role of virtual attachment and travel intention. A total of 390 respondents were invited to participate in a virtual tourism experience, and provide feedback through a questionnaire. SmartPLS 3.3.2 was used to validate the causal model, and most of the study hypotheses were supported. The findings show that virtual tourism significantly promotes travel intention. Specifically, content quality, system quality, and interaction quality positively affect tourists' travel intention through the complementary mediations of tourism experience and virtual attachment; and system quality even directly promotes travel intention. However, tourism experience does not affect virtual attachment. The present study extends prior studies on virtual tourism with SOR as a general model for field tourism experience research, while demonstrating the effectiveness of virtual tourism in promoting tourists’ travel intention. The results are useful in assisting governments with developing relevant policies and services, as well as helping tourism companies understand virtual tourism as an enhancement for tourist travel intention, thus contributing to the recovery of the tourism industry in the post-COVID-19 era.
Keywords
Virtual tourism
Stimulus-organism-response model
Travel intention
PLS-SEM
Post-COVID-19
China
==== Body
pmcIntroduction
The COVID-19 pandemic was the most significant public health emergency with the most rapid spread and broadest infection range since the founding of the People's Republic of China in 1949 [114]. During the outbreak, economic activities in China were basically at a standstill [79], except for certain industries that still functioned to meet the basic needs of the public. The outbreak significantly disrupted the entire tourism industry due to the allopatric and clustered nature of tourism activities [110], a series of activities undertaken by tourists who travel from the source to the destination via the tourism corridor. In response to COVID-19, Chinese cities have primarily closed communities and surrounding villages and towns to restrict the movement of people. As a result of this response, demand for tourism activities was passively reduced to zero. Travel agencies, which were the gateways, were asked to suspend their operations. In contrast, air carriers and railroads, the gateways, changed their usual "change and refund" policy and offered free refunds to passengers who had already purchased tickets. Accordingly, passenger traffic and revenue in the transportation industry plummeted during the pandemic [92]. In tourist destinations, tourist attractions were almost completely shut down. Hotels were either shut down, or provided temporary housing for medical staff or used as temporary isolation sites to receive patients [16]. Arguably, the tourism supply was essentially at zero.
The sudden onslaught of COVID-19 kept people at home [3], which was a big blow to the offline tourism industry. Since the pandemic was alleviated in China, the offline tourism industry still faces a significant challenge [25]. In this course, the tourism industry has also explored new ways of development, i.e., "virtual tourism [116]," "live-streaming tourism [80]," and other new ways of relaunching the tourism industry with the help of the internet and innovative technologies. The segmentation of "smart technology + tourism" is reflected in integrating virtual reality (VR) technology and tourism products [61]. The development of virtual reality technology has broken through past limitations, which could only simulate the natural environment through pictures or videos and provided participants with a better "immersion" experience [40]. As the influence of VR technology expands and its attention increases, academic research on VR technology is also increasingly available. The pioneering research mainly combines virtual reality technology with landscape design, medical education, and disease treatment from the research content. Nevertheless, a further literature search revealed that the existing literature on "VR and tourism" is sparse. Scholars have now focused on the marketing value of VR technology [57, 60, 87] and destination image building [4, 12, 59]. Although the methods used in these studies are relatively homogeneous and empirical studies are lacking, they still provide vital tools, methods, and mindsets for studying VR technology and tourism. The theory has guiding significance for practice, and VR technology, as a new communication medium, has an immeasurable impact on the tourism industry's development. Hence, it is imperative to strengthen academic research in virtual tourism.
The decisive intervention of virtual technology has given new connotations, characteristics, and forms to the relationship between people and places [28]. The traditional binary space of tourist and physical tourist places shifts into the ternary connections between virtual place, tourist, and real place. The interrelationship of such ternary spatial continuity has become an essential topic in the study of the people and the place in the digital era. At present, scholars have paid attention to constructing the ternary space [50]. However, the study of tourists' behavioral intentions in the ternary space still needs to be advanced. Virtual tourism may affect users' experiences, attitudes, and behaviors [20, 43, 98]. However, the way virtual tourism affects users' travel experience and emotional attachment, as well as the role of virtual tourism on future travel intention in the field, is still unclear. Therefore, starting from the SOR theory, this paper constructs a model of the influencing factors of virtual tourism experience (3D reconstruction of the real tourism place)—the inner psychology of users in the virtual tourism process—the travel intention in the field. We explore how virtual tourism uses digital technology to construct users' attachment to real tourism places and then influence their travel intention in the field.
This study, therefore, starts from the characteristics of virtual tourism, draws on the research theories and experiences of established social media platforms such as live streaming, short video, and online shopping, and introduces two mediating variables of tourism experience and virtual attachment based on the psychological perspective, with the “Stimulus-Organism-Response” theory (referred to as SOR model) as the theoretical framework, constructs a conceptual model of the impact of virtual tourism on tourists' travel intention, and carries out empirical analysis with virtual tourism to explore how virtual tourism affects their field travel intention through users’ intrinsic state. This study contributes to a comprehensive and in-depth understanding of the new tourism human–ground relationship in the information age. It clarifies the mechanism of human-ground emotional attachment in virtual tourism. At the same time, the present study helps to clarify the mechanism of virtual tourism influence on users’ travel intention in the field, deepening the study of the human–place relationship in the ternary space. The results provide a feasible direction for promoting the integration of tourism locations and people in the information age. Furthermore, this study has important practical implications for the experience design of virtual tourism, tourism destination marketing innovation, and enhancement of tourist loyalty in the post-COVID-19 era. Therefore, this paper empirically examines the impact of virtual tourism on tourists' travel intention in the field through the questionnaire method, specifically by addressing the following research questions:
RQ1: Does the virtual tourism experience increase users' travel intention in the field?
RQ2: What factors contribute to tourists' travel intention in the field of virtual tourism?
RQ3: How do important influencing factors of virtual tourism experience (content quality, system quality, interaction quality), tourism experience, virtual attachment, and travel intention interact with each other?
Theoretical background and research hypothesis
Stimulus-organism-response (SOR) model
The Stimulus-Organism-Response (SOR) model is a model of human cognitive behavior, first proposed by [63]. It reflects the “stimulus-perception-response” process of human behavior. The SOR theory model suggests that stimuli from the external environment influence individuals' behavioral decisions by affecting their emotions. The term "stimulus" refers to the factors that stimulate and cause individuals to act [85]. The term “organism” refers to an individual's interior psychological state [85]. The term "response" refers to the individual's numerous behaviors or behavior intentions in response to the stimuli and the organism [85]. In the SOR framework, the stimulus is generally used as the independent variable, the organism as the mediating variable, and the response as the dependent variable. SOR began as a cognitive model used in psychology and is now frequently utilized to examine Internet user behavior [14, 46, 99, 118].
In tourism research, the SOR model was used to study travel experience [15, 64], travel intention [64, 91, 97], and user engagement [2, 96, 111]. Researchers have found various factors capable of influencing tourist travel intention, including tourists' internal reasoning and the influence of certain external factors, such as tourists' reference information, personal perceptions, and perceived risks [69]. Various social, economic, and psychological factors may impact tourists' travel intentions. Among them, both the reference group and the individual's subjective perceptions play a vital role in forming travel intentions. The reference group of tourists and their subjective knowledge psychologically form emotional preferences and inherent impressions of a tourist destination, which affects the willingness and choice of tourists to travel to that destination. SOR theory suggests that the external environment stimulates the individual's perceptions and emotions, which affects the individual's behavior [88]. Therefore, in studies of consumer travel intentions or behaviors, giving subjects “S” (e.g., VR environment) stimuli before measuring “R” (e.g., questionnaire results) tend to reveal more valid findings when travelers use digital technology such as virtual reality (VR), artificial intelligence (AI), and augmented reality (AR) to immerse themselves in a highly immersive virtual tourism experience. The process of content stimulation produces the individual's organism, i.e., psychological state. The content quality, system quality, and interaction quality of virtual tourism will directly impact the tourists' tourism experience and virtual attachment, which will affect the tourists' travel intention in the field.
The impact of virtual tourism experience on tourists' intrinsic state
According to previous research findings, content quality, system quality, and interaction quality are essential variables that entice tourists to use digital devices for virtual tourism experiences [43, 67, 90]. According to the SOR theory, tourists experienced content quality, system quality, and interaction quality of the virtual travel as stimuli in the experience process. The organism, i.e., the internal process that mediates the external stimulus and behavioral response received by the individual, is in this study expressed as the tourist experience and virtual attachment of tourists.
Virtual tourism and tourism experience
Experience is an objectively existing psychological need. Essentially, the tourism experience is an individualized feeling of an individual responding to certain stimuli [73]. Morrison et al. [68] discovered that physical environmental factors such as music, lighting, and facilities were positively associated with customer mood. Lee, park, and Han [56] found that the quality of online content affects user engagement and acceptance. Due to the attractiveness of the web, uploading high-quality images or videos can influence user satisfaction. Ghose and Huang [33] found that the higher the availability of modern technology, the more companies can promote product quality through personalized services and products. In this way, service quality is improved by increasing satisfaction. In the context of virtual tourism, its interaction quality mainly refers to the ability to provide personalized information, understand tourists' needs and preferences, and personalized interactions. Chang [13] designed an AR-based cultural heritage tour system. He found through a questionnaire that tourists had a strong experience with this system. Jung's [43] study found that the content of augmented reality technology, personalized services, and system quality affect tourists' experience and thus their satisfaction. Based on this, this study proposes the following hypotheses:
H1:
The content quality of virtual tourism positively affects the tourism experience;
H2:
The system quality of virtual tourism positively affects the tourism experience;
H3:
The interaction quality of virtual tourism positively affects the tourism experience.
Virtual tourism and virtual attachment
The specific and deep connections people make to a place by assigning meaning to it are called "place attachments" [71]. Attachment refers to "the human tendency to develop strong emotional ties to specific people and objects [7]." With the development of technology, an individual's attachment is not necessarily to a real place. However, it can be extended to a broader scope [81]. For example, with the rapid development of new media, people with common interests or experiences communicate and interact through computer networks, creating emotional attachments to the virtual communities they form [10]. When experiencing social isolation or loneliness, some users develop an emotional attachment to the Internet and social media [104].
Moreover, in the context of the current COVID-19 pandemic, virtual tourism rekindles tourists' travel confidence, helps them gain a sense of control and security, and regains their travel rhythm. Attachment arises from experience [35, 42], and the perceived value of different experiences has different effects on attachment production [31, 75]. When the virtual world is more realistic, the user's visual senses are more actively involved in aesthetic perception, resulting in a more memorable experience and encouraging attachment to virtual tourism [46]. In addition, the dual stimulation of content quality (virtual scenery) and interactive format (virtual experience) allows users to perceive the experience value of entertainment and enjoyment [71]. When users' own emotional needs are satisfied, they will have a positive emotional evaluation of the virtual scenery, which leads to the enhancement of virtual attachment [71]. This study contends that while utilizing digital devices for virtual tourism, tourists perceived positive content quality, system quality, and interaction quality would drive their connection to virtual tourism. Therefore, the following hypotheses are proposed in the present study:
H4:
The content quality of virtual tourism positively affects virtual attachment;
H5:
The system quality of virtual tourism positively affects virtual attachment;
H6:
The interaction quality of virtual tourism positively affects virtual attachment.
The impact of tourists' intrinsic state on travel intention
Tourism experience and travel intention
The organism is an internal cognitive process in response to external stimuli. In this process, the information generated by the organism as a result of the stimulus, such as sensations, perceptions, and emotions, will become the basis for subsequent travel intentions and travel behaviors. The emotions generated by an individual stimulated by the external environment will lead the user to approach or avoid behaviors toward the environment and mediate between environment and behavior [113]. Ekanayake [27] found that tourists' experiences positively influenced their travel intentions by studying tourists in the eastern province of Sri Lanka. Kim [47] studied tourists of virtual reality tourism based on the diffusion of innovation theory. He found that users' authentic tourism experience positively affects their travel intention. Kim and Jung studied the effect of tourists' authentic experience on their travel intention based on SOR theory. The study finds that tourists' behavioral intentions are influenced by their sense of authentic virtual reality tourism experiences [46]. Raouf Ahmad Rather [82] investigated the impact of experiential marketing activities on tourists' behavioral intentions in tourist destinations. It was found that tourists' experiences of marketing activities in tourist destinations affect their travel intention. Based on this, this paper proposes the following hypothesis:
H7: Tourism experience positively affects travel intention.
Virtual attachment and travel intention
The prior studies have found that with the intervention of network information, the traditional two-dimensional perspective of the human–territory relationship is extended to a three-dimensional space. New human-territory links and interaction forms of (real)person–(virtual)person, (virtual)person–(virtual)place, and (real)place–(virtual)place emerge [17]. The cognition, emotion, and behavior of virtual and real spaces influence each other and intermingle. Users' purchase intentions in the virtual world are consistent with those in the real world [93], and Ren et al. [83] demonstrated that attachment to virtual communities affects the frequency with which individuals visit virtual communities. Kim et al. [46] also found that their attachment to VR influenced VR tourism users' intentions to visit real places. Therefore, based on the above literature, this study argues that users' awareness and behavior in the virtual world somehow represent real-world awareness and behavior. Virtual attachment enhances tourists' desire for real-world travel destinations. Therefore, the virtual attachment will stimulate the travel intention in the field. Based on this, the following hypothesis is proposed:
H8: Virtual attachment positively affects travel intention.
Virtual attachment and tourism experience
Regarding the relationship between residents' activity participation and local attachment, many empirical studies point out that residents' involvement in local activities positively affects their local attachment [36, 84, 89]. Furthermore, through an empirical study, Kyle et al. and Hwang et al. [41, 53] found that tourists' leisure activity participation had a significant positive effect on their place attachment. In the study on the relationship between the cognitive gap, affective experience, and place attachment, scholars also confirmed that affective experience positively affects place attachment [38]. The higher the quality of tourists' emotional experience, the higher their level of attachment. It is hypothesized that the higher the positive tourism experience, the greater the likelihood that the tourist will develop a virtual attachment to the destination. Based on this, the following hypothesis was proposed:
H9: Tourism experiences positively influence virtual attachment.
The mediating role of the tourists' intrinsic state
The SOR theory emphasizes the mediating role of the response between the stimulus and the individual's behavior. By comparing the results of prior studies, it was found that tourism experience and virtual attachment play a connecting role between various influencing factors and behavioral intention. Chae, Kuo et al. [9, 52] found that service quality positively impacts customer recommendation and continued usage behavior. Wang and Chen found that the presentation of information systems positively affected consumers' intention to use them [101]. Jung [44] points out that the quality of content affects tourists' attitudes toward AR applications. N Zhang et al. [115] found that brand attachment exerts a partial mediating effect on the relationship between effective CC and brand commitment. Jenny Lee et al. [55] found a mediating effect between place attachment on satisfaction and loyalty. Based on this, the following hypotheses were formulated:
H10:
Tourism experience has a mediating role in influencing travel intention on content quality;
H11:
Tourism experience has a mediating role in influencing travel intention on system quality;
H12:
Tourism experience has a mediating role in influencing travel intention on interaction quality;
H13:
Virtual attachment has a mediating role in influencing travel intention on content quality;
H14:
Virtual attachment has a mediating role in influencing travel intention on system quality;
H15:
Virtual attachment has a mediating role in influencing travel intention on interaction quality.
Methods
Research model
In summary, this study employs SOR theory as the research framework. Content quality, system quality, and interaction quality of virtual tourism as the antecedent variables of tourism experience, virtual attachment, and tourists' travel intention in the field. The construction of a research model of the factors influencing virtual tourism on tourists' travel intention in the field, as shown in Fig. 1.Fig. 1 Research model
Usability test design and data collection
Virtual tourism is a new type of tourism as well as a new industry. Currently, Chinese tourists have less exposure and knowledge about it [62, 107]. With universal access to such new consumer technology, we implemented a real usage scenario to get participants actually and bodily experience virtual tourism in the field. The few users who already experience virtual tourism products were not directly selected for this study. In this way, researchers can measure users' temporal-sensitive experience and concurrent travel intention after the tourists have touched and used the virtual tourism device. The specific method can be referred to as "usability testing [26]." The concept of usability testing was first introduced in 1981 [49]. Usability testing evaluates the usability and unavailability of a product based on certain usability criteria. Usability testing is also a programmed process used to identify problems that may occur during user–product interaction. In a narrow sense, usability testing refers to user testing methods that allow a representative group of users to perform typical operations on a product. At the same time, observers and design developers observe, listen, and record. It is a process used to evaluate a product or system's external form, functional operation, and interaction patterns [26].
Previous studies investigated the research and application of usability tests from multiple aspects and dimensions in tourism research. Researchers discuss usability test methods and applications from different perspectives, including usability evaluation methods such as user testing [72], questionnaire survey [78], and eye-tracking technology [29]. And the researchers also make practical exploration of usability test research. We can consider usability tests an effective verification method for analyzing user behavior and experience. This lays an important theoretical foundation for this study.
Participants
Our research team obtained the research sample (Table 1) in two ways: First, we recruited virtual tourism experience participants (52% of the total number) through the online social media platform Sina Weibo. With over 300 million monthly active users, Sina Weibo is the largest public social media platform in China. The researcher recruited participants by posting recruitment posts on the Sina Weibo platform. The research team then used the Weibo-targeted promotion service to promote the recruitment posts to users who were potentially willing to participate in the virtual tourism experience. Second, the research team recruited users who were willing to participate in virtual tourism through offline in-person networks (48% of the total population). In the data collected, we found that the number of users aged 18–25 who had experienced virtual travel was higher than those in other age groups. This is consistent with Li and Chen's [58] findings. Therefore, we believe that the research sample obtained is highly credible, reliable, and reproducible in terms of feedback effects. Our research team was trained to strictly follow the test procedures to ensure data quality. The participants were also asked to follow the test procedures strictly.Table 1 Distribution of participants (N = 390)
Measure Category Frequency Percentage (%)
Gender Male 209 53.6
Female 181 46.4
Age < 18 108 25.1
18–25 142 33.8
26–34 79 19.2
35–45 70 17.4
> 45 17 4.4
Education Junior college 167 17.2
Bachelor 182 64.6
Master 59 15.1
Doctor 8 3.1
SPSS 25.0 was used to perform descriptive statistical analysis on the sample data to investigate the sample's demographic characteristics, as shown in Table 1. A total of 390 subjects were recruited to participate in the virtual tourism experience, of which 53.6% and 46.4% were male and female, respectively. In terms of age, users aged 18–25 account for the most significant proportion, 33.8 percent, followed by users younger than 18, accounting for 25.1 percent; in terms of education, bachelor's education is the mainstay, accounting for 64.6 percent.
Material and Procedure
Quan Jing Wang (http://www.vra.cn/#/home) has rich virtual tourism scenic spots resources and contains pictures, videos, texts, audio, and other forms. Therefore, we chose the virtual tourism of "VR Panorama of the Palace Museum" launched by Quan Jing Wang as the tourists' tour material, as shown in Fig. 2. The usability testing process was broken down into four main steps, as shown in Fig. 3. In the first step, the researcher introduced the participants to the content of this usability test and the VR equipment that they would need to use to conduct the virtual tourism. In the second step, the researcher showed the participants an instructional video on using the VR device for virtual tourism. In the third step, the researcher asked the participants if they had learned how to use the VR device for virtual tourism. If the participant answered yes, the participant moved on to the next step. If the participant answered no, the participant had to go back to the second step and continue learning until the participant learned to move on to the next step. In the fourth step, the participant begins to experience virtual tourism for 40 min. In the fifth step, the research team personnel collected feedback from all participants utilizing a questionnaire and randomly selected participants for a brief interview. In this way, the participants were urged to complete the test as required until the end of the test and start filling in the feedback questionnaire.Fig. 2 VR Panorama of the Palace Museum
Fig. 3 Tourist experience process
This research focuses on virtual tourists' psychological processes and behavioral intentions. The question scale content was developed using correlation theories and a conceptual model. The questionnaire includes two parts: basic information about the subjects and feelings about using virtual tourism. The relevant scales in this study were all fine-tuned using mature scales or based on mature scales to fit this paper's virtual tourism research context. The stimulus factors were measured using the variables of content quality (three items), system quality (four items), and interaction quality (three items) adopted from [1, 5, 9, 19, 23, 45, 51, 112]. The organism factors were measured using the variable of tourism experience (four items) and virtual attachment (five items) adopted from [43, 44, 65, 70, 76], 100, 103, 106, 109]. The response factor was measured using the variable of travel intention (three items) adopted from [105].
To ensure the accuracy and scientific validity of the questionnaire, the research team first conducted a small-scale questionnaire pre-test. The test results were also given back to five researchers in virtual tourism, information behavior, and news communication. The questionnaire questions followed the Delphi method to adjust, correct, and calibrate the questionnaire repeatedly until the error was controlled within the appropriate range.
Descriptive analysis of demographic characteristics
SPSS 25.0 was used in this study to perform descriptive statistical analysis on the sample data to investigate the sample's demographic characteristics. According to demographic characteristics, as shown in Table 1. A total of 390 subjects were recruited to participate in the virtual tourism experience in this study, of which 53.6% and 46.4% were male and female, respectively. In terms of age, users aged 18–25 account for the most significant proportion, 33.8 percent, followed by users younger than 18, accounting for 25.1 percent; in terms of education, bachelor's education is the mainstay, accounting for 64.6 percent.
Results
Measurement model analysis
Common method variance
The question of common method variance in the sample was first tested for correlation. To prevent the effect of common method variance on the sample, we hid the names of variables and measurement items in the questionnaire beforehand and randomly assigned the measurement items in the questionnaire. The Harman one-way test for common method variance was used. The results of the unrotated factor analysis showed that there were five factors with characteristic roots greater than one, explaining a total of 75.598% of the variance variation. The first factor explained 36.722% of the method variance. Therefore, there is no significant common method bias problem.
Measurement model
This study performed structural equation modeling on 390 samples using "PLS-SEM." The SmartPLS 3.3.2 software settings were: "weighting scheme" using "path weighting scheme," "maximum number of iterations" = 300, and end criterion = 1*10 − 7. The Bootstrapping test for significance of each indicator: subsample = 5000, confidence interval method using Bias-Corrected and Accelerated (BCA) Bootstrap, test type using the two-tailed test, significance level = 0.05.
The reliability, convergent validity, and VIF values of the measurement models are shown in Table 2: in terms of reliability, Cronbach's α were above the critical value of 0.7 [30], from 0.868 (SQ) to 0.931 (VA),the Composite Reliability is also more significant than the critical value of 0.7 [37], from 0.910 (SQ) to 0.948 (VA). Thus, the measurement model has good reliability. In terms of validity, the factor loadings (λ) for all measurement questions were more significant than the critical value of 0.7 [37], from 0.795 (SQ4) to 0.921 (CQ3). In addition, the AVE for all variables were more significant than the critical value of 0.5 [37], from 0.718 (SQ) to 0.822 (CQ). Thus, the measurement model has good convergent validity. In terms of VIF, the Outer VIF Value was all less than the standard value of 5 [37], from 1.821 (SQ4) to 3.629 (TE3). The Inner VIF Value was also less than the standard value of 5, from 1.217 (VA) to 1.531 (SQ), so the structural model did not have serious multicollinearity problems. The square root of each variable AVE is greater than the correlation coefficient of that variable with any other variable by the discriminant validity test, and the measurement model has good discriminant validity, as shown in Table 3.Table 2 Reliability and convergent validity analysis (N = 390)
Construct Items Factor Loading Cronbach's α CR AVE Outer VIF Value Inner VIF Value
Content quality (CQ) CQ1 0.917 0.892 0.933 0.822 2.856 1.321
CQ2 0.882 2.338
CQ3 0.921 2.893
System quality (SQ) SQ1 0.850 0.868 0.910 0.718 2.291 1.531
SQ2 0.840 2.129
SQ3 0.901 2.849
SQ4 0.795 1.821
Interaction quality (IQ) IQ1 0.899 0.870 0.920 0.794 2.384
IQ2 0.881 2.180
IQ3 0.894 2.364
Tourism experience (TE) TE1 0.870 0.918 0.942 0.804 2.475 1.262
TE2 0.880 2.760
TE3 0.919 3.629
TE4 0.916 3.540
Virtual attachment (VA) VA1 0.906 0.931 0.948 0.784 3.495 1.217
VA2 0.869 2.765
VA3 0.879 2.920
VA4 0.874 2.836
VA5 0.900 3.521
Travel intention (TI) TI1 0.912 0.883 0.928 0.810 2.766 /
TI2 0.889 2.239
TI3 0.899 2.595
Table 3 Discriminant validity analysis (N = 390)
CQ IQ SQ TE TI VA
CQ 0.907
IQ 0.346 0.891
SQ 0.409 0.458 0.847
TE 0.306 0.303 0.410 0.896
TI 0.326 0.352 0.493 0.678 0.900
VA 0.310 0.345 0.303 0.102 0.351 0.885
The diagonal elements (in bold) are the square root of variance shared between the AVEs, whereas the off-diagonal elements are correlations among constructs
Structural model analysis
Path analysis
In terms of path analysis, as shown in Table 4 and Fig. 4, CQ had a significant positive effect on TE with a path coefficient of 0.144 (p < 0.05); SQ had a significant positive effect on TE with a path coefficient of 0.298 (p < 0.001); IQ had a significant positive effect on TE with a path coefficient of 0.116 (p < 0.05); CQ had a significant positive effect on VA with a path coefficient of 0.195 (p < 0.05); SQ has a significant positive effect on VA with a path coefficient of 0.153 (p < 0.05); IQ has a significant positive effect on VA with a path coefficient of 0.235 (p < 0.001); TE has a significant positive effect on TI with a path coefficient of 0.577 (p < 0.001); VA has a significant positive effect on TI with a path coefficient of 0.577 (p < 0.001); SQ has a significant positive effect on TI with a path coefficient of 0.180 (p < 0.01). Therefore, H1-H8 were established. On the other hand, there is no significant effect of TE on VA; there is no significant effect of CQ and IQ on TI. Therefore, H9 was not valid. Meanwhile, when performing model fitting and hypothesis testing, the researchers found that system quality also had a positive effect on travel intention to some extent. So, this study added hypothesis H16: System quality positively affects travel intention.Table 4 Path coefficients and research hypothesis testing results
Path coefficients Mean T-value P-value Hypothesis testing results
CQ → TE 0.144 0.143 2.493 0.013 H1 establish
SQ → TE 0.298 0.299 5.053 0.000 H2 establish
IQ → TE 0.116 0.117 2.164 0.031 H3 establish
CQ → VA 0.195 0.194 3.477 0.001 H4 establish
SQ → VA 0.153 0.153 2.524 0.012 H5 establish
IQ → VA 0.235 0.235 4.193 0.000 H6 establish
TE → TI 0.577 0.579 10.737 0.000 H7 establish
VA → TI 0.233 0.231 4.339 0.000 H8 establish
TE → VA -0.092 -0.092 1.850 0.064 H9 not valid
CQ → TI -0.002 -0.002 0.041 0.967 /
SQ → TI 0.180 0.180 3.285 0.001 H16 establish
IQ → TI 0.015 0.015 0.345 0.730 /
***p < 0.001, **p < 0.01, *p < 0.05
Fig. 4 PLS results of the structural model. Note The dotted line indicates that the relationship is not valid. **p < 0.001; **p < 0.01; *p < 0.05
Mediating effect analysis
For the mediating effects, TE and VA acted as mediating variables in the relationship between the effects of CQ on TI, with indirect effects of 0.083 (p < 0.05) and 0.045 (p < 0.01) and total effects of 0.081 (p < 0.001) and 0.043 (p < 0.001), respectively; TE and VA acted as mediating variables in the relationship between the effects of SQ on TI, with indirect effects of 0.172 (p < 0.001) and 0.036 (p < 0.05) and total effects of 0.352 (p < 0.001) and 0.216 (p < 0.05), respectively. indirect effects of 0.172 (p < 0.001) and 0.036 (p < 0.05), respectively, and total effects of 0.352 (p < 0.001) and 0.216 (p < 0.001), respectively; TE and VA acted as mediating variables in the relationship between the effects of IQ on TI, with indirect effects of 0.067 (p < 0.05) and 0.055 (p < 0.01), with total effects of 0.082 (p < 0.001) and 0.070 (p < 0.001), respectively. Since all the above indirect effects were positive, the mediating effects were all Complementary mediation [37], and H10–H15 were established, as shown in Table 5.Table 5 Mediating effect
Indirect effect value Indirect effect P-value Direct effect value Direct effect P-value Total effect value Total effect P-value Hypothesis testing results
CQ → TE → TI 0.083 0.014 -0.002 0.967 0.081 0.000 H10 establish
CQ → VA → TI 0.045 0.006 -0.002 0.967 0.043 0.000 H11 establish
SQ → TE → TI 0.172 0.000 0.180 0.001 0.352 0.000 H12 establish
SQ → VA → TI 0.036 0.025 0.180 0.001 0.216 0.000 H13 establish
IQ → TE → TI 0.067 0.027 0.015 0.730 0.082 0.000 H14 establish
IQ → VA → TI 0.055 0.006 0.015 0.730 0.070 0.000 H15 establish
Predictive power assessment
The predictive power assessment showed R2 = 0.201 for TE, R2 = 0.566 for TI, and R2 = 0.178 for VA. Q2 was calculated based on Blindfolding, and Q2 > 0 indicates that the structural model has predictive relevance for the endogenous variables and vice versa [37]. The calculated Q2 = 0.158 for TE, Q2 = 0.450 for TI, and Q2 = 0.135 for VA, all of which are greater than 0. This indicates that the structural model has predictive relevance for TE, TI, and VA.
Discussion and conclusions
Key findings
This study adopts a psychologically-derived SOR theoretical framework to explore the impact of virtual tourism on tourists' travel intention in the field. The findings and implications are described below.
RQ1 was clearly answered (Does virtual tourism experience increase users' travel intention in the field?). This research revealed a significant positive relationship between content quality, system quality, and interaction quality of virtual tourism on tourists' tourism experience, virtual attachment, and travel intention through model testing and mediating effect testing, which is consistent with the findings of previous studies [33]. During home quarantine, the recurrence of the pandemic and the disclosure of various negative information led to a surge in stress and anxiety [108]. Many people were compelled to cancel their travel plans. With the intrinsic motivation to travel [22] and the negative information about the pandemic, people's psychological need to escape from the real world and be free from bondage is triggered to a great extent. The magnificent scenery of virtual tourism attractions provides people with the opportunity to have a short escape from reality. It brings a visual feast of beauty, thus facilitating the creation of virtual attachment. In addition, the quality of content, system quality, and interaction of virtual tourism [24] enable tourists to obscure the boundaries between the virtual and physical realities [6, 24], which generates a higher degree of authenticity in the tourism experience [46]. In the virtual environment created by virtual tourism, tourists develop virtual attachments and experience unique tourism experiences distinct from natural tourism. These factors ultimately impact tourists' travel intention in the field. As a new type of tourism, virtual tourism has novel and unique characteristics that attract tourists and inspire their curiosity. As a result, virtual tourism strengthens tourists' intention to run to the destinations promoted in virtual tourism by providing immersive tourism experiences and realistic virtual attachments.
The following findings can be reported for RQ2 (What factors change tourists' travel intention in the field of virtual tourism?). Tourism experience and virtual attachment are critical variables that influence tourists' travel intention in the field and are crucial factors in whether virtual tourism can attract more tourists. This empirical analysis revealed that tourism experience and virtual attachment positively impact tourists' travel intention in the field. Tourism experience has the most significant impact on travel intention in the field, followed by virtual attachment, which is mainly relevant to the form and nature of virtual tourism. As a new form of tourism, the ultimate goal of virtual tourism is to provide tourists with an ideal tourism experience. Thus, virtual tourism will be designed to consider tourists' needs to meet the tourism experience. Tourists gain their desired experience in virtual tourism; therefore, the tourism experience has the most significant impact on the travel intention in the field. In addition, tourists may develop an emotional attachment to the actual tourist place through the interaction between the virtual space and the 3D reconstructed scenery, thus creating an emotional connection with the place they have not visited, which further influences the travel intention in the field.
The following findings can be reported for RQ3 (How do influencing factors of virtual tourism experience <content quality, system quality, interaction quality>, tourism experience, virtual attachment, and travel intention interact with each other?). System quality may directly and significantly affect travel intention positively. Tourism experience and virtual attachment play a complementary mediating role in influencing relationships of content quality, system quality, and interaction quality. As shown by the relative effect values of the mediating effects, the indirect effects are significant in proportion. Tourists' knowledge of the scenery presented in virtual tourism drives their travel intention while granting them real tourism experience and virtual attachment during the tour, further motivating their travel intention. In addition, there was no significant positive impact of tourism experience on virtual attachment. The results of this study contradict the findings of [36, 89]. The emergence of virtual attachment requires a deeper connection between the tourist and the destination. The positive perceptions and memorability developed from a single or a few tourism experiences do not allow for the emotional connection and psychological identification with the destination that tourists accumulate to give rise to the virtual attachment. Therefore, tourism experiences and virtual attachment do not have a direct correlation.
Theoretical implications
This study's theoretical contributions are categorized into three parts.
First, we studied the three contributing dimensions of virtual tourism experience and travel motivation under the immersive "S-O-R" research design. Namely, content quality, system quality, and interaction quality [34, 48, 74]. Further, we expanded their meaning under the scope of VR tourism. Content quality refers to the intrinsic values, i.e., objectivity, credibility, and the amount of information the virtual tourism provides. System quality refers to the quality of the tourist's use of virtual tourism and its design in terms of structure, presentation, and connectivity. Interaction quality is the ability of virtual tourism to provide interactive features to tourists. With these delimitations, this study constructs a relational model which suggests a positive relationship between those qualities and tourism intention through the mediation of virtual content characteristics. Together, the results should be carefully considered in building virtual tourism interactions, providing information to tourists, and understanding tourists' preferences.
Second, prior Studies have conducted empirical tourism research for VR only in a limited number of areas: first, web-based visual images, such as 360°panoramic images [8, 39, 62]. Second, virtual environments are based on fictional worlds in VR games [21, 32]. Third, virtual environments are displayed on a two-dimensional computer screen [18, 95]. In addition, few tourism studies have used VR devices to conduct usability tests on live-action virtual tourism products. By focusing on the sensory stimuli evoked by VR technology-based virtual tourism for tourists, this work examines the psychological processes of tourists' travel experiences and reactions to virtual tourism products. Since VR technology is constantly evolving, this study will be a good reference paradigm for future usability studies of virtual tourism products based on VR technology.
Third, scholars' prior studies on travel intention have focused on influencing factors of travel intention [86, 102], relationship studies [66, 77], and place attachment [54, 94]. Researchers' work on travel intention was first extended from social behavior, where it was generally accepted that "travel" is also social behavior. The research methods used by scholars for research in this area are mainly quantitative methods, such as factor analysis and constructive modeling. In terms of research content, scholars have expanded from the factors influencing its formation to the study of relationships. Later scholars have focused more on the emotional expression of individual tourists and explored the relationship between tourism intention and place attachment. This study builds on the prior studies and further explores a research model that is more consistent with Chinese tourists' travel intention to virtual tourism destinations in the post-COVID-19 era.
Finally, this study identified virtual attachment as an antecedent factor influencing tourists' travel intention in the field, rarely seen in virtual tourism research. The study revealed that virtual attachment is a prominent variable influencing tourists' travel intention in the field when users are experiencing virtual tourism. Virtual attachment differs from other variables because it incorporates more of the tourists' affective experiences. Virtual tourism is a new way of tourism; tourists' emotions are evident in their subsequent behavior of tourists. Tourists develop emotional attachments to real tourist places through interactions with three-dimensional reconstructions of scenic areas in virtual space, thus creating emotional connections to places they have not visited and further influencing their travel intention in the field. The findings empirically develop the ternary space theory of tourist places, i.e., the traditional two-dimensional perspective of the tourist-place relationship is extended to a three-dimensional ternary space, further illustrating the interrelated, interactive, and dynamic response characteristics of the ternary area [17]. Since the individual variability of tourist groups, the influence of virtual attachment on tourists' travel intention in the field cannot be neglected.
Practical implications
First, this study discovered that tourism experience positively impacts travel intention (p = 0.001). The path coefficient of the variable of the tourism experience is 0.577, which is greater than other variables. This indicates the magnitude of tourists' perceptions of their experiences during virtual tours on their subsequent travel intentions in the field. Virtual tourism changes the way tourists visit a scenic area by providing them with more opportunities to interact, contextualize, and share with others. This study finds that content quality, system quality, and interaction quality can affect tourists' tour experience. Hence, three perspectives of content, system, and interaction can be adopted to enhance their experience.
From a system perspective, the system quality of virtual tourism impacts the tourist experience, virtual attachment, and travel intention. Hence, the design of the presentation form as the carrier of the system quality is essential. The current way of virtual tourism, in addition to the traditional AR, VR, as a powerful means of augmented reality—holographic projection technology also has a vast space for development. This technology is not unattainable, and holographic projection technology has been used in many scenarios, such as virtual idol concerts. However, holographic projection technology has rarely been used in the tourism industry. Future virtual tourism technology researchers can make this technology universal so that tourists' homes become distant scenic spots.
From an interactivity perspective, the interactivity quality affects tourists' tourism experience and virtual attachment, which affects their travel intention. Currently, virtual tourism has a lot of room for development in the interactivity aspect. Granted, the rise of the Internet has brought a completely different experience to virtual tourism interactivity than in the past. Tourists can communicate and interact through real-time voice and other means. When tourists travel in the field, on the one hand, they can interact with people; on the other hand, the interaction between people and scenery, people and things are also essential for tourists when traveling in the field. Photographing is usually one of how people interact with the scenery. Developers of virtual tourism can use cinematic stunts to allow users to pose for photos with scenes in virtual scenes. Techniques such as AR photography and photoshop (PS) can also increase tourists' engagement. In terms of human–object interaction, virtual tourism does not allow tourists to taste the specialties of the tourist destination as much as onsite tourism. Therefore, virtual tourism developers can incorporate the function of take-out recommendation in the course of the virtual tourism scenario setting. Tourists can order a local specialty dish at home, for example, Hangzhou dish, during virtual tourism in Hangzhou. In addition, developers can also add restaurant recommendations. For example, the system can recommend local Sichuan cuisine restaurants in Sichuan virtual tourism. Users can go straight to the restaurants in the tourist destination after virtual tourism to satisfy the tourists' desire for food, which is one of the most effective ways to increase tourists' travel intention in the field. Finally, future virtual tourism developers can optimize virtual tourism from other perspectives, such as growing tourists' olfactory or tactile experience.
From a content perspective, content quality affects tourists' tourism experience and virtual attachment, affecting their travel intention. At the beginning of the pandemic, virtual tourism took up a large amount of demand for outbound travel. While domestic travel in China has thawed after the pandemic has been alleviated, outbound travel remains ice-bound. Content quality has become almost the only growth area in the post-pandemic era that can increase tourists' travel intentions onsite. The quality of content for virtual tourism requires inspiring tourists and providing practical travel information, and generating real value for tourist attractions and local government departments for tourism. The pandemic has further deepened the importance of tourism content quality. From the perspective of tourists' needs, in the post-pandemic era, faced with new information gaps, tourists increasingly need tourism information related to COVID-19 precautionary measures, such as the need for Polymerase Chain Reaction (PCR) test reports and unique business hours schedules for attractions. This information needs to be updated by virtual tourism operators in the first instance in the virtual tourism scenario. Virtual tourisms need to provide tourists with the fastest and most accurate information related to the pandemic.
Finally, virtual attachment directly and positively affects the travel intention, highlighting the importance of constructing a person–place emotional identity in virtual tourism [46]. When tourists develop heartfelt and emotional attachments (e.g., desire for spiritual purification and pilgrimage experiences [11]) to tourist places during virtual tours, and these needs are challenging to satisfy through the mirroring experience of virtual tourism, travel intention in the field are significantly enhanced in the future. In this regard, the virtual tourism design process should focus on the realistic simulation of beautiful scenery and stimulating tourists' spiritual resonance and emotional identification with the tourist destination, thus enhancing the effectiveness of virtual tourism as a marketing method. In addition, virtual tourism must not develop independently due to the inseparable relationship between online and offline. Specifically, virtual tourism will not wholly replace offline tourism because offline tourism is the root of virtual tourism. Without the development of offline tourism, virtual tourism will also lose the creative material. Although virtual tourism is not the pure auxiliary tool it used to be, an essential role of virtual tourism in the post-pandemic era is still to provide a reference for people to travel offline. After all, there is no substitute for offline travel that brings tourists a natural feeling.
Limitations and future research
There are some shortcomings in this research on the effect of virtual tourism on tourists' travel intention on site. First, the quantitative dimensions of virtual tourism can be further enriched and deepened. The quality of virtual tourism in this study was mainly manifested in three aspects: interactivity quality, content quality, and system quality. With the development of technology however, new quantitative dimensions of virtual tourism display are bound to emerge, so in future research, new quantitative dimensions of virtual tourism can be explored through methods such as grounded theory and big data mining. Second, since the data collection of this study was completed during the pandemic, the virtual attachment of virtual tourists may be affected by the pandemic. When there is a risk of infection during the trip, tourists in a housebound state may generate higher attachment levels than during the unusual period. In other words, attachment to the virtual tourism experience may be weakened when tourists are free to go out and experience the beauty of the travel destination firsthand. Therefore, the transformational relationship between tourists' virtual attachment and travel intention in the field when going out is unrestricted and can be further investigated in the future. Third, the data collected in this study come from virtual tourists in China. The high collectivism and high-power distance, and relational culture characteristics of China [117] may also impact interpersonal relationships and social interactions, which may be very different from the perceptions of virtual tourism by tourists in other countries and cultures. Therefore, considering the impact of cultural differences, cross-cultural comparisons can be enhanced in the context of future work, to fully understand the effects and impact mechanisms of virtual tourism on tourists from different cultural backgrounds.
Declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Aladwani AM Palvia PC Developing and validating an instrument for measuring user-perceived web quality Inf. Manag. 2002 39 6 467 476 10.1016/s0378-7206(01)00113-6
2. Ali F Terrah A Wu C Ali L Wu H Antecedents and consequences of user engagement in smartphone travel apps J. Hosp. Tour. Technol. 2021 12 2 355 371 10.1108/jhtt-09-2020-0221
3. Altena E Baglioni C Espie CA Ellis J Gavriloff D Holzinger B Riemann D Dealing with sleep problems during home confinement due to the COVID-19 outbreak: practical recommendations from a task force of the European CBT-I academy J. Sleep Res. 2020 29 4 e13052 10.1111/jsr.13052 32246787
4. Alyahya M McLean G Examining tourism consumers’ attitudes and the role of sensory information in virtual reality experiences of a tourist destination J. Travel Res. 2021 10.1177/00472875211037745
5. Babin LA Burns AC A modified scale for the measurement of communication-evoked mental imagery Psychol. Mark. 1998 15 3 261 278 10.1002/(sici)1520-6793(199805)15:3<261::Aid-mar4>3.0.Co;2-8
6. Biocca F Communication within virtual reality: creating a space for research J. Commun. 1992 42 4 5 22 10.1111/j.1460-2466.1992.tb00810.x
7. Bowlby J The making and breaking of affectional bonds. I. Aetiology and psychopathology in the light of attachment theory. An expanded version of the Fiftieth Maudsley Lecture, delivered before the Royal College of Psychiatrists, 19 November 1976 Br. J. Psychiatry 1977 130 201 210 10.1192/bjp.130.3.201 843768
8. Caciora T Herman GV Ilieș A Baias Ș Ilieș DC Josan I Hodor N The use of virtual reality to promote sustainable tourism: a case study of wooden churches historical monuments from Romania Remote Sens. 2021 10.3390/rs13091758
9. Chae M Kim J Kim H Ryu H Information quality for mobile internet services: a theoretical model with empirical validation Electron. Mark. 2002 12 1 38 46 10.1080/101967802753433254
10. Chan KW Li SY Understanding consumer-to-consumer interactions in virtual communities: the salience of reciprocity J. Bus. Res. 2010 63 9–10 1033 1040 10.1016/j.jbusres.2008.08.009
11. Chang AY-P Li M Vincent T Development and validation of an experience scale for pilgrimage tourists J. Destin. Mark. Manag. 2020 10.1016/j.jdmm.2019.100400
12. Chang HH Virtual reality, youtube, or social media? Assessing promotional effects on tourism destination J. Vacat. Mark. 2021 10.1177/13567667211038960
13. Chang Y-L Hou H-T Pan C-Y Sung Y-T Chang K-E Apply an augmented reality in a mobile guidance to increase sense of place for heritage places J. Educ. Technol. Soc. 2015 18 2 166 178
14. Chen C-C Yao J-Y What drives impulse buying behaviors in a mobile auction? The perspective of the stimulus-organism-response model Telemat. Inform. 2018 35 5 1249 1262 10.1016/j.tele.2018.02.007
15. Chen G So KKF Hu X Poomchaisuwan M Travel for affection: a stimulus-organism-response model of honeymoon tourism experiences J Hosp. Tour. Res. 2021 10.1177/10963480211011720
16. Chen S Zhang Z Yang J Wang J Zhai X Bärnighausen T Wang C Fangcang shelter hospitals: a novel concept for responding to public health emergencies Lancet 2020 395 10232 1305 1314 10.1016/s0140-6736(20)30744-3 32247320
17. Chen, Z., Zhenfang, H.: Constructing the theoretical model of "tri-space" interaction in the tourism context. Geographical Research. 39(2), 232–242 (2020)
18. Chiquet S Martarelli CS Mast FW Eye movements to absent objects during mental imagery and visual memory in immersive virtual reality Virtual Real. 2020 25 3 655 667 10.1007/s10055-020-00478-y
19. Chu K-M Yuan J The effects of perceived interactivity on e-trust and e-consumer behaviors: the application of fuzzy linguistic scale J. Electron. Commer. Res. 2013 14 1 124
20. Chung N Han H Joun Y Tourists’ intention to visit a destination: The role of augmented reality (AR) application for a heritage site Comput. Hum. Behav. 2015 50 588 599 10.1016/j.chb.2015.02.068
21. Coghlan, A., Carter, L.: New product design: creating a digital VR game to promote the conservation of nature-based tourism attractions. In: The Emerald Handbook of ICT in Tourism and Hospitality (2020)
22. Crompton JL Motivations for pleasure vacation Ann. Tour. Res. 1979 6 4 408 424 10.1016/0160-7383(79)90004-5
23. DeLone WH McLean ER Information systems success: the quest for the dependent variable Inf. Syst. Res. 1992 3 1 60 95 10.1287/isre.3.1.60
24. Deng X Unnava HR Lee H “Too true to be good?” When virtual reality decreases interest in actual reality J. Bus. Res. 2019 100 561 570 10.1016/j.jbusres.2018.11.008
25. Donthu N Gustafsson A Effects of COVID-19 on business and research J. Bus. Res 2020 117 284 289 10.1016/j.jbusres.2020.06.008 32536736
26. Dumas, J. S., Dumas, J. S., Redish, J.: A practical guide to usability testing. In: Intellect Books. Published in Great Britain (1999)
27. Ekanayake IA Gnanapala AC Travel experiences and behavioural intentions of the tourists: a study on eastern province of Sri Lanka Tour. Leis. Glob. Chang. 2016 3 50 61
28. El-Said O Aziz H Virtual tours a means to an end: an analysis of virtual tours’ role in tourism recovery post COVID-19 J. Travel Res. 2021 61 3 528 548 10.1177/0047287521997567
29. Font X English R Gkritzali A Tian W Value co-creation in sustainable tourism: a service-dominant logic approach Tour. Manag. 2021 10.1016/j.tourman.2020.104200
30. Fornell C Larcker DF Evaluating structural equation models with unobservable variables and measurement error J. Mark. Res. 2018 18 1 39 50 10.1177/002224378101800104
31. Frost RO Hartl TL Christian R Williams N The value of possessions in compulsive hoarding: patterns of use and attachment Behav. Res. Ther. 1995 33 8 897 902 10.1016/0005-7967(95)00043-w 7487849
32. Gao BW Zhu C Song H Dempsey IMB Interpreting the perceptions of authenticity in virtual reality tourism through postmodernist approach Inf. Technol. Tour. 2022 24 1 31 55 10.1007/s40558-022-00221-0
33. Ghose A Huang K-W Personalized pricing and quality customization J. Econ. Manag. Strategy 2009 18 4 1095 1135 10.1111/j.1530-9134.2009.00239.x
34. Gorla N Somers TM Wong B Organizational impact of system quality, information quality, and service quality J. Strateg. Inf. Syst. 2010 19 3 207 228 10.1016/j.jsis.2010.05.001
35. Gross MJ Brown G An empirical structural model of tourists and places: progressing involvement and place attachment into tourism Tour. Manag. 2008 29 6 1141 1151 10.1016/j.tourman.2008.02.009
36. Guest AM Lee BA Sentiment and evaluation as ecological variables Sociol. Perspect. 2016 26 2 159 184 10.2307/1389089
37. Hair Jr, J. F., Hult, G. T. M., Ringle, C. M., Sarstedt, M.: A primer on partial least squares structural equation modeling (PLS-SEM). Sage publications (2021)
38. Hidalgo MC HernÁNdez B Place attachment: conceptual and empirical questions J. Environ. Psychol. 2001 21 3 273 281 10.1006/jevp.2001.0221
39. Hu Y Sun W Liu X Gan Q Shi J Tourism demonstration system for large-scale museums based on 3D virtual simulation technology Electron. Libr. 2020 38 2 367 381 10.1108/el-08-2019-0185
40. Hudson S Matson-Barkat S Pallamin N Jegou G With or without you? Interaction and immersion in a virtual reality experience J. Bus. Res. 2019 100 459 468 10.1016/j.jbusres.2018.10.062
41. Hwang S-N Lee C Chen H-J The relationship among tourists’ involvement, place attachment and interpretation satisfaction in Taiwan’s national parks Tour. Manag. 2005 26 2 143 156 10.1016/j.tourman.2003.11.006
42. Io M-U Wan PYK Relationships between tourism experiences and place attachment in the context of casino resorts J. Qual. Assur. Hosp. Tour. 2017 19 1 45 65 10.1080/1528008x.2017.1314801
43. Jung T Chung N Leue MC The determinants of recommendations to use augmented reality technologies: the case of a Korean theme park Tour. Manag. 2015 49 75 86 10.1016/j.tourman.2015.02.013
44. Jung, T., tom Dieck, M. C., Lee, H., Chung, N.: Effects of virtual reality and augmented reality on visitor experiences in museum. In: Information and Communication Technologies in Tourism 2016, pp. 621–635. Springer (2016)
45. Kim M-K Park M-C Jeong D-H The effects of customer satisfaction and switching barrier on customer loyalty in Korean mobile telecommunication services Telecommun. Policy 2004 28 2 145 159 10.1016/j.telpol.2003.12.003
46. Kim MJ Lee C-K Jung T Exploring consumer behavior in virtual reality tourism using an extended stimulus-organism-response model J. Travel Res. 2018 59 1 69 89 10.1177/0047287518818915
47. Kim MJ Lee C-K Preis MW The impact of innovation and gratification on authentic experience, subjective well-being, and behavioral intention in tourism virtual reality: the moderating role of technology readiness Telemat. Inform. 2020 10.1016/j.tele.2020.101349
48. Korfiatis N García-Bariocanal E Sánchez-Alonso S Evaluating content quality and helpfulness of online product reviews: the interplay of review helpfulness vs. review content Electron. Commer. Res. Appl. 2012 11 3 205 217 10.1016/j.elerap.2011.10.003
49. Krahmer Thinking about thinking aloud: a comparison of twoverbal protocols for usability testing IEEE Trans. Prof. Commun. 2004 47 2 105 117 10.1109/tpc.2004.828205
50. Kunz W Seshadri S From virtual travelers to real friends: relationship-building insights from an online travel community J. Bus. Res. 2015 68 9 1822 1828 10.1016/j.jbusres.2015.01.009
51. Kuo Y-F A study on service quality of virtual community websites Total Qual. Manag. Bus. Excell. 2010 14 4 461 473 10.1080/1478336032000047237a
52. Kuo Y-F Wu C-M Deng W-J The relationships among service quality, perceived value, customer satisfaction, and post-purchase intention in mobile value-added services Comput. Hum. Behav. 2009 25 4 887 896 10.1016/j.chb.2009.03.003
53. Kyle G Graefe A Manning R Bacon J An examination of the relationship between leisure activity involvement and place attachment among hikers along the Appalachian Trail J. Leis. Res. 2017 35 3 249 273 10.1080/00222216.2003.11949993
54. Lalicic L Garaus M Tourism-induced place change: the role of place attachment, emotions, and tourism concern in predicting supportive or oppositional behavioral responses J. Travel Res. 2020 61 1 202 213 10.1177/0047287520967753
55. Lee J Kyle G Scott D The mediating effect of place attachment on the relationship between festival satisfaction and loyalty to the festival hosting destination J. Travel Res. 2012 51 6 754 767 10.1177/0047287512437859
56. Lee S Park D-H Han I New members’ online socialization in online communities: the effects of content quality and feedback on new members’ content-sharing intentions Comput. Hum. Behav. 2014 30 344 354 10.1016/j.chb.2013.09.015
57. Leung XY Lyu J Bai B A fad or the future? Examining the effectiveness of virtual reality advertising in the hotel industry Int. J. Hosp. Manag. 2020 10.1016/j.ijhm.2019.102391
58. Li T Chen Y Will virtual reality be a double-edged sword? Exploring the moderation effects of the expected enjoyment of a destination on travel intention J. Destin. Mark. Manag. 2019 12 15 26 10.1016/j.jdmm.2019.02.003
59. Lin L-P Huang S-C Ho Y-C Could virtual reality effectively market slow travel in a heritage destination? Tour. Manag. 2020 10.1016/j.tourman.2019.104027
60. Lo WH Cheng KLB Does virtual reality attract visitors? The mediating effect of presence on consumer response in virtual reality tourism advertising Inform. Technol. Tour. 2020 22 4 537 562 10.1007/s40558-020-00190-2
61. Loureiro SMC Guerreiro J Ali F 20 years of research on virtual reality and augmented reality in tourism context: a text-mining approach Tour. Manag. 2020 10.1016/j.tourman.2019.104028
62. Lu J Xiao X Xu Z Wang C Zhang M Zhou Y The potential of virtual tourism in the recovery of tourism industry during the COVID-19 pandemic Curr. Issue Tour. 2021 25 3 441 457 10.1080/13683500.2021.1959526
63. Mehrabian, A., Russell, J. A.: An Approach to Environmental Psychology. The MIT Press (1974)
64. Min Z Jie Z Xiao X Mengyuan Q Youhai L Hui Z Meng H How destination music affects tourists’ behaviors: travel with music in Lijiang, China Asia Pac. J. Tour. Res. 2019 25 2 131 144 10.1080/10941665.2019.1683046
65. Mintz R Litvak S Yair Y 3D-virtual reality in science education: an implication for astronomy teaching J. Comput. Math. Sci. Teach. 2001 20 3 293 305
66. Moon H Han H Destination attributes influencing Chinese travelers' perceptions of experience quality and intentions for island tourism: a case of Jeju Island Tour. Manag. Perspect. 2018 28 71 82 10.1016/j.tmp.2018.08.002
67. Moon H Han H Tourist experience quality and loyalty to an island destination: the moderating impact of destination image J. Travel Tour. Mark. 2018 36 1 43 59 10.1080/10548408.2018.1494083
68. Morrison M Gan S Dubelaar C Oppewal H In-store music and aroma influences on shopper behavior and satisfaction J. Bus. Res. 2011 64 6 558 564 10.1016/j.jbusres.2010.06.006
69. Mullet GM Karson MJ Analysis of purchase intent scales weighted by probability of actual purchase J. Mark. Res. 2018 22 1 93 96 10.1177/002224378502200110
70. Oh H Fiore AM Jeoung M Measuring experience economy concepts: tourism applications J. Travel Res. 2016 46 2 119 132 10.1177/0047287507304039
71. Oleksy T Wnuk A Catch them all and increase your place attachment! The role of location-based augmented reality games in changing people-place relations Comput. Hum. Behav. 2017 76 3 8 10.1016/j.chb.2017.06.008
72. Othman MK Nogoibaeva A Leong LS Barawi MH Usability evaluation of a virtual reality smartphone app for a living museum Univ. Access Inf. Soc. 2021 10.1007/s10209-021-00820-4
73. Packer J Ballantyne R Conceptualizing the visitor experience: a review of literature and development of a multifaceted model Visit. Stud. 2016 19 2 128 143 10.1080/10645578.2016.1144023
74. Pelau C Dabija D-C Ene I What makes an AI device human-like? The role of interaction quality, empathy and perceived psychological anthropomorphic characteristics in the acceptance of artificial intelligence in the service industry Comput. Hum. Behav. 2021 10.1016/j.chb.2021.106855
75. Pelled LH Hill KD Employee work values and organizational attachment in North Mexican maquiladoras Int. J. Hum. Resour. Manag. 1997 8 4 495 505 10.1080/095851997341577
76. Pine, B. J., Pine, J., Gilmore, J. H: The Experience Economy: Work is Theatre & Every Business a Stage. Harvard Business Press (1999)
77. Piramanayagam S Rathore S Seal PP Destination image, visitor experience, and behavioural intention at heritage centre Anatolia 2020 31 2 211 228 10.1080/13032917.2020.1747234
78. Pratidina, T. M., Setyohadi, D. B.: Usability experience on tourism website using the use questionnaire approach. In: 2021 international conference on electrical engineering and informatics (ICEEI) (2021)
79. Qarnain SS Muthuvel S Bathrinath S Review on government action plans to reduce energy consumption in buildings amid COVID-19 pandemic outbreak Mater. Today Proc. 2020 10.1016/j.matpr.2020.04.723
80. Qiu Q Zuo Y Zhang M Can live streaming save the tourism industry from a pandemic? A study of social media ISPRS Int. J. Geo-Inf. 2021 10.3390/ijgi10090595
81. Rakić T Chambers D Rethinking the consumption of places Ann. Tour. Res. 2012 39 3 1612 1633 10.1016/j.annals.2011.12.003
82. Rather RA Customer experience and engagement in tourism destinations: the experiential marketing perspective J. Travel Tour. Mark. 2019 37 1 15 32 10.1080/10548408.2019.1686101
83. Yuqing R., Maxwell Harper, F., Sara, D., Loren, T., Sara K., John R., Robert EK.: Building member attachment in online communities: applying theories of group identity and interpersonal bonds. MIS Q (2011). 10.2307/41703483
84. Rollero C De Piccoli N Place attachment, identification and environment perception: an empirical study J. Environ. Psychol. 2010 30 2 198 205 10.1016/j.jenvp.2009.12.003
85. Sherman E Mathur A Smith RB Store environment and consumer purchase behavior: mediating role of consumer emotions Psychol. Mark. 1997 14 4 361 378 10.1002/(sici)1520-6793(199707)14:4<361::Aid-mar4>3.0.Co;2-7
86. Shin H Nicolau JL Kang J Sharma A Lee H Travel decision determinants during and after COVID-19: the role of tourist trust, travel constraints, and attitudinal factors Tour. Manag. 2022 88 104428 10.1016/j.tourman.2021.104428 34539019
87. Song H Kim J Nguyen TPH Lee KM Park N Virtual reality advertising with brand experiences: the effects of media devices, virtual representation of the self, and self-presence Int. J. Advert. 2020 40 7 1096 1114 10.1080/02650487.2020.1834210
88. Song S Yao X Wen N What motivates Chinese consumers to avoid information about the COVID-19 pandemic? The perspective of the stimulus-organism-response model Inf. Process. Manag. 2021 58 1 102407 10.1016/j.ipm.2020.102407 33041437
89. St. John, C., Austin, D. M., Baba, Y.: The question of community attachment revisited. Sociol. Spectr. 6(4), 411–431 (1986) 10.1080/02732173.1986.9981800
90. Stylidis D Woosnam KM Tasci ADA The effect of resident-tourist interaction quality on destination image and loyalty J. Sustain. Tour. 2021 10.1080/09669582.2021.1918133
91. Su L Yang Q Swanson SR Chen NC The impact of online reviews on destination trust and travel intention: the moderating role of online review trustworthiness J. Vacat. Mark. 2021 10.1177/13567667211063207
92. Suau-Sanchez P Voltes-Dorta A Cuguero-Escofet N An early assessment of the impact of COVID-19 on air transport: just another crisis or the end of aviation as we know it? J Transp Geogr 2020 86 102749 10.1016/j.jtrangeo.2020.102749 32834670
93. Kil-Soo, S., Hongki K., Eung KS.: What if your avatar looks like you? Dual-congruity perspectives for avatar use. MIS Q. (2011) 10.2307/23042805
94. Tasci ADA Uslu A Stylidis D Woosnam KM Place-oriented or people-oriented concepts for destination loyalty: destination image and place attachment versus perceived distances and emotional solidarity J. Travel Res. 2021 61 2 430 453 10.1177/0047287520982377
95. Tian F Tsai S-B Immersive 5G virtual reality visualization display system based on big-data digital city technology Math. Probl. Eng. 2021 2021 1 9 10.1155/2021/6627631
96. Tian Z Lu M Cheng Q The relationships among mobile travel application attributes, customer engagement, and brand equity Soc. Behav. Personal. Int. J. 2021 49 7 1 10 10.2224/sbp.10353
97. Turulja L Činjarević M How customer-driven and vendor-driven information cues shape a travel app user behaviour? J. Hosp. Tour. Technol. 2021 12 4 745 761 10.1108/jhtt-02-2020-0044
98. Tussyadiah, I. P., Wang, D., Jung, T. H., tom Dieck, M. C.: Virtual reality, presence, and attitude change: empirical evidence from tourism. Tour. Manag. 66, 140–154. (2018) 10.1016/j.tourman.2017.12.003
99. Ul Islam J Rahman Z The impact of online brand community characteristics on customer engagement: an application of stimulus-organism-response paradigm Telemat. Inform. 2017 34 4 96 109 10.1016/j.tele.2017.01.004
100. van der, H.: User acceptance of hedonic information systems. MIS Q. (2004) 10.2307/25148660
101. Wang, H., Chen, C.-Y.: System quality, user satisfaction and perceived net benefits of mobile broadband services. In: Proceedings of 8th international telecommunication society Asia-Pacific regional conference Taiwan (2011)
102. Wang L-H Yeh S-S Chen K-Y Huan T-C Tourists’ travel intention: revisiting the TPB model with age and perceived risk as moderator and attitude as mediator Tour. Rev. 2022 10.1108/tr-07-2021-0334
103. Wang P-C Yu C-Y Aesthetic experience as an essential factor to trigger positive environmental consciousness Sustainability 2018 10.3390/su10041098
104. Wang X Wong YD Yuen KF Rise of 'lonely' consumers in the post-COVID-19 era: a synthesised review on psychological, commercial and social implications Int. J. Environ. Res. Public Health 2021 78 98 10.3390/ijerph18020404
105. Woodside AG Lysonski S A general model of traveler destination choice J. Travel Res. 2016 27 4 8 14 10.1177/004728758902700402
106. Wu H-C Ai C-H Cheng C-C Virtual reality experiences, attachment and experiential outcomes in tourism Tour. Rev. 2019 75 3 481 495 10.1108/tr-06-2019-0205
107. Wu, W.: Analysis of digital tourism, virtual tourism and wisdom tourism. In: The international conference on cyber security intelligence and analytics (2020)
108. Xi C Junyi L The impact of virtual tourism experience on stress and emotion under the stay-at-home restrictions due to COVID-19 epidemic Tour. Trib./Lvyou Xuekan 2020 7 35 13 23
109. Xiaozhou D Zhuanqing C Empirical research on the effect of attachment to e-loyalty: the third path towardse-loyalty J. Mark. Sci. 2015 10 4 68 84
110. Xie, C.-S., Sha, C.-L.: Discussion on the essence and characteristics of tourism. Tourism Forum. 5(2), 1–5 (2012)
111. Yadav N Verma S Chikhalkar RD eWOM, destination preference and consumer involvement: a stimulus-organism-response (SOR) lens Tour. Rev. 2021 10.1108/tr-10-2020-0506
112. Yang Z Cai S Zhou Z Zhou N Development and validation of an instrument to measure user perceived service quality of information presenting web portals Inform. Manag. 2005 42 4 575 589 10.1016/s0378-7206(04)00073-4
113. Yüksel A Tourist shopping habitat: effects on emotions, shopping value and behaviours Tour. Manag. 2007 28 1 58 69 10.1016/j.tourman.2005.07.017
114. Zhang H The influence of the ongoing COVID-19 pandemic on family violence in China J. Fam. Violence 2020 10.1007/s10896-020-00196-8
115. Zhang N Zhou ZM Su CT Zhou N How do different types of community commitment influence brand commitment? The mediation of brand attachment Cyberpsychol. Behav. Soc. Netw. 2013 16 11 836 842 10.1089/cyber.2012.0456 23768073
116. Zhang S-N Li Y-Q Ruan W-Q Liu C-H Would you enjoy virtual travel? The characteristics and causes of virtual tourists’ sentiment under the influence of the COVID-19 pandemic Tour. Manag. 2022 10.1016/j.tourman.2021.104429
117. Zhao X Huo B Flynn BB Yeung JHY The impact of power and relationship commitment on the integration between manufacturers and customers in a supply chain J. Oper. Manag. 2008 26 3 368 388 10.1016/j.jom.2007.08.002
118. Zhu L Li H Wang F-K He W Tian Z How online reviews affect purchase intention: a new model based on the stimulus-organism-response (S-O-R) framework Aslib J. Inf. Manag. 2020 72 4 463 488 10.1108/ajim-11-2019-0308
| 36466580 | PMC9707137 | NO-CC CODE | 2022-12-01 23:20:21 | no | Univers Access Inf Soc. 2022 Nov 29;:1-17 | utf-8 | Univers Access Inf Soc | 2,022 | 10.1007/s10209-022-00952-1 | oa_other |
==== Front
Behav. Soc. Iss.
Behavior and Social Issues
1064-9506
2376-6786
Springer International Publishing Cham
109
10.1007/s42822-022-00109-y
Original Paper
Rule-Governed Behavior and Climate Change: Why Climate Warnings Fail to Motivate Sufficient Action
http://orcid.org/0000-0002-7187-5336
Pietras Cynthia J. [email protected]
grid.268187.2 0000 0001 0672 1122 Western Michigan University, 1903 W. Michigan Ave, Kalamazoo, MI 49008-5439 USA
28 11 2022
2022
31 1 373417
18 10 2022
© Association for Behavior Analysis International 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Climate scientists warn of dire consequences for ecological systems and human well-being if significant steps to mitigate greenhouse gas emissions are not taken immediately. Despite these warnings, greenhouse gas concentrations continue to rise, indicating that current responses are inadequate. Climate warnings and reactions to them may be analyzed in terms of rules and rule-governed behavior. The literature on rule-governed behavior in behavior analysis has identified a variety of factors that can reduce rule following, including insufficient rule exposure, insufficient learning history and rule complexity, incomplete rules, instructed behavior not sufficiently learned, rules having weak function-altering effects, conflicting rules, lack of speaker credibility, rule plausibility and inconsistency with prior learning, and insufficient reinforcement for rule following. The present paper aims to analyze how these factors might impact responses to climate change, and possible solutions and strategies are discussed. Much of the theory and research on climate-change communication has come from outside of behavior analysis. Thus, the paper also aims to integrate findings from this literature with a behavior-analytic approach to rule control. Interpreting climate warnings and climate solutions in terms of rule-governed behavior may improve our understanding of why such rules are not more effective, and aid in the development of verbal and nonverbal strategies for changing behavior and reducing greenhouse gas emissions.
Keywords
Rule-governed behavior
Climate change
Verbal behavior
Environmental sustainability
issue-copyright-statement© Association for Behavior Analysis International 2022
==== Body
pmcHuman activities, mainly fossil-fuel use and agriculture and forestry practices, have warmed the planet by about 1 °C from pre-industrial times, and evidence shows that the planet will continue to warm unless mitigation steps are undertaken (IPCC, 2018). Global warming produces a cascade of catastrophic impacts that include dangerous high temperatures, sea level rise, flooding, drought, wildfires, poor air quality, acidification of oceans, the collapse of ecosystems, and species extinctions (Jay et al., 2018). Climate change will have major impacts on food production, water supplies, urban infrastructure, and security. Vulnerable populations will experience the greatest impacts. A 2021 Lancet Countdown on health and climate change found that the negative impacts of climate change on health are rising, and that the impacts are distributed inequitably—those in developing countries in economically disadvantaged communities experience worse health impacts (Romanello et al., 2021). Climate change disasters are also expensive; the number of billion-dollar weather and climate disaster events has increased significantly (NOAA, 2022). Climate change will have major impacts on human societies. As climate changes, vulnerable groups will be increasingly driven from places that can no longer sustain human communities (IPCC, 2018). If global warming increases to 4 °C, it is predicted that social systems will fail, and people will need to move to places on the globe that remain capable of sustaining human populations (Lynas, 2020).
There are actions that can limit the worst of these impacts. The IPCC states, “Future climate- related risks would be reduced by the upscaling and acceleration of far-reaching, multilevel and cross-sectoral climate mitigation and by both incremental and transformational adaptation” (2018, p. 5). Addressing the climate crisis therefore necessitates significant changes in human behavior. As stated in the fourth National Climate Assessment, “Future impacts and risks from climate change are directly tied to decisions made in the present, both in terms of mitigation to reduce emissions of greenhouse gases (or remove carbon dioxide from the atmosphere) and adaptation to reduce risks from today’s changed climate conditions and prepare for future impacts” (Jay et al., 2018, p. 60). People must cut carbon emissions and it must occur on a global scale. The United Nations Secretary General Antonio Guterres has stated, “Nothing less than our future and the fate of humankind depends on how we rise to the climate challenge” (Guterres, 2018, 22:22) and that “Every day we fail to act is a day that we step a little closer towards a fate that none of us wants—a fate that will resonate through generations in the damage done to humankind and life on earth” (26:57).
Climate scientists have known about the negative effects of carbon emissions for decades (Weart, 2008), as have fossil fuel companies (see Oreskes & Conway, 2010), but not enough has been done to combat the growing crisis (Thompson, 2010). A landmark step occurred in 2015, as countries from across the globe pledged to reduce emissions and keep temperature rise well below 2 °C in an international treaty known as the Paris Agreement (UNFCCC, 2015). Although the agreement was a monumental step, to date, few countries have taken adequate steps to meet their commitments. In a review of national policies to accomplish pledges, Roelfsema et al. (2020) conclude, “Unless governments increase ambition, the collective effort of current national policies significantly stays short of the objectives of the Paris Agreement” (p. 6). An acceleration of local efforts to reduce greenhouse gas emissions is also needed. The fourth US National Climate Assessment states that since 2014, “a growing number of states, cities, and businesses have pursued or expanded upon initiatives aimed at reducing greenhouse gas emission, and the scale of adaptation implementation across the country has increased. However, these efforts do not yet approach the scale needed to avoid substantial damages to the economy, environment, and human health expected over the coming decades” (Jay et al., 2018, p. 60). As Leiserowitz concluded in a 2006 study on climate-change perceptions, and which still appears largely true today, “climate change currently lacks a sense of urgency” (p. 64).
For climate scientists it is puzzling why, given the dire nature of the problem (climate change is an existential threat), climate-change warnings do not motivate greater action. Much research by psychologists, often in the domain of environmental risk perception, social and cognitive psychology, environmental psychology, and communications, has sought to answer this question (e.g., Corner, & Clarke, 2016; Gifford, 2011; Hornsey & Fielding, 2020; Moser, 2016; Markowitz, & Guckian, 2018; Stoknes, 2015; Weber & Stern, 2011; and see Center for Research on Environmental Decisions [CRED], 2009). Many factors thought to influence risk perception have been proposed, including cognitive, emotional, social, cultural, experiential, and socio-demographic factors, yet there have been few attempts to integrate the understanding of climate-change risk perception into a single framework (see van der Linden, 2017). Behavior analysts could contribute more to this work. Alavosius and colleagues (2016) argue that interpreting the influence of climate warnings in terms of the operant and associative conditioning, including relational responding, may help improve climate messaging. Moreover, behavior analysts can provide a cohesive theoretical framework for understanding the human response to climate change.
The present paper aims to apply a behavior-analytic account of rules and rule-governed behavior to responses to climate warnings to help understand reasons why such warnings fail to motivate sufficient action. While doing so, it aims to integrate findings from diverse research areas into an operant framework. Behavior analysts have a long history of analyzing verbal behavior and rules, with advancements evident in applications (e.g., Fryling et al., 2020; Greer & Ross, 2007). Furthermore, psychologists have shown that verbal stimuli can be important in altering pro-environmental behaviors (see Lehman & Geller, 2004; Osbaldiston, & Schott, 2012). Analyzing reactions to climate warnings as rule-governed behavior may help us understand the failure of climate warnings to motivate behavior, and aid the development of better behavior-change strategies, including verbal strategies (see Alavosius et al., 2016).
Indirect Contingencies and Rule-Governed Behavior
The accumulation of greenhouse-gas (GHG) emissions in the atmosphere can be understood as a macrocontingency, or the cumulative outcome of behavior of many individuals and organizations (i.e., macrobehavior) (Glenn et al., 2016), mainly fossil fuel use by people in developed countries. Part of the challenge to understanding and addressing climate change is that, as a macrocontingency, GHG concentration is not a direct outcome of the behavior of any one individual and organization. Actions of individuals have no noticeable climate impact. Another challenge to the public understanding of climate change comes from the fact that greenhouse gases are invisible (Weber & Stern, 2011), and that climate change is not a discrete, salient event. As noted by Weber (2010), because climate change is gradual and embedded in random weather fluctuations (and because remembering is fallible), “climate change is not easily detected by personal experience” (p. 333). Even when people observe climate trends or directly experience climate disasters, for those who are not climate scientists, the connection to human action seems impossible to discern.
Verbal behavior is what provides the “link” between climate-relevant actions and the aggregate product of climate change (see Glenn et al., 2016). As Weber and Stern note, “for most Americans, exposure to ‘climate change’ has been almost entirely indirect, mediated by news coverage, Internet postings, informal conversations, and documentaries and video footage of events in distant regions (such as melting glaciers in Greenland) that describe these events in relation to climate change” (2011, p. 320). Thus, the human response to climate change may be understood as a social, or more specifically, as a verbal process. Verbal responses that link human behavior to climate-change may be described as rules, and responses to them may be described as rule-governed behavior.
Rules and Rule-Governed Behavior
B.F. Skinner (1957) proposed that verbal behavior could be analyzed in terms of the roles of speakers and listeners. Speakers engage in verbal behavior as a result of reinforcement provided by listeners, who collectively are called the verbal community. Listeners (which include readers) react to verbal stimuli produced by speakers, and their reactions reinforce speaker behavior. Specific forms of verbal behavior by speakers may be effective in altering the behavior of listeners because they specify consequences for the listener for engaging in action. Skinner (1969) described these verbal responses as rules or contingency-specifying stimuli. Rules may take the form of commands, instructions, warnings, or advice. Rule following in listeners is reinforced by the verbal community, and as listeners learn stimulus relations and experience many forms of reinforcement for rule following, rule following becomes generalized and effective rule following can occur even under novel verbal stimuli.
Many benefits for speakers and listeners result from establishing a repertoire of rule following (Skinner, 1969). For example, rules can alter behavior without direct exposure to contingencies, and can establish behavior quickly without the need for prolonged shaping. Rule control may also hasten contact with reinforcement, and allow listeners to avoid punishing stimuli without the need to contact those stimuli. Rule following allows listeners to respond adaptively when consequences are too temporally remote, small, or probabilistic to directly control responding (see Malott, 1988). One potential disadvantage of rule following though, is that behavior may show rigid rule control and insensitivity to consequences (i.e., is inflexible) (e.g., Fox & Kyonka, 2017; Hackenberg & Joker, 1994; Shimoff et al., 1981).
Rules allow behavior of speakers who are socially, physically, and temporally distant to impact listeners. Rules also allow verbal behavior of speakers to impact multiple listeners. Thus, through rule control, effective behavior can spread across members of a culture (Baum, 1995; Palmer, 2012; Skinner, 1969).
Once a person has learned to respond as both a speaker and listener, a person may react as a listener to their own verbal behavior (Skinner, 1957). Speakers may learn to emit self-rules which influence their own subsequent behavior (Skinner, 1969). Self-rules may be learned from others, may be occasioned by exposure to contingencies, or occasioned by stimulus associations (derived relational responding) (Pelaez, 2013). Self-rules may be useful in problem solving (Skinner, 1969), but also can lead to insensitivity to contingencies and contribute to maladaptive behaviors (see McEnteggart, 2018).
Learning by Rules
Skinner (1969) proposed that rules alter behavior in listeners as discriminative stimuli, signaling consequences for action (p. 148). For example, a rule such as “Turning down your thermostat will save electricity and money on your electric bill” may alter behavior by signaling reinforcement for adjusting the temperature on the thermostat. Citing limitations of a discriminative-stimulus account of rules, other conceptualizations have been offered. Schlinger (1993) and Schlinger and Blakely (1987) proposed that rules may be better understood as function-altering stimuli, or verbal stimuli that change the discriminative, reinforcing, eliciting, or motivating effect of stimuli. Consider again the rule, “Turning down your thermostat will save electricity and money on your electric bill.” The rule may influence behavior by altering the stimulus function of the actual thermostat (such as by establishing the thermostat as a conditioned reinforcer), changing the stimulus function of the current temperature reading (establishing the numerical value as a discriminative stimulus for lowering the setting), and changing the stimulus function of the new temperature reading (establishing the lower numerical value as a conditioned reinforcer for the adjustment).
Schlinger and Blakely’s (1987) account provides a functional definition of rules—verbal stimuli are defined as rules if they alter the functional relationship between stimuli and behavior. In a similar vein, Palmer (2007) has argued that many verbal sequences can “condition the behavior of the listener with respect to some stimulus, complex of stimuli, condition, or state of affairs” (p. 167). The function of the structure of verbal responses, he suggests, is the modification of the behavior of listeners. Similarly, Pelaez (2013) states that the primary function of a rule is to alter the behavior of the listener in a way specified by the verbal stimuli provided by the speaker. Conditioning produced by verbal stimuli appears to occur even when the stimuli do not explicitly specify all parts of a reinforcement contingency. Thus, rules may be better described as verbal stimuli that influences listener behavior, rather than verbal stimuli having a specific structure, i.e., contingency-specifying (see discussions in Baum, 1995; Palmer, 2007; Schlinger, 1993; Törneke, 2010).
The mechanisms behind the conditioning effect proposed by Schlinger and Blakely (1987) was not detailed, but as Skinner (1957) speculated in a section of Verbal Behavior titled “Conditioning the behavior of the listener” (p. 357), conditioning by verbal stimuli appears to require a specialized history of reinforcement within a verbal community. Schlinger (2008) has argued that when verbal stimuli or rules evoke discriminative or motivated behavior then indeed no special account seems warranted, but a different analysis is needed when listeners engage in behaviors that are verbal, such as when they emit echoic and intraverbal responses, verbal conditioning can occur.
Two other theoretical approaches have offered accounts for complex rule control, Naming (e.g., Horne & Lowe, 1996) and Relational-Frame Theory (RFT; Hayes et al., 2001). These accounts share the assumption that rule control stems from a history of reinforcement within a verbal community that establishes generalized operant behavior that underlie rule control (i.e., Naming, relational responding). Naming theory assumes that verbal communities establish bi-directional speaker-listener responding, such that the establishment of one type of responding enables the other. Relational-Frame Theory assumes that verbal communities establish patterns of relating stimuli (relational responding), and that contextual cues come to evoke specific patterns of relating stimuli and transformations of the function of those stimuli (see Barnes-Holmes et al., 2004; Harte et al., 2020; Hughes & Barnes-Holmes, 2016; O'Hora & Barnes-Holmes, 2004). These learning histories allow listeners to react appropriately to even novel rules.
Consider the rule on energy use described above, “Turning down your thermostat will save electricity and money on your electric bill.” A RFT account might assume that this rule will alter behavior if there is a previously learned stimulus association (frame of coordination) between the words and objects (e.g., thermostat and actual thermostat; money and actual money), money is reinforcing, and the behavior of changing thermostats previously learned. The rule may invoke a previously learned “if–then” relational frame such that the reinforcing effect of money establishes the behavior of lowering the actual thermostat as reinforcing.
Although much remains to be understood about these verbal processes and stimulus–stimulus relations, an account of rule control that conceptualizes rules as verbal stimuli that alter behavior–environment relations in listeners in ways that depend on prior social learning can provide a framework for analyzing human responses to climate change.
Categories of Rule Following
Although a rule may alter stimulus–behavior functions, this does not necessarily imply that a listener will behave in ways that correspond to the rule (e.g., O’Hora et al., 2014). Rule following may also depend on the reinforcement experienced by the listener for rule following, which may come from various sources (Pelaez, 2013). Skinner (1984) argued that rules may be followed because of reinforcement provided by the natural contingencies for behavior. He states, “The reinforcers contingent on prior stimulation from maxims, rules, or laws are sometimes the same as those which directly shape behavior” (Skinner, 1984, p. 587). Once rule following is reinforced, it may occur in the presence of similar but novel stimuli. Skinner (1984) noted, “We tend to follow advice because previous behavior in response to similar verbal stimuli has been reinforced” (p. 587). Alternatively, he notes that rules may be followed because of consequences provided by the speaker. Such additional consequences may be needed when natural consequences are deficient, such as when outcomes are delayed or probabilistic. In those cases, additional consequences, such as engineered aversive stimuli (punishers), may be necessary.
Zettle and Hayes (1982) (and see Hayes et al., 1989), proposed three categories of rule following: pliance, tracking, and augmentals (or augmenting, see Törneke 2010). Pliance refers to rule following produced by social reinforcement (or more specifically, speaker-mediated consequences, see Kissi et al., 2017). Thus, a person may follow a rule because doing so has produced social approval, whereas failing to do so has produced disapproval. For example, if a person is a member of an environmental organization in which members state, “Using public transportation is better for the environment,” the listener may take the bus instead of driving because doing so produces praise from group members. The rule may be called a ply. Pliance is likely to come under stimulus control such that stimuli signaling enforcement and nonenforcement of rule following may influence action.
Tracking refers to rule following produced by reinforcement from the correspondence between the rule and the environment. For example, if your friend tells you “Using public transportation will save money from parking” you may follow a rule because in the past following your friend’s advice resulted in better financial outcomes. Rules that control behavior via natural contingencies are called tracks. Tracking relates to what Pelaez (2013) refers to as accuracy of a rule. Accurate rules that correspond to contingencies are likely to generate greater rule following.
Pliance and tracking bear some resemblance to concepts developed by social psychologists: injunctive and descriptive social norms (Cialdini et al., 1991). Injunctive norms refer to evidence (such as by rules) of what behavior is socially approved. Behavior change produced by injunctive norms resembles pliance. Cialdini (2003) gives an example of an injunctive norm for park preservation, a sign stating “Please don’t remove the petrified wood from the Park, in order to preserve the natural state of the Petrified Forest,” along with a picture of a visitor taking wood under a red circle-and-bar. The phrase “Please don’t” likely has been correlated with social disapproval for noncompliance, and thus rule following may be socially mediated. Descriptive norms refer to evidence of what people commonly do. Descriptive norms may indicate what behavior typically produces reinforcers, and thus behavior change produced by descriptive norms resembles tracking. Cialdini (2003) gives an example of a descriptive norm, a sign stating “Many past visitors have removed petrified wood from the Park changing the natural state of the Petrified Forest” with a picture of three visitors taking wood. The latter sign was less effective at deterring wood theft. If the rule encouraged behavior because of a history of nonverbal reinforcement for imitating others, then the behavior may be described as tracking.
Augmenting refers to rule-governed behavior controlled by the altered extent to which stimuli function as reinforcers or punishers. That is, rules may function as motivational operations that have both value-altering and behavior-altering effects (see Michael, 1993). Rules with motivational effects are called augmentals. Augmenting can occur in conjunction with tracking and pliance (Törneke, 2010). Augmentals can be formative or motivative (Hayes & Wilson, 1993). Formative augmentals create new reinforcers or punishers, whereas motivative augmentals alter the relative reinforcing or punishing value of consequences. Consider the rule “To reduce greenhouse gas emissions, use public transportation.” If greenhouse gas (GHG) emissions has been established as aversive stimuli, then the rule, as a motivative augmental, may increase the use of public transportation as a reinforcer and behavior of using public transportation may increase in probability.
Rule-governed behavior maintained by social reinforcers obtained by failing to comply with a rule may be viewed as a form of counterpliance (Hayes et al., 1989). Rules may motivate counterpliance (including doing the opposite of the rule) if the rules are overly prescriptive and aversive, or come from disliked authority figures, such that breaking the rule would produce reinforcing social disapproval in the speaker. For example, rules may generate counterpliance if they instruct listeners to engage in non-preferred activities or to emit effortful behavior. This motivating effect on behavior has been described by other psychologists as reactance (e.g., Brehm, 1966).
Some scholars have questioned whether these categories of rule control improve prediction and control of rule-governed behavior (Harte & Barnes-Holmes, 2022; Kissi et al., 2017), with some suggesting that the effects may be more productively incorporated into alternative conceptual frameworks (e.g., Harte et al., 2020). Nonetheless, the terms may be helpful here in distinguishing sources of rule control.
Climate-Change Knowledge as Rules
Skinner (1957) argued that scientific knowledge can be understood as verbal behavior. According to Skinner (1974), scientific knowledge is a “corpus of rules for effective action” (p. 235). Scientists tact relations observed in data, or produced through intraverbal processes, and construct rules about relationships (Schnaitter, 1980). The rules specify contingencies, including if–then relations that characterize scientific activities (see Lee, 1985). Some of the relationships tacted by scientists may be transformed into graphs or models, and such behavior is also reinforced by the altered behavior of listeners. Verbal reports, graphs and models created by climate scientists can thus be interpreted as rules and reactions to them described as rule-governed behavior.
Climate-change reports, warnings, and climate models, like other scientific knowledge, may be analyzed as rules. They may have the structure of contingency-specifying stimuli if they describe outcomes for continuing to release GHGs into the atmosphere or for curbing carbon emissions, and more importantly, may serve as function-altering stimuli if they change the psychological function of climate-relevant verbal and nonverbal stimuli and alter listener behavior.
Consider the following statement from the IPCC (2022): “Global warming, reaching 1.5 °C in the near-term, would cause unavoidable increases in multiple climate hazards and present multiple risks to ecosystems and humans (very high confidence)” (p. 15). The sentence associates global warming (caused by humans burning fossil fuels) with dangerous outcomes, and thus may alter the function of burning fossil fuels. The sentence, “Near-term actions that limit global warming to close to 1.5 °C would substantially reduce projected losses and damages related to climate change in human systems and ecosystems…” (IPCC, 2022, p. 15) relates pro-climate actions to harm reduction and may therefore establish pro-climate actions as reinforcing. The IPCC 2021 report describes various actions that can reduce emissions, including electrification and switching from fossil fuels, retrofitting buildings for energy efficiency, reducing travel demand, carbon storage, and changing land use, among others. The evaluation that “Regions and people with considerable development constraints have high vulnerability to climatic hazards (high confidence)” (IPCC 2022, p. 14) may alter the function of stimulus relations concerning climate justice.
Nonfiction climate-change books and extended narratives also may have a rule function if they condition reader behavior. Books such as The Sixth Extinction, The Uninhabitable Earth, and Our Final Warning: Six Degrees of Climate Emergency detail the human causes and consequences of climate change and may condition behavior, whereas others may alter behavior by offering solutions (e.g., Drawdown; Saving Us). Documentaries, films, podcasts, and TEDtalks on climate change (e.g., Breaking Boundaries: The Science of Our Planet; Before the Flood, An Inconvenient Sequel:Truth to Power, Don’t Look Up, Years of Living Dangerously, Great Thunberg’s Ted Talk, 2018, etc.), can establish new stimulus functions by associating climate-relevant behaviors with many kinds of ecological and health impacts.
Graphs and figures included in climate reports may also have a function-altering effect. The well-known “hockey stick” figure (Mann et al., 1999), illustrating the exponential increase in northern hemisphere temperature rise, linked with carbon emissions, may have altered the psychological function of the words “carbon-emissions” and associated stimuli. Figures from IPCC reports, such as those that directly illustrate the near-linear relationship between cumulative carbon emissions and global surface temperature (i.e., SPM. 10, IPCC 2021, p. 28), may have a comparable function-altering effects. Graphs that show projections of sea level rise, drought severity, storm frequency with temperature increases may also alter behavioral relations with respect to climate-relevant stimuli.
As described above, rule following has many benefits for listeners, and these benefits apply to warnings about the climate crisis. That is, climate rules have the potential to influence the behavior of multiple people (i.e., cultural behaviors) and guide actions to limit future global warming and avoid its worst impacts.
Climate-Change Warnings as Ineffective Rules
Years of climate warnings, however, have not produced the necessary behavior change. As early as 1896, Svante Arrhenius predicted the greenhouse effect resulting from burning fossil fuels (Arrhenius, 1896/2012). Concern about CO2 concentrations increased in the 1960s and 1970s as evidence increased, and greater public awareness followed the media coverage of James Hansen’s 1988 congressional testimony on global warming (BoyKoff & BoyKoff, 2004; Weart, 2008). To address growing concerns, in 1988 the IPCC was formed, and in 1990 it released its First Assessment Report indicating that the planet was warming and that the warming could be caused by human activities. Each successive IPCC report has attributed global temperature increase to human activities with greater certainty and warnings became stronger. The Summary for Policy Makers from Climate Change 2021: The Physical Science Basis (IPCC, 2021) notes, “It is unequivocal that human influence has warmed the atmosphere, ocean and land” (p. 4) and that, “Global warming of 1.5 °C and 2 °C will be exceeded during the 21st century unless deep reductions in CO2 and other greenhouse gas emissions occur in the coming decades” (p. 14). In 2017, Over 15,000 scientists jointly signed “World scientists’ warning to humanity: a second notice” (Ripple et al., 2017) in which the authors reflected on the lack of progress in solving environmental challenges, especially climate change.
Across the globe, there are individuals (e.g., Johnson & Wilkinson, 2021), cities (see Markolf et al., 2020), and nations (see Climate Action Tracker, 2021) who are taking action and making changes to lead communities and countries toward net-zero carbon emissions. Yet, progress is inadequate. Concentrations of greenhouse gases continue to steadily rise (NOAA Global Monitoring Laboratory, n.d.). Markoff et al. report that slightly less than half of the largest US cities have established GHG reduction targets. Emission pledges by nations are insufficient, GHGs in the atmosphere are increasing, and current estimates place temperature increases by the end of the century at 2.7 °C (Climate Action Tracker, 2021).
Climate-change rules are failing to generate sufficient action. To help understand this failure of rule control, the following section seeks to apply a behavioral approach to rule following to climate-change warnings and solutions and identifies factors that may influence rule following. The paper draws from prior scholarship on variables impacting rule following (Barnes-Holmes et al., 2001; Pelaez, 2013; Stapleton, 2020; Törneke, 2010), and like the approach taken by Stapleton (2020) to understand responses to COVID-19 rules, applies these to climate-change messages. The factors sometimes overlap, and represent only some of the variables likely to impact rule following. Possible solutions and strategies for increasing rule control are also discussed. The paper incorporates research from environmental psychology, risk and decision making, cognitive and social psychology, and climate-change communication, as the bulk of climate-psychology research has come from outside of behavior analysis. The hope is to help connect diverse research areas and theoretical approaches.
1. Insufficient Exposure to the Rule
Climate warnings may fail to motivate action because people are not sufficiently exposed to accurate information about the causes of climate change, its risks, and the steps needed to mitigate climate disasters. Gifford (2011) calls this the barrier of “ignorance” and is part of what Hornsely and Fielding (2020) describe as the “deficit model” of climate communication—that inaction is attributable to a lack of exposure to the problem. Climate change surveys in the United States (US) suggest, however, that there is growing awareness of anthropogenic climate change. Polls suggest that about 76% of adult Americans believe that climate change is happening and a majority (60%) believe it is human caused (Leiserowitz et al., 2021a). Leiserowitz et al. (2022) have found that in the US, 33% of the population can be categorized as alarmed, 25% as concerned, 17% as cautious, 5% as disengaged, 10% as doubtful, and 9% as dismissive of climate change. Thus, over half of all Americans can be categorized as alarmed or concerned about the climate crisis. Leiserowitz et al. (2021a) showed, however, that Americans rarely discuss (61%) climate change, and a little more than half (57%) hear about climate change each month. People also significantly underestimate the scientific consensus on climate change (only 59% of those polled believe scientists agree that climate change is happening) and underestimate personal risk from climate-change (Leiserowitz et al., 2021a). A study by Ranney and Clark (2016) found that only 12% of participants showed a partial understanding of the mechanism of global warming. An international survey found that most people reported that they needed a little more information to form an opinion about climate change (Leiserowitz et al., 2021). Thus, although a growing number of people are aware of climate change, there may be insufficient exposure to information about the severity or urgency of the problem.
A lack of media coverage on climate change may be partly responsible. A study investigating media reporting on climate change found that across 10 countries sampled, only 0.53% of newspaper articles in top papers in a given month (2006–2018) focused on climate change (0.63% in the US; Hase et al., 2021). Even with increasing frequencies and severity of extreme weather events and environmental disasters, reports of those events may not heighten concern if they are not associated with climate change. Hassol et al. (2016) have argued that, “Even as occurrences of certain classes of extreme events have increased, the media in some countries have not kept pace in communicating the scientific understanding of the connection between climate change and extremes” (p. 5).
People may also have insufficient exposure to information about climate-harming activities or pro-climate solutions. For example, people may not know about the high greenhouse-gas emissions of red meat production (e.g., Camilleri et al., 2019), or know about benefits of reducing air travel compared to actions such as waste reduction (e.g., Wynes et al., 2020).
Possible Solutions and Strategies for Change
One solution to the lack of information and insufficient rule exposure is to increase opportunities for people to learn about causes, risks, and solutions to the climate crisis. Even brief information about climate change has been shown to influence climate-change acceptance (Ranney & Clark, 2016). Ballew et al. (2019) argue that greater media coverage, increasing interpersonal conversations, and enhanced educational strategies, may improve awareness and concern.
Media reports that describe climate change may motivate action. The framing of such media stories, however, may be important to its impact. Reports that focus on benefits for climate action may improve engagement, but reports that focus on uncertainties are likely to reduce action (see discussion in Bolsen & Shapiro, 2018). A challenge to greater media reporting, however, is the increasing polarization of news sources and the growing use of online sources that present inaccurate or biased climate coverage.
Increasing informal conversations on climate change may require interventions that counteract punishment. People may avoid discussing climate change because of the risk of social disapproval (punishment) from listeners with opposing views (e.g., Maibach et al., 2016). Skinner (1957) defined the audience as listeners who serve a discriminative function over speaker behavior. Punishment may establish listeners as negative audiences (discriminative stimuli for punishment, see Skinner, 1957). Mixed (positive + negative) or novel audiences also may weaken the likelihood of climate-change conversation. Thus, increasing discussions about climate change may require educational interventions that alter audience control and create opportunities for reinforcement. A study by Geiger and Swim (2016) showed that when participants who were concerned about climate change were told that an audience was also concerned about climate change (i.e., was not a negative audience), the reported tendency to talk about climate change increased. Thus, sharing information about the proportion of people who are concerned about climate change may increase talking about the topic. Listeners who effectively challenge speaker’s statements on climate change may punish discussion. Geiger and colleagues (2017a) showed, however, that when participants were educated about climate change, they reported that they were more capable and willing to discuss it.
Studies have shown that educating people about causes and impacts of climate change also can increase engagement (e.g., Geiger et al., 2017b; Ranney & Clark, 2016). Efforts have been made to promote teaching climate change in schools across content courses from the sciences to the humanities (e.g., Beach et al., 2017; Shepardson et al., 2017). The teaching of climate change, however, has been attacked by political interests (e.g., Branch et al., 2016). Research has found that middle and high school teachers do not devote much time to the topic, and about a third emphasize both human and natural causes (Plutzer et al., 2016). To enhance education on climate change, advocacy for teaching climate-change content in schools and universities is needed, as well as resources for teachers wanting to include climate change in their classes (for examples see Beach et al., 2017; Fretz, 2015, and for behavioral approaches see Sustainability & Behavior Analysis Sample Course Units at the Behaviorists for Social Responsibility website https://bfsr.abainternational.org/).
Educating the public about the individual actions that are likely to have the most impact on carbon emissions is also needed. People may have poor “carbon numeracy” in that they do not know the relative GHG emissions produced by different activities (e.g., Wynes et al., 2020) (see Rules are Incomplete, below).
As noted by Geller (1992) and many others (e.g., Blake, 1999; Bulkeley, 2000; (McKenzie-Mohr, 2000; Staats et al., 1996; and see Nisa et al., 2019), simply educating or informing the public on environmental problems may be insufficient to promote significant or lasting behavior change. Van der Linden (2017) notes that knowledge about human-caused climate change may be necessary but not sufficient for climate concern (and action). Educational strategies may need to be combined with other interventions.
2. Insufficient Learning History (Lack of Understanding by Listeners) and Rule Complexity
Climate scientists, in presenting data and results of climate models, use technical and scientific jargon which may be unfamiliar to lay audiences (Hassol, 2008; Stoknes, 2015). Consider the following statement drawn from a passage in the Summary for Policymakers by the IPCC (2007), “It is more likely than not that anthropogenic forcing has increased the risk of heat waves” (p.10). Although terms are carefully defined within reports, statements such as these (especially taken out of context), may fail to generate effective listener behavior. The general public may not react appropriately to the unfamiliar term “anthropogenic forcing” nor the probability of the statement “more likely than not.” Budescu and colleagues (2009) found, for example, that when people read sentences from the 2007 IPCC report, they often misinterpreted the probability statements of climate scientists.
Broadly speaking, what may be called, “understanding” of a rule is dependent on a specialized history of socially mediated reinforcement. Pre-requisite skills for rule control may include the establishment of: listener behavior (e.g., auditory-to-visual conditional discriminations), echoic control, observing behavior, or arbitrarily applicable relational responding (see Tarbox et al., 2020). Rule following depends on the prior establishment of discriminative, conditioned reinforcing, and eliciting functions of verbal and nonverbal stimuli. Scientists, including climate scientists, are taught to tact scientific observations using precise, technical terms, and emit intraverbals in accordance with scientific principles. They are also taught listener skills necessary to respond appropriately to statements by other scientists. Statements made by climate scientists may fail to generate appropriate responses in lay audiences because they lack a comparable verbal reinforcement history. The view that inaction on climate change is attributable to a lack of scientific background in listeners is the second part of the “deficit model” of climate communication (Hornsey & Fielding, 2020).
The role of a listener’s learning history in rule control relates to Pelaez’s (2013) discussion of rule complexity: Rules vary in complexity of stimulus dimensions and derived relations from lower to higher. Compared to simpler rules, rule control by complex rules requires more advanced listener skills and as a result, complex rules may be less likely to produce effective listener behavior. To illustrate, compare the rules, “To reduce carbon emissions, people should reduce air travel” to the rule, “Flying produces greater carbon emissions than riding a bus, taking a train, and driving except when the car has only a single passenger in which case flying may produce lower emissions per person, depending on whether or not the plane is flown near capacity.” The second is more complex than the first because it involves a greater number of stimulus relations, including a higher-order class of relations, where relations are conditional upon other relations (see Robertson & Pelaez, 2018). From a relational frame perspective, it might be said that the former rule involves fewer derived relational operants. As a result of its greater complexity, the second rule may fail to alter travel behavior. Rules about climate change may fail to generate appropriate behavior because of their complexity. Exposure to complex rules may even be aversive. De Vries, Rietkerk, and Kooger (2020) argue that if information about sustainable actions is complex or difficult to understand, it may generate a negative stress reaction leading people to ignore it.
Possible Solutions and Strategies for Change
Solutions to the problem of insufficient verbal history may include educating listeners and establishing audience control of scientific verbal behavior. Educating listeners will require a multi-faceted approach that includes expanding climate-change curriculum in K–12 schools and universities and expanding climate-science coverage in the media (see above for challenges of bringing climate change education to schools). Educational efforts have been shown, for example, to produce more willingness to discuss climate change (e.g., Geiger et al. 2017a).
Scientists could also be better trained in public outreach so that climate communication comes under audience control. Skinner (1957) argued that one function of audiences was to serve as discriminative stimuli for word choice or jargon. The verbal behavior of climate scientists, like all scientists, has been shaped for precision and has come under control of scientific audiences. Effective repertoires for lay audiences, though, may be lacking. Climate scientists could be trained to translate climate expressions or describe the science with terms familiar to lay audiences (Hassol, 2008), such as by replacing “anthropogenic” with “human-caused.” Budescu et al. (2009) recommended that climate reports use exact percentages instead of words to describe likelihoods. Consistent with this, Myers, Maibach, Peters, and Leiserowitz (2015) showed that participants’ ratings of the scientific consensus on climate change was greater when consensus was first described numerically (e.g., “97%”) rather than non- numerically, (e.g., “an overwhelming majority”). Behavior analysts have made analogous arguments for communicating the science of behavior analysis to the public, noting that non-technical terms are better understood and may generate more positive reactions in listeners (e.g., Cihon et al., 2016; Critchfield et al., 2017; Jarmolowicz et al., 2008). Using stories may also be effective by presenting information in formats that are not only engaging, but that allow listeners to react more effectively to the content (Grant, 2007).
Simplifying rules may increase climate-relevant actions. For example, researchers have shown that clear labels on commodities to highlight their GHG emissions can increase people’s understanding of the relative GHG emissions of purchases and improve choices (e.g., Camilleri et al., 2019). There are also many ways to reduce carbon emissions, and presenting many at once may reduce rule following. Gardner and Stern (2008) argue that pro-environmental information should focus on a small number of specific actions that have a large impact on emissions.
Improved education and better science communication alone, however, may be unable to generate sufficient climate action or engagement. Correlational work, for example, has found that education is not a strong predictor of belief in climate change (Hornsey & Fielding 2020), suggesting that scientific illiteracy is unlikely to fully explain why climate warnings do not motivate action.
3. Incomplete Rules
The specificity of a rule may influence rule following. (Pelaez, 2013; Robertson & Pelaez, 2018). Explicit or complete rules specify all parts of a reinforcement contingency (antecedent, behavior, and consequence), whereas implicit rules fail to include some part. The more explicit the rule is, the more likely it will produce rule following. For instance, Kaufman, Baron, and Kopp (1966) showed that specific instructions about reinforcement contingencies (even if inaccurate) produced different response rates than minimal instructions. Similarly, Presbie (1970) showed that specific (vs. vague) instructions produced more effective patterns of avoidance in humans.
Lack of rule completeness or specificity may hinder action on climate change. Climate warnings may fail to describe how people can mitigate climate change, or which actions are most effective. Gifford (2011) describes this barrier to climate action as another element of ignorance—people may not know what to do to address the climate crisis. Consider the statement, “Failure to address global warming will result in more severe droughts.” Although the statement associates droughts with global warming, the rule lacks concrete steps listeners can take to avoid warming. Gardner and Stern (2008) note that the documentary An Inconvenient Truth, although effective in highlighting the urgency of addressing climate change, failed to provide much guidance to viewers about what to do. De Vries (2020) makes a similar point about the documentary Our Planet. Media coverage of behaviors that can impact climate change has also been lacking. Hart and Feldman (2014), in an analysis of network news stories between 2005 and June 30, 2011, found that TV news rarely discussed impacts from climate change together with actions that could mitigate climate change. A comment to a New York Times article by reader #BirdsAreNotReal (2022) stated, “I’ve trusted and believed the climate science for three decades, for the last decade and a half I’ve been asking, so what do you want ME to do about it?”
De Vries (2020) argues that some action recommendations for mitigating climate-change are too vague. Simply telling people that they should reduce their carbon emissions provides insufficient stimuli for action. Garder and Stern (2008) note, “From a householder’s perspective, a desire to reduce carbon emissions, even combined with knowledge that doing so has net financial and environmental benefits, is insufficient to yield effective action unless that person knows which actions will produce the benefits” (p. 14).
Actions also vary in their climate impact, and information about actions that are most effective may be lacking. For example, people may be more likely to (incorrectly) believe that recycling is more effective at mitigating climate change than energy use or meat consumption (Whitmarsh, 2009). If rules do not clarify high-impact actions, people may engage in behaviors that are most convenient. This effect is related to the problem Gifford (2011) calls tokenism: Individuals who are motivated to act on climate change may simply choose actions that are easiest to do, even if they have a low impact.
Information about effective community interventions is also needed. Commenting on initiatives by cities to limit green-house gas emissions, Markolf and colleagues (2020), note that, “Whereas top-level abstract targets are available, what is equally or even more needed is detailed information about which policies are being put into place, which initiatives are working and which are failing” (p. 22).
Rules may also fail to establish effective or enduring stimulus control. For example, a rule that states, “Reducing home energy use can reduce your carbon emissions” is not explicit about what nonverbal stimuli should function as discriminative stimuli for energy saving actions and about what actions are required to reduce home energy consumption. Similarly, “Support businesses with net- zero emission targets” may be too vague to generate climate-friendly purchases.
Possible Solutions and Strategies for Change
Climate communication should clearly specify reinforcement contingencies, including the most effective actions for mitigating climate change. Hart and Feldman (2016) found that exposure to news stories that described actions one could take to mitigate climate change increased a person’s stated efficacy to fight climate change and willingness to act, compared to exposure to stories that only described climate impacts. Kotcher et al. (2021) found that descriptions of climate impacts, a solution, and appeals for calls to action (positive social norms, which could signal social approval) were all influential features of advocacy messages, although solution information was the most important.
Information that is specific or tailored to an individual, group, or setting may be especially effective (see Lehman & Geller, 2004). For example, information about reducing energy use that is tailored to a household’s energy use patterns and installed appliances (see Abrahamse et al., 2007) may be more useful then general recommendations on energy saving.
Rules about solutions, such as those promoting individual action, should be explicit about which actions should be prioritized (e.g. Gardner & Stern, 2008). Rare’s Center for Behavior and the Environment identified 30 of the 80 most impactful actions to address the climate crisis listed in the book Drawdown (Hawkins, 2017) that involved behavior change (Heller, 2019) and created a list of seven behaviors with the greatest potential to reduce emissions (namely purchase electric vehicles, reduce air travel, eat a plant rich diet, offset carbon, reduce food waste, tend carbon-sequestering soil, and purchase green energy) (Rare and California Environmental Associates, 2019). Actions with the greatest emissions impact could be given priority in climate communications, and given greater emphasis in psychological research (see Nielsen et al., 2021). Electrification, for example, is key to reducing GHG emissions (Griffith, 2022). Thus, a message campaign to reduce transportation emissions could state, “To reduce your household carbon emissions your next vehicle purchase should be an electric vehicle.”
Rules about the effectiveness of climate-change behavioral interventions also need to be widely communicated. Markolf et al. (2020) suggest that new platforms and mechanisms could be developed for sharing effective city-wide interventions. One information-sharing website for community interventions can be found at the Community-Based Social Marketing website https://cbsm.com/.
Rules should also specify actions that can change practices of organizations that produce GHG emissions. Twenty fossil fuel companies are responsible for a third of all carbon emissions (see Matthew & Watts, 2019). Yet, Mann (2021) has argued that fossil fuel industries have made significant attempts to shift blame from corporate actions to individual behaviors to avoid regulations and revenue loss. The concept of the personal carbon footprint gained popularity after a large advertising campaign by the oil company BP (see Yoder, 2020). Thus, rules could describe how to be a climate voter and support politicians who will create policies to regulate fossil-fuel extraction and GHG emissions in corporations. Wynes and colleagues (2021) have proposed the development of statements that indicate the impact on GHG emissions from voting for different political candidates (given the policies they support) and that indicate an individual’s GHG impact from voting. Rules could also clarify how to engage in collective action (see Ardila Sánchez et al., 2020; Ardila Sánchez et al., 2020; Mattaini, 2013). McKibben (cited in Hayhoe, 2021) has argued that, “The most important thing an individual can do now is not be an individual” (p. 185).
4. Instructed Behavior Not Sufficiently Learned
As Barnes-Holmes et al. (2001) suggest, the extent to which instructed behavior has been shaped and reinforced in a listener will influence rule following. A person may be able to react as a listener to rules such as, “Weathering your home will reduce GHG emissions,” or “Switching to green energy can protect the environment,” but may not engage in the instructed behavior because the necessary behaviors are not sufficiently learned. People may not know how to weatherize a home or how to switch to green energy. A campaign in the Netherlands designed to increase energy efficiency with home insulation failed in part because of confusion in the public about how (and when) to act (Schalkwijk, 2017 as cited in de Vries, 2020).
Possible Solutions and Strategies for Change
Community-based social-marketing research indicates that an important step in designing an intervention is to identify the barriers (as well as benefits) to action (McKenzie-Mohr, 2011). One possible barrier is skill deficits. Climate communicators and those designing climate-relevant interventions should evaluate whether listeners know how to engage in the target behavior. If the target behavior has not been learned, the instructed behavior may need to be taught via further instructions, modeled, or shaped via successive approximations. For example, home owners could be given free energy audits and home visits could be used to demonstrate how to best weatherize their homes. Advertisements on TV, influencers on social media, or teachers in schools could model green actions. Incentives for climate-friendly actions (e.g., taking public transportation, becoming involved in environmental groups, composting, etc.,) could be provided for increasingly complex behaviors.
5. Rules Have a Weak Function-Altering Effect (Rules do not Function as Augmentals)
As described above, many events obtain their reinforcing (or punishing) function from rules (see Torneke, 2010), an effect described in terms of their augmentive function. Climate-change rules may fail to motivate action because the stimuli specified in the rule do not alter the reinforcing value of stimuli (the rules are not motivative augmentals), or because the rules fail to create new reinforcers (are not formative augmentals). For example, the statement, “To reduce the melting of the Greenland ice sheet, we need to cut GHG emissions” may fail to establish steps to reduce GHG emissions as negative reinforcers if melting ice sheets are not aversive stimuli.
Studies on risk perception have identified “emotion” as an important factor in risk judgements (see van der Linden, 2017), and climate researchers have analyzed the motivating effect of climate-relevant stimuli in terms of their ability to engage the “rational” versus “emotional” brain (e.g., Marshall, 2015). Engaging the emotional brain is said to motivate climate action. In behavior analysis, emotional reactions are not interpreted as causes but changes in operant and respondent behavior produced by motivating variables and eliciting stimuli (Skinner, 1953). Rules that associate emotional-stimuli with other climate-relevant stimuli can therefore produce function-altering effects. For example, if coral reefs are reinforcers for a listener, then rules about how climate change will cause mass die-offs of coral reefs (an eliciting stimulus and negative reinforcer) might establish climate-change mitigation as reinforcing and motivate action.
Lack of an augmenting effect of climate-change rules and warnings may stem from the different learning histories of scientists and nonscientists with verbal stimuli included in the rule. That is, if words in climate rules are unfamiliar, have not been associated with other stimuli or reinforcers (are not part of established relational networks), or have little evocative effect, they may fail to motivate action. Weber and Stern (2011) provided a health-related example from Sinaceur, Health, and Cole (2005) showing that reports about “bovine spongiform encephalitis (BSE)” or “Creutzfeld-Jacob disease” elicited less fear (and presumably were less motivating) than reports using more familiar terms, “mad cow disease” (p. 320).
One example of terms lacking motivating effect may be the words “climate-change” itself. Although the words “climate change” may be preferred to “global warming” by scientists because “climate change” is more descriptive of the broad changes produced by GHG concentrations, the words “climate change” may have a lower motivating effect. In a memo about recommended communication strategies, the Republican strategist Luntz wrote that Republicans should use the term “climate change” instead of “global warming” because,” “climate change” is less frightening than “global warming,” and that “While global warming has catastrophic connotations attached to it, climate change suggests a more controllable and less emotional challenge” (Luntz, 2002, p. 142).
The stimulus function (or relational networks) of climate-change terminology may also differ between climate scientists and lay audiences. If a climate scientist, for example, describes an event as causing a “positive feedback loop” (such as melting Artic permafrost), it may not function as an aversive stimulus to the lay public because the term “positive feedback” may be associated with good events (Hassol, 2008). Thus, warnings about positive feedback loops may not have an augmentive function.
Weber (2016) also argues that climate change risk may fail to motivate action in politicians and the general public due to its statistical nature. Consider a statement from the IPCC (2018) that notes, “Model-based projections of global mean sea level rise (relative to 1986–2005) suggest an indicative range of 0.26 to 0.77 m by 2100 for 1.5 °C of global warming, 0.1 m (0.04–0.16 m) less than for a global warming of 2 °C (medium confidence).” For nonscientists, the quantitative nature of the comparisons of sea level rise at 1.5 °C and 2 °C degrees, may fail to motivate avoidance behavior. Numbers alone may not be very compelling.
Counterpliance is also a risk if rules come from government or partisan speakers (for a health- related example see Irmak et al., 2020). For example, government regulations about fossil-fuel use or rules from environmentalists about eating less meat may heighten energy use or meat consumption in listeners for whom these behaviors are highly reinforcing (and speakers distrusted).
The framing of climate impacts may also reduce their augmentive function. Rules can specify immediate or temporally remote outcomes, certain or probabilistic ones, and consequences for the self or distant others. The delay, probability, and social (and spatial) closeness of an event (i.e., who it impacts) has to be described as its psychological distance (Trope & Liberman, 2010). Outcomes that are more immediate, likely, and personal/local are likely to be more potent as consequences. The temporal discounting that can occur with instructed reinforcers relates to what Pelaez (2013) identifies as the timing of the contingencies specified in the rule. Rules can specify immediate or delayed contingencies. A large body of research on discounting shows that as reinforcers become more delayed from choice, more probabilistic, or the recipient more socially distant, the lower their value (e.g., Green, & Myerson, 2004; Jones & Rachlin, 2006; Madden, & Bickel, 2010; Rachlin et al., 1991). Furthermore, research has shown that delayed and probabilistic environmental events (and environmental impacts on others) are discounted in ways similar to that of other types of consequences (e.g., Hardisty & Weber, 2009; Kaplan et al., 2014; McKerchar et al., 2019).
Climate change is often described as having its most dire outcomes for future generations, as being probabilistic, and impacting only select populations (e.g., low-lying nation states). Thus, lack of effectiveness of climate warnings may be attributed in part to temporal, probability, and social discounting (e.g., Hirsh et al., 2015; Gifford, 2011; Weber & Stern, 2011). Stoknes (2015) calls this barrier to climate action the barrier of distance. Spence, Poortinga, and Pidgeon (2012) found a significant correlation between concern about climate change and dimensions of psychological: those rating climate change impacts as having lower psychological distance (sooner, more likely, local and with personal impacts) showed greater concern about climate change (and see Maiella et al., 2020).
Consider a climate statement, “By the end of this century, what have been once-in-20-year extreme heat days (one-day events) are projected to occur every two or three years over most of the nation” (Melillo et al., 2014, p. 39). Such a temporally distant outcome may fail to motivate much action. Polar bears are the most iconic image of climate-change messaging, yet polar bears and melting Arctic ice are socially and spatially distant from most people. Thus, verbal stimuli linking ice loss and declining polar bear populations to climate change may not be especially motivating.
Climate warnings about sea level rise for island nations, or warnings about prolonged droughts decreasing food production in the Global South, may fail to generate significant action in some listeners in more developed countries because of the social/spatial distance of those impacts. It is interesting to note that a US climate survey that found that only 50% of the public thought that climate change would harm themselves personally (Leiserowitz et al., 2021a). This suggests that in the US, climate-impacts may often be framed as socially distant in the media.
Impacts of climate change are described as probabilistic, and thus warnings may fail to raise concern. A key strategy of climate-change denialists aiming to protect the financial interests of fossil fuel corporations has been to frame the presence of climate change and its human cause as scientifically uncertain (see discussion in Weber & Stern, 2011). Cushman (1998) described a leaked draft of a proposal by industry groups (especially fossil fuel corporations) to oppose a treaty on climate change. The proposal was to train specially recruited scientists to “help convince journalists, politicians and the public that the risk of global warming is too uncertain to justify controls on greenhouse gases.” By emphasizing the uncertainty, industry groups hoped to dissuade action on climate change. Research has shown that when people are told that there is scientific disagreement about an environmental hazard, even if a very small proportion of scientists disagree, support for regulation decreases (Aklin & Urpelainen, 2014).
Images used in the media may also enhance discounting of climate change. O'Neill (2013) examined 1500 images attached to climate stories in 2010 in US, UK and Australian papers and found that images not only focused on people, especially political figures, thereby highlighting climate change as a political and contested issue, but also depicted climate change as geographically distant (e.g., ice imagery) or personally distant (e.g. smokestack images).
Possible Solutions and Strategies for Change
The motivating or augmental function of climate-change warnings could be improved to maximize reinforcer or (punishment) value of climate-relevant stimuli and reduce psychological distance of climate impacts and solutions to motivate action. Van der Linden et al. (2015) offer several “best practices” about climate change communication, and conclude that, “in order to improve public engagement with the issue, policymakers should emphasize climate change as an experiential, local and present risk; define and leverage relevant social group norms; highlight the tangible gains associated with immediate action; and last, but certainly not least, appeal to long-term motivators of pro- environmental behavior and decision making” (p. 761). The first point is about discounting: the augmenting function of climate rules may be enhanced by using stimuli that reduce temporal, social and geographical, and probability discounting (i.e., reduce psychological distance). Spence, Poortinga, and Pidgeon (2012) argue that “…in order to promote concern about climate change, risk communications should focus on making climate change psychologically closer and make potential climate change impacts relevant to individuals’ social group, locality, and lifetime” (p. 13). Similarly, climate communicators suggest that it is important to make the distant, global, and nebulous threat of climate change personally relevant (e.g., CRED, 2009; Hayhoe, 2021; Van Leuvan et al., 2022; Weber, 2016; Stoknes, 2015).
Behavioral strategies that reduce temporal discounting (see review in Rung & Madden, 2018) may also be effective at improving climate-relevant decisions. One strategy is having people imagine their future selves, i.e., episodic future thinking (see Kaplan et al., 2016). Nalau and Cobb (2022) describe a variety of future visioning manipulations designed to increase engagement on climate change. Other tools for reducing discounting of climate impacts are films and stories. People have reported that reading climate-change fiction reduced temporal distance of climate impacts (Schneider-Mayerson, 2018).
Emphasizing local impacts can reduce social and geographical discounting. Research has shown that describing local impacts was more effective in increasing concern about climate change than describing impacts in distant places (e.g., Jones et al., 2017). Visualization tools (i.e., edited images) that show possible climate-impacts for local places and landscapes can improve engagement and encourage action (Sheppard et al., 2011).
One way to reduce probability discounting is to use messages that clarify the scientific certainty of climate change (i.e., consensus messaging). Studies have shown that consensus messaging can increase climate-change concern and climate-change policy support (van der Linden et al., 2015; Van der Linden, Leiserowitz, & Maibach, 2019).
Using emotionally engaging text might increase the motivating impact of climate messages (e.g., CRED, 2009; Weber, 2016). Descriptions of indigenous communities losing sacred land to sea level rise or a community’s successful installation of a community solar garden may be more persuasive than climate statistics. Vivid imagery may also motivate behavior (O'Neill & Smith, 2014). For example, images of coral reef bleaching, shrinking glaciers, superstorm damage, or perhaps images of new windfarms or solar arrays might engage behavior (see O'Neill et al., 2013). Analogies could also improve motivation by using familiar stimuli. Gonzales and colleagues (1988) give an example for encouraging weatherstripping for energy conservation. Energy auditors, they state, instead of just describing the cost-effectiveness of weatherstripping, could say, “If you were to add up all the cracks around and under the doors of your home, you’d have the equivalent of a hole the size of a football in your living room wall” (p. 1052).
Creating social norms can also promote engagement (e.g., CRED, 2009; Van der Linden et al., 2015; Van Leuvan et al., 2022). As noted above, injunctive norms may impact behavior by signaling social consequences for behavior, whereas descriptive norms may impact behavior by signaling what behavior is effective under prevailing contingencies (see Huber et al., 2018). As an example, people provided with information about their energy consumption relative to their neighbors (descriptive norms) and given smiley faces for lower usage (injunctive norms) showed energy-use reductions (see Allcott, 2011). Huber et al. found that people who were given information about government policy on corporate carbon offsets and information about social norms (descriptive and injunctive norm) showed greater willingness to pay, and payment of carbon offsets, for driving compared to a control group. Sparkman and Walton (2017) showed that dynamic descriptive norms about a growing number of people eating less red meat reduced meat consumption.
Some have suggested that climate-change warnings that describe dire outcomes for failures to act—“fear” messaging—may cause disengagement (e.g., CRED, 2009; Van Leuvan et al., 2022). Mann (2021) makes a similar point about “doomist” messages that imply it is too late to avoid climate catastrophe. Positive frames (e.g., Morton et al., 2011) and humorous messages may also increase engagement (see Kaltenbacher & Drews, 2020). A recent meta-analysis, however, found that negative or loss framing (i.e., losses that can be prevented) was more likely to lead to behavior change than positive framing (gains from action) (Homar & Cvelbar, 2021). Given the mixed findings in the literature, Hornsey and Fielding (2020) argue that a diversity of messages may be warranted. It seems likely that the impact of framing may depend on the audience, what is being lost or gained, and whether effective action is proposed.
Rules that associate climate-relevant actions with established reinforcers may also motivate action (i.e., augmentals). Honsey and Fielding (2020) argue, “…rather than persisting with data-driven arguments that do not speak to people’s underlying motivations, the goal is to identify what their motivations are, and to present arguments that align with them,” (p. 15) a strategy they call value-based messaging. Interventions focused on value statements have been shown to impact decision making (e.g., Jackson et al., 2016), and can be applied to climate-relevant behaviors (see discussion in Newsome & Alavosius, 2011).
Climate communicators have promoted connecting climate change solutions to values such as community, equity, security, or health (e.g., Corner & Clarke, 2016; CRED, 2009; Hayhoe, 2021; Stoknes, 2015; Van Leuvan et al., 2022). Climate solution statements could also describe co-benefits of addressing climate change, such as improved housing and employment (e.g., Jennings et al. 2020). A study by Wolstenholme et al. (2020) investigated the impacts on meat consumption of messages about improved health, environmental benefits, both, and control messages, and found that all but the control group showed reduced consumption. What functions as a reinforcer (what one values) of course differs across individuals, and thus climate communication should come under audience control. Republicans, who tend to show less concern about the climate crisis, have been shown to react positively to climate statements when messages are targeted to their values, e.g., security (Goldberg et al., 2021). Religious leaders may motivate action in faith-based groups by connecting climate action to religious values. Highlighting the connection of climate change to social justice also can motivate action (see Moser, 2016).
Stories may also influence climate action (Grant & Forrest, 2020). Stories present problems or disruptors that function as motivating operations for resolution or return to equilibrium (Grant, 2007). This motivative function of climate fiction (e.g., Parable of the Sower, Water Knife, The Ministry for the Future) may be used to increase climate-change engagement. Jones (2014) investigated the effects of vignettes (short fiction stories) on support for climate policy. The study found that narratives were more persuasive than statements of climate facts, and that the level of positive affect for the hero was positively associated with increases in support for the policy solution described in each narrative. Stories and films that use satire and humor, such as Don’t Look Up, may also increase climate action.
Messaging strategies can minimize the aversiveness of climate rules and reduce counterpliance. Climate researchers could borrow communication strategies from reactance research on health-related persuasive messaging such as by using empathy, providing choices for action, or using narratives and other-referencing statements (Reynolds-Tylus, 2019). For example, Gardner and Leshner (2016) found that using stories describing the benefits for other people (loved ones) for rule following increased compliance with health messages.
Corner and Clarke (2016) and others (e.g., de Vries, 2020; Dietz, 2013) have argued that to increase public engagement with climate change, climate communication should be less “top- down” and more participatory. “Top-down” engagement strategies may fail because they do not consider the unique values (i.e., motivators) of listeners. Instead, climate communication should be an ongoing dialogue. Climate-change conversations should be inclusive to allow diverse individuals to formulate shared values. Such participatory public engagement, might lead to the creation of rules based on shared community values that could guide broad classes of climate-relevant actions.
6. Conflicting Rules
Rules may be ineffective if listeners are exposed to multiple, conflicting rules. For example, the function-altering impacts of the statement “climate change is human caused,” oppose those of the statement by climate-change denialists, “climate change is a natural process.” People may have been told that global warming is not happening, it is a hoax, there is no scientific consensus that it is human caused, or that it is too expensive to address. These rules are inconsistent with rules of climate experts. Climate messages may also conflict with self-generated rules. For example, people may state rules such as, “There is nothing I can do to stop climate change,” which conflicts with rules by climate scientists about climate-change mitigation.
When rules conflict, the control exerted by one rule may supersede the other. One rule may be more consistent with a listener’s prior learning history, or the listener may have experienced greater reinforcement for following rules from one of the speakers. Effects of conflicting rules also may sum algebraically (see discussion on multiple control in Skinner, 1953), resulting in a weaker response to the more influential rule than if presented alone. A study by van der Linden, Leiserowitz, Rosenthal, and Maibach (2017) measured belief in the scientific consensus on climate change in a pre-post design. They found that in participants given a consensus message on climate change (a pie chart with the message, “97% of climate scientists have concluded that human-caused climate change is happening”), belief in the scientific consensus on climate change increased. In participants given a counter-message (that there was disagreement among scientists about the evidence for human caused climate change), belief decreased. In participants shown the consensus message and then counter-message, there was no change in belief, indicating that the effects of the two statement counteracted each other.
Conflicting rules by Republican and Democratic political leaders may have impacted climate-change concern. Research by Brulle and colleagues (2012) indicated that in the US, when partisanship of climate change decreased in 2006 and 2007 and Republicans and Democrats both supported addressing climate change, climate concern increased, but after 2008 when Republican anti- environmental voting increased, concern for climate change decreased.
The use of conflicting (or false) rules to minimize the impact of climate warnings has been explicit and purposeful. Leaders of fossil fuel corporations and associated think tanks designed and implemented messaging campaigns to cast doubt on climate science by using corporate-hired scientists who described the science as uncertain (Oreskes & Conway, 2010; Weber & Stern, 2011). Some of the same individuals who were hired to create doubt about the evidence linking tobacco to cancer and heart disease were involved. This misinformation about the scientific consensus reduced engagement. The distinction between actual scientific agreement on human-caused climate change (97%) and people’s belief in the scientific consensus (67%) is known as the consensus gap (Cook et al., 2018). The consensus gap is a result of the conflicting information from climate scientists and scientists who aim to create controversy and doubt.
News media have also been part of presenting conflicting information. Boykoff and Boykoff (2004) describe how the media, in an attempt to provide “balanced” reporting, gave equal time to climate-change deniers and climate scientists, despite the high consensus in the scientific community. They examined climate reporting by four prestigious US newspapers (from 1988 to 2002) and concluded that “[A]dherence to the norm of balanced reporting leads to informationally biased coverage of global warming. This bias, hidden behind the veil of journalistic balance, creates both discursive and real political space for the US government to shirk responsibility and delay action regarding global warming” (p. 134).
Economic contingencies were likely responsible for the false statements about climate-change. Skinner (1957) described lies as distorted (or impure) tacts. Distorted tacts are those which are not entirely under discriminative stimulus control, but rather are impacted by motivational variables or added generalized conditioned reinforcement. Tacts may be distorted by the operant behavior of listeners. In other words, people may lie because doing so alters the listeners response in a way that is reinforcing. Lies about causes of climate change by corporations may have been maintained by decreased support for climate-mitigation policy in voters and politicians. Tacts may also be distorted by generalized conditioned reinforcement from listeners (attention). Lies about the reality of climate change may have generated greater attention and approval from listeners than climate facts, perhaps because the topic or solutions are aversive.
Possible Solutions and Strategies for Change
Consensus messaging is one method of overcoming the weakening effect of misinformation about scientific consensus on climate change on climate change action (Cook et al., 2018). For example, van der Linden, Leiserowitz, and Maibach (2019) found that telling participants that “97% of climate scientists have concluded that human-caused global warming is happening” increased belief and worry about climate change, and support for action. Van der Linden, Leiserowitz, Feinberg, and Maibach (2015) argue that belief in scientific agreement is, a “gateway belief” that underlies public opinion about climate change and influences support for action (but see Hornsey & Fielding, 2020). Using satire and humor may also increase belief. A study by Brewer and McKnight (2017) found that participants who watched a satirical video segment from Last Week Tonight with John Oliver, “A statistically representative climate change debate,” in which three climate change skeptics debated 97 climate scientists (instead of inaccurate balanced reporting) showed an increase in belief in global warming.
Another proposed solution to conflicting messages is to warn people that they may be told lies. This approach has been described as inoculation (van der Linden, Leiserowitz, Rosenthal, & Maibach, 2017). Van der Linden et al. showed that when participants were given a consensus message, warned that politically motivated groups were trying to convince them that there was no agreement, and then shown a countermessage (that there was no scientific consensus), belief in scientific consensus on climate change still increased.
7. Lack of Speaker Credibility
Speaker credibility will influence rule following. Speaker credibility may be influenced by factors such as the speaker’s expertise and trustworthiness (e.g., Wiener & Mowen, 1986). Skinner (1957) notes that a listener’s belief in a speaker will “vary between speakers (to reflect the listener’s judgment of the speaker’s accuracy, honesty, and so on)” ( p. 88). Considerable research in communication has shown that source credibility can influence persuasive messaging (Pornpitakpan, 2004).
Barnes-Holmes et al. (2001) argue that credibility of a speaker can be established directly through reinforcing experiences, but also indirectly via generalization from similar speakers, or through verbal processes, such as through rules given by others. For example, simply noting a person’s political affiliation may impact credibility. Similarly, Törneke et al. (2010) argued that “The speaker’s credibility can be based on the listener’s actual experiences of following rules uttered by this speaker, or on derived stimulus functions. An example of the former would be following a piece of advice from a lifelong partner or close friend who has earlier given counsel that was helpful. An example of the latter is the way in which we normally follow rules given by various experts or, for that matter, when we do not follow this type of rule because we follow the rule ‘So-called experts are usually wrong’” (p. 117).
Research has shown that the experience following the rule of a speaker can influence rule following. For example, Hackenberg and Joker (1994) found that participants followed rules when the instructed response pattern maximized reinforcement, and continued when instructions were made increasingly inaccurate, but that rule following eventually decreased when benefits decreased. An example of possible generalization comes from a study by Penner et al. (1973) that found that when the experimenter appeared more competent, rule following increased. Hale and colleagues (2018) exposed participants to speakers who described themselves in ways indicating that they were reliable or irresponsible. They found that research participants were more likely to follow the advice of the speaker described as reliable, likely because of prior experience following rules from similar speakers.
Public trust in the risk communicator is a critical determinant of responses to risk (e.g., Slovic, 1999). Thus, the credibility of climate scientists and climate communicators can determine the impact of climate-rules. Trust is especially relevant for communicating climate change because public understanding comes from verbal reports (Weber, 2010). In support of this, Malka et al. (2009) found that the relation between knowledge and concern about global warming depended upon trust in scientists, and political party identification. In the US, people report distrust in climate scientists, particularly Republican voters, and therefore are less likely to believe their warnings (e.g., Hamilton et al., 2015).
Part of the polarization of climate-change in the US has been attributed to people following “elite” cues—rules and actions of group leaders. These people may serve as trustworthy rule sources. For example, Republican voters may trust rules from Republican political leaders. A study by Brulle and colleagues (2012) reported that, “The most important factor in influencing public opinion on climate change, however, is the elite partisan battle over the issue. The two strongest effects on public concern are Democratic Congressional action statements and Republican roll-call votes, which increase and diminish public concern, respectively” (p. 185). Rinscheid et al. (2021) also showed that elite cues, endorsements of climate-mitigation policy, influenced policy support; when cues came from a trusted source (political leaders from the respondent’s political party), climate policy support increased, but when cues came from an untrusted source (political leaders from the opposite party), support for climate policy decreased. This finding supports suggestions by Merkley and Stecula (2018) that Republican party members formed negative opinions on climate science because Democrat elites supported it.
Possible Solutions and Strategies for Change
Many have suggested that to enhance the efficacy of climate-change messages, messages should come from sources the audience trusts (e.g., CRED, 2009). Brewer and Ley (2013) found that science media (T.V., websites) were trusted more than news sources. Medical professionals can also be a trusted sources on health impacts of climate change (e.g., Romanello et al., 2021). Messages could come from speakers whose political ideologies are congruent with the audience. Bolsen et al. (2019) found that Republicans were more likely to view climate change as a national security threat when that threat was communicated by either Republicans or military leaders (trusted speakers). When the messages came from Democrats, they showed an increase in belief that climate change was a hoax (see also Motta et al., 2021). Faith-based groups may trust messages from religious leaders or from speakers with shared religious values. Hayhoe, Bloom, and Webb (2019) found that a lecture on climate change framed from an evangelical Christian perspective shifted climate opinions in undergraduate students at evangelical institutions.
Celebrities, such as Leonardo DiCaprio, as role models, may function as trusted sources of climate information (i.e., a source of stimulus control). Greta Thunberg, a climate activist who began protesting inaction on climate change when she was 15 years old, gained media attention with the “Fridays for Future” campaign and inspired further climate activism (see, for example, Sabherwal et al., 2021). Several documentaries have featured celebrities with the goal of encouraging climate action (e.g., Years of Living Dangerously, Before the Flood). After reviewing the literature on celebrity endorsements, however, Olmedo et al. (2020) argued that better evaluations of the efficacy of celebrity involvement in environmental campaigns are needed.
A challenge of this messaging strategy is determining who is a trusted source for a specific audience. Bolsen and Shapiro (2018) argue that one solution to source credibility is to bring together scientists and communicators who hold diverse values and ideologies to communicate climate change facts. Others have recommended that climate communicators build trust through social interactions, such as by being transparent, making themselves vulnerable, and acknowledging limitations (Goodwin & Dahlstrom, 2014).
8. Implausibility of Rule and Inconsistency with Prior Learning
Törneke (2010) noted that rules may not be followed even if understood because “…the rule is incoherent or contradictory in relation to the listener’s learning history” (p. 117). Barnes-Holmes et al. (2001) describe this effect on rule following as the plausibility of the rule. Grant and Forrest (2020) make a similar point about stories, arguing that the stories that are most influential are those that have coherence (internally consistent) and fidelity (similar to other stories established as true).
Plausibility of a rule may be described as people’s belief in a rule. Belief stems from a history of reinforcement. Skinner (1957) has noted that “Frequency of effective action accounts in turn for what we may call the listener’s ‘belief’-the probability that he will take effective action with respect to a particular verbal stimulus. In general this will vary between speakers . . . and between responses (depending upon the plausibility of the response in connection with the rest of a given situation)” (p. 88).
Climate warnings may fail to generate action because warnings or solutions are implausible, i.e., they contrast with prior learning experiences. For example, people may disbelieve climate change warnings that describe catastrophic events such as mass extinction, 30 m sea level rise, or civilization collapse, because they contrast with everyday experiences. Similarly, rules about global warming may be inconsistent with observations of local weather patterns which show little obvious change.
Denial messages may have been effective because, in some verbal communities, climate warnings likely oppose reinforced stimulus associations (relational networks) and stimulus functions. A common finding from climate-change risk perception research is that individuals who identify as Democrats show more concern about climate change than Republicans (e.g., Hornsey & Fielding, 2020; Leiserowitz et al., 2021b). For Republicans, climate-rules may conflict with existing stimulus associations. For example, if a political verbal community has established opposing relations between “free-markets” (as good) and “government regulation” (as bad), then climate change warnings, which are associated with calls for increased government regulations, are also likely to motivate avoidance (cf. Lewandowsky et al., 2013). Similarly, if the word “tax” is equated with economic harm, then “carbon taxes” are likely to be rejected.
Environmental psychologists investigating climate-change perceptions have identified multiple ideological factors (cultural values) that are correlated with climate skepticism, including individualistic and hierarchical values, and free-market ideologies (e.g., Lewandowsky et al., 2013; and see Hornsey & Fielding, 2020). Such beliefs and attitudes may be understood as verbal responses shaped by the reinforcing practices of verbal communities (Guerin, 1994). Climate rules that contrast with learned verbal associations are unlikely to be followed. In a study investigating the relationship between climate change skepticism and beliefs across different countries, Hornsey et al. (2018; and see Hornsey & Fielding, 2020) found that only in some countries (namely Brazil, Australia and Canada and most strongly, the US) was climate skepticism correlated with ideologies such as political conservatism, individualistic values (individuals over society), hierarchical values (hierarchical over egalitarian power structures), free-market ideologies, and belief in conspiracies. Hornsey et al. (2018) found that the relationship between climate-change skepticism and ideology was most pronounced in countries with considerable fossil fuel emissions, suggesting a possible influence on verbal behavior of associated economic costs of addressing climate change. A study by McCright and colleagues (2013) indicated that partisan rejection of science was not generalized to all science, but only to science that warned of negative impacts of economic development. Although climate skepticism is linked to contrasting verbal relations in political conservatives, research has shown that in the US, both liberals and conservatives deny scientific information that contrasts with their prior beliefs (Washburn & Skitka, 2018; and see Ditto et al., 2019).
The implausibility of addressing climate change through individual action may hinder action. That is, if people may have learned that climate change is a global problem requiring global solutions, then rules about actions individuals can take to address climate change may be doubted. For members of religious communities who are told that nature is controlled by supernatural forces, rules about human interventions also may be viewed as implausible. Environmental psychologists have discussed such barriers to climate action in terms of “self-efficacy” or “perceived behavioral control” (e.g., Gifford, 2011). If people do not believe that their actions will be effective, they will not act.
Climate-relevant rules which imply loss of specific valued reinforcers may also be rejected. For example, in verbal communities in which gasoline cars are highly valued, rules about replacing gasoline cars with electric cars may fail to influence behavior. A study by Vainio and Hartikainen (2018) found that messages about climate and health benefits for consuming more plant-based meat were ineffective in those with positive beliefs about red meat. One might also predict that people who rely on fossil-fuel revenue for income would be more likely to reject climate change science and climate-change mitigation strategies. Consistent with this, a survey of mining company employees and local government employees in Australia found that government employees had greater concern and belief in climate change (Loechel et al., 2013).
Possible Solutions and Strategies for Change
Climate rules could be targeted to audiences in such a way that the rules align with prior learning. One strategy is context framing (e.g., CRED, 2009). For example, in groups for whom “doomist” messages of climate change are implausible, messages that frame climate-change mitigation as “insurance” or “risk management” may be more impactful (see discussion in Stoknes, 2015). Rules that use words with more positive associations may also be effective. Hardisty et al. (2009) found that framing a carbon tax as an “offset” increased climate-friendly purchase preferences in Republicans and Independents.
As discussed above, scholars have argued that climate communications may be more effective if messages correspond rather than conflict with existing values (e.g., CRED, 2009; Hayhoe, 2021; Hornsey & Fielding, 2020; Stoknes, 2015). In the US, climate messages could be designed to align with liberal or conservative values, depending on the audience. For example, although climate warnings about dire ecological impacts of global temperature rise may be viewed as implausible in some conservative group, rules about economic benefits of green energy transitions may be persuasive. Emphasizing free-market solutions to reducing GHG emissions might be effective in groups who oppose government regulation of industry. Rules about the health impacts of addressing GHG emissions might be motivating across broad audiences. Fortunately, in the US, climate skepticism has decreased across time, and Leiserowitz et al. (2022) have reported that in the latest US poll to date only 9% of responders were categorized as dismissive of climate change. However, they also find that 29% still say global warming is either “not too” (15%) or “not at all” (15%) personally important (Leiserowitz et al., 2021a).
9. Insufficient Reinforcement for Rule Following
Rule following will depend on the short-term social and nonsocial consequences of compliance. Laboratory studies have shown for example, that rule following is influenced by the relative reinforcement for following or not following rules (e.g., Galizio, 1979; Hackenberg & Joker, 1994), and reinforcement for compliance and punishment (response cost) for noncompliance (Fox & Pietras, 2013; Nergaard & Couto, 2021). Rule following may be unlikely if reinforcement for compliance is relatively small in magnitude, probabilistic, or delayed, or if punishment for noncompliance is small or probabilistic, or has no specified time of occurrence (see Malott, 2010). Furthermore, rule following may be unlikely if behavior has high costs/effort, or produces social disapproval. Thus, climate warnings and rules for climate solutions may fail to generate sufficient action because the reinforcement contingencies do not support responding, or because they support opposing actions.
Contingencies impacting responses to climate warnings may be analyzed at both the individual and group (or organizational) level. Lack of rule following by individuals may occur because contingencies may be insufficient to support rule following. As described above, rule following may be maintained by nonsocial (tracking) and/or social (pliance) consequences. Decisions to following rules about taking public transportation to work, for instance, may vary as a function of reinforcer delays, effort, and reinforcer quality for each transportation mode. For people who enjoy red meat, rules about reducing meat consumption may be ignored if plant-based meat substitutes are lower-valued reinforcers. Financial barriers may limit adoption of energy-saving technologies (Gardner & Stern, 2008). For many consumers, solar panels, energy-saving appliances, or electric cars are unaffordable. Social reinforcers for compliance with climate mitigation strategies, such as praise for switching to renewable energy, carpooling, or diet shifts may be lacking or insufficient compared to the effort/costs. Alternatively, social disapproval for such actions may suppress rule following, such as in verbal communities whose values statements (about personal freedoms, automobiles, or red meat) conflict with climate-change mitigation actions.
The physical infrastructure, economic systems, and social systems may create barriers to climate action. For example, the sprawl of cities and communities create travel requirements. Renewable energy sources may be unavailable. Limited housing or food options may restrict sustainable choices. Social and economic contingencies may encourage high levels of resource consumption and energy use. Individuals in developed countries may have a considerable reinforcement history for GHG-producing behaviors (especially in wealthier groups). Whitmarsh (2009) has noted that energy consumption patterns are habitual and (at least in developed countries) an integral part of people’s lives. Rules derived from these contingencies likely also play a role in maintaining harmful behaviors.
Contingencies operating at the group level have been described as metacontingencies (Glenn, 2004; Glenn et al., 2016). A metacontingency describes a relationship between (a) interacting contingencies of individuals (interlocking behavioral contingencies, IBCs) functioning as a unit and yielding a product, and (b) environmental conditions (or the product itself) which influences the future probability of the IBCs. Climate warnings may be ineffective in the context of metacontingencies that favor practices that oppose climate action.
For example, consider responses by fossil-fuel corporations to warnings by climate science to cut GHG emissions. By some estimates, to keep within the a 1.5 °C carbon budget, by 2050, 60% of oil and fossil methane gas and 90% of coal must remain in the ground (Welsby et al., 2021). Yet, fossil-fuel production is a billion-dollar industry: In 2021, 28 of the largest oil and gas companies made $183.9 bn in profits (Milman, 2022). Corporations that halt operations will face enormous losses, as will their investors. The economics of fossil fuel extraction illustrate why actions of oil and gas companies and related industries are resistant to efforts to address climate change. Reductions in fossil fuel production are unlikely to happen without changes to these metacontingencies.
Rapid declines in fossil fuel production may have economic repercussions for national economies (Halttunen et al., 2022), which may dissuade politicians from acting. That is, political actions by citizens and corporations faced with economic losses from government climate policies may function as a metacontingency selecting political actions of governments that are too incremental or unsubstantial to adequately address the climate crisis.
Transitioning to green energy may also be cost-prohibitive for some communities or countries. Under the Paris Agreement, wealthy countries pledged to provide financial assistance to developing countries to aid with green energy development (and climate adaptation) given its costs, although they have not yet met their commitments (Timperley, 2021). Few contingencies at the international level support compliance with such pledges.
Possible Solutions and Strategies for Change
Contingencies of reinforcement for climate-relevant behaviors need to be aligned with recommendations of energy and climate experts to increase individual actions and group practices (cultural behaviors) that limit global warming. That is, reinforcers for following climate-relevant rules and punishers for breaking them need to be established.
Changing contingencies for individuals and households can have significant impacts on GHG emissions. In what they call the “behavioral wedge,” Dietz and colleagues (2009) argued that 7.4% of national emissions could be reduced by behavioral interventions at the household level (home weatherization, using fuel efficient vehicles, changing driving behaviors, and using energy-saving appliances). To overcome financial barriers to energy-use reduction, Gardner and Stern (2008) proposed a short list of curtailment (rather than purchase) actions that individuals could take to reduce carbon emissions that involved little cost. Examples include carpooling, turning down the water heater temperature, line-drying clothes, etc.
Applied behavior analysis, community-based social marketing, and environmental psychology have developed successful interventions for changing environmentally relevant behavior (see Abrahamse et al., 2007; Cone & Hayes, 1980; Lehman & Geller, 2004; McKenzie-Mohr, 2011; McKenzie-Mohr, Lee, Schultz, & Kotler, 2011; Steg & Vlek, 2009; Van Leuvan et al., 2022). Strategies include increasing reinforcement (e.g., feedback and incentives) and social reinforcement (praise) for pro-environmental behaviors and creating “nudges” that decrease effort and cost for those actions. Commitment responses or goal-setting have been used to create conditioned reinforcers and punishers. Social norms have been used to signal and motivate pro-environmental action. Schneider and Sanguinetti (2021) describe a wide variety of learning principles that are applicable to climate-relevant behavior change, including shaping and intermittent reinforcement schedules. Better dissemination of these strategies is needed, however, to produce greater levels of behavior change (see Biglan, 1995). Communication strategies using value-based messaging could be implemented along with contingency manipulations to motivate engagement with interventions and maintain public support.
Although the behavioral sciences have identified effective methods to change individual and household behaviors by manipulating individual-level reinforcement contingencies, altering conditions responsible for the replication of contingencies across a group, contingencies that yield problematic macrobehavior, may produce large-scale behavior change (Biglan & Glenn, 2013). The distinction between individual versus replicated contingencies is what Malagodi and Jackson (1989) describe as troubles (problems for individuals) versus issues (contingencies experienced by many across society). Analyses are needed of the environments that support macrobehaviors that generate GHGs and metacontingencies that influence problematic group behavior, and those environments and metacontingencies could be altered through a process of intentional cultural design (e.g., Biglan & Glenn, 2013).
For example, carbon pricing or carbon taxes have been proposed to decrease GHG emissions. Such laws can change macrobehaviors and metacontingencies by increasing costs for consumers of fossil-fuel use and lowering profits for corporations that produce fossil fuels and generate GHG emissions. Thus, they can promote the adoption of renewable energies. Laws could also be put in place to penalize actions (such as methane release or deforestation) that increase GHG concentration. Co-benefits of such actions, such as better air quality and improved public health, may help maintain such practices.
As with rule following more generally, monitoring and strict enforcement of laws will be needed to ensure compliance, and penalties for noncompliance need to be significant. As laboratory research has shown (e.g., Azrin & Holz, 1966), to maximize punishment efficacy, punishment should be certain and large in magnitude. Rules describing monetary penalties for methane emissions from gas wells or illegal deforestation, for example, are unlikely to be followed if there is little oversight or if the fine is small (for an analysis of punishment for environmental crimes, see Lynch et al., 2016). Similarly, carbon pricing may be ineffective if the price is too low.
To maintain public support and prevent counterpliance, policies that establish penalties for GHG emissions must be fair and equitable (see the Citizens’ Climate Lobby Carbon Fee and Dividend, https://citizensclimatelobby.org). Policies and regulations also could be coupled with strategies to help employees in fossil-fuel and related industries minimize losses. Public participation in policy development may be important.
Contingencies that promote pro-climate actions across a culture could be identified and manipulated. Such manipulations may be less likely to produce counterpliance. Examples include incentives for organizations to switch to clean energy or to develop renewable-energy technologies, and subsidies for household or organizational purchases of electric appliances can change macrobehaviors. Infrastructural investments, such as for community solar or public transportation, could lower costs and effort and increase their adoption and use. Griffith (2022) has argued that electrifying everything through renewable energy will simplify the behavior changes needed to reduce emissions and allow for rapid reductions in GHGs. Thus, policies could subsidize and support widespread electrification. Unchecked consumption and continuous economic growth are unsustainable (see Raworth, 2017) and contribute to the degradation of the climate. Thus, contingencies are also needed to support cultural behaviors that do not involve buying or consuming goods (see Grant, 2010). Again, messaging campaigns employing augmentals that alter the value of climate-relevant actions could work in conjunction with those contingency changes to establish new verbal practices that support pro-climate macrobehaviors (see examples in Biglan & Glenn, 2013).
Although global institutions may be able to create rules and create infrastructures that alter contingencies for climate action that are broad in scope and influence, to date, progress by such institutions has been slow. Thus, Ostrom (2009) has argued for a polycentric approach in which smaller organizational units act at various scales. Their cumulative effect could yield significant reductions in GHG emissions. For example, smaller political and organizational units (e.g., communities, county and state governments, universities, corporations), could create rules for guiding climate action and establish contingencies supporting pro-climate behaviors. Groups with practices that support collaboration and coordination with related groups may be especially effective at producing change (see Wilson et al., 2013).
Leaders of organizations have an important role in changing pro-climate behaviors. They can help construct rules that guide climate-relevant actions of organizations (e.g., mission statements), and ensure that organizational actions are socially responsible (see Alavosius et al., 2016; Houmanfar et al., 2015). Promoting prosocial decisions in organizational leaders will require analyzing contingencies that impact their decision making (Houmanfar et al., 2015).
Behavior analysts also can have an important role in the process of cultural change. They can contribute to efforts to promote climate action by working with organizations to analyze nonverbal and verbal contingencies influencing climate-relevant behaviors, create measurement systems, and empirically evaluate interventions designed to change behavior on a broad scale (e.g., Biglan, 1995; Biglan, 2016; Cihon & Mattaini, 2020; Alavosius & Houmanfar, 2020).
An example in the United States of a government policy that creates incentives and penalties for climate-relevant individual behaviors and organizational practices is the Inflation Reduction Act (Inflation Reduction Act, 2022). Among other actions, this legislation establishes incentives (e.g., tax credits or rebates) for individual consumers who make green purchases (e.g., heat pumps, electric vehicles, solar panels), grants to communities for carbon reduction plans, as well as incentives for companies to construct carbon capture technologies and to produce renewable energy, including in underserved communities. The legislation also adds a penalty for methane emissions by oil and gas companies in the form of a methane fee that increases across time. By changing individual contingencies and metacontingencies across the nation, laws such as this can produce large shifts in climate-relevant actions.
A challenge to changing contingencies through laws and policies, such as a carbon price, are actions of powerful special-interest groups who are able to sway governmental climate policy. Altering those dynamics and motivating government action may require political mobilization and activism. Behavior analysts (e.g., Ardilla Sánchez et al., 2020; Ardila Sánchez et al., 2020; Mattaini, 2013) have described how the science of behavior analysis can be applied to collective action, and how behavior analysts can become involved in collective action to promote the application of the science to societal problems, including climate change. Greta Thunberg has argued, “Today we use 100 million barrels of oil every single day. There are no politics to change that. There are no rules to keep that oil in the ground. So we can’t save the world by playing by the rules because the rules have to be changed. Everything needs to change and it has to start today” (2018, 10:29).
Conclusion
As the world prepares for the 6th IPCC report, it is uncertain whether the significant efforts of its authors will finally generate necessary climate action. If past reports are an indication though, it may not. Researchers from behavioral and social sciences have sought to understand why climate communication is not more effective at promoting necessary public engagement. This work has produced valuable insights and strategies for action, and shows that effective climate communication requires attention to psychological variables. The approach in the present paper here has been to highlight some of that work and, following Newsome and Alavosius (2011), show how diverse literatures can be integrated through the framework provided by verbal behavior and rule-governed behavior. It is hoped that such an approach might clarify factors impacting the human response to climate warnings and aid the development of new strategies for changing climate-relevant behaviors.
As Weber (2010) concludes, direct experience with climate catastrophes will likely motivate action on climate change, but a better strategy is to try to avoid those dire consequences before they occur. Avoiding worse-case climate-change scenarios will therefore require, among many other strategies, rules and verbal stimuli that effectively guide and motivate climate action, and reinforcement contingencies that support such behavior.
The author would like to thank John Esch, Barb Esch, Bob Dlouhy, and Wayne Fuqua, for their helpful conversations on verbal behavior and the contributions of behavior analysis to addressing climate change.
Data availability
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.
Declarations
Conflict of interest
The author declares no conflict of interest.
==== Refs
References
#BirdsAreNotReal. (2022, March 1). I've trusted and believed the climate science for three decades, for the last decade and a half I've been asking, so what do you want ME to do about it? [Comment on the article "These climate scientist are fed up and ready to go on strike," R. Zhong]. New York Times, online version). https://www.nytimes.com/2022/03/01/climate/ipcc-climate-scientists-strike.html
Abrahamse W Steg L Vlek C Rothengatter T The effect of tailored information, goal setting, and tailored feedback on household energy use, energy-related behaviors, and behavioral antecedents Journal of Environmental Psychology 2007 27 4 265 276 10.1016/j.jenvp.2007.08.002
Aklin M Urpelainen J Perceptions of scientific dissent undermine public support for environmental policy Environmental Science & Policy 2014 38 173 177 10.1016/j.envsci.2013.10.006
Alavosius MP Houmanfar RA Cihon TM Mattaini MA Global warming: Analysis of behavior and organizational practices on climate change Behavior science perspectives on culture and community 2020 Springer 221 256
Alavosius MP Newsome WD Houmanfar R Biglan A Hayes S Barnes-Homes D Zettle R Biglan A A functional contextualist analysis of the behavior and organizational practices relevant to climate change Handbook for contextual behavioral science 2016 Wiley 513 530
Allcott H Social norms and energy conservation Journal of Public Economics 2011 95 9–10 1082 1095 10.1016/j.jpubeco.2011.03.003
Ardila Sánchez, J. G., Cihon, T. M., Malott, M. E., Mattaini, M. A., Rakos, R. F., Rehfeldt, R. A., Richling, S. M., Roose, K. M., Seniuk, H. A., & Watson-Thompson, J. (2020). Collective editorial: Ten guidelines for strategic social action. Behavior and Social Issues, 29(1), 15–30. 10.1007/s42822-020-00038-8
Ardila Sánchez, J. G., Richling, S. M., Benson, M. L., & Rakos, R. F. (2020). Activism, advocacy, and accompaniment. In T. C. Cihon & M. A. Mattaini (Eds.), Behavior science perspectives on culture and community (pp. 413–436). Springer. 10.1007/978-3-030-45421-0_17
Arrhenius, S. (2012). On the influence of carbonic acid in the air upon the temperature on the ground. In B. McKibben (Ed.), The global warming reader (pp. 19–30). Penguin. (Reprinted from "On the influence of carbonic acid in the air upon the temperature on the ground," 1896, The Philosophical Magazine 41, 237–276.)
Azrin NH Holz WC Honig WK Punishment Operant behavior: areas of research and application 1966 Appleton-Century-Crofts 213 270
Ballew, M. T., Leiserowitz, A., Roser-Renouf, C., Rosenthal, S. A., Kotcher, J. E., Marlon, J. R., Lyon, E., Goldberg, M. H., & Maibach, E. W. (2019). Climate change in the American mind: Data, tools, and trends. Environment: Science and Policy for Sustainable Development, 61(3), 4–18.
Barnes-Holmes D Hora D Roche B Hayes SC Bissett RT Lyddy F Hayes SD Barnes-Holmes D Roche B Understanding and verbal regulation Relational frame theory: A post-Skinnerian account of human language and cognition 2001 Kluwer/Plenum 103 117
Barnes-Holmes Y Barnes-Holmes D McHugh L Hayes SC Relational frame theory: Some implications for understanding and treating human psychopathology International Journal of Psychology and Psychological Therapy 2004 4 355 375
Baum WM Rules, culture, and fitness The Behavior Analyst 1995 18 1 1 21 10.1007/BF03392688 22478201
Beach R Share J Webb A Teaching climate change to adolescents: Reading, writing, and making a difference 2017 Taylor & Francis
Biglan, A. (1995). Changing cultural practices: A contextualist framework for intervention research. Context Press.
Biglan A The need for a more effective science of cultural practices The Behavior Analyst 2016 39 1 97 107 10.1007/s40614-016-0051-z 27606190
Biglan A Glenn S Madden GJ Dube WV Hackenberg TD Hanley GP Lattal KA Toward prosocial behavior and environments: Behavioral and cultural contingencies in a public health framework APA Handbook of Behavior Analysis, Vol. 2: Translating Principles Into Practice 2013 American Pscyhological Association 255 275
Blake J 0vercoming the 'value-action gap' in environmental policy: Tensions between national policy and local experience Local Environment 1999 4 3 257 278 10.1080/13549839908725599
Bolsen T Shapiro MA The US news media, polarization on climate change, and pathways to effective communication Environmental Communication 2018 12 2 149 163 10.1080/17524032.2017.1397039
Bolsen T Palm R Kingsland JT The impact of message source on the effectiveness of communications about climate change Science Communication 2019 41 4 464 487 10.1177/1075547019863154
Boykoff MT Boykoff JM Balance as bias: Global warming and the U.S. prestige press Global Environmental Change 2004 14 2 125 136 10.1016/j.gloenvcha.2003.10.001
Branch G Rosenau J Berbeco M Climate education in the classroom: cloudy with a chance of confusion Bulletin of the Atomic Scientists 2016 72 2 89 96 10.1080/00963402.2016.1145906
Brehm JW A theory of psychological reactance 1966 Academic Press
Brewer PR Ley BL Whose science do you believe? Explaining trust in sources of scientific information about the environment Science Communication 2013 35 1 115 137 10.1177/1075547012441691
Brewer PR McKnight J "A statistically representative climate change debate": Satirical television news, scientific consensus, and public perceptions of global warming Atlantic Journal of Communication 2017 25 166 180 10.1080/15456870.2017.1324453
Brulle RJ Carmichael J Jenkins JC Shifting public opinion on climate change: an empirical assessment of factors influencing concern over climate change in the US, 2002–2010 Climatic Change 2012 114 2 169 188 10.1007/s10584-012-0403-y
Budescu DV Broomell S Por HH Improving communication of uncertainty in the reports of the Intergovernmental Panel on Climate Change Psychological Science 2009 20 3 299 308 10.1111/j.1467-9280.2009.02284.x 19207697
Bulkeley H Common knowledge? Public understanding of climate change in Newcastle Australia. Public Understanding of Science 2000 9 3 313 10.1177/096366250000900301
Camilleri AR Larrick RP Hossain S Patino-Echeverri D Consumers underestimate the emissions associated with food but are aided by labels Nature Climate Change 2019 9 1 53 58 10.1038/s41558-018-0354-z
Center for Research on Environmental Decisions. (2009). The psychology of climate change communications: A guide for scientists, journalists, educators, political aides, and the interested public. New York. http://guide.cred.columbia.edu/pdfs/CREDguide_full-res.pdf
Cialdini RB Crafting normative messages to protect the environment Current Directions in Psychological Science 2003 12 4 105 109 10.1111/1467-8721.01242
Cialdini RB Kallgren CA Reno RR A focus theory of normative conduct: A theoretical refinement and reevaluation of the role of norms in human behavior Advances in experimental social psychology 1991 Academic Press 201 234
Cihon TM Mattaini MA Behavior science perspectives on culture and community 2020 Springer
Cihon, T. M., Cihon, J. H., & Bedient, G. M. (2016). Establishing a common vocabulary of key concepts for the effective implementation of applied behavior analysis. International Electronic Journal of Elementary Education, 9(2), 337–348.
Climate Action Tracker. (2021). Glasgow’s 2030 credibility gap: net zero’s lip service to climate action. https://climateactiontracker.org/publications/glasgows-2030-credibility-gap-net-zeros-lip-service-to-climate-action/
Cone JD Hayes SC Environmental problems/behavioral solutions 1980 Cambridge University Press
Cook, J., van der Linden, S., Maibach, E., & Lewandowsky, S. (2018). The consensus handbook. 10.13021/G8MM6P. Available at http://www.climatechangecommunication.org/all/consensus-handbook/
Corner A Clarke J Talking climate: From research to practice in public engagement 2016 Springer
Critchfield, T. S., Doepke, K. J., Epting, L. K., Becirevic, A., Reed, D. D., Fienup, D. M., ... & Ecott, C. L. (2017). Normative emotional responses to behavior analysis jargon or how not to use words to win friends and influence people. Behavior Analysis in Practice, 10(2), 97–106.
Cushman, J. H. (1998). Industrial group plans to battle climate treaty. New York Times, New York, NY, April 26, Section 1, p. 1.
De Vries G Public communication as a tool to implement environmental policies Social Issues and Policy Review 2020 14 1 244 272 10.1111/sipr.12061
De Vries G Rietkerk M Kooger R The hassle factor as a psychological barrier to a green home Journal of Consumer Policy 2020 43 2 345 352 10.1007/s10603-019-09410-7
Dietz T Bringing values and deliberation to science communication Proceedings of the National Academy of Sciences 2013 110 Supplement 3 14081 14087 10.1073/pnas.1212740110
Dietz T Gardner GT Gilligan J Stern PC Vandenbergh MP Household actions can provide a behavioral wedge to rapidly reduce US carbon emissions Proceedings of the National Academy of Sciences 2009 106 44 18452 18456 10.1073/pnas.0908738106
Ditto PH Liu BS Clark CJ Wojcik SP Chen EE Grady RH Celniker JB Zinger JF At least bias is bipartisan: A meta-analytic comparison of partisan bias in liberals and conservatives Perspectives on Psychological Science 2019 14 2 273 291 10.1177/1745691617746796 29851554
Fox AE Pietras CJ The effects of response-cost punishment on instructional control during a choice task Journal of the Experimental Analysis of Behavior 2013 99 3 346 361 10.1002/jeab.20 23408327
Fox AE Kyonka EG Searching for the variables that control human rule-governed “insensitivity” Journal of the Experimental Analysis of Behavior 2017 108 2 236 254 10.1002/jeab.270 28776677
Fretz EJ Climate change across the curriculum 2015 Lexington books
Fryling M Rehfeldt RA Tarbox J Hayes LJ (Eds.). Applied behavior analysis of language and cognition: Core concepts and principles for practitioners 2020 New Harbinger Publications
Galizio, M. (1979). Contingency-shaped and rule-governed behavior: Instructional control of human loss avoidance. Journal of the Experimental Analysis of Behavior, 31, 53–70. 10.1901/jeab.1979.31-53
Gardner L Leshner G The Role of Narrative and Other-Referencing in Attenuating Psychological Reactance to Diabetes Self-Care Messages Health Communication 2016 31 6 738 751 10.1080/10410236.2014.993498 26528578
Gardner GT Stern PC The short list: The most effective actions U.S. households can take to curb climate change Environment: Science and Policy for Sustainable Development 2008 50 12 25
Geiger N Swim JK Climate of silence: Pluralistic ignorance as a barrier to climate change discussion Journal of Environmental Psychology 2016 47 79 90 10.1016/j.jenvp.2016.05.002
Geiger, N., Swim, J. K., & Fraser, J. (2017a). Creating a climate for change: Interventions, efficacy and public discussion about climate change. Journal of Environmental Psychology, 51, 104–116.
Geiger, N., Swim, J. K., Fraser, J., & Flinner, K. (2017b). Catalyzing public engagement with climate change through informal science learning centers. Science Communication, 39(2), 221–249.
Geller ES It takes more than information to save energy American Psychologist 1992 47 6 814 815 10.1037/0003-066X.47.6.814
Gifford R The dragons of inaction: psychological barriers that limit climate change mitigation and adaptation American Psychologist 2011 66 4 290 10.1037/a0023566 21553954
Glenn SS Individual behavior, culture, and social change The Behavior Analyst 2004 27 133 151 10.1007/BF03393175 22478424
Glenn SS Malott ME Andery MAPA Benvenuti M Houmanfar RA Sandaker I Todorov JC Tourinho EZ Vasconcelos LA Toward consistent terminology in a behaviorist approach to cultural analysis Behavior and Social Issues 2016 25 1 11 27 10.5210/bsi.v25i0.6634
Goldberg MH Gustafson A Rosenthal SA Leiserowitz A Shifting Republican views on climate change through targeted advertising Nature Climate Change 2021 11 7 573 577 10.1038/s41558-021-01070-1
Gonzales MH Aronson E Costanzo MA Using social cognition and persuasion to promote energy conservation: a quasi-experiment 1 Journal of Applied Social Psychology 1988 18 12 1049 1066 10.1111/j.1559-1816.1988.tb01192.x
Goodwin J Dahlstrom MF Communication strategies for earning trust in climate change debates Wiley Interdisciplinary Reviews: Climate Change 2014 5 1 151 160
Grant LK The veils of Clio: Dimensions of a behavioral narratology The Analysis of Verbal Behavior 2007 23 1 57 69 10.1007/BF03393047 22477381
Grant LK Sustainability: From excess to aesthetics Behavior and Social Issues 2010 19 1 7 47 10.5210/bsi.v19i0.2789
Grant L Forrest M Cihon TM Mattaini MA Can stories influence sustainable behavior? Behavior science perspectives on culture and community 2020 Springer 283 306
Green L Myerson J A discounting framework for choice with delayed and probabilistic rewards Psychological Bulletin 2004 130 5 769 10.1037/0033-2909.130.5.769 15367080
Greer RD Ross DE Verbal behavior analysis 2007 Pearson Education
Griffith S Electrify: an optimist’s playbook for our clean energy future 2022 MIT Press
Guerin B Attitudes and beliefs as verbal behavior The Behavior Analyst 1994 17 1 155 163 10.1007/BF03392661 22478181
Guterres, A. (2018, September 10). Antonio Guterres (Secretary-General) delivers speech on Climate Change and his vision for the 2019 Climate Change Summit [video] UN Web TV. https://media.un.org/en/asset/k1i/k1in57tn2m
Hackenberg TD Joker VR Instructional versus schedule control of humans' choices in situations of diminishing returns Journal of the Experimental Analysis of Behavior 1994 62 3 367 383 10.1901/jeab.1994.62-367 16812747
Hale J Payne ME Taylor KM Paoletti D De C Hamilton AF The virtual maze: A behavioural tool for measuring trust Quarterly Journal of Experimental Psychology 2018 71 4 989 1008 10.1080/17470218.2017.1307865
Halttunen K Slade R Staffell I What if we never run out of oil? From certainty of "peak oil" to "peak demand" Energy Research & Social Science 2022 85 102407 10.1016/j.erss.2021.102407
Hamilton LC Hartter J Saito K Trust in scientists on climate change and vaccines Sage Open 2015 5 3 2158244015602752 10.1177/2158244015602752
Hardisty DJ Weber EU Discounting future green: money versus the environment Journal of Experimental Psychology: General 2009 138 3 329 10.1037/a0016433 19653793
Hardisty D Johnson E Weber E Samu S Vaidyanathan R Chakravarti D Framing interacts with political affiliation to predict environmentally-relevant purchase preferences AP - Asia-Pacific Advances in Consumer Research 2009 Association for Consumer Research 379 380
Hart PS Feldman L Threat without efficacy? Climate change on US network news Science Communication 2014 36 3 325 351 10.1177/1075547013520239
Hart PS Feldman L The Impact of Climate Change–Related Imagery and Text on Public Opinion and Behavior Change Science Communication 2016 38 4 415 441 10.1177/1075547016655357
Harte C Barnes-Holmes D The status of rule-governed behavior as pliance, tracking and augmenting within relational frame theory: Middle-level rather than technical terms The Psychological Record 2022 72 1 145 158 10.1007/s40732-021-00458-x
Harte C Barnes-Holmes D Barnes-Holmes Y Kissi A The study of rule-governed behavior and derived stimulus relations: Bridging the gap Perspectives on Behavior Science 2020 43 361 385 10.1007/s40614-020-00256-w 32647787
Hase V Mahl D Schafer MS Keller TR Climate change in news media across the globe: An automated analysis of issue attention and themes in climate change coverage in 10 countries(2006–2018) Global Environmental Change 2021 70 102353 10.1016/j.gloenvcha.2021.102353
Hassol SJ Improving how scientists communicate about climate change Eos, Transactions American Geophysical Union 2008 89 11 106 107 10.1029/2008EO110002
Hassol SJ Torok S Lewis S Luganda P (Un)natural disasters: communicating linkages between extreme events and climate change WMO Bull 2016 65 2 2 9
Hawken P Drawdown: the most comprehensive plan ever proposed to reverse global warming 2017 Penguin Books
Hayes SC Wilson KG Some applied implications of a contemporary behavior-analytic account of verbal events The Behavior Analyst 1993 16 2 283 10.1007/BF03392637 22478159
Hayes SC Barnes-Holmes D Roche B Relational frame theory: A post-Skinnerian account of human language and cognition 2001 Plenum
Hayes, S. C., Zettle, R. D., & Rosenfarb, I. (1989). Rule-following. In S. C. Hayes (Ed.), Rule-governed behavior: Cognition, contingencies, and instructional control (pp. 191–220). Plenum.
Hayhoe K Saving us: a climate scientist's case for hope and healing in a divided world 2021 Simon and Schuster
Hayhoe D Bloom MA Webb BS Changing evangelical minds on climate change Environmental Research Letters 2019 14 2 024016 10.1088/1748-9326/aaf0ce
Heller K Solution search: climate change needs behavior change. Local solutions for a global challenge 2019 Rare
Hirsh JL Costello MS Fuqua RW Analysis of delay discounting as a psychological measure of sustainable behavior Behavior and Social Issues 2015 24 1 187 202 10.5210/bsi.v24i0.5906
Homar AR Cvelbar LK The effects of framing on environmental decisions: A systematic literature review Ecological Economics 2021 183 106950 10.1016/j.ecolecon.2021.106950
Horne PJ Lowe CF On the origins of naming and other symbolic behavior Journal of the Experimental Analysis of Behavior 1996 65 1 185 241 10.1901/jeab.1996.65-185 16812780
Hornsey MJ Fielding KS Understanding (and reducing) inaction on climate change Social Issues and Policy Review 2020 14 1 3 35 10.1111/sipr.12058
Hornsey MJ Harris EA Fielding KS Relationships among conspiratorial beliefs, conservatism and climate scepticism across nations Nature Climate Change 2018 8 7 614 620 10.1038/s41558-018-0157-2
Houmanfar RA Alavosius MP Morford ZH Herbst SA Reimer D Functions of organizational leaders in cultural change: Financial and social well-being Journal of Organizational Behavior Management 2015 35 1–2 4 27 10.1080/01608061.2015.1035827
Huber RA Anderson B Bernauer T Can social norm interventions promote voluntary pro environmental action? Environmental Science & Policy 2018 89 231 246 10.1016/j.envsci.2018.07.016
Hughes S Barnes-Holmes D Zettle RD Hayes SC Barnes-Holmes D Biglan A Relational frame theory: The basic account The Wiley handbook of contextual behavioral science 2016 Wiley-Blackwell 129 178
Inflation Reduction Act of 2022, H.R.5376 - 117th Congress (2021–2022). https://www.congress.gov/bill/117th-congress/house-bill/5376/text
IPCC Solomon S Qin D Manning M Chen Z Marquis M Averyt KB Tignor M Miller HL Summary for Policymakers Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change 2007 Cambridge University Press 1 18
IPCC. (2018). Summary for Policymakers. In Masson-Delmotte, V., P. Zhai, H.-O. Portner, D. Roberts, J. Skea, P.R. Shukla, A. Pirani, W. Moufouma-Okia, C. Pean, R. Pidcock, S. Connors, J.B.R. Matthews, Y. Chen, X. Zhou, M.I. Gomis, E. Lonnoy, T. Maycock, M. Tignor, and T. Waterfield (Eds.), Global Warming of 1.5°C. An IPCC Special Report on the impacts of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty (pp. 3–24). Cambridge University Press. 10.1017/9781009157940.001
IPCC Masson-Delmotte V Zhai P Pirani A Connors SL Pean C Berger S Caud N Chen Y Goldfarb L Gomis MI Huang M Leitzell K Lonnoy E Matthews JBR Maycock TK Waterfield T Yeleki O Yu R Zhou B Summary for Policymakers Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change 2021 Cambridge University Press 3 32
IPCC Pörtner H-O Roberts DC Poloczanska ES Mintenbeck K Tignor M Alegría A Craig M Langsdorf S Löschke S Möller V Okem A Summary for Policymakers Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change 2022 Cambridge University Press
Irmak C Murdock MR Kanuri VK When consumption regulations backfire: The role of political ideology Journal of Marketing Research 2020 57 5 966 984 10.1177/0022243720919709
Jackson ML Williams WL Hayes SC Humphreys T Gauthier B Westwood R Whatever gets your heart pumping: The impact of implicitly selected reinforcer-focused statements on exercise intensity Journal of Contextual Behavioral Science 2016 5 1 48 57 10.1016/j.jcbs.2015.11.002
Jarmolowicz DP Kahng S Ingvarsson ET Goysovich R Heggemeyer R Gregory MK Taylor SJ Effects of conversational versus technical language on treatment preference and integrity Intellectual and Developmental Disabilities 2008 46 3 190 199 10.1352/2008.46:190-199 18578577
Jay A Reidmiller DR Avery CW Barrie D DeAngelo BJ Dave A Dzaugis M Kolian M Lewis KLM Reeves K Winner D Reidmiller DR Avery CW Easterling DR Kunkel KE Lewis KLM Maycock TK Stewart BC Overview Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment, Volume II 2018 U.S. Global Change Research Program
Jennings N Fecht D De Matteis S Mapping the co-benefits of climate change action to issues of public concern in the UK: a narrative review The Lancet Planetary Health 2020 4 9 e424 e433 10.1016/S2542-5196(20)30167-4 32918888
Johnson AE Wilkinson KK All we can save: truth, courage, and solutions for the climate crisis 2021 One World/Ballantine
Jones MD Cultural characters and climate change: How heroes shape our perception of climate science Social Science Quarterly 2014 95 1 1 39 10.1111/ssqu.12043
Jones BA Rachlin H Social discounting Psychological Science 2006 17 4 283 286 10.1111/j.1467-9280.2006.01699.x 16623683
Jones C Hine DW Marks AD The future is now: Reducing psychological distance to increase public engagement with climate change Risk Analysis 2017 37 2 331 341 10.1111/risa.12601 26989845
Kaltenbacher M Drews S An inconvenient joke? A review of humor in climate change communication Environmental Communication 2020 14 6 717 729 10.1080/17524032.2020.1756888
Kaplan BA Reed DD Jarmolowicz DP Effects of episodic future thinking on discounting: Personalized age-progressed pictures improve risky long-term health decisions Journal of Applied Behavior Analysis 2016 49 1 148 169 10.1002/jaba.277 26679531
Kaplan BA Reed DD McKerchar TL Using a visual analogue scale to assess delay, social, and probability discounting of an environmental loss The Psychological Record 2014 64 2 261 269 10.1007/s40732-014-0041-z
Kaufman A Baron A Kopp RE Some effects of instructions on human operant behavior Psychonomic Monograph Supplements 1966 1 243 250
Kissi A Hughes S Mertens G Barnes-Holmes D De Houwer J Crombez G A systematic review of pliance, tracking, and augmenting Behavior Modification 2017 41 5 683 707 10.1177/0145445517693811 28776429
Kotcher J Feldman L Luong KT Wyatt J Maibach E Advocacy messages about climate and health are more effective when they include information about risks, solutions, and a normative appeal: evidence from a conjoint experiment The Journal of Climate Change and Health 2021 3 100030 10.1016/j.joclim.2021.100030
Lee VL Scientific knowledge as rules that guide behavior The Psychological Record 1985 35 2 183 192 10.1007/BF03394924
Lehman PK Geller ES Behavior analysis and environmental protection: Accomplishments and potential for more Behavior and Social Issues 2004 13 1 13 33 10.5210/bsi.v13i1.33
Leiserowitz A Climate change risk perception and policy preferences: The role of affect, imagery, and values Climatic Change 2006 77 1 45 72 10.1007/s10584-006-9059-9
Leiserowitz A Carman J Buttermore N Wang X Rosenthal S Marlon J Mulcahy K International Public Opinion on Climate Change 2021 Yale Program on Climate Change Communication and Facebook Data for Good
Leiserowitz, A., Maibach, E., Rosenthal, S., Kotcher, J., Carman, J., Neyens, L., Marlon, J., Lacroix, K., & Goldberg, M. (2021a). Climate change in the American mind, September 2021. Yale University and George Mason University. Yale Program on Climate Change Communication. Communication. https://climatecommunication.yale.edu/publications/climate-change-in-the-american-mind-september-2021/
Leiserowitz A Maibach E Rosenthal S Kotcher J Carman J Neyens L Marlon J Lacroix K Goldberg M Public support for climate action by the president and Congress is surging. Yale University and George Mason University 2021 Yale Program on Climate Change Communication
Leiserowitz A Maibach E Rosenthal S Kotcher J Neyens L Marlon J Carman J Lacroix K Goldberg M Global Warming's Six Americas, September 2021. Yale University and George Mason University 2022 Yale Program on Climate Change Communication
Lewandowsky S Oberauer K Gignac GE NASA faked the moon landing- therefore, (climate) science is a hoax: An anatomy of the motivated rejection of science Psychological Science 2013 24 5 622 633 10.1177/0956797612457686 23531484
Loechel B Hodgkinson J Moffat K Climate change adaptation in Australian mining communities: comparing mining company and local government views and activities Climatic Change 2013 119 2 465 477 10.1007/s10584-013-0721-8
Luntz F The environment: a cleaner, safer, healthier America 2002 Luntz Research
Lynas M Our final warning: Six degrees of climate emergency 2020 Harper Collins UK
Lynch MJ Barrett KL Stretesky PB Long MA The weak probability of punishment for environmental offenses and deterrence of environmental offenders: A discussion based on USEPA criminal cases, 1983–2013 Deviant Behavior 2016 37 10 1095 1109 10.1080/01639625.2016.1161455
Madden GJ Bickel WK Impulsivity: The behavioral and neurological science of discounting 2010 American Psychological Association
Maibach E Leiserowitz A Rosenthal S Roser-Renouf C Cutler M Is there a climate "spiral of silence" in America: March, 2016. Yale University and George Mason University 2016 Yale Program on Climate Change Communication
Maiella, R., La Malva, P., Marchetti, D., Pomarico, E., Di Crosta, A., Palumbo, R., Cetara, L., Di Domenico, A., & Verrocchio, M. C. (2020). The psychological distance and climate change: A systematic review on the mitigation and adaptation behaviors. Frontiers in Psychology, 11, 568899.
Malagodi EF Jackson K Behavior analysts and cultural analysis: Troubles and issues The Behavior Analyst 1989 12 1 17 33 10.1007/BF03392474 22478014
Malka A Krosnick JA Langer G The association of knowledge with concern about global warming: Trusted information sources shape public thinking Risk Analysis: An International Journal 2009 29 5 633 647 10.1111/j.1539-6924.2009.01220.x
Malott RW Rule-governed behavior and behavioral anthropology The Behavior Analyst 1988 11 2 181 203 10.1007/BF03392471 22478012
Malott RW I'll Save the World from Global Warming-Tomorrow: Using Procrastination Management to Combat Global Warming The Behavior Analyst 2010 33 2 179 10.1007/BF03392214 22532710
Mann ME The new climate war: The fight to take back our planet 2021 PublicAffairs
Mann ME Bradley RS Hughes MK Northern hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations Geophysical Research Letters 1999 26 6 759 762 10.1029/1999GL900070
Markolf, S.A., Azevedo, I.M.L., Muro, M. & Victor, D.G. (2020). Pledges and progress: Steps toward greenhouse gas emissions reductions in the 100 largest cities across the United States. Brookings. https://www.brookings.edu/wp-content/uploads/2020/10/FP_20201022_ghg_pledges_v4.pdf.
Markowitz EM Guckian ML Clayton S Manning C Climate change communication: Challenges, insights, and opportunities Psychology and Climate Change: Human perceptions, impacts, and responses 2018 Elsevier 35 63
Marshall G Don't even think about it: Why our brains are wired to ignore climate change 2015 Bloomsbury Publishing USA
Mattaini M Strategic nonviolent power: The science of satyagraha 2013 AU Press, Athabasca University
Matthew, T., & Watts, J. (2019). Revealed: The 20 firms behind a third of all carbon emissions. The Guardian. https://www.theguardian.com/environment/2019/oct/09/revealed-20-firms-third-carbon-emissions
McCright AM Dentzman K Charters M Dietz T The influence of political ideology on trust in science Environmental Research Letters 2013 8 4 044029 10.1088/1748-9326/8/4/044029
McEnteggart C A brief tutorial on acceptance and commitment therapy as seen through the lens of derived stimulus relations Perspectives on Behavior Science 2018 41 1 215 227 10.1007/s40614-018-0149-6 31976394
McKenzie-Mohr D Fostering sustainable behavior through community-based social marketing. American Psychologist 2000 55 5 531 537 10.1037/0003-066X.55.5.531 10842434
McKenzie-Mohr D Fostering sustainable behavior: An introduction to community-based social marketing 2011 New Society Publishers
McKenzie-Mohr D Lee N Schultz W Kotler P Social marketing to protect the environment: What works 2011 Sage
McKerchar TL Kaplan BA Reed DD Suggs SA Franck CT Discounting environmental outcomes: Temporal and probabilistic air-quality gains and losses Behavior Analysis: Research and Practice 2019 19 3 273
McKibben B The global warming reader: A century of writing about climate change 2012 Penguin
Melillo, J.M., Richmond, T.T.C., & Yohe, G.W. (Eds.) (2014). Climate change impacts in the United States: The third national climate assessment. U.S. Global Change Research Program. 10.7930/J0Z31WJ2
Merkley E Stecula DA Party elites or manufactured doubt? The informational context of climate change polarization Science Communication 2018 40 2 258 274 10.1177/1075547018760334
Michael J Establishing operations. The Behavior Analyst 1993 16 2 191 206 10.1007/BF03392623 22478146
Milman, O. (2022, May 13). Largest oil and gas producers made close to $100bn in first quarter of 2022. The Guardian. https://www.theguardian.com/business/2022/may/13/oil-gas-producers-first-quarter-2022-profits
Morton TA Rabinovich A Marshall D Bretschneider P The future that may (or may not) come: How framing changes responses to uncertainty in climate change communications Global Environmental Change 2011 21 1 103 109 10.1016/j.gloenvcha.2010.09.013
Moser SC Reflections on climate change communication research and practice in the second decade of the 21st century: what more is there to say? WIREs Climate Change 2016 7 3 345 369 10.1002/wcc.403
Motta M Ralston R Spindel J A call to arms for climate change? How military service member concern about climate change can inform effective climate communication Environmental Communication 2021 15 1 85 98 10.1080/17524032.2020.1799836
Myers TA Maibach E Peters E Leiserowitz A Simple messages help set the record straight about scientific agreement on human-caused climate change: The results of two experiments PloS one 2015 10 3 e0120985 10.1371/journal.pone.0120985 25812121
Nalau J Cobb G The strengths and weaknesses of future visioning approaches for climate change adaptation: A review Global Environmental Change 2022 74 102527 10.1016/j.gloenvcha.2022.102527
Nergaard SK Couto KC Effects of reinforcement and response‐cost history on instructional control Journal of the Experimental Analysis of Behavior 2021 115 3 679 701 10.1002/jeab.680 33634471
Newsome WD Alavosius MP Toward the Prediction and Influence of Environmentally Relevant Behavior: Seeking Practical Utility in Research Behavior and Social Issues 2011 20 1 44 71 10.5210/bsi.v20i0.3234
Nielsen KS Clayton S Stern PC Dietz T Capstick S Whitmarsh L How psychology can help limit climate change American Psychologist 2021 76 1 130 10.1037/amp0000624 32202817
Nisa CF Belanger JJ Schumpe BM Faller DG Meta-analysis of randomised controlled trials testing behavioural interventions to promote household action on climate change Nature Communications 2019 10 1 1 13 10.1038/s41467-019-12457-2
NOAA Global Monitoring Laboratory. (n.d.). Trends in atmospheric carbon dioxide. National Oceanic and Atmospheric Administration. Retrieved February 18, 2021, from https://www.esrl.noaa.gov/gmd/ccgg/trends/
NOAA National Centers for Environmental Information (NCEI) U.S. Billion-Dollar Weather and Climate Disasters. (2022). Retrieved November 22, 2022, from https://www.ncei.noaa.gov/access/monitoring/billions/. 10.25921/stkw-7w73
O’Neill SJ Smith N Climate change and visual imagery WIREs Climate Change 2014 5 73 87 10.1002/wcc.249
O'Hora D Barnes-Holmes D Instructional control: Developing a relational frame analysis International Journal of Psychology and Psychological Therapy 2004 4 2 263 284
O'Hora D Barnes-Holmes D Stewart I Antecedent and consequential control of derived instruction-following Journal of the Experimental Analysis of Behavior 2014 102 1 66 85 10.1002/jeab.95 24977339
Olmedo, A., Milner-Gulland, E. J., Challender, D. W., Cugniere, L., Dao, H. T. T., Nguyen, L. B., Nuno, A., Potier, E., Ribadeneira, M., Thomas-Walters, L., Wan, A. K. Y., Wang, Y., & Verissimo, D. (2020). A scoping review of celebrity endorsement in environmental campaigns and evidence for its effectiveness. Conservation Science and Practice, 2(10), e261.
O'Neill SJ Image matters: Climate change imagery in US, UK and Australian newspapers Geoforum 2013 49 10 19 10.1016/j.geoforum.2013.04.030
O'Neill SJ Boykoff M Niemeyer S Day SA On the use of imagery for climate change engagement Global Environmental Change 2013 23 2 413 421 10.1016/j.gloenvcha.2012.11.006
Oreskes N Conway EM Merchants of doubt: How a handful of scientists obscured the truth on issues from tobacco smoke to global warming 2010 Bloomsbury Publishing
Osbaldiston R Schott JP Environmental sustainability and behavioral science: Meta- analysis of proenvironmental behavior experiments Environment and Behavior 2012 44 2 257 299 10.1177/0013916511402673
Ostrom E A polycentric approach for coping with climate change 2009 The Word Bank
Palmer DC Verbal behavior: What is the function of structure? European Journal of Behavior Analysis 2007 8 2 161 175 10.1080/15021149.2007.11434280
Palmer DC Skinner BF Foreword Contingencies of reinforcement: A theoretical analysis 2012 BF Skinner Foundation
Pelaez M Dimensions of rules and their correspondence to rule-governed behavior European Journal of Behavior Analysis 2013 14 2 259 270 10.1080/15021149.2013.11434459
Penner LA Hawkins HL Dertke MC Obedience as a function of experimenter competence Memory & Cognition 1973 1 3 241 245 10.3758/BF03198103 24214552
Plutzer E McCaffrey M Hannah AL Rosenau J Berbeco M Reid AH Climate confusion among US teachers Science 2016 351 6274 664 665 10.1126/science.aab3907 26912845
Pornpitakpan C The persuasiveness of source credibility: A critical review of five decades' evidence Journal of applied social psychology 2004 34 2 243 281 10.1111/j.1559-1816.2004.tb02547.x
Presbie RJ Effects of instructions on human free-operant avoidance conditioning with white noise Psychological Reports 1970 27 3 895 898 10.2466/pr0.1970.27.3.895
Rachlin H Raineri A Cross D Subjective probability and delay Journal of the Experimental Analysis of Behavior 1991 55 2 233 244 10.1901/jeab.1991.55-233 2037827
Ranney MA Clark D Climate change conceptual change: Scientific information can transform attitudes Topics in Cognitive Science 2016 8 1 49 75 10.1111/tops.12187 26804198
Rare and California Environmental Associates. (2019). Changing behaviors to reduce U.S. emissions: Seven pathways to achieve climate impact. Rare. https://rare.org/wp-content/uploads/2019/07/Changing-behaviors-to-reduce-U.S.-emissions-digital.pdf
Raworth K Doughnut economics: seven ways to think like a 21st-century economist 2017 Chelsea Green Publishing
Reynolds-Tylus T Psychological reactance and persuasive health communication: A review of the literature Frontiers in Communication 2019 4 56 10.3389/fcomm.2019.00056
Rinscheid A Pianta S Weber EU What shapes public support for climate change mitigation policies? The role of descriptive social norms and elite cues Behavioural Public Policy 2021 5 4 503 527 10.1017/bpp.2020.43
Ripple, W. J., Wolf, C., Newsome, T. M., Galetti, M., Alamgir, M., Crist, E., . . . & 15,364 Scientist Signatories from 184 Countries. (2017). World scientists' warning to humanity: a second notice. BioScience, 67(12), 1026–1028.
Robertson DL Pelaez M Rules, rule-governed behavior, and organizational change in a large metropolitan research university Behavioral Development 2018 23 1 1 10.1037/bdb0000066
Roelfsema, M., van Soest, H. L., Harmsen, M., van Vuuren, D. P., Bertram, C., den Elzen, M., ,Höhne, N., Iacobuta, G., Krey, V., Kriegler, E., Luderer, G., Riahi, K., Ueckerdt, F., Després, J., Drouet, L., Emmerling, J., Frank, S., Fricko, O., Gidden, M., . . . Vishwanathan, S. S. (2020). Taking stock of national climate policies to evaluate implementation of the Paris Agreement. Nature Communications, 11(1), 1–12.
Romanello M McGushin A Di Napoli C Drummond P Hughes N Jamart L Kennard H Lampard P Rodriguez BS Arnell N Ayeb-Karlsson S The 2021 report of the Lancet Countdown on health and climate change: code red for a healthy future The Lancet 2021 398 10311 1619 1662 10.1016/S0140-6736(21)01787-6
Rung JM Madden GJ Experimental reductions of delay discounting and impulsive choice: A systematic review and meta-analysis Journal of Experimental Psychology: General 2018 147 9 1349 10.1037/xge0000462 30148386
Sabherwal A Ballew MT van Der Linden S Gustafson A Goldberg MH Maibach EW Kotcher JE Swim JK Rosenthal SA Leiserowitz A The Greta Thunberg Effect: Familiarity with Greta Thunberg predicts intentions to engage in climate activism in the United States Journal of Applied Social Psychology 2021 51 4 321 333 10.1111/jasp.12737
Schalkwijk M Energie besparen doe je nu: Eindrapportage campagne-effectonderzoek 2017 Kantar Public
Schlinger HD Separating discriminative and function-altering effects of verbal stimuli The Behavior Analyst 1993 16 1 9 23 10.1007/BF03392605 22478128
Schlinger HD Conditioning the behavior of the listener International Journal of Psychology and Psychological Therapy 2008 8 3 309 322
Schlinger H Blakely E Function-altering effects of contingency-specifying stimuli The Behavior Analyst 1987 10 1 41 45 10.1007/BF03392405 22477959
Schnaitter R Science and verbal behavior Behaviorism 1980 8 2 151 160
Schneider SM Sanguinetti A Positive reinforcement is just the beginning: Associative learning principles for energy efficiency and climate sustainability Energy Research & Social Science 2021 74 101958 10.1016/j.erss.2021.101958
Schneider-Mayerson M The influence of climate fiction: an empirical survey of readers Environmental Humanities 2018 10 2 473 500 10.1215/22011919-7156848
Shepardson DP Roychoudhury A Hirsch AS Teaching and learning about climate change 2017 Routledge
Sheppard SR Shaw A Flanders D Burch S Wiek A Carmichael J Robinson J Cohen S Future visioning of local climate change: a framework for community engagement and planning with scenarios and visualisation Futures 2011 43 4 400 412 10.1016/j.futures.2011.01.009
Shimoff E Catania AC Matthews BA Uninstructed human responding: Sensitivity of low-rate performance to schedule contingencies Journal of the Experimental Analysis of Behavior 1981 36 2 207 220 10.1901/jeab.1981.36-207 16812240
Sinaceur M Heath C Cole S Emotional and deliberative reactions to a public crisis Psychological Science 2005 16 3 247 254 10.1111/j.0956-7976.2005.00811.x 15733207
Skinner, B. F. (1953). Science and human behavior. Macmillan.
Skinner BF Verbal behavior 1957 Appleton-Century-Crofts
Skinner BF Contingencies of reinforcement: A theoretical analysis 1969 Appleton-Century Crofts
Skinner BF About behaviorism 1974 Knopf
Skinner BF An operant analysis of problem solving Behavioral and Brain Sciences 1984 7 4 583 613 10.1017/S0140525X00027412
Slovic P Trust, emotion, sex, politics, and science: Surveying the risk-assessment battlefield Risk Analysis 1999 19 4 689 701 10.1111/j.1539-6924.1999.tb00439.x 10765431
Sparkman G Walton GM Dynamic norms promote sustainable behavior, even if it is counternormative Psychological Science 2017 28 11 1663 1674 10.1177/0956797617719950 28961062
Spence A Poortinga W Pidgeon N The psychological distance of climate change Risk Analysis: An International Journal 2012 32 6 957 972 10.1111/j.1539-6924.2011.01695.x
Staats HJ Wit AP Midden CYH Communicating the greenhouse effect to the public: Evaluation of a mass media campaign from a social dilemma perspective Journal of Environmental Management 1996 46 2 189 203 10.1006/jema.1996.0015
Stapleton Alison Choosing not to follow rules that will reduce the spread of COVID-19 Journal of Contextual Behavioral Science 2020 17 73 78 10.1016/j.jcbs.2020.07.002 32834967
Steg L Vlek C Encouraging pro-environmental behaviour: An integrative review and research agenda Journal of Environmental Psychology 2009 29 3 309 317 10.1016/j.jenvp.2008.10.004
Stoknes PE What we think about when we try not to think about global warming: Toward a new psychology of climate action 2015 Chelsea Green Publishing
Tarbox J Campbell V Pio S Fryling MJ Rehfeldt RA Tarbox J Hayes LJ Rule-governed behavior and verbal regulation Applied behavior analysis of language and cognition: Core concepts and principles for practitioners 2020 Context Press
Thompson LG Climate change: The evidence and our options The Behavior Analyst 2010 33 2 153 170 10.1007/BF03392211 22532707
Thunberg G. (2018, December). School strike for climate - save the world by changing the rules. [video]. TED Conferences. https://youtu.be/EAmmUIEsN9A
Timperley J How to fix the broken promises of climate finance Nature 2021 598 400 402 10.1038/d41586-021-02846-3 34671142
Torneke N Learning RFT: An introduction to relational frame theory and its clinical application 2010 New Harbinger Publications
Trope Y Liberman N Construal-level theory of psychological distance Psychological Review 2010 17 2 440 10.1037/a0018963
United Nations Framework Convention on Climate Change (UNFCCC). (2015). Paris agreement. https://unfccc.int/sites/default/files/english_paris_agreement.pdf
Vainio A Irz X Hartikainen H How effective are messages and their characteristics in changing behavioural intentions to substitute plant-based foods for red meat? The mediating role of prior beliefs Appetite 2018 125 217 224 10.1016/j.appet.2018.02.002 29447995
van der Linden S Nisbet MC Schafer M Markowitz E Ho S O'Neill S Thaker J Determinants and measurement of climate change risk perception, worry, and concern The Oxford Encyclopedia of Climate Change Communication 2017 Oxford University Press 1 53
van der Linden SL Leiserowitz AA Feinberg GD Maibach EW The scientific consensus on climate change as a gateway belief: Experimental evidence PloS one 2015 10 2 e0118489 10.1371/journal.pone.0118489 25714347
van der Linden S Maibach E Leiserowitz A Improving public engagement with climate change: Five "best practice" insights from psychological science Perspectives on Psychological Science 2015 10 6 758 763 10.1177/1745691615598516 26581732
van der Linden S Leiserowitz A Rosenthal S Maibach E Inoculating the public against misinformation about climate change Global Challenges 2017 1 2 1600008 10.1002/gch2.201600008 31565263
van der Linden S Leiserowitz A Maibach E The gateway belief model: A large-scale replication Journal of Environmental Psychology 2019 62 49 58 10.1016/j.jenvp.2019.01.009
van Leuvan N Highleyman L Fujita R Kellerman A Making shift happen: designing for successful environmental behavior change 2022 New Society Publishers
Washburn AN Skitka LJ Science denial across the political divide: Liberals and conservatives are similarly motivated to deny attitude-inconsistent science Social Psychological and Personality Science 2018 9 8 972 980 10.1177/1948550617731500
Weart SR The discovery of global warming: Revised and expanded edition 2008 Harvard University Press
Weber EU What shapes perceptions of climate change? Wiley Interdisciplinary Reviews: Climate Change 2010 1 3 332 342
Weber EU What shapes perceptions of climate change? New research since 2010 Wiley Interdisciplinary Reviews: Climate Change 2016 7 1 125 134
Weber EU Stern PC Public understanding of climate change in the United States American Psychologist 2011 66 4 315 10.1037/a0023253 21553956
Welsby D Price J Pye S Ekins P Unextractable fossil fuels in a 1.5 C world Nature 2021 597 7875 230 234 10.1038/s41586-021-03821-8 34497394
Whitmarsh L Behavioural responses to climate change: Asymmetry of intentions and impacts Journal of Environmental Psychology 2009 29 1 13 23 10.1016/j.jenvp.2008.05.003
Wiener JL Mowen JC Source credibility: On the independent effects of trust and expertise Advances in Consumer Research 1986 13 1 306 310
Wilson DS Ostrom E Cox ME Generalizing the core design principles for the efficacy of groups Journal of Economic Behavior & Organization 2013 90 S21 S32 10.1016/j.jebo.2012.12.010
Wolstenholme E Poortinga W Whitmarsh L Two birds, one stone: The effectiveness of health and environmental messages to reduce meat consumption and encourage pro-environmental behavioral spillover Frontiers in Psychology 2020 11 577111 10.3389/fpsyg.2020.577111 33117243
Wynes S Zhao J Donner SD How well do people understand the climate impact of individual actions? Climatic Change 2020 162 3 1521 1534 10.1007/s10584-020-02811-5
Wynes S Motta M Donner SD Understanding the climate responsibility associated with elections One Earth 2021 4 3 363 371 10.1016/j.oneear.2021.02.008
Yoder, K. (2020). Footprint fantasy: Is it time to forget about your carbon footprint? Grist. https://grist.org/energy/footprint-fantasy/
Zettle RD Hayes SC Kendall PC Rule governed behavior: A potential theoretical framework for cognitive behavior therapy Advances in cognitive behavioral research and therapy 1982 Academic Press 73 118
| 0 | PMC9707142 | NO-CC CODE | 2022-12-16 23:17:51 | no | Behav. Soc. Iss.. 2022 Nov 28; 31(1):373-417 | utf-8 | null | null | null | oa_other |
==== Front
J Solid State Electrochem
J Solid State Electrochem
Journal of Solid State Electrochemistry
1432-8488
1433-0768
Springer Berlin Heidelberg Berlin/Heidelberg
5330
10.1007/s10008-022-05330-8
Original Paper
Electrochemical immunosensor nanoarchitectonics with the Ag-rGO nanocomposites for the detection of receptor-binding domain of SARS-CoV-2 spike protein
Wang Aiping 234
Li Yuya 24
You Xiaojuan 24
Zhang Shoutao 24
Zhou Jingming 24
Liu Hongliang 24
Ding Peiyang 24
Chen Yumei 24
Qi Yanhua 24
Liu Yankai 24
Liang Chao 24
Zhu Xifang 24
Zhang Ying 24
Liu Enping 24
Zhang Gaiping [email protected]
1234
1 grid.11135.37 0000 0001 2256 9319 School of Advanced Agricultural Sciences, Peking University, Beijing, 100871 China
2 grid.207374.5 0000 0001 2189 3846 School of Life Sciences, Zhengzhou University, Zhengzhou, 450001 Henan China
3 Longhu Laboratory of Advanced Immunology, Zhengzhou, Henan China
4 grid.108266.b 0000 0004 1803 0494 College of Animal Science and Veterinary Medicine, Henan Agricultural University, Zhengzhou, 450002 Henan China
29 11 2022
111
27 9 2022
27 10 2022
1 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
As the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) poses a grave threat to human life and health, it is essential to develop an efficient and sensitive detection method to identify infected individuals. This study described an electrode platform immunosensor to detect SARS-CoV-2-specific spike receptor-binding domain (RBD) protein based on a bare gold electrode modified with Ag-rGO nanocomposites and the biotin-streptavidin interaction system. The Ag-rGO nanocomposites was obtained by chemical synthesis and characterized by electrochemistry and scanning electron microscope (SEM). Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were used to record the electrochemical signals in the electrode modification. The differential pulse voltammetry (DPV) results showed that the limit of detection (LOD) of the immunosensor was 7.2 fg mL−1 and the linear dynamic detection range was 0.015 ~ 158.5 pg mL−1. Furthermore, this sensitive immunosensor accurately detected RBD in artificial saliva with favorable stability, specificity, and reproducibility, indicating that it has the potential to be used as a practical method for the detection of SARS-CoV-2.
Keywords
Electrochemical immunosensor
Ag-rGO nanocomposites
Biotin-streptavidin system
SARS-CoV-2
Receptor-binding domain
http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 32072944 Wang Aiping COVID-19 Emergency Project of the National Key Research and Development Program of China2022YFC0867900 Wang Aiping Emergency Scientific Research on Epidemic Prevention and Control of Henan Province211100310200 Wang Aiping
==== Body
pmcIntroduction
Since the end of 2019, coronavirus disease 2019 (COVID-19), an infectious disease caused by SARS-CoV-2, rapidly swept the world, seriously affecting the normal running of work and life and endangering human health. Although many COVID-19 vaccines have been approved for marketing [1], this is not enough to fully safeguard the health of the population. It is essential to take measures, such as enhancing case detection and contacting tracing to stop the further spread of the virus [2]. Rapid detection technology is another important technology in addition to interception measures such as vaccines. Accurate, simple, cheap, and rapid detection methods can help people find cases early. Therefore, providing accurate and efficient testing methods is necessary for pandemic prevention and control.
There are some measures to detect the SARS-CoV-2: the use of quantitative reverse transcription polymerase chain reaction (RT-PCR) for nucleic acid detection [3], and the use of immune response between antigens and antibodies for serological detection. RT-PCR is undoubtedly the gold standard technique utilized. However, the detection procedure is complex and it has high requirements for professional researchers and experimental equipment [4]. What is more serious is that the RT-PCR has a high false-negative rate [5]; for example, virus nucleic acids were undetectable in a significant number of cases with clinical features of COVID-19 [6], which complicated the prevention and control of the pandemic. Common immunological tests are lateral flow immune assays (LFIA) and electrochemical immunosensors. In the two methods mentioned above, LFIA operation is relatively simple [7]. However, due to its shortcoming in quantitative detection, it is more suitable as an auxiliary diagnostic tool compared to other tests. In contrast, the electrochemical immunosensor based on antigen–antibody response can accurately detect the low concentration of antigen in a short time, which has not only a wide detection range but also a low detection limit [8].
In these 3 years, electrochemical immunosensors have been increasingly used in the detection of SARS-CoV-2. Shimaa Eissa [9] demonstrated a type of electrochemical immunosensor of swabs without transferring or pretreatment of samples. This method effectively improved safety and convenience and the sensitivity of this method can reach 10 pg mL−1. Elif Burcu Aydın [10] prepared a sensitive sandwich immunosensor for detection of RBD, which can be reused, and the minimum detection limit can reach 0.577 fg mL−1. In the construction of the electrochemical immunosensor, the selection of the target is one of the favorable factors to increase the sensitivity and specificity of the immunosensor. SARS-CoV-2 is a highly diverse envelope-positive single-stranded RNA virus belonging to the coronavirus family [11, 12]. It has four main structural antigens: nucleocapsid (N), spike (S), matrix (M), and envelope (E) [13]. Among them, the S protein is divided into two functionally distinct parts S1 and S2 [14], and the RBD in the S1 can specifically bind to the host cell receptor: angiotensin-converting enzyme 2 (ACE2) [15, 16]. Therefore, screening of RBD as a target for SARS-CoV-2 detection is a priority.
In addition, the electrode material is also an important part for the construction of immunosensor, since it has great biocompatibility [17] and can achieve signal amplification [18]. Graphene is a nanomaterial in which sp2 hybridized carbon atoms are tightly stacked into a single-layer, two-dimensional honeycomb-like lattice structure, with unique lamellar structure [19]. Functionalizing graphene such as reduced graphene oxide (rGO) is widely used in the material field because of its large specific surface area and good electrical conductivity as well as thermal stability [20]. In this study, rGO was introduced as the electrode nanomaterial. However, rGO tends to form flaky agglomerates in water, thus hindering electron transport. Therefore, silver nanoparticles were introduced into the rGO for amplifying signals, enhancing biocompatibility [21], reducing the accumulation of graphene sheets, and facilitating electron transfer [22, 23]. Furthermore, compared with the precious nanometals such as Au [24], Pd [25], and Pt [26], the introduction of silver nanoparticles can effectively reduce the difficulty and cost of preparation. At the same time, nafion was used as a dispersant to overcome aggregation. Eventually, we developed Ag-rGO nanocomposites as the electrode material to achieve the signal amplification.
What’s more, studies have shown that the accession of biotin-streptavidin system is an efficient way to immobilize more biotin-modified capture probes [27] and improve LOD and selectivity [28]. The advantage of this system is that one molecule of SA can provide four biotin-binding sites [27, 29]. Therefore, in this study, the biotin was conjugated to the anti-RBD monoclonal antibody at first. Then, the biotin-streptavidin system was utilized as the bridge to link the Ag-rGO nanocomposites and RBD antibody in this immunosensor.
Given the evolution of today’s global pandemic, there is an urgent need for faster and more sensitive methods to detect this rapidly spreading deadly disease to achieve fast control and diagnosis of confirmed cases of the virus and close contracts. In this study, an efficient and accurate biosensor was made by Ag-rGO nanocomposites and biotin-streptavidin system for RBD detection.
Materials and methods
Chemicals
SARS-CoV-2 RBD antibody, SARS-CoV-2 RBD (His Tag), and pseudotyped virus were prepared in the molecular immunology laboratory, Zhengzhou University. Biotin labeling kit was acquired from Elabscience Biotechnology Co., Ltd. (Wuhan, China). SA and artificial saliva were obtained from Beijing Solarbio Science & Technology Co., Ltd (Beijing, China). GO was provided by XFNANO Materials Tech. Co., Ltd (Nanjing, China). Nafion (Nf) and BSA were purchased from Sigma Chemicals (St Louis, USA). AgNO3 was brought from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). NaBH4 was acquired from Shanghai Macklin Biochemical Co., Ltd (Shanghai, China). Phosphate buffer solution (PBS, pH 7.4) was prepared from 0.1 M Na2HPO4, 0.1 M KH2PO4, and 0.1 M KCl, which was used as an electrolyte in the electrochemistry measurements.
Apparatus
All electrochemistry measurements were performed with the CHI 660E electrochemical workstation (Shanghai Chenhua Instrument Co., Ltd., Shanghai, China). The electrochemical workstation consists of the working electrode (gold electrode), the counter electrode (platinum electrode), and the reference electrode (saturated calomel electrode, SCE) (Gaoss Union Co., Ltd, Wuhan, China). Thermo Nanodrop 2000c spectrophotometer (Thermo Fisher Scientific Co., Ltd., USA), scanning electron microscope (Zeiss Gemini 300), and X-ray diffractometer (X’Pert PRO, Netherland) were used to characterize the nanomaterials. Ultrapure water in all experiments was obtained by Barnstead Nanopure ultrapure water meter. SB25-12DTD ultrasonic cleaning machine (Scientz Co., Ltd., Ningbo, China) was used for material dispersion and related cleaning.
Preparation of Ag-rGO nanocomposites
The preparation process of Ag-rGO is shown in Fig. 1A with reference to previous studies [20]. Fifty milligrams of GO was dispersed with 30 mL of ethylene glycol and stirred at high speed for 1 h. After that, 200 mg of AgNO3 was added in 15 mL of ethylene glycol with 5 mL of ultrapure water. Then, the AgNO3 dispersion was slowly added to the GO dispersion, and stirred at 50℃ for 2 h. After a full reaction, 20 mL of NaBH4 (0.1 mol mL−1) was added drop by drop and the mixture was stirred at 110℃ for 2 h. After that, the solution was centrifuged at 6124 × g for 10 min, and the collected precipitation was washed with ultrapure water and dried in an oven at 80℃. Finally, the material was ground into powder and stored at 4℃. Before experiment, 3 mL of the mixture consisting of ultrapure water and 0.4%Nafion (Nf) in a ratio of 3:1 was added to 0.9 mg Ag-rGO powder. The uniform Ag-rGO-Nf solution (0.3 mg mL−1) was obtained under sonicated treatment for 2 h.Fig. 1 Schematic illustrations of the preparation of Ag-rGO (A). The preparation of biotin-anti-RBD antibody (B). The preparation of electrochemical immunosensor (C)
Preparation of biotin-anti-RBD antibody
Anti-RBD antibody was labeled with biotin following the kit instructions and the diagram is shown in Fig. 1B. The antibody was mixed with NH2-Reactive Biotin at a ratio of 1:20 molecular weight, incubated for 30 min protected from light, and then collected by centrifugation to obtain the biotinylated monoclonal antibody.
Preparation of electrochemical immunosensor
Before preparing the sensors, the gold electrode was polished with a powder of 0.3 μm and 0.05 μm. After that, the electrode was washed with anhydrous ethanol and ultrapure water respectively, and then dried with nitrogen. Figure 1C shows the construction process of the electrochemical immunosensor. First, the 10 μL of Ag-rGO-Nf nanocomposites was dropped on the surface of the gold electrode. After drying at room temperature, the 20 μL SA and biotin-anti-RBD antibody were dropped respectively to immobilize the antibody. During this process, the biotin-anti-RBD was linked to the modified electrode through streptavidin–biotin interactions [30]. Eventually, the remaining active sites were blocked with 10 mL 0.1%(w/t) BSA. During the immunosensor construction, all incubations were performed at 37℃ and each incubation was followed by a rinse with ultrapure water.
Electrochemical measurement
The construction process of the immunosensor was characterized by CV and EIS. The CV measurements were made in the range of − 0.3 ~ 0.5 V at the scanning rate of 50 mV s−1 in 0.1 M PBS (0.1 M, pH 7.4) to characterize immunosensor development. The EIS measurements were performed in the PBS solution containing 5 mM [Fe(CN)6]3-/4- with an initial potential of 0.226 V, an amplitude of 0.005 V, and a frequency range of 0.1 ~ 105 Hz. Moreover, the response performance of the immunosensor was measured in PBS solution (0.1 M, pH 7.4) by DPV with a voltage range of − 0.05 ~ 0.15 V, an amplitude of 0.05 V, a pulse width of 0.05 s, a sampling width of 0.0167, and a pulse period of 0.5 s. Different concentrations of RBD were added to the modified electrode for measurement, and the antigen–antibody immune complex formed after the antigen bind to antibody hindered electron transfer, thereby reducing the electrochemical signal. This allows us to track changes in the current signal for sensor performance analysis.
Preparation of simulated samples
Simulated samples were prepared by mixing commercialized artificial saliva and the SARS-CoV-2 spike protein pseudotyped virus. The simulated sample was diluted by artificial saliva to the concentration of 0.315 ~ 315 TCID50/mL.
Result and discussion
Characterization of Ag-rGO nanocomposites
In the preparation process of Ag-rGO nanocomposites, the oxygen functional groups of GO promoted the initial attachment of free Ag+ in solution through electrostatic interactions. Afterwards, Ag+ was reduced to silver nanoparticles in the presence of ethylene glycol and NaBH4 and attached to GO. In this process, GO provides the attachment area for the silver nanoparticles and the formation of AgNPs also avoids the lamellar stacking of GO. The Ag-rGO nanocomposite formed when GO was reduced to rGO by the strong reducing agent. Following the successful preparation of the material, its structure and morphology were characterized by XRD and SEM respectively.
The UV–Visible spectra of GO are shown in Fig. 2A, where GO showed absorption peaks at around 230 nm and 300 nm, which was probably due to the π → π* transition of the C–C bond and the n → π* transition of the C = O bond. After the reduction of GO by NaBH4, the peak was red-shifted to around 260 nm, which indicated that the electronic conjugation within the graphene sheet was restored [31] and GO was reduced to rGO. Furthermore, the characteristic peak of AgNPs appeared at 410 nm [23], suggesting that the AgNPs were successfully formed on the rGO. The X-ray diffraction (XRD) patterns of GO and Ag-rGO are exhibited in Fig. 2B. The GO showed a characteristic diffraction peak at 10.95 [32], which is caused by the oxygen functional groups in GO. The Ag-rGO demonstrated four diffraction peaks at 2θ = 38.09°, 44.20°, 64.39°, and 77.42°, which could be corresponded to Ag (111), Ag (200), Ag (220), and Ag (311) on the central cubic (fcc) crystal [33]. In contrast to the spectra of GO, the disappearance of GO diffraction peaks in the synthetic composites indicates the presence of rGO exfoliation [20].Fig. 2 The UV–vis spectra (A) and XRD pattern (B) of GO and Ag-rGO; SEM images of GO (C) and Ag-rGO (D)
Eventually, the morphology of the nanocomposites was characterized using SEM. GO showed an obvious folded lamellar state in Fig. 2C, and these corrugated sheets were due to the disruption of the planar sp2 carbon sheets by the introduction of sp3-hybridized carbon during oxidation [34]. The morphology of Ag-rGO is shown in Fig. 2D; the rGO had a large and compact surface, showing curved and folded sheets, and was uniformly covered with AgNPs, indicating that AgNPs were successfully reduced to the surface of rGO. The above results demonstrated that the preparation of Ag-rGO was successful.
Electrochemical behavior of the immunosensor
CV was used to confirm each modification of the electrode. As shown in Fig. 3A, the bare gold electrode (curve a) had no redox peak at a given voltage due to the lack of electroactive substances. After Ag-rGO-Nf modification (curve b), it showed obvious redox current response, which was due to AgNPs possessing great ability to gain or lose electrons [35], indicating that it can act as a redox probe for electron transfer with the electrode surface. When SA (curve c) was added to the electrode surface, it can be fixed on the modified electrode by Ag–N bond [36], while the decrease in current indicated that SA as an insulating biomolecule impeded electron transfer. After immobilization of biotin-anti-RBD antibody (curve d), the redox peak was decreased owing to the bioactive material possessing the ability of impeding electron transfer [30]. With the addition of BSA (curve e), the non-specific binding site was blocked and the electron transfer rate was reduced, leading to a further decrease in current signal intensity. After the addition of RBD (curve f), the current intensity reached a minimum due to the formation of immune complexes.Fig. 3 CVs (A) and EIS (B) of the same electrode at different stages. (a) bare gold electrode, (b) Ag-rGO-Nf/gold electrode, (c) SA/Ag-rGO-Nf/gold electrode, (d) biotin-anti-RBD antibody/SA/Ag-rGO-Nf/gold electrode, (e) BSA/biotin-anti-RBD antibody/SA/Ag-rGO-Nf/gold electrode, and (f) RBD/BSA/biotin-anti-RBD antibody/SA/Ag-rGO-Nf/gold electrode
EIS was also used for electrochemical characterization of each step of the electrode modification. The Nyquist representation contains a linear portion at low frequencies and a semicircle portion at high frequencies. The semicircle portion is related to the electrochemical process subject to electron transfer. And the diameter of the semicircle represents the electrode transfer resistance (Rct), the larger the diameter, the larger the Rct. As shown in Fig. 3B, the gold electrode (curve a) had a small Rct, and the Rct decreased after the incubation of Ag-rGO-Nf nanocomposites (curve b), which meant that the nanomaterial had better electrochemical activity and could enhance the electron transfer rate. With the addition of bioactive substances such as SA (curve c), biotin-anti-RBD antibody (curve d), and BSA (curve e), the Rct of the Nyquist plot gradually increased, indicating that the successful immobilization of proteins hindered the electron transfer efficiency [37]. Eventually, Rct reached the maximum after incubation with RBD (curve f), indicating that the immunocomplex formed by biotin-anti-RBD antibody and RBD further impeded electron transport. The results of EIS were consistent with CV, suggesting that the fabrication of electrochemical immunosensor was successful.
In addition, the immunosensor was further electrochemically characterized by CV at different scan rates. As shown in Fig. 4A, the peak anode peak current (Ipa) and the peak cathode current (Ipc) of the immunosensor gradually increased with the scan rates ranging from 25 to 250 mV s−1. The linear regression equations for Ipa and Ipc were obtained in Fig. 4B as Ipa = 25.69v1/2 + 8.843 with a correlation coefficient of 0.9995 and Ipc = − 25.23v1/2 + 15.66 with a correlation coefficient of 0.9983, and the square root of the scan rate (v1/2) was linear in the range of 25 ~ 250 mV s−1. The results demonstrated that the redox reaction process of the immunosensor was controlled by diffusion [38].Fig. 4 (A) CVs of the electrochemical immunosensor at the different scan rates: 25, 50, 75, 100, 125, 150, 175, 200, 225, 250 mV s−1. (B) The relationship between the square root of the sweep rate and the redox peak current
Optimization of experimental conditions
In order to obtain the best experimental results, some experimental conditions were optimized by CV. Firstly, four concentrations (1, 5, 15, and 25 μg mL−1) and four incubation times (40, 60, 80, and 100 min) of SA were set, respectively. As shown in Fig. 5A, the changes in current (ΔI) gradually increased with increasing incubation concentration and time, reached a highest value when 15 μg mL−1 of SA was incubated for 80 min, and then remained stable with the increasing of time or concentration. Therefore, to save material and time, the optimal incubation concentration and time were determined to be 15 μg mL−1 and 80 min.Fig. 5 Optimization of the incubation concentration and time of SA (A). The incubation time of biotin-anti-RBD antibody (B). The incubation time of RBD (C) (error bar = SD, n = 3)
In addition, the biotin-anti-RBD antibody incubation time also affected the performance of the immunosensor. As shown in Fig. 5B, the reaction time was investigated ranging from 20 to 100 min. Due to the poor electrical conductivity of the biological active substances, ΔI gradually increased and reached a maximum at 40 min, then remained stable. Therefore, the optimal incubation time was determined to be 40 min.
Finally, the effect of the RBD reaction time on the performance of the electrochemical sensor was investigated, with reaction time ranging from 20 to 100 min. Figure 5C shows that ΔI gradually increased with increasing incubation time and reached a maximum at 60 min, followed by a slight decrease. The results indicated that the immunocomplexes formed on the immunosensor surface were saturated. Hence, 60 min was determined as the optimal incubation time for subsequent experiments.
Analytical performance of the immunosensor
The DPV current values of the immunosensor were measured after incubation with different concentrations of RBD in 0.1 M PBS under optimal conditions. As shown in Fig. 6A, the DPV curve gradually decreased as the RBD concentration increased from 0.015 to 158.5 pg mL−1. In addition, the logarithmic calibration plot in Fig. 6B showed that the current response was linearly related to the logarithmic value of the RBD concentration with the linear equation of Y = 12.67X + 76.24 (Y represents ΔI and X represents the logarithmic value of RBD protein concentration) and a correlation coefficient of 0.9821. The DPV detection results denoted that the linear range of the immunosensor was 0.015 ~ 158.5 pg mL−1, and the LOD was calculated as 7.2 fg mL−1 (LOD = 3SB/m, SB stands for standard deviation without RBD, m stands for slope of calibration curve, n = 10 [39, 40]).Fig. 6 (A) The DPV current values of the immunosensor at different concentrations (0.015, 0.1, 0.63, 3.98, 25.1, 158.5 pg mL−1) of RBD. (B) Calibration plot of ΔI versus the logarithm of RBD concentrations (error bar = SD, n = 3)
As shown in Table 1, the LOD of this electrochemical immunosensor (7.2 fg mL−1) has a significant advantage over the studies by Basca Wadramani (0.7 nM), Birjana Moisoska (260 nM), and Yang Yi (0.2 ng mL−1). In addition, compared with the studies of Basca Vadramani (14 ~ 1400 nM) and Arzum Eldem (0.5 ~ 10 ng mL−1), the proposed method has a wider linear detection range (0.015 ~ 158.5 pg mL−1). The results may be attributed to several factors: (i) The excellent electrical conductivity and high biocompatibility of the Ag-rGO nanocomposite. rGO provides a large specific surface area for the attachment of AgNPs, and the formation of AgNPs enables the material to transfer electrons efficiently. (ii) The introduction of the biotin-streptavidin system enables the streptavidin to act as a sensitive trap and efficiently capture the biotin-labeled monoclonal antibody. (iii) The high affinity and specificity of the antibody for the RBD allow it to sensitively capture the antigen and bind to it to form an immune complex.Table 1 Comparison of different analytical methods for SARS-CoV-2 detection
Materials Detection technology Linear range Limit of detection Refs
Cobalt-functionalized TiO2 nanotubes Amperometric immunoassay 14 to 1400 nM ~ 0.7 nM [41]
Graphene Square wave voltammetry - 260 nM [42]
- Fluorescence 2.0–14.6 ng mL−1 0.2 ng mL−1 [43]
- Amperometric immunoassay 0.5–10 ng mL−1 0.19 ng mL−1 [44]
Ag-rGO nanocomposites Differential pulse voltammetry 0.015 ~ 158.5 pg mL−1 7.2 fg mL−1 This work
Stability, reproducibility, and specificity of the immunosensor
To characterize the stability of the BSA/biotin-anti-RBD antibody/Ag-rGO immunosensor, one immunosensor was continuously scanned 20 cycles using CV. The result is shown in Fig. 7A, where the current response value of the immunosensor was reduced by 6.4% compared to the initial value. In addition, the same batch of treated immunosensors was placed at 4℃ and tested every 4 days. As shown in Fig. 7B, the current response values of the sensor were 90.71% and 86.07% on day 8 and day 16, respectively, suggesting the good stability of the electrochemical immunosensor in this study.Fig. 7 (A) The stability of the immunosensor after 20 scanning cycles. (B) The stability of the immunosensor at various storage periods. (C) The reproducibility of the immunosensor after incubation of 100 pg mL−1 of RBD using five identical sensors. (D) The selectivity of the immunosensor for 200 pg mL−1 HPV protein, 200 pg mL−1 VZV protein, 200 pg mL−1 SARS-CoV protein, 100 pg mL−1 RBD protein, and 100 pg mL−1 mixture of above proteins with 100 pg mL−1 RBD (n = 3)
The reproducibility of an immunosensor is also an important indicator to evaluate the applicability of the immunosensor. The current variation of the immunosensor was tested using five identical sensors. Figure 7C shows that the relative standard deviation (RSD) of the five immunosensors was 6.1%. The result indicated that the immunosensor was reliable.
Specificity is another important feature to demonstrate the suitability of the sensor. To test the cross-reactivity of the sensor, the current responses of HPV, VZV, and SARS-CoV structural proteins incubated on the immunosensor were recorded. The specificity of the sensor was measured by dropping 200 pg mL−1 interfering agents, 100 pg mL−1 RBD protein, and a mixture of RBD and interfering agents on the same treated immunosensors. As shown in Fig. 7D, the current of the immunosensor was essentially unchanged with the addition of interfering substances only. The change in current with the addition of a mixture of interferent and RBD was essentially the same as with the addition of RBD only, indicating that the immunosensor had good selectivity for RBD and no cross-reactivity caused by HPV, VZV, and SARS-CoV proteins.
Antigen protein and simulated virus spiked sample analysis
The immunosensor was applied to the measure SARS-CoV-2 RBD protein samples in artificial saliva, and the RBD content was determined by DPV. Artificial saliva samples containing 1 pg mL−1, 10 pg mL−1, and 100 pg mL−1 RBD proteins were measured respectively. As shown in Table 2, the recovery rates were 94.69 ~ 98.71%, and the relative standard deviations were 5.16 ~ 9.08%, which indicated that the accuracy of the proposed method was acceptable.Table 2 The results of measurement of RBD protein samples in artificial saliva (n = 3)
The added amount of RBD (pg mL−1) RBD found (mean ± standard deviation; pg mL−1) Recovery (mean ± standard deviation; %) Relative standard deviation (%)
1 0.960 ± 0.087 96.00 9.08
10 9.871 ± 0.510 98.71 5.16
100 94.690 ± 8.131 94.69 8.59
To verify the effectiveness of the BSA/biotin-anti-RBD antibody/SA/Ag-rGO-Nf/gold electrode immunosensor, the simulated virus diluted by commercial artificial saliva was incubated on the modified electrode and detected by DPV. The voltammetry in Fig. 8 showed that the current signal decreased after incubation of the simulated virus, and ΔI increased with the increase of the simulated virus concentration. The result indicated that the immunosensor could successfully detect the RBD on the simulated virus diluted with artificial saliva. Suggested that the immunosensor has the potential to detect viruses in real saliva samples.Fig. 8 (A) The DPV current value of the immunosensor at different concentrations (315, 31.5, 3.15, 0.315 TCID50/mL) of simulated virus incubation. (B) The relationship between the different concentrations of simulated virus and the ΔI
Conclusion
In this study, an efficient and ultrasensitive biosensor was prepared successfully based on BSA/biotin-anti-RBD antibody/SA/Ag-rGO-Nf/gold electrode platform to detect SARS-CoV-2 RBD. The large specific surface area of rGO increased the loading of AgNPs, and the good electrical conductivity of AgNPs promoted electron transfer and signal amplification. The utilization of the biotin-streptavidin system increased the amount capture of RBD, which made this immunosensor sensitive and selective. The results suggested that this biosensor had a linear dynamic range of 0.015 to 158.5 pg mL−1 and the LOD was 7.2 fg mL−1. As the results from CV and DPV, this immunosensor can detect the RBD in artificial saliva with recovery rates ranging from 94.69 to 98.71% with great reproducibility and stability, which may provide ideas for the quality control of recombinant subunit vaccine by detecting the content of spike protein. Furthermore, the biosensor was capable of successfully detecting the RBD on the simulated viral environment diluted with artificial saliva, suggesting its potential as a practical detection method in the future.
Funding
This research was funded by the National Natural Science Foundation of China (32072944), the COVID-19 Emergency Project of the National Key Research and Development Program of China (2022YFC0867900), and the Emergency Scientific Research on Epidemic Prevention and Control of Henan Province (211100310200).
Declarations
Conflict of interest
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Dong Y, Dai T, Wei Y, Zhang L, Zheng M, Zhou F (2020) A systematic review of SARS-CoV-2 vaccine candidates. Signal Transduct Target Ther 5(1):237. 10.1038/s41392-020-00352-y
2. Bedford J, Enria D, Giesecke J, Heymann DL, Ihekweazu C, Kobinger G, Lane HC, Memish Z, Oh MD, Schuchat A, Ungchusak K, Wieler LH (2020) COVID-19: towards controlling of a pandemic. The Lancet 395(10229):1015–1018. 10.1016/s0140-6736(20)30673-5
3. Panpradist N, Wang Q, Ruth PS, Kotnik JH, Oreskovic AK, Miller A, Stewart SW, Vrana J, Han PD, Beck IA, Starita LM, Frenkel LM, Lutz BR (2021) Simpler and faster Covid-19 testing: strategies to streamline SARS-CoV-2 molecular assays. EBioMedicine 64:103236. 10.1016/j.ebiom.2021.103236
4. Osmanodja B, Budde K, Zickler D, Naik MG, Hofmann J, Gertler M, Hülso C, Rössig H, Horn P, Seybold J, Lunow S, Bothmann M, Barrera-Pesek A, Mayrdorfer M (2021) Accuracy of a novel SARS-CoV-2 antigen-detecting rapid diagnostic test from standardized self-collected anterior nasal swabs. J Clin Med 10(10). 10.3390/jcm10102099
5. Kucirka LM, Lauer SA, Laeyendecker O, Boon D, Lessler J (2020) Variation in false-negative rate of reverse transcriptase polymerase chain reaction-based SARS-CoV-2 tests by time since exposure. Ann Intern Med 173(4):262–267. 10.7326/M20-1495
6. Zhang Z, Wang X, Wei X, Zheng SW, Lenhart BJ, Xu P, Li J, Pan J, Albrecht H, Liu C (2021) Multiplex quantitative detection of SARS-CoV-2 specific IgG and IgM antibodies based on DNA-assisted nanopore sensing. Biosens Bioelectron 181:113134. 10.1016/j.bios.2021.113134
7. Liu H, Dai E, Xiao R, Zhou Z, Zhang M, Bai Z, Shao Y, Qi K, Tu J, Wang C, Wang S (2021) Development of a SERS-based lateral flow immunoassay for rapid and ultra-sensitive detection of anti-SARS-CoV-2 IgM/IgG in clinical samples, Sens Actuators B Chem 329:129196. 10.1016/j.snb.2020.129196
8. Wang A, You X, Liu H, Zhou J, Chen Y, Zhang C, Ma K, Liu Y, Ding P, Qi Y, Zhang G (2022) Development of a label free electrochemical sensor based on a sensitive monoclonal antibody for the detection of tiamulin. Food Chem 366:130573. 10.1016/j.foodchem.2021.130573
9. Eissa S, Zourob M (2021) Development of a low-cost cotton-tipped electrochemical immunosensor for the detection of SARS-CoV-2. Anal Chem 93(3):1826–1833. 10.1021/acs.analchem.0c04719
10. Aydın EB, Aydın M, Sezgintürk MK (2021) Highly selective and sensitive sandwich immunosensor platform modified with MUA-capped GNPs for detection of spike Receptor Binding Domain protein: a precious marker of COVID 19 infection. Sens Actuators B Chem 345:130355. 10.1016/j.snb.2021.130355
11. V’kovski P, Kratzel A, Steiner S, Stalder H, Thiel V (2021) Coronavirus biology and replication: implications for SARS-CoV-2. Nat Rev Microbiol 19(3):155–170. 10.1038/s41579-020-00468-6
12. Paludan SR, Mogensen TH (2022) Innate immunological pathways in COVID-19 pathogenesis. Sci Immunol 7(67). 10.1126/sciimmunol.abm5505
13. Komiyama M (2022) Molecular mechanisms of the medicines for COVID-19. Bullet Chem Soc Japan 95(9):1308–1317. 10.1246/bcsj.20220179
14. Komiyama M (2021) Molecular-level anatomy of SARS-CoV-2 for the battle against the COVID-19 pandemic. Bullet Chem Soc Japan 94(5):1478–1490. 10.1246/bcsj.20210030
15. Lan J, Ge J, Yu J, Shan S, Zhou H, Fan S, Zhang Q, Shi X, Wang Q, Zhang L, Wang X (2020) Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor. Nature 581(7807):215–220. 10.1038/s41586-020-2180-5
16. van Eijk LE, Binkhorst M, Bourgonje AR, Offringa AK, Mulder DJ, Bos EM, Kolundzic N, Abdulle AE, van der Voort PH, Olde Rikkert MG, van der Hoeven JG, den Dunnen WF, Hillebrands JL, van Goor H (2021) COVID-19: immunopathology, pathophysiological mechanisms, and treatment options. J Pathol 254(4):307–331. 10.1002/path.5642
17. Cho IH, Kim DH, Park S (2020) Electrochemical biosensors: perspective on functional nanomaterials for on-site analysis. Biomater Res 24:6. 10.1186/s40824-019-0181-y
18. Yu S, Cao X, Yu M (2012) Electrochemical immunoassay based on gold nanoparticles and reduced graphene oxide functionalized carbon ionic liquid electrode. Microchem J 103:125–130.10.1016/j.microc.2012.02.002
19. Sha Y, Guo Z, Chen B, Wang S, Ge G, Qiu B, Jiang X (2015) A one-step electrochemiluminescence immunosensor preparation for ultrasensitive detection of carbohydrate antigen 19–9 based on multi-functionalized graphene oxide. Biosens Bioelectron 66:468–73.10.1016/j.bios.2014.12.013
20. Shen J, Shi M, Li N, Yan B, Ma H, Hu Y, Ye M (2010) Facile synthesis and application of Ag-chemically converted graphene nanocomposite. Nano Res 3(5):339–349. 10.1007/s12274-010-1037-x
21. Huang X, Deng X, Zhu H, Qi W, Wu D (2018) Ag@Fe2O3-graphene oxide nanocomposite as a novel redox probe for electrochemical immunosensor for alpha-fetoprotein detection. J Solid State Electrochem 23(2):335–343. 10.1007/s10008-018-4139-0
22. Han L, Liu CM, Dong SL, Du CX, Zhang XY, Li LH, Wei Y (2017) Enhanced conductivity of rGO/Ag NPs composites for electrochemical immunoassay of prostate-specific antigen. Biosens Bioelectron 87:466–472. 10.1016/j.bios.2016.08.004
23. Salgado S, Pu L, Maheshwari V (2012) Targeting chemical morphology of graphene oxide for self-assembly and subsequent templating of nanoparticles: a composite approaching capacitance limits in grapheme. J Phys Chem C 116(22):12124–12130. 10.1021/jp3023875
24. Xiao H, Wei S, Gu M, Chen Z, Cao L (2021) A sandwich-type electrochemical immunosensor using rGO-TEPA-Thi-Au as sensitive platform and CMK-3@AuPtNPs as signal probe for AFP detection. Microchem J 170. 10.1016/j.microc.2021.106641
25. Yan Q, Cao L, Dong H, Tan Z, Hu Y, Liu Q, Liu H, Zhao P, Chen L, Liu Y, Li Y, Dong Y (2019) Label-free immunosensors based on a novel multi-amplification signal strategy of TiO2-NGO/Au@Pd hetero-nanostructures. Biosens Bioelectron 127:174–180. 10.1016/j.bios.2018.12.038
26. Li J, Jin H, Wei M, Ren W, Wang J, Zhang Y, Wu L, He B (2021) Dual mode competitive electrochemical immunoassay for dibutyl phthalate detection based on PEI functionalized nitrogen doped graphene-CoSe2/gold nanowires and thionine-Au@Pt core-shell. Sens Actuators B Chem 331. 10.1016/j.snb.2020.129401
27. Chen J, Yu C, Zhao Y, Niu Y, Zhang L, Yu Y, Wu J, He J (2017) A novel non-invasive detection method for the FGFR3 gene mutation in maternal plasma for a fetal achondroplasia diagnosis based on signal amplification by hemin-MOFs/PtNPs. Biosens Bioelectron 91:892–899. 10.1016/j.bios.2016.10.067
28. Lakshmipriya T, Gopinath SC, Tang TH (2016) Biotin-streptavidin competition mediates sensitive detection of biomolecules in enzyme linked immunosorbent assay. PLoS One 11(3):e0151153. 10.1371/journal.pone.0151153
29. Wang Y, Zhao G, Wang H, Zhang Y, Zhang N, Wei D, Feng R, Wei Q (2020) Label-free electrochemical immunosensor based on biocompatible nanoporous Fe3O4 and biotin-streptavidin system for sensitive detection of zearalenone, Analyst 145(4):1368–1375. 10.1039/c9an02543j
30. Ji R, Chen S, Xu W, Qin Z, Qiu JF, Li CR (2018) A voltammetric immunosensor for clenbuterol based on the use of a MoS2-AuPt nanocomposite. Mikrochim Acta 185(4):209. 10.1007/s00604-018-2746-1
31. Gurunathan S, Han JW, Dayem AA, Eppakayala V, Kim JH (2012) Oxidative stress-mediated antibacterial activity of graphene oxide and reduced graphene oxide in Pseudomonas aeruginosa. Int J Nanomed 7:5901–14. 10.2147/IJN.S37397
32. Dikin DA, Stankovich S, Zimney EJ, Piner RD, Dommett GH, Evmenenko G, Nguyen ST, Ruoff RS (2007) Preparation and characterization of graphene oxide paper, Nature 448(7152):457–460. 10.1038/nature06016
33. Huang L, Wang Z, Zhang J, Pu J, Lin Y, Xu S, Shen L, Chen Q, Shi W (2014) Fully printed, rapid-response sensors based on chemically modified graphene for detecting NO2 at room temperature. ACS Appl Mater Interfaces 6(10):7426–33. 10.1021/am500843p
34. McAllister MJ, Li JL, Adamson DH, Schniepp HC, Abdala AA, Liu J, Herrera-Alonso M, Milius DL, Car R, Prud'homme RK, Aksay IA (2007) Single sheet functionalized graphene by oxidation and thermal expansion of graphite. Chem Mater 19(18):4396–4404. 10.1021/cm0630800
35. Das TR, Sharma PK (2020) Hydrothermal-assisted green synthesis of Ni/Ag@rGO nanocomposite using Punica granatum juice and electrochemical detection of ascorbic acid. Microchem J 156. 10.1016/j.microc.2020.104850
36. Hou L, Huang Y, Hou W, Yan Y, Liu J, Xia N (2020) Modification-free amperometric biosensor for the detection of wild-type p53 protein based on the in situ formation of silver nanoparticle networks for signal amplification. Int J Biol Macromol 158:580–586. 10.1016/j.ijbiomac.2020.04.271
37. Lee SJ, Anandan V, Zhang G (2008) Electrochemical fabrication and evaluation of highly sensitive nanorod-modified electrodes for a biotin/avidin system. Biosens Bioelectron 23(7):1117–24. 10.1016/j.bios.2007.10.025
38. Yin Z, Wu J, Yang Z (2011) Amperometric sensors based on Ni/Al and Co/Al layered double hydroxides modified electrode and their application for hydrogen peroxide detection. Biosens Bioelectron 26(5):1970–4. 10.1016/j.bios.2010.08.049
39. Wang P, Pei F, Ma E, Yang Q, Yu H, Liu J, Li Y, Liu Q, Dong Y, Zhu H (2020) The preparation of hollow AgPt@Pt core-shell nanoparticles loaded on polypyrrole nanosheet modified electrode and its application in immunosensor. Bioelectrochemistry 131:107352. 10.1016/j.bioelechem.2019.107352
40. Winefordner JD, Long GL (1983) Limit of detection. Anal Chem 55(7):712A-724A. 10.1021/ac00258a001
41. Vadlamani BS, Uppal T, Verma SC, Misra M (2020) Functionalized TiO2 nanotube-based electrochemical biosensor for rapid detection of SARS-CoV-2. Sensors (Basel) 20(20). 10.3390/s20205871
42. Mojsoska B, Larsen S, Olsen DA, Madsen JS, Brandslund I, Alatraktchi FA (2021) Rapid SARS-CoV-2 detection using electrochemical ummunosensor. Sensors (Basel) 21(2). 10.3390/s21020390
43. Yang Y, Zhao R, Wang Y, Song D, Jiang B, Guo X, Liu W, Long F, Song H, Hao R (2022) Rapid and universal detection of SARS-CoV-2 and influenza A virus using a reusable dual-channel optic fiber immunosensor. J Med Virol 94(11):5325–5335. 10.1002/jmv.28015
44. Erdem A, Senturk H, Yildiz E, Maral M (2022) Amperometric immunosensor developed for sensitive detection of SARS-CoV-2 spike S1 protein in combined with portable device. Talanta 244:123422. 10.1016/j.talanta.2022.123422
| 36466035 | PMC9707143 | NO-CC CODE | 2022-12-01 23:20:21 | no | J Solid State Electrochem. 2022 Nov 29;:1-11 | utf-8 | J Solid State Electrochem | 2,022 | 10.1007/s10008-022-05330-8 | oa_other |
==== Front
Graefes Arch Clin Exp Ophthalmol
Graefes Arch Clin Exp Ophthalmol
Graefe's Archive for Clinical and Experimental Ophthalmology
0721-832X
1435-702X
Springer Berlin Heidelberg Berlin/Heidelberg
36445446
5909
10.1007/s00417-022-05909-x
Cornea
The usefulness of lutein/trypan blue vital dye for the staining of corneal endothelium: a pilot study on DMEK pretreated tissues
Gisoldi Rossella Colabelli 1
Lodato Gemma 1
Balzamino Bijorn Omar 2
Esposito Graziana 2
Micera Alessandra [email protected]
2
Pocobelli Augusto 1
1 grid.415032.1 0000 0004 1756 8479 San Giovanni Addolorata Hospital, UOC Oftalmologia, Banca degli Occhi, Rome, Italy
2 grid.414603.4 Research and Development Laboratory for Biochemical, Molecular and Cellular Applications in Ophthalmological Sciences, IRCCS, Fondazione Bietti, Rome, Italy
29 11 2022
19
28 6 2022
3 11 2022
12 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose
The study aims to evaluate the usefulness of lutein/trypan blue vital dye for the staining of corneal tissues and endothelium–Descemet membrane (EDM) for Descemet membrane endothelial keratoplasty (DMEK).
Methods
Sixteen human corneal tissues (Eye Bank, Rome, Italy) were used. Corneal endothelium was tested at 25 s (T0), 1 min (T1), 2 min (T2), and 4 min (T4) from dye addition. Staining intensity and cell counting were compared. Stripped EDM was analyzed for selected apoptotic (AP, caspases, BCL2, BAX) and differentiation (VEGF-A, TGF-β1RI, SMAD3/7, SMA) targets and changes in target expression. Protein extracts were analyzed through SDS-PAGE/IB.
Results
Although trypan blue staining produced the same color intensity of lutein/trypan blue dye in half the time, lutein/trypan blue reached a good and adequate color intensity at T4, which persisted even on excised and washed EDM grafts. Lutein/trypan blue-stained EDM showed a reduced number of blue-stained cells and AP immunoreactivity was significantly reduced in the same samples. An increased BCL2 transcript and a reduced BAX transcript were detected in lutein/trypan blue-stained EDM. No significant changes were observed for the main effector caspases (3/9) upon both treatments and the target genes representative of endothelial cell trans-differentiation (TGF-β1RI, SMAD3/7, SMA). A trend in vascular endothelial growth factor (VEGF-A) regulation was observed in lutein/trypan blue-treated EDM grafts.
Conclusion
Obtained results suggest that lutein/trypan blue dye deserves attention in the DMEK field and support the potential routine use of this dye as a valid alternative to trypan blue for all procedures devoted to the assessment of endothelial cell viability and visualization of EDM graft before DMEK grafting.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00417-022-05909-x.
Keywords
Vital dyes
Corneal endothelial cells
EDM
DMEK
Keratoplasty
Eye bank
Toxicity
==== Body
pmc
Introduction
Descemet membrane endothelial keratoplasty (DMEK) is the gold standard procedure for treating corneal endothelial diseases [1, 2]. The preparation of pre-stripped DMEK grafts requires the staining of the endothelium–Descemet membrane (EDM) [3].
The trypan blue vital dye is routinely used to allow good visualization of corneal endothelium for quick excision, vital cell assessment, and easy intracameral unfolding EDM for positioning [4, 5]. Unfortunately, some studies displayed that trypan blue dye can exert toxic effects on corneal cells and retinal pigment epithelium, forcing the definition of appropriate dosage and exposure times to reduce cytotoxic effects [6, 7]. Since endothelial cell loss can occur (over 30%) after DMEK surgery, even in a standardized procedure, new vital dyes deserve attention in the DMEK field [8, 9].
Phacodyne™ and Phaco Lutein™ (Alfa Instruments s.r.l., Italy) are vital intraocular dyes consisting of 1% soluble lutein, a major component of the macular pigment, in combination with 0.04% trypan blue (hereafter termed lutein/trypan blue dye). Lutein/trypan blue dye was tested as a good, useful, and safe dye during capsulorhexis in phacoemulsification. No signs of toxicity or side effects were detected during follow-up (30 days) [10]. This protective effect was associated with the properties of lutein, as suggested by clinical, histological, and electroretinographic analyses performed to evaluate the safety profile of solutions containing lutein in rabbit eyes after intravitreal injections, implying good protection of vitreoretinal interphase and underlined retina [11].
Since lutein-based dyes showed a good safety profile and optimized outcomes in some ophthalmic surgery, this pilot study aimed to test lutein-based dye performance to prepare EDM grafts before DMEK surgery. Both lutein/trypan blue dye and the trypan blue dye were tested in parallel and compared by means of staining, immunofluorescent (acidic phosphatases), and molecular (caspases 3/9, BCL2, BAX, VEGF-A, TGF-β1RI, SMAD3/7, SMA) target expression.
Materials and methods
Reagents and ethics
The lutein/trypan blue dye (1% lutein in combination with 0.04% trypan blue; Phacodyne™ and Phaco Lutein™; Alfa Instruments s.r.l., Italy) and the trypan blue (0.1%) dye were used for these comparative studies.
All tissue procedures were carried out according to standardized operating procedures belonging to the Eye Bank protocols, in line with the Declaration of Helsinki and the ARVO statements for management of ocular biosamples. Written informed consent was collected from donors in life in order to use corneal tissues for transplantation and/or research purposes.
Specimens: tissue dissection and subgrouping
A total of 16 age- and sex-matched corneoscleral tissues (38% females/62% males; mean age 65 ± 5 years; age range: 47–79 years) were obtained within 24 h from death, and tissue handling was carried out under sterile conditions. Human tissues were preserved in tissue-C medium (Alchimia, Ponte San Nicolò, Padova, Italy) at 2–8 °C for up to 7 days or at 31 °C for up to 30 days [12, 13]. Tissues were deemed unsuitable for transplantation because of insufficient endothelial density (<1500 cells/mm2), severe endothelial polymorphism or severe endothelial dystrophy, and stromal alteration, in line with proper exclusion criteria [14].
Light microscopy analysis, digital acquisitions, and cell counting
Whole corneas (eight samples/four time points) were placed upside-down on Petri dishes under the stereomicroscope (STEMI 2000-c; Carl Zeiss, Oberkochen, Germany). Specimens were equally divided between the lutein/trypan blue and trypan blue analyses. Dye drops were gently added, and the blue staining intensity was registered at each time point under a stereomicroscope (SZW1000 Nikon, Tokyo, Japan), according to a standard grading scale for T0 (25 s), T1 (1 min), T2 (2 min), and T4 (4 min) [15]. T0 was defined as control (25 s) and used for single ratio calculations. After that, blue-stained nuclei were counted from dye addition for each time point. Optical fields (10 × 10 mm2) were used for counting the number of blue cells. Cell mortality was assessed on flat-mounted EDM by counting the number of blue nuclei at each time point under a direct light microscope (Eclipse TE200; Nikon). Optic fields (areas; 5 central squares, 1 mm2) were selected to analyze the most central portion of each sample by using a 10 × 10 ocular grid (×100 magnification). Blue cell ratios (mean ± SD) were produced for each time point with respect to T0 value (cell mortality).
Acidic phosphatase staining on stripped EDM
Corneal tissues (four specimens) were stained at T4, subjected to the whole EDM stripping, and graded EDM for color intensity under a stereomicroscope equipped with optic fibers (Nikon). Over time, the persistence of lutein/trypan blue staining was assessed by placing free-floating EDM in agitation (physiological solution for 15 min).
Excised EDM was post-fixed in buffered formaldehyde, placed on glass slides, and equilibrated in 20 mM phosphate-buffered saline (PBS). Flat-mounted EDMs were probed with anti-acid phosphatase (AP) antibody (1:1000 in PBS; ab58688, Abcam, Cambridge, UK) for 1 h at room temperature. After gentle washing in 0.05% Tween 20-PBS, slides were covered with FC-conjugated secondary antibodies (1:400 in PBS; Vector Laboratories Inc., CA, USA) for 30 min at room temperature. Finally, slides were mounted in an aqueous mounting medium (Vectamount; Vector) containing propidium iodide (PI; 1:1000 in PBS; Sigma-Aldrich, PA, USA) for nuclear staining. Images were acquired using the epifluorescent direct microscope (Ni-Eclipse; Nikon, Tokyo, Japan) equipped with a UV lamp, digital camera (Axiocam 208 color), and ZEN 3.1 (free available blue edition) acquisition software (both from Carl Zeiss, Jena, Germany). Single specific acquisitions were carried out at ×40 objectives and merged according to a standard procedure (8-tiff format). Mean fluorescence intensities (MFI) of stained areas (1000 × 1000 square) were calculated for comparison between subgroups.
Biomolecular analysis on stripped EDM
Corneal tissues (four specimens) were stained at T4 and subjected to whole EDM stripping. Peeled-off EDM was processed according to the TRIfast technique (EuroClone SpA, Milan, Italy) to extract total RNA and denatured proteins simultaneously, as below described in detail.
Real-time PCR and REST analysis
Total RNA was quantified spectrophotometrically (N1000 Spectrophotometer; NanoDrop; Celbio, Milan, Italy) after rehydration in DEPC-treated Milli-Q water. cDNA synthesis was carried out using the GoScript kit (Promega, Madison, WI, USA) in a programmable thermocycler (LifePro Thermal Cycler; EuroClone) in the presence of 50 ng total RNA (template), random primers, and dNTPs. Amplification procedures were performed in a real-time thermocycler (ECO-48 Illumina, San Diego, CA, USA) using the SYBR Green PCR mixture (Applied Biosystems, CA, USA) and cDNA (3 μL for target or /1 μL for referring genes) as template and primers. Supplementary Table 1 shows the primer set as designed one-intron spanning (Primer3 free-software) and synthesized by Eurofins Genomics GmbH (Ebersberg, Germany). The amplification profile comprised one cycle of 95 °C/5-min initial denaturation followed by 35 cycles at 95 °C/15 s (denaturation), 59 °C/5 s (annealing), and 72 °C/15 s (elongation), followed by melting analysis. Normalized RNAs were used for amplification, and cycle threshold (Ct) values from good melting curves were utilized for the calculation of fold-changes (FC; 2log expression) with the REST software (www.gene-quantification.de). The relative gene expression for each target was obtained with respect to referring genes (H3 and/or GAPDH). Fold-changes inside the range [−2 to +2] were not considered different for statistical analysis.
SDS-PAGE analysis and IntDen analysis
Total protein contents were A280 quantified (NanoDrop), and normalized samples (15 μg/lane) were separated by electrophoresis on 12% SDS-polyacrylamide gel under reducing conditions (130 V/frontline; Miniprotean apparatus; Bio-Rad Laboratories, Inc., Hercules, CA, USA). Subsequently, the separated bands were transferred from SDS-PAGE to Hy-Bond membranes (GE Healthcare, Buckinghamshire, UK) under semi-dry conditions (10 V/55 min; semi-dry condition, transblotting apparatus; Bio-Rad) in the presence of 48 mM Tris-HCl (pH 6.8), 39 mM glycine, 0.0037% SDS, and 20% methanol solution. Membranes were stained with Sypro blot stain assay (Invitrogen, MA, USA) for image acquisition in a B-BOX Blue Light LED epi-Illuminator (SMOBIO Technology Inc., Hsinchu City, Taiwan). Specific immunoblotting was performed to verify Integrin (anti-Integrin β1/ITGB1 antibody [A-4]: predicted/observed band size at 138/100 kDa; 1:1000; sc-374429; Santa Cruz Biotech, Santa Cruz, CA), F-actin (anti-F-actin antibody, [4E3.adl], ab130935; predicted/observed band size at 42/37 kDa; 1:1000; Abcam, Cambridge, UK), and VEGF (anti-VEGF 165A antibody [6B7]; ab69479; predicted/observed band size at 23/20 kDa; 1:1000) proteins at related size separation. Bands were quantified with the ImageJ available at the imagej.nih.gov/ij, and data for graphical representation were provided as Integrated Density values (IntDen; mean ± SD).
Statistical analysis
All data were provided as mean ± SD or mean ± standard error from the mean (SEM). Prism 9 software (GraphPad Software Inc., CA, USA) was used for statistical analysis and graphical representation. After the normality check (Shapiro–Wilk test), the parametric ANOVA coupled Tukey-Kramer post hoc comparison was used to estimate differences between groups after the normality check (Shapiro–Wilk test). p < 0.05 or p < 0.01 was considered statistically significant.
Results
This ex vivo pilot study was conducted to compare lutein/trypan blue and trypan blue dyes’ staining performance in terms of staining intensity and cell viability (number, apoptosis, and differentiation).
Lutein/trypan blue and trypan blue staining are comparable
A pilot study was carried out on whole corneal specimens for both dyes in parallel before EDM excision and specific analysis to define the best staining protocol. As shown in Fig. 1, a time-dependent increase of blue staining was monitored for both dyes showing a staining intensity directly proportional to exposure times (trypan blue vs. lutein/trypan blue; Fig. 1A–C vs. D–F). The quantification showed that at each time point the trypan blue had a one-degree higher blue staining than lutein/trypan blue one (Fig. 1C vs. F). Lutein/trypan blue dye induced a grade 3 color intensity at T4-processed as observed after EDM excision (Fig. 1G, H). Of interest, grade 3 lutein/trypan blue intensity of excited-EDM persisted after extensive washing in physiological solution for up to 15 min (Fig. 1I). The color-grid scale used for scoring is shown (Fig. 1J).Fig. 1 Corneal endothelial staining. Representative images of whole upside-down corneas stained with trypan blue (A–C) and lutein/trypan blue (D–F) dyes according to timelines (T1, T2, T4). Representative free-floating lutein/trypan blue-stained EDM (T4; G), folded EDM (T4; H), and flat-mounted EDMs after washing (T4; I). Magnification: ×200. Staining intensity was graded according to the color grid palette (J): grade 1 (hexadecimal (HEX)#E0E0FF), grade 2 (HEX#AFB1FE), grade 3 (HEX#898CFD), grade 4 (HEX#4F53FC), and grade 5 (HEX#0000E0). K Graphical representation of blue-stained cells counted in whole corneal endothelia from each group (lutein/trypan blue, white bar and trypan blue, gray bar) at 1 min (T1), 2 min (T2), and 4 min (T4), with respect to 25 s (T0; control). **p < 0.01, lutein/trypan blue vs. trypan blue at T4 (intergroup analysis)
Endothelial mortality was assessed on whole stained corneal specimens placed upside-down on Petri dishes to allow endothelial cell counting at increasing time points (T0–T4). The results are shown in Fig. 1K. The counting of blue-stained nuclei per optic field showed that cell mortality was significantly higher after trypan blue staining with respect to T0 (25 s, control) and depending on time exposure (T0 vs. T1: p < 0.05; T0 vs. T2 and T4: p < 0.01; see Supplementary Fig. 1). Lutein/trypan blue dye showed a comparable number of blue-stained cells at all times examined (T1, T2, T4 vs. T0; p > 0.05). The number of blue-stained cells at T4 was significantly reduced in the presence of lutein/trypan blue compared to trypan blue (p < 0.05; Fig. 1K). No statistically significant difference in cell mortality was detected intragroup for lutein/trypan blue dye, at all examined times.
Lutein/trypan blue displayed lower AP immunoreactivity with respect to trypan blue
As lutein/trypan blue dye displayed good staining of endothelium at T4 (Fig. 1I), this time point was selected for the subsequent immunofluorescent analyses. Epifluorescent analysis carried out on flat-mounted stained and prefixed-EDM showed a lower AP immunoreactivity in lutein/trypan blue samples with respect to trypan blue ones over a nuclear counterstaining (p < 0.05; Fig. 2A–C). Images in the figure are representative of a marginal zone, showing cells characterized by an elongated morphological appearance. The densitometric analysis (IntDen) was carried out on single-channel images (fluorescent channel; 8-tiff) (Fig. 2A).Fig. 2 Acidic phosphatase immunoreactivity. Flat-mounted EDMs were processed for immunofluorescent analysis of acidic phosphatases (AP/green). Graphical representation of AP Integrated Density (IntDen) expression from specific quantification on single-channel images (ImageJ). Difference between groups was *p < 0.05 (ANOVA analysis; A). Representative digital images of green AP immunoreactive cells over a red nuclear staining (PI/red) in EDM exposed to lutein/trypan blue (B) and trypan blue (C). Note the higher immunoreactivity in trypan blue-stained EDM. Magnification: ×400
Lutein/trypan blue dye did not influence the expression of apoptotic transcripts
Next, we checked for transcript expression of some genes involved in the apoptotic pathway (caspases 3/9 and BCL2/BAX). As shown in Fig. 3A, no significant changes between groups were observed for caspase 3 and 9 subtypes. Interestingly, the analysis of BCL2 (anti-apoptotic target) and BAX (pro-apoptotic target) resulted in a low BCL2/BAX ratio in trypan blue group, as it shifted toward the apoptotic activation. Noteworthy, a high BCL2/BAX ratio was detected in lutein/trypan blue group due to the BCL2 upregulation with respect to BAX, which was significantly downregulated (p < 0.05).Fig. 3 Real-time PCR analysis of lutein/trypan blue and trypan blue-stained extracts. Real-time PCR analysis specific for apoptotic (A) and EndMT (B) patterns performed on stained-EDM extracts. A Bar diagram showing the caspase 3/9 and BCL2/BAX target gene expression. Note the inverted BCL2/BAX ratio in lutein/trypan blue-stained vs. trypan blue ones. B Bar diagram showing the specific amplification of VEGF-A, TGF-β1RI, SMAD3–7, and SMA transcripts in lutein/trypan blue and trypan blue-stained EDM. Note the absence of specific amplification for VEGFR targets with respect to the huge deregulation of EndMT ones. Statistical significance is indicated by asterisk (*) in bar graphs: ***p < 0.001
Lutein/trypan did not influence proper endothelial cell parameters
Some targets belonging to the endothelial-to-mesenchymal transformation (EndMT) were also investigated in lutein/trypan blue and trypan blue groups. REST analysis showed no significant changes in both groups for TGF-β1/SMAD3–7 and SMA transcript expression (Fig. 3B). Target gene expression was highly deregulated upon both treatments. Unchanged VEGF mRNA expression was also observed in the lutein/trypan blue group compared to the trypan blue one, even if a trend for regulation can be observed in lutein/trypan blue group.
Lutein/trypan blue displayed no changes in total protein profile
To characterize the total protein profile of stained EDM, preliminary SDS-PAGE analyses were carried out on protein extracts. As shown in Fig. 4A, a similar total protein expression and separation were observed for both groups. Although upon further immunoblotting, no IntDen changes were detected for 100-kDa (Integrin) and 37-kDa (F-actin) protein fractions between groups, a trend to an increase for 20-kDa (VEGF-A) proteins was observed for lutein/trypan blue extracts, as compared to trypan blue and control ones (p < 0.001; Fig. 4B).Fig. 4 SDS-PAGE analysis. A EDM specimens were extracted in lysis buffer under reducing conditions and equal protein amounts were separated according to their molecular weights. B Specific immunolabeled 20-kDa (VEGF), 37-kDa (F-actin), and 100-kDa (Integrin β1) observed band sizes from total protein fractions were subjected to densitometric analysis (Integrated Density (IntDen); ImageJ). Related IntDen analysis showed a trend to an increase for VEGF (***p < 0.001) in lutein/trypan blue-stained EDM protein extracts. CRT: control; tryp + lut: lutein/trypan blue; tryp: trypan
Discussion
In the present study, we compared the staining properties of lutein/trypan blue and trypan blue vital dyes in order to provide data to characterize the use of lutein/trypan blue vital dye for pre-surgery EDM preparation.
Since endothelial cell mortality might be a limitation for DMEK transplantation, corneal endothelial cell viability assessment by trypan blue dye represents an obligatory phase and a useful tool for EDM grafts [16]. Trypan blue also allows a good visualization and positioning of blue-stained EDM during DMEK [12, 13]. Increasing concentrations (0.1–0.4%), as well as appropriate time exposures (up to 7 min) for trypan blue staining, have been tested to visualize tissue better and allow cell counting [12, 16]. Although standardized, this trypan blue procedure is not completely “free from cytotoxic effects,” so the search for alternative vital dyes with higher performance and lower cellular toxicity is still ongoing [12, 13].
New-generation vital dyes composed of 0.04% trypan blue in solution with 1% lutein are now available for ophthalmic surgery (Phacodyne™, Phaco Lutein™, manufactured by Alfa Instruments s.r.l., Italy) [10].
Herein, we investigated, discussed, and implemented the literature data on lutein/trypan blue dye specifically for EDM preparation before DMEK. Although trypan blue dye produced the same staining intensity as lutein/trypan blue dye at half the time, the lutein/trypan blue protocol was also effective at 4 min. Lutein/trypan blue staining persisted even if excited EDM tissues were extensively washed before DMEK grafting. This finding represents a great value as DMEK surgery involves the manipulation of EDM-stained grafts inside the anterior chamber for at least 7 min, the mean time required for the unrolling of the flap and its correct positioning in contact with the recipient’s cornea [17, 18].
Endothelial cell mortality can represent a gap in the DMEK procedure. Herein, the stain intensity analysis showed that even if trypan blue dye displayed a one-degree higher score than lutein/trypan blue dye at all exposure times, the number of blue-stained cells displayed by lutein/trypan blue dye was lower if compared to those observed by trypan blue at the same exposure times and this reduction was statistically significant at T4 (p < 0.05). The reduced endothelial cell mortality highlights the safety profile of lutein/trypan blue dye combination. Of interest, the reduced immunoreactivity for AP in stripped EDM further supports it. This latter observation is noteworthy as AP is a protein involved in the early stages of apoptosis, representing initial cell membrane perturbation [19].
The absence of transcript regulation specific for caspase 3/9 (few apoptotic effector genes) in both groups supports the absence of any apoptotic event for both combined and single dyes [19]. More interesting, lutein/trypan blue RNA extracts displayed downregulation of BAX and upregulation of BCL2 transcripts, supporting the presence of some protective anti-apoptotic mechanisms for lutein exerted in the lutein/trypan blue combination, as previously reported [20]. Indeed, although still controversial, BCL2 gene has been associated with an anti-apoptotic action, as suggested by different studies on the crosstalk between autophagy and apoptosis [21, 22]. Noteworthy, lutein has been described as protective for both macula and the entire retina, most probably due to its ability to work as a “scavenger” in preventing the formation of free radicals (reactive oxygen species) [15]. Furthermore, the lutein’s property to absorb light at 446 nm (blue region) might appear of great utility also during EDM manipulation, as the blue light was found to trigger reactive oxygen species generation at least in retinal tissue (mainly by photoreceptors) [23].
Any tissue (including cornea) can respond to vital dyes depending on intrinsic characteristics and/or the type of iatrogenic insult (cadaveric sampling/storage), so the possibility that cells can undergo further stimulation cannot be excluded for different donor specimens [24]. A panel of previously reported targets associated with exogenous-driven differentiation of endothelial cells (EndMT) was further investigated at the molecular level [24–27]. The findings on the absence of any upregulation of TGF-β1RI, SMAD3/7, and SMA imply that both lutein/trypan blue and trypan blue do not affect endothelial cell function or trigger the expression of other molecules/factors related to endothelial differentiation [24]. Although investigated mainly in the proangiogenic and inflammatory process, the VEGF/VEGFR represents an essential pathway for the functional activity of corneal endothelial cells, acting primarily as a survival factor at physiological levels [28, 29]. Our biomolecular data suggests that lutein/trypan blue does not affect this pathway showing data comparable to trypan blue dye. Since corneal specimens selected for EDM excision were in accordance with exclusion criteria, we cannot exclude tissue/cell changes due to other growth factors’ imbalance in storage media.
The main limitation of this study is the small number of human corneoscleral specimens to test ex vivo for both vital dyes (further reduced by the COVID-19 pandemic) and the missing data from in vivo models to verify lutein/trypan blue-stained EDM grafts at follow-up.
Nevertheless, taken together, our findings are aligned with literature data and prospect the lutein/trypan blue use as a valid alternative to trypan blue to better counteract endothelial cell mortality, which represents a limitation for the DMEK procedure [30]. These promising results need further in vitro investigation before considering full validation of this staining procedure as a propaedeutic for DMEK surgical practice.
Supplementary Information
Figure S1. Time-point cell mortality (PNG 79 kb)
High Resolution Image (TIF 7968 kb)
Supplementary Table 1 (DOCX 14 kb)
Acknowledgements
The authors would like to thank Dr. Anna Rita Blanco (Medical Liaison, Alfa Intes) for her technical assistance and scientific support. Editorial and graphical assistance was provided by Simonetta Papa, PhD; Massimiliano Pianta; and Aashni Shah (Polistudium SRL, Milan, Italy). This editorial assistance was supported by Alfa Instruments s.r.l. Alfa Instruments s.r.l. was not involved in the study design, collection, analysis, or interpretation of data, in the writing of this article or in the decision to submit it for publication. The study was partially supported by the Italian Ministry of Health (RC 2765948). Bijorn Omar Balzamino, Graziana Esposito and Alessandra Micera are grateful to Fondazione Roma (Italy) for continuous support.
Author contribution
Study conception and design: Rossella Colabelli Gisoldi, Alessandra Micera, and Augusto Pocobelli; data collection: Gemma Lodato and Bijorn Omar Balzamino; experiments/statistics: all; data interpretation: all; manuscript drafting: Rossella Colabelli Gisoldi and Alessandra Micera; manuscript editing: all; and approval to submit: all.
Data availability
Data can be made available upon request.
Declarations
Ethics approval
This experimental study (Jan–Dec 2020), involving human specimens not suitable for transplantation, was approved by Università Tor Vergata (G. L.). Donors gave the appropriate written consent in life to donate for corneal tissue transplantation or research purposes in case of unsuitability for transplantation.
Consent to participate
NA.
Consent for publication
NA.
Competing interests
The authors declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Deng SX Lee WB Hammersmith KM Descemet membrane endothelial keratoplasty: safety and outcomes: a report by the American Academy of Ophthalmology Ophthalmology 2018 125 2 295 310 10.1016/j.ophtha.2017.08.015 28923499
2. Cursiefen C Descemet membrane endothelial keratoplasty: the taming of the shrew JAMA Ophthalmol 2013 131 88 89 10.1001/jamaophthalmol.2013.609 23307214
3. Bachmann B Schaub F Cursiefen C Treatment of corneal endothelial disorders by DMEK and UT-DSAEK: indications, complications, results and follow-up [in German] Ophthalmologe 2016 113 196 203 10.1007/s00347-016-0221-0 26868828
4. Menassa N Pagano L Gadhvi KA Coco G Kaye SB Levis HJ Romano V Free-floating DMEK in the host anterior chamber: surgical management Cornea. 2020 39 11 1453 1456 10.1097/ICO.0000000000002380 32452994
5. Fogla R Thazethaeveetil IR A novel technique for donor insertion and unfolding in Descemet membrane endothelial keratoplasty Cornea. 2021 40 8 1073 1078 10.1097/ICO.0000000000002698 33797465
6. Rezai KA Farrokh-Siar L Gasyna EM Trypan blue induces apoptosis in human retinal pigment epithelial cells Am J Ophthalmol 2004 138 492 495 10.1016/j.ajo.2004.03.033 15364242
7. Hisatomi T Enaida H Matsumoto H Staining ability and biocompatibility of brilliant blue G: preclinical study of brilliant blue G as an adjunct for capsular staining Arch Ophthalmol 2006 124 514 519 10.1001/archopht.124.4.514 16606877
8. Bachmann BO Laaser K Cursiefen C A method to confirm correct orientation of Descemet membrane during Descemet membrane endothelial keratoplasty Am J Ophthalmol 2010 149 922 925.e2 10.1016/j.ajo.2010.01.005 20346440
9. Kruse FE Laaser K Cursiefen C A stepwise approach to donor preparation and insertion increases safety and outcome of Descemet membrane endothelial keratoplasty Cornea 2011 30 580 587 10.1097/ICO.0b013e3182000e2e 21598430
10. Vianna LM Cohen MJ Muccioli C Efficacy of a lutein-based dye (Phacodyne™) for visualizing anterior capsulorhexis during cataract surgery by phacoemulsification Arq Bras Oftalmol 2014 77 3 173 177 10.5935/0004-2749.20140044 25295905
11. Furlani BA Barroso L Sousa-Martins D Lutein and zeaxanthin toxicity with and without brilliant blue in rabbits J Ocul Pharmacol Ther 2014 30 7 559 566 10.1089/jop.2013.0171 24901374
12. Park S Fong AG Cho H Zhang C Gritz DC Mian G Herzlich AA Gore P Morganti A Chuck RS Protocol for vital dye staining of corneal endothelial cells Cornea. 2012 31 12 1476 1479 10.1097/ICO.0b013e31824d0dda 22531436
13. van Dooren BT Beekhuis WH Pels E Biocompatibility of trypan blue with human corneal cells Arch Ophthalmol 2004 122 5 736 742 10.1001/archopht.122.5.736 15136322
14. Lo HM Tsai YJ Du WY A naturally occurring carotenoid, lutein, reduces PDGF and H2O2 signaling and compromised migration in cultured vascular smooth muscle cells J Biomed Sci 2012 19 18 10.1186/1423-0127-19-18 22313606
15. Sujak A Gabrielska J Grudziński W Lutein and zeaxanthin as protectors of lipid membranes against oxidative damage: the structural aspects Arch Biochem Biophys 1999 371 2 301 307 10.1006/abbi.1999.1437 10545218
16. Singh G, Böhnke M, von-Domarus D, Draeger J, Lindstrom RL, Doughman DJ (1985-1986) Vital staining of corneal endothelium. Cornea. 4(2):80-91
17. Majmudar PA Johnson L Enhancing DMEK success by identifying optimal levels of trypan blue dye application to donor corneal tissue Cornea. 2017 36 2 217 221 10.1097/ICO.0000000000001074 28060070
18. Ling JJ Kyrillos R Burckart KA Aldrich BT Skeie JM Schmidt GA Conwell C Ramirez T Reed CR Zimmerman MB Greiner MA Li JY Optimizing visualization of Descemet membrane endothelial keratoplasty tissue: assessing the impact of trypan blue exposure on stain duration and corneal endothelial cell function Cornea. 2021 40 3 292 298 10.1097/ICO.0000000000002440 32732698
19. Zhou F Yang Y Xing D Bcl-2 and Bcl-xL play important roles in the crosstalk between autophagy and apoptosis FEBS J 2011 278 3 403 413 10.1111/j.1742-4658.2010.07965.x 21182587
20. Siebelmann S Matthaei M Hörster R Cursiefen C Bachmann B Lutein and brilliant blue-based dye for donor preparation and transplantation in Descemet membrane endothelial keratoplasty Cornea. 2017 36 4 440 444 10.1097/ICO.0000000000001140 28079691
21. Yang J Yao S JNK-Bcl-2/Bcl-xL-Bax/Bak pathway mediates the crosstalk between matrine-induced autophagy and apoptosis via interplay with Beclin 1 Int J Mol Sci 2015 16 10 25744 25758 10.3390/ijms161025744 26516844
22. Sheu SJ Chen JL Bee YS Differential autophagic effects of vital dyes in retinal pigment epithelial ARPE-19 and photoreceptor 661W cells PLoS One 2017 12 3 e0174736 10.1371/journal.pone.0174736 28358857
23. Roberts JE Dennison J The photobiology of lutein and zeaxanthin in the eye J Ophthalmol 2015 2015 687173 10.1155/2015/687173 26798505
24. Khodapasand E Jafarzadeh N Farrokhi F Is Bax/Bcl-2 ratio considered as a prognostic marker with age and tumor location in colorectal cancer? Iran Biomed J 2015 19 2 69 75 10.6091/ibj.1366.2015 25864810
25. Cruz-Solbes AS Youker K Epithelial to mesenchymal transition (EMT) and endothelial to mesenchymal transition (EndMT): role and implications in kidney fibrosis Results Probl Cell Differ 2017 60 345 372 10.1007/978-3-319-51436-9_13 28409352
26. Díez M Musri MM Ferrer E Barberà JA Peinado VI Endothelial progenitor cells undergo an endothelial-to-mesenchymal transition-like process mediated by TGFbetaRI Cardiovasc Res 2010 88 3 502 511 10.1093/cvr/cvq236 20631156
27. Noseda M McLean G Niessen K Chang L Pollet I Montpetit R Shahidi R Dorovini-Zis K Li L Beckstead B Durand RE Hoodless PA Karsan A Notch activation results in phenotypic and functional changes consistent with endothelial-to-mesenchymal transformation Circ Res 2004 94 7 910 917 10.1161/01.RES.0000124300.76171.C9 14988227
28. Eichmann A Simons M VEGF signaling inside vascular endothelial cells and beyond Curr Opin Cell Biol 2012 24 2 188 193 10.1016/j.ceb.2012.02.002 22366328
29. Philipp W Speicher L Humpel C Expression of vascular endothelial growth factor and its receptors in inflamed and vascularized human corneas Invest Ophthalmol Vis Sci 2000 41 9 2514 2522 10937562
30. Yoon YC Byun YS Kim P Ha MJ Whang WJ Na KS Kim EC Kim HS Hwang HS Quantitative analysis of cornea endothelial cell damage from enucleation, corneal buttoning, and storage in donor corneas using trypan blue dye staining Medicine 2022 101 36 e30430 10.1097/MD.0000000000030430 36086769
| 36445446 | PMC9707145 | NO-CC CODE | 2022-12-01 23:21:38 | no | Graefes Arch Clin Exp Ophthalmol. 2022 Nov 29;:1-9 | utf-8 | Graefes Arch Clin Exp Ophthalmol | 2,022 | 10.1007/s00417-022-05909-x | oa_other |
==== Front
AI Soc
AI Soc
Ai & Society
0951-5666
1435-5655
Springer London London
1599
10.1007/s00146-022-01599-5
Open Forum
The worst mistake 2.0? The digital revolution and the consequences of innovation
http://orcid.org/0000-0002-5626-3729
O’Lemmon Matthew [email protected]
grid.412434.4 0000 0004 1937 1127 School of Global Studies, Thammasat University, Rangsit, Thailand
28 11 2022
110
26 7 2022
14 11 2022
© The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The invention of agriculture 12,000 years ago has been called the worst mistake in human history. Alongside the social, political, and technological innovations that stemmed from it, there came a litany of drawbacks ranging from social inequality, a decline in human health, to the concentration of power in the hands of a few. Millennia after the invention of agriculture, another revolution—the digital revolution—is having a similar impact on humanity, albeit at a scale and speed measured in decades. Despite the tremendous advances brought about by this revolution, there is today a rapidly expanding gulf within and between societies along technological lines; alarming effects on sociality and cognition due to a persistent online presence; and a concentration of power in the form of wealth and data within a handful of tech companies, the likes of which has never been seen before. While the effects of agriculture can now be discerned with thousands of years of hindsight, those of the digital revolution can be witnessed in real time. Is the digital revolution paving the way for a more equitable and stable world, or is it leading humanity down a road that will prove to be more detrimental the more ensconced in technology we become?
Keywords
Agricultural revolution
Digital revolution
Digital divide
Inequality
Cognition
Social media
==== Body
pmcIntroduction
In 1987, the geographer and physiologist Jared Diamond wrote an article titled “The Worst Mistake in the History of the Human Race”, outlining the consequences of our shift from a hunter-gatherer existence to subsistence agriculture. A shift he described as “a catastrophe from which we have never recovered” (1987). A move away from a nomadic, hunter-gatherer lifestyle roughly 12,000 years ago to a settled, farming one brought a stable source of calories, increase in population, and laid the foundation for trade and innovation, all of which would have made the world in which we live in today.
And yet, it also brought untold changes that proved irreversible: inequality through divisions of labor and social inequality (Shin et al. 2012); deforestation and soil erosion (Wright and White 1996); an array of diseases and malnutrition that locked societies in a bondage of reliance on a far more limited diet than their hunter-gatherer ancestors enjoyed (Martin and Goodman 2002); as well as the concentration of political and ritual authority (Laneri et al. 2015).
In a later work, Diamond noted, “If you could choose between being a middle-class American, a Bushman hunter, and peasant farmer in Ethiopia, the first choice would undoubtedly be the healthiest, but the third choice might be the least healthy” (2002). What he was pointing to is the conundrum of innovation: there is no such thing as a free lunch in evolution, be it biological, social, or technological. With the accrued benefits from one invention, so rise a litany of drawbacks that come with it. The so-called ‘progress trap’ (Wright 2004; O’Leary 2007) where material innovation leads to the rise of problems and uncertainty which societies are incapable of solving, resulting in stagnation and possible collapse.
This essay examines the developments stemming from another paradigm shift in technological innovation—the digital revolution—or what is popularly called the third industrial revolution. A revolution that is perhaps as profound as its ancient agricultural counterpart with equally far-reaching implications. Diamond identified three areas where agriculture has made a significant and negative impact: class division and inequality; adverse effects on health; and the concentration of power in the hands of the few.
The digital revolution is producing a similar impact along associated lines but at a far more rapid rate given the advances in artificial intelligence (AI). Specifically, this includes the digital divide along sociodemographic lines as well as between the developed and developing world; the social and physical effects of a near constant online existence; and the centralization and control over data (personal and otherwise) by a handful of technology corporations, and the perhaps inevitable transfer of that control over to the state.
This is not the first essay to raise such issues. Similar concerns about the geometric growth of technology and our ability to control it have been voiced over the years by people such as the mathematician Irving John Good (1966) and philosopher Martin Heidegger (1977). There are also issues that simply cannot be covered within a single essay. Two, specifically, include the looming threat of mass unemployment due to automation, and the potential (some argue likelihood) for the rise of sentient machines, or what is often what is often called the technological singularity. What this essay aims to do, however, is examine the immediate and potentially long-term effects of the digital revolution against those of agriculture which, with thousands of years of hindsight, we know forever changed humanity.
Despite the concerns over technology, there is also no denying that the digital revolution has brought about remarkable changes over the past half a century in areas, such as the democratization of higher education, personalized medicine, to advances in defense, logistics, and transportation (Scott 2006; Hui-Chen et al. 2021; Hamet and Tremblay 2017; Lau and Haugh 2018). But along with these developments have come an array of drawbacks from online ‘doxing’, Internet addiction, to the threats to individual liberties posed by the concentration of power in the hands of a few tech firms.
The first nomadic tribes experimenting with planting seeds could not foresee the physical and social detriments that arose from them, and it has only been through archaeological excavations and forensic analyses over the past several decades that we can see the true impact of agriculture. The changes unleashed by the digital revolution, by contrast, are occurring in real time. Societal destiny is not always in the hands of individuals, though, and the dramatic environmental shifts that contributed to the demise of agriculture-based civilizations of the past such as the Anasazi in southwestern US (Benson and Berry 2009), Angkor in Cambodia (Buckley et al. 2010), and possibly the Maya in Central America (Kennett et al. 2012) may await us still. But outside of such calamities and our own reckoning with a changing climate, it is highly unlikely that the digital revolution will ever reverse itself with societies choosing to adopt a pre-digital lifestyle, just as most societies never willingly gave up agriculture. If Diamond is correct in his description of the peasant farmer as worse off compared to hunter-gatherers and those in economically advanced societies, are the bulk of those whose lives revolve around digital technology his equivalent?
Class divisions and a perpetual developing world
The adoption of agriculture varied widely around the world and involved diverse crops, ranging from the cultivation of cereals in the Middle East more than 12,000 years before present (B.P.); the cultivation of rice in China around 9000 B.P.; domesticated maize in Mexico at 8700 B.P.; bananas and taro roughly 7000 B.P. in New Guinea; and up to 4500 B.P. with the cultivation of wild seed crops in the Eastern Woodlands of North America (Fuller et al. 2014). The picture that emerges is one of incremental change as societies moved from a mobile, hunter-gatherer existence to a more settled one that focused on food production. And while there are historical examples of societies abandoning farming due to the effects of war, disease, famine, or natural disasters, for most, once they became locked into a dependence on agriculture, there was no going back.
As populations grew, so, too, arose an expansion in inequality. Accumulated resources seen in an increase in grave goods and the size of dwellings in the Near East or control over land tenureship by individuals and powerful families in Polynesia, came with attendant dominance over economic and political matters (Price and Bar-Yosef 2010; Hayden and Villeneuve 2010). In the larger city-states of the Basin of Mexico, elites could hold sway over vast territories, acting as the bond between centers of power and outlying areas through intermarriage, trade, or other alliances (Smith 1986, 2021).
Status differentiation in and of itself is not unknown in non-agricultural societies (Borgerhoff Mulder et al. 2009) and has been witnessed in non-human primates for decades (Di Bitetti 1997; Sparks 2006 [1967]). However, it was the monopolization of predictable resources and intergenerational wealth that played a role in persistent inequality (Mattison et al. 2016). A commanding individual could rise to prominence through displays of power and prosperity, acquiring followers along the way through patron–client relationships, and eventually establishing a hereditary rule through passing that ledger of reciprocal relationships on to one’s heirs. Over time, social inequalities became ‘baked in’ and part of the social fabric from which individuals could rarely move out of.
Unlike agriculture, the digital revolution was not a set of experiments separated by thousands of years in varied locations around the world. Rather, it was spearheaded by those countries already at the forefront of technology within the span of a few decades following the end of WWII. However, whereas those monumental shifts initiated by agriculture were slow to accumulate, the inequality brought about by the digital revolution has been ushered in quickly given the global transition to digitization. The post-war interconnectedness of nations in terms of trade, defense, transportation, and diplomacy, to name a few, ushered in this transition. But so, too, did Cold War-era fears and the potential of being left behind in the steady march toward modernization.
This new form of technological inequality has centered on the so-called ‘digital divide’, an issue that has been around for decades and originally described as the advantage conferred to those with access to computers compared to those without, as well as socioeconomic variables of residence (urban versus rural), education, income, English language capabilities, and access to pre-existing technology (Rao et al. 1999; Kleinman 2001). Critics, however, have argued that issues of differential use along with deterministic assumptions that the presence of technology facilitates learning are more salient features of the divide, while the more basic needs of literacy may be the real determinant as to whether access to the Internet has any real meaning (Warschauer 2003; Chun-Yao and Hau-Ning 2010).
As technological innovations such as smartphones became more ubiquitous, particularly among developing nations that saw a greater increase across technologies (Dewan et al. 2010), the access divide decreased. Yet, globally, there continues to be a gap between sexes, age, education, and economic stability among lower paying jobs that do not require the use of the Internet and those that do (Calderón-Gómez et al. 2020); the COVID-19 pandemic making this latter distinction all the more obvious. Similarly, are continuing issues regarding inequality and class division with regard to Internet access and social networks with significant social capital (Winseck 2017; Zhao and Elesh 2007).
The divide also falls along generational lines. The degradation of historical knowledge may be inevitable the more distant in time one gets from historical events and those who lived through them, but the rapid development of the Internet and its outgrowths such as social media have accelerated this process to an unprecedented extent. The result has been a shift in the transfer of cultural knowledge away from generational sources (parents, grandparents, teachers, and so on) and toward those found online. Those born in the mid-90s and onward (Millennials and Generation Z) are less likely to be raised in a religious household compared to their parents, but also less likely to be familiar with the more seminal events of the twentieth century, and more likely to form their identity and opinions via influence from social media (Sultan 2017).
Looking at the early twentieth century by way of comparison, the generation of Americans that fought in WWI was 50 years removed from the US Civil War, but the cause of the latter continued to be vociferously debated (Ramsdell 1937). Today, more than 50 years after the collapse of South Vietnam, younger generations in the US and Vietnam express little interest or knowledge of the Vietnam War (or the American War as it is called in Vietnam), despite the momentous ramifications it had on both countries. As with their American counterparts, younger Vietnamese instead turn to social media to form their opinions around contemporary issues (Rosen 2015).
However, it is between nations that the divide may be of greater concern for countries on the lower end of the socioeconomic ladder. According to the International Telecommunications Union (ITU), urban populations and younger generations have greater access to the Internet, while developed countries have nearly double the number of users as developing countries, and four and a half times the amount of least developed countries (LDCs) (ITU 2020). Social media, likewise, has become a ubiquitous and even mundane part of our online experience with 3 billion of the 3.8 billion Internet users engaged with social media at any point in time (Kemp 2017).
The digital age has, thus, led to inequality on two levels: individually and nationally. Individuals with greater skills/know-how regarding the use and access of information (generally middle-class and college educated) will have a distinct advantage in terms of employment, social networks, and economic advancement. Nationally, the situation is potential even more dire. While access within developing countries is still an issue, the real challenge in the future and one that will impact developmental trajectory is the divide between countries with technological prowess and those without.
A country such as Niger with 6.62 children per woman will have far different needs in terms of supportive care and employment than countries such as Singapore and Taiwan that hover around one child per woman. However, as poorer nations do develop and generally become more prosperous compared to their former state, that also leads to greater health outcomes such as infant survivability and longer lives. That is, greater numbers of people striving for a better life.
Thus, it is countries such as Nigeria whose population has increased tenfold since 1950 that will likely be the wellsprings from which future migrants hail: countries that have made it out of the bottom rungs of development but not so prosperous that their populations would have no reason to leave. Throw into the mix ongoing conflicts and environmental change, and those rising populations will likely be less inclined to stay at home and more likely—and fiscally capable—of making the journey to more prosperous regions of the world (King 2017). But given the backlash over globalization throughout the Western world and severe restrictions on immigration in many countries in Asia, that journey has become all the more difficult.
A far more desperate future could just as well be a technical and economic chasm the likes of which have not been seen since the Age of Discovery. Within richer nations, the social divide may lie between those whose jobs can be conducted online, and a mass of humanity that has been made redundant by automation; the so-called ‘non-essential’ workers of the COVID-19 pandemic. Poorer countries, meanwhile, could become forever chained to the bottom rung of development, dependent on the developed world acting as paternalistic overseers that dole out technology according to the latter’s estimation of which country is capable of handling what and when, or in exchange for whatever resources that they can extract from the ground. Indeed, an even more ominous scenario could see poorer nations cease to functionally develop, becoming ever more dependent on powerful patrons who guard their technological secrets as ancient China once guarded the manufacture of silk.
Social and cognitive change
In his 1987-piece, Diamond wrote “From the progressivist perspective on which I was brought up, to ask ‘Why did almost all our hunter-gatherer ancestors adopt agriculture?’ is silly. Of course they adopted it because agriculture is an efficient way to get more food for less work…While the case for the progressivist view seems overwhelming, it's hard to prove.” Indeed, the opposite is the case. The amount of work hunter-gatherers spend acquiring food is less than that required for agricultural production (Sahlins 1972; Dyble et al. 2019). The fruits of their labor, so to speak, are also more diversified, leading to a far healthier diet and lower food-related pathologies compared to their agriculturally dependent counterparts.
Research from around the world has demonstrated a general decline in dental health as well as an overall increase in morbidity with the shift to agriculture and the sedentism that followed (Ubelaker 1992; Gualandi 1992; Lukacs 1996). The narrowing of diets and reliance on domesticated plants compared to animal protein led to a retardation in bone length, thickness, and robusticity (Larsen 1995), not to mention the dangers of relying on a limited number of crops should a bad harvest or disease wipe them out. A carbohydrate-based diet satisfied the need to feed a growing number of hungry mouths, but the lack of nutritional value weakened the bodies attached to them.
And so, too, for our dopamine-driven clicks on the Internet. Chamath Palihapitiya, a former vice president for user growth at Facebook, has publicly stated that the short-term, dopamine effects of social media are having long-term consequences on civil society through the manipulation and overproduction of shared information. At a talk at the Stanford Graduate School of Business in 2017, Palihapitiya said that “Bad actors can now manipulate large swaths of people to do anything you want. And we compound the problem. We curate our lives around this perceived sense of perfection, because we get rewarded in these short-term signals—hearts, likes, thumbs up—and we conflate that with value and we conflate it with truth” (Wang 2017).
While there is evidence that moderate use of social media may not pose a risk (Przybylski and Weinstein 2017), there is also evidence that heavy use of social media (5 + hours a day) among adolescents leads to higher rates of unhappiness and suicidal risk factors (Twenge and Campbell 2019). The risk for teenage girls in particular is much higher than for boys, with rates of depression, suicide-related outcomes, and suicide rates spiking after 2012 when use of social media among teens became common (Twenge et al. 2018; Haidt and Allen 2020). The explosion in growth of teenage girls coming out as transgender—70 times the expected prevalence rate—and increased use of social media has also been cited by parents of girls who belong to peer groups where one or even all friends within the group become gender dysphoric within the same timeframe (Littman 2018).
The primacy of place of social media for many has led to the rise of digital narcissistic behavior (Faucher 2018) through the amplification of one’s profile to millions online where the latest Tweet, Instagram photo, or YouTube video elicits an almost immediate response and cycles of engagement that occupy, for many, most of their waking hours. As the use of digital media has more than doubled from close to 3 h per day in 2009 to more than 6 h per day in 2018, it is perhaps no surprise that living arrangements have begun to reflect this trend, with upscale “physical social network” (Hansen-Bundy 2018) residential apartment blocks offering all the services one needs. Obviating the necessity of venturing out into other parts of a city to complete daily tasks becomes, interacting with others outside of one’s immediate sphere of influence likewise becomes obsolete.
But while these physical spaces are designed to bring individuals together and provide an antidote for the impersonality of social media, a remedy is only needed when there is a problem. Specifically, the dwindling importance many people, especially young people, place on face-to-face interaction. Instead, there is the increasing importance of immersing oneself in a sea of images, videos, and text that can be digested and curated quickly to make room for the next in an endless stream of social consciousness from all points of the globe.
An increasingly isolated populace also leads to broader repercussions for the future demographic health of a society. W. Bradford Wilcox (2018) of the National Marriage Project at the University of Virginia cites online porn and video games as contributing to the decline in birthrates in the US, but also a decline in an interest in sex among younger people. Although the US is nowhere near Japan where one-third of Japanese adults between the ages of 18–34 have never had sex, movements such as the #MeToo movement have the capacity to drive down sexual activity even further as men fear the aftermath of sexual encounters in an age where definitions of what constitutes consent remain ambiguous (ibid), and online ‘doxing’ campaigns can quickly upend lives.
The necessity of the Internet and handheld devices has also given rise to new ailments and an expanding catalog of acronyms to account for them. Internet Addiction Disorder (IAD) and problematic smartphone use (PSU) involve primarily visual stimuli, but those stimuli share similar characteristics with substance-use disorders such as rapid impulsive processes with little reflection by the user (Roh et al. 2018). PSU has been shown to be positively related to anxiety, time spent on one’s phone, and, interestingly, the number of selfies taken, but negatively related to a connection with nature. The latter, by contrast, has been positively related to photos of nature (Richardson et al. 2018). In other words, the more time spent on smartphones, the less one is likely to take a photo of, say, the Grand Canyon for the sake of its own natural splendor, but a photo of oneself at the Grand Canyon as if it were simply the background on which one’s persona is layered and validated through likes, retweets, and shares.
Our reliance on AI for tasks that were once part of the normal repertoire of human activities has saturated modern societies. These include the well-known ‘Google effect’, or relying on the Internet in place of our own memory, and a reduction in the desire to engage in demanding mental tasks and encode new information in our brains (Sparrow et al. 2011; Bohannon 2011; Storm et al., 2017). Where once our friends and family members were part of a network of transactive memory partners with whom we shared information, the Internet increasingly fills that role and has become for many the sole arbiter of knowledge (Wegner and Ward 2013). Surveys of university students have also found a similar reliance on the search engine over library services, even when students are physically present in university libraries (O’Connor and Lundstrom 2011).
These trends stand alongside more serious issues such as the increase addictive disorders and associated complications. Massively multiplayer online role-playing games (MMORPGs) can lead to Internet gaming disorder (IGD) which has been linked to negative perceptions of past experiences (So-Kum Tang et al. 2017; Lukavská 2018), as well as lower densities of gray matter in the cortex and corresponding behaviors such as impulsivity, distorted decision-making capabilities, depression, and anxiety disorders (Lee et al. 2018). And what occurs at the individual level possibly translates to the population level, meaning that greater amounts of information consumed at greater rates are indeed resulting in lower attention spans for the public at large (Firth et al. 2020).
And it is these latter developments that are perhaps the most alarming of all. As Diamond noted of average human height within skeletons found in Greece and Turkey pre- and post-agriculture, whereas hunter-gatherers reached 5′ 9″ and 5′ 5″ for men and women, respectively, their agricultural successors saw those heights plummet to 5′ 3″ and 5′ (1987). And this drop in height was in addition to associated increases in malnutrition, infectious diseases, and degenerative conditions of the spine (ibid). From 1990 to today, use of the Internet went from close to 0 to nearly 50% of the world’s population, with increases in IAD and IGD seeing similar exponential growth. As with agriculture’s impact on height, the digital revolution is having an equally profound effect on those with IAD and IGD who exhibit a decrease in frontal lobe functions and gray-matter volume of (Jun et al. 2013; Pan et al. 2018). And when considering that more than 50% of US children aged 8–18 have smartphones and spend between roughly 5–7 h a day on them (5 h a day for those 8–12; 7 h a day for those 13–18) (Rideout and Robb 2019), the long-term consequences for the cognitive abilities of younger generations are an open question.
However, acronyms and clinical analyses have a way of blinding us to the generality of such conditions, and the true number of those living in modern societies who may accurately be described as having IAD or IGD is unknown. A simple observational experiment can bear this out. Look around the coffee shop, train, airport terminal, or other public space you happen to be in now—what do you see? More than likely, it is individuals immersed in the soft glow of a screen next to others who are similarly wrapped in a constellation of pixels. A community of people who have walled their minds off from a shared reality which they dip in and out of between clicks and swipes. Our switch to agriculture and what we consumed with our mouths did not make our bodies bigger and stronger but smaller and weaker. The transition to the digital age and our sensory immersion into what for many is a perpetual online existence, is well on its way to having an equally profound effect on our neurological health.
Toward a universal social credit system?
The conglomeration of power through the invention of agriculture in many ways can be said to be self-fulfilling. A reliable and increased source of calories means a greater number of individuals that can be fed and survive. An increase in the number of individuals making it into adulthood means a greater number of offspring which will likely be born to those people in the following years. As agriculture requires competent organization if a community has any reasonable chance of reaping its benefits year after year, those who have the skills and charisma necessary to lead will usually rise to the top of the social hierarchy. But what historically solidified their position was an overarching religious system and cadre of specialists capable of reading the signs of nature and the heavens, thus, legitimizing the power of the sovereign, and ultimately the political and economic system in which commoner and elite all were entwined. For the average peasant, the workings of those who could divine the whims of the gods were clouded in secrecy, hidden by sacred incantations and magical objects. But above all else, they were not to be questioned.
The digital sorcery of today has its own veil that shrouds its intrigues. Social media companies, at least within the United States, fall under a protected class of companies that are not held as responsible for the content that can be accessed on them. Section 230, a provision in the Communications Decency Act passed by the US Congress in 1996, states that as platforms and not disseminators for information, they are protected from lawsuits over users’ posts while allowing them to moderate posts that express violent or misleading content. With the growth of the Internet and social media since then, this has also meant that companies, such as Google, Facebook, and Twitter, have become the defacto mediums which everyone from ordinary citizens to world leaders use to express their views. The result has been the unprecedented acquisition of power and influence by private companies over what individuals read, see, and share online (Lee 2018).
Although bemoaned in the West for its totalitarian underpinnings, the Chinese Communist Party’s (CCP) rollout of its Social Credit System may be a hallmark of things to come. The system quantifies all online actions from opinions posted online, shopping habits, to the amount of time spent watching videos or playing video games. Planning documents for the new system cite that it would “allow the trustworthy to roam everywhere under heaven while making it hard for the discredited to take a single step” (Chin 2016). As Denyer writes, “your score becomes the ultimate truth of who you are—determining whether you can borrow money, get your children into the best schools or travel abroad; whether you get a room in a fancy hotel, a seat in a top restaurant—or even just get a date” (2016).
A high score means greater ease when making online transactions; a low score means expensive purchases via Alipay (a Chinese online payment system also connected to the government) could be blocked. Those with low scores may also be prohibited from boarding planes and high-speed trains (Wade 2017). Perhaps, the most pernicious aspect of the system is that an individual’s social credit score can be lowered through the actions of his friends. You may be a firm believer in the CCP, but should your friend say something that could be construed as less than positive, your score would be in jeopardy for being associated with him.
But how different is this from the ‘court of Twitter opinion’ or other methods used by groups to silence and shame those with whom they disagree, or pressure social media to adopt their definitions of what constitutes hate (Dibble 2020; Perman 2021; Capatides 2020)? The banning of individuals from Twitter, Facebook, and YouTube has become routine, while ‘sandboxing’ (limiting access to videos that are deemed controversial or hateful) is used against ordinary people and world leaders alike (YouTube 2017; Alba et al. 2021). Such bans have been criticized across the political spectrum by politicians around the world (Brennan 2021), raising concerns of the specter of a “digital oligarchy” that in many ways has greater power than the state to silence individuals with whom it disagrees (Jennen and Nussbaum 2021).
There are those who argue that far from being a limit on information, online platforms have democratized information more than at any other time in history. To that end, they are partially correct. While it is true that foreseeably anyone is capable of putting anything online, should search engines or social media sites choose not to index it (or in the case of Amazon, refuse to let companies use its servers), that material will likely never be seen, or at least be seen be a consequential number of people. It is akin to the US Library of Congress filing a book within its vast depository but never assigning it a reference number. Sure, the information is there—but who will ever find it?
The information that does get the approval of corporate gatekeepers is recycled at a speed and degree never before seen, meaning that a post, comment, or picture can be rediscovered and repackaged, gaining a life of its own and demonstrating information’s pathogenic quality. Information comes to light and becomes ‘viral’, spreading throughout a population only to die down before being rediscovered once again, sometimes years after the fact, and resurfacing with perhaps even greater virulence than before. Appropriate behavior and acceptable conduct are policed by a vigilant and unforgiving online community who judge a person’s past deeds against the shifting sands of what is and is not considered appropriate. Authoritarianism, thus, becomes the purview of everyone online, or what Jared Lanier, one of the original creators of virtual reality, calls “digital Maoism” (in Appleyard 2011).
These threats to individual liberty are eased along by the migration by government and businesses in developed economies to an almost wholly online existence, meaning that some form of online activity is a necessity for most people. The individual is thus forced to collaborate in his slow transformation into a ‘datavidual’: An aggregation of online information accessed by codes, the combination of which is unlikely expressed in anyone else (van der Meulen and Bruinsma 2018). One could envision these trends only continuing, such as the calls for ‘vaccine passports’ and tracking of infected individuals via their smartphones during the COVID-19 pandemic, inevitably becoming part of our everyday lives. An ID card or social security number could be replaced by an individualized Internet protocol, or IP address, without which one would have no way of conducting basic transactions from buying milk at the corner store to registering to vote. A continuous online presence—which many people essentially already have via their smartphones—may soon not be an option but a necessary part of everyday life.
Primates are extremely adaptable species with humans being the most adaptable among them. Countless individuals walking among us would not be here if not for advances in everything from eyeglasses to antibiotics. However, our reliance on technology to fill the gaps of our own physical limitations also allows for a continued redefinition of what it means to be human. If we accept the premise that a continuous online presence will be more rather than less likely to occur in the future, then the probability that invasive procedures to augment those limitations may similarly become routine. As Maguire and McGee write, “As enhancements become more widespread, enhancement becomes the norm, and there is increasing social pressure to avail oneself of the ‘benefit.’ Thus, even those who initially shrink from the surgery may find it a necessity” (1999).
Technology has already moved from smartphones to smart glasses, with smart contact lenses just over the horizon. How farfetched is the idea of implanted chips becoming a necessity of daily life, or worse, even required via state decree, much as it is required of pet owners to implant a chip in the family dog?
Conclusion
And, so, we are left with the question posed in the title of this essay: Is the digital revolution our second worst mistake (assuming that Diamond was correct about the first)? The answer, I believe, is that both agriculture and the digital revolution are not exceptions to a rule but necessary outcomes of a rule itself. That rule being that human beings have an inexhaustible capacity to tinker and will do so inexhaustibly. Our extinct cousin, Homo erectus, lived from around two million to 100 thousand years ago; more than six times as long as our species has been in existence. However, during his time on Earth, his toolkit never advanced beyond rudimentary stone tools. By contrast, within the past 40 thousand years, our species, Homo sapiens, went from cave paintings to putting a man on the moon, with the most advanced innovations occurring after the adoption of agriculture.
The transformations brought about by the digital revolution are and will continue to have far-reaching ramifications in terms of our social and political structures, but also on education, the way we consume information, and the digitization of economic exchange. However, as with agriculture, it is also leading to changes in our physiology, reproduction, and a new range of ailments stemming from our novel online existence. And these are in addition to and likely compounded by longer working hours as the divide between home and work becomes increasingly blurred. It is the young, though, which will bear the brunt of any long-term negative effects of innovation. The impact on brains alone is enough to reconsider the steady entrenchment of younger generations in a persistent online presence, and what this may mean for societies writ large in the years to come.
The digital revolution has so thoroughly reorganized modern life, that is also similar to agriculture in that it is extremely difficult to conceive of a time in the future when an online presence would be obviated, much less optional. However, as the COVID-19 pandemic’s impact on childhood literacy has shown (World Bank 2022), the pitfalls that come with an interconnected world can chip away at the very fundamentals on which that world rests, regardless of how technologically sophisticated a country may be. And the poorer the nation, the more impoverished the learning, meaning that succeeding generations and the societies they live in may be left even further behind the digital divide.
And while democratic nations in the West lament the restrictions on freedoms implemented by China with its Social Credit System, those societies have also come to rely on technology platforms to surveil and collect data on their own citizens as never before. Where once a slip of the tongue or casual observation would be as quickly forgotten, today a post on social media can be dissected and reinterpreted—weaponized—even years after the fact. Freedom of speech and association are the most at risk, ironically through the use of the very online platforms created for sharing ideas and associating with likeminded individuals.
Thousands of years ago, human beings simply could not foresee all that would unfold in the distant future with the planting of the first seeds. Millennia after its creation, we can look back on the invention of agriculture and clearly see the costs and benefits of that turn in our collective development. Agriculture led to a settled way of life, greater populations, and the centralization of power in political and religious institutions; the two latter of which, until the modern age, were quite often of the oppressive variety. Our digital revolution is no less radical in the changes that it is bringing to our species and the societies we live in, only at a speed and scale that dwarfs agriculture.
But even if agriculture was the worst mistake in history, that does not mean that everything flowing downstream from it—including the digital revolution—is necessarily a mistake as well, any more than is the offspring of a doomed marriage. What we can say with certainty is that inequality, human health, and individual liberties have been impacted by the digital revolution both positively and negatively. How these three spheres will develop in 50 or 100 years is anyone’s guess, but there are three possible outcomes: they will improve, go unchanged, or worsen. If the drawbacks stemming from the digital revolution continue on their current trajectory as digital technology becomes even more embedded within our lives, then perhaps we can be equally certain of the latter.
Data availability
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
Declarations
Conflict of interest
The author has no financial or non-financial interests directly or indirectly related to this work.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Alba D, Koeze E, Silver J (2021) What happened when Trump was banned on social media. The New York Times 7 June 2021. https://www.nytimes.com/interactive/2021/06/07/technology/trump-social-media-ban.html. Accessed 28 June 2021
Appleyard B The digital generation RSA J 2011 157 5548 18 23
Benson LV Berry MS Climate change and cultural response in the prehistoric American southwest Kiva 2009 75 1 87 117 10.1179/kiv.2009.75.1.005
Bohannon J Searching for the Google effect on people’s memory Science 2011 333 277 10.1126/science.333.6040.277 21764724
Borgerhoff Mulder M Bowles S Hertz T Intergenerational wealth transmission and the dynamics of inequality in small-scale societies Science 2009 326 682 688 10.1126/science.1178336 19900925
Brennan D (2021) Donald Trump's Twitter ban concerns world leaders, officials. Newsweek 1 Dec. 2021. https://www.newsweek.com/donald-trump-twitter-ban-concerns-world-leaders-officials-1560771. Accessed 1 May 2022
Buckley BM Anchukaitis KJ Penny D Fletcher R Cook ER Sano M Le NC Wichienkeeo A Ton MT Hong HM Marcus J Climate as a contributing factor in the demise of Angkor, Cambodia Proc Natl Acad Sci USA 2010 107 15 6748 6752 10.1073/pnas.0910827107 20351244
Calderón-Gómez D Casas-Mas B Urraco-Solanilla M Carlos Revilla J The labour digital divide: digital dimensions of labour market segmentation Work Organ Labour Glob 2020 14 2 7 30
Capatides C (2020) White silence on social media: why not saying anything is actually saying a lot. CBS News, 03 June 2020. https://www.cbsnews.com/news/white-silence-on-social-media-why-not-saying-anything-is-actually-saying-a-lot/. Accessed 10 Nov 2021
Chin J (2016) China’s new tool for social control: a credit rating for everything. The Wall Street Journal, 28 November 2016. https://www.wsj.com/articles/chinas-new-tool-for-social-control-a-credit-rating-for-everything-1480351590. Accessed 22 May 2020
Chun-Yao H Hau-Ning C Global digital divide: a dynamic analysis based on the Bass model J Public Policy Mark 2010 29 2 248 264 10.1509/jppm.29.2.248
Denyer S (2016) China’s plan to organize its society relies on ‘big data’ to rate everyone. The Washington Post 22 October 2016. https://www.washingtonpost.com/world/asia_pacific/chinas-plan-to-organize-its-whole-society-around-big-data-a-rating-for-everyone/2016/10/20/1cd0dd9c-9516-11e6-ae9d-0030ac1899cd_story.html?noredirect=on&utm_term=.9191ee7391dd. Accessed 22 May 2021
Dewan S Ganley D Kraemer KL Complementarities in the diffusion of personal computers and the Internet: implications for the global digital divide Inf Syst Res 2010 21 4 925 940 10.1287/isre.1080.0219
Di Bitetti MS Evidence for an important social role of allogrooming in a platyrrhine primate Anim Behav 1997 54 199 211 10.1006/anbe.1996.0416 9268450
Diamond J The rise and fall of the third chimpanzee: how our animal heritage affects the way we live 2002 London Vintage Books
Diamond J (1987) The worst mistake in the history of the human race. Discover Mag, May, 64–66
Dibble M (2020) Dean of Massachusetts nursing school fired after saying ‘everyone’s life matters’. Washington Examiner 2 July 2020. https://www.washingtonexaminer.com/news/dean-of-massachusetts-nursing-school-fired-after-saying-everyones-life-matters. Accessed 15 June 2021.
Dyble M Thorley J Page AE Smith D Bamberg Migliano A Engagement in agricultural work is associated with reduced leisure time among Agta hunter-gatherers Nat Hum Behav 2019 3 792 796 10.1038/s41562-019-0614-6 31110340
Faucher KX Fuchs C Alienation 2.0—symptoms of narcissism and aggression Social capital online: alienation and accumulation 2018 London University of Westminster Press 87 108
Firth JA Torous J Firth J Exploring the impact of Internet use on memory and attention processes Int J Environ Res Public Health 2020 17 9481 1 12
Fuller DQ Denham T Arroyo-Kalin M Lucas L Stevens CJ Qin L Allaby RG Purugganan MD Convergent evolution and parallelism in plant domestication revealed by an expanding archaeological record PNAS 2014 111 17 6147 6152 10.1073/pnas.1308937110 24753577
Good IJ Speculations concerning the first ultraintelligent machine Adv Comput 1966 6 31 88 10.1016/S0065-2458(08)60418-0
Gualandi PB Food habits and dental disease in an iron-age population Anthropol Anz 1992 50 1/2 67 82 10.1127/anthranz/50/1992/67 1637149
Haidt J Allen N Scrutinizing the effects of digital technology on mental health Nature 2020 578 226 227 10.1038/d41586-020-00296-x 32042091
Hamet P Tremblay J Artificial intelligence in medicine Metab Clin Exp 2017 69 S36 S40 10.1016/j.metabol.2017.01.011
Hansen-Bundy B (2018) A week inside WeLive, the utopian apartment complex that wants to disrupt city living. GQ 27 February 2018. https://www.gq.com/story/inside-welive. Accessed 17 May 2020
Hayden B Villeneuve S Feinman GM Price TD Who benefits from complexity? A view from Futuna Pathways to power: new perspectives on the emergence of social inequality 2010 New York Springer 95 145
Heidegger M The question concerning technology The question concerning technology and other essays 1977 New York Harper & Rowe Inc 3 35
Hui-Chen L Yun-Fang T Gwo-Jen H Hsin H From precision education to precision medicine Educ Technol Soc 2021 24 1 123 137
International Telecommunication Union (ITU) (2020) Measuring digital development, facts and figures, 2020. International Telecommunication Union Development Sector. https://www.itu.int/en/ITU-D/Statistics/Pages/ff2020interactive.aspx. Accessed 29 May 2021
Jennen B, Nussbaum A (2021) Germany and France oppose Trump’s Twitter exile. Bloomberg 11 January 2021. https://www.bloomberg.com/news/articles/2021-01-11/merkel-sees-closing-trump-s-social-media-accounts-problematic. Accessed 28 June 2021
Jun L Esmail F Lingjiang L Zhifeng K Decreased frontal lobe function in people with internet addiction disorder Neural Regen Res 2013 5 34 3225 3232
Kemp S (2017) Three billion people now use social media. Wearesocial.com. https://wearesocial.com/blog/2017/08/three-billion-people-now-use-social-media. Accessed 29 May 2021
Kennett DJ Breitenbach SFM Aquino VV Asmerom Y Awe J Baldini JUL Bartlein P Culleton BJ Ebert C Jazwa C Macri MJ Marwan N Polyak V Prüfer KM Ridley HE Sodemann H Winterhaider B Haug GH Development and disintegration of Maya political systems in response to climate change Science 2012 338 6108 788 791 10.1126/science.1226299 23139330
King SD Grave new world: the end of globalization, the return of history 2017 New Haven Yale University Press
Kleinman SS Understanding the digital divide: implications for college teaching Transform J Incl Sch Pedagog 2001 12 2 51 67
Laneri N Schwartz M Ur J Ritual and identity in rural Mesopotamia: Hirbemerdon Tepe and the upper Tigris River Valley in the middle Bronze Age Am J Archaeol 2015 119 4 533 564 10.3764/aja.119.4.0533
Larsen CS Biological changes in human populations with agriculture Annu Rev Anthropol 1995 24 185 213 10.1146/annurev.an.24.100195.001153
Lau C G, Haugh B A (2018) Megatrend issues in artificial intelligence and autonomous systems. Institute for Defense Analyses. http://www.jstor.org/stable/resrep22645. Accessed 20 June 2022
Lee D Park J Namkoong K Kim IY Jung Y Gray matter differences in the anterior cingulate and orbitofrontal cortex of young adults with Internet gaming disorder: surface-based morphometry J Behav Addict 2018 7 1 21 30 10.1556/2006.7.2018.20 29529887
Lee D (2018) Facebook’s data-sharing deals exposed. BBC News 19 December 2018. https://www.bbc.com/news/technology-46618582. Accessed 20 Dec 2021
Littman L Parent reports of adolescents and young adults perceived to show signs of a rapid onset of gender dysphoria PLoS ONE 2018 13 8 1 44 10.1371/journal.pone.0202330
Lukacs JR Sex differences in dental caries rates with the origin of agriculture in South Asia Curr Anthropol 1996 37 1 147 153 10.1086/204481
Lukavská K The immediate and long-term effects of time perspective on Internet gaming disorder J Behav Addict 2018 7 1 44 51 10.1556/2006.6.2017.089 29313730
Maguire GQ Jr McGee EM Implantable brain chips? Time for debate Hastings Cent Rep 1999 29 1 7 13 10.2307/3528533
Martin DL Goodman AH Health conditions before Columbus: paleopathology of native North Americans West J Med 2002 176 1 65 68 10.1136/ewjm.176.1.65 11788545
Mattison SM Smith EA Shenk MK Cochrane EE The evolution of inequality Evol Anthropol 2016 25 184 199 10.1002/evan.21491 27519458
O’Leary D B (2007) Escaping the progress trap. Geozone Communications, Montreal
O'Connor L Lundstrom K The impact of social marketing strategies on the information seeking behaviors of college students Ref User Serv Q 2011 50 4 351 365 10.5860/rusq.50n4.351
Pan N Yongxin Y Xin D Brain structures associated with Internet addiction tendency in adolescent online game players Front Psychiatry 2018 9 67 1 8 10.3389/fpsyt.2018.00067.Accessed19June2022 29410632
Perman S (2021) After NBC calls on HFPA to oust ex-president, citing ‘racist rhetoric,’ Phil Berk is out. Los Angeles Times 20 April 2021. https://www.latimes.com/entertainment-arts/business/story/2021-04-20/nbc-dick-clark-productions-condemn-former-hfpa-presidents-actions-ask-for-his-ouster. Accessed 1 May 2021
Price DT Bar-Yosef O Feinman GM Price TD Traces of inequality at the origins of agriculture in the ancient Near East Pathways to power: new perspectives on the emergence of social inequality 2010 New York Springer 147 168
Przybylski AK Weinstein N A large-scale test of the Goldilocks hypothesis: quantifying the relations between digital-screen use and the mental well-being of adolescents Psychol Sci 2017 28 2 204 215 10.1177/0956797616678438 28085574
Ramsdell CW The changing interpretation of the civil war J South Hist 1937 3 1 3 27 10.2307/2192113
Rao M Sanjib RB Iqbal SM Struggling with the digital divide: Internet infrastructure, policies and regulations Econ Pol Wkly 1999 34 46/47 3317 3320
Richardson M Hussain Z Griffith MD Problematic smartphone use, nature connectedness, and anxiety J Behav Addict 2018 7 1 109 116 10.1556/2006.7.2018.10 29415553
Rideout V, Robb M B (2019) The common sense census: media use by tweens and teens, 2019. Common Sense Media. https://www.commonsensemedia.org/sites/default/files/research/report/2019-census-8-to-18-full-report-updated.pdf. Accessed 15 June 2022
Roh D Bhang SY Choi JS The validation of implicit association test measures for smartphone and Internet addiction in at-risk children and adolescents J Behav Addict 2018 7 1 79 87 10.1556/2006.7.2018.02 29383939
Rosen E (2015) How young Vietnamese view the Vietnam War. The Atlantic 30 April 2015. https://www.theatlantic.com/international/archive/2015/04/youth-vietnam-war-fall-saigon/391769/. Accessed 3 June 2021
Sahlins M (1972) The original affluent society. In: Stone age economics. Aldine-Atherton, Inc, Chicago, pp 1–39
Scott JC The mission of the university: medieval to postmodern transformations J High Educ 2006 77 1 1 39 10.1353/jhe.2006.0007
Shin SC Rhee SN Aikens CM Chulmun Neolithic intensification, complexity, and emerging agriculture in Korea Asian Perspect 2012 51 1 68 109 10.1353/asi.2012.0008
Smith ME The role of social stratification in the Aztec empire: a view from the provinces Am Anthropol 1986 88 70 91 10.1525/aa.1986.88.1.02a00050
Smith ME Durable inequality in Aztec society J Anthropol Res 2021 77 2 162 186 10.1086/713639
So-Kum Tang C Yee WK Gan YQ Addiction to Internet use, online gaming, and online social networking among young adults in China, Singapore, and the United States Asia Pac J Public Health 2017 29 8 673 682 10.1177/1010539517739558 29191049
Sparks J Morris D [1967]) Allogrooming in primates: a review Primate ethology 2006 Piscataway Aldine Transaction 148 175
Sparrow B Liu J Wegner DM Google effects on memory: cognitive consequences of having information at our fingertips Science 2011 333 6043 776 778 10.1126/science.1207745 21764755
Storm BC Stone SM Benjamin AS Using the Internet to access information inflates future use of the Internet to access other information Memory 2017 25 6 717 723 10.1080/09658211.2016.1210171 27424847
Sultan O Combatting the rise of ISIS 2.0 and terrorism 3.0 Cyber Defense Rev 2017 2 3 41 50
Twenge JM Campbell KW Media use is linked to lower psychological well-being: evidence from three datasets Psychiatr Q 2019 90 311 331 10.1007/s11126-019-09630-7 30859387
Twenge JM Joiner TE Rogers ML Martin GN Increases in depressive symptoms, suicide-related outcomes, and suicide rates among US adolescents after 2010 and links to increased new media screen time Clinical Psychological Science 2018 6 3 17 10.1177/2167702617723376
Ubelaker DH Temporal trends of dental disease in ancient Ecuador Anthropologie 1992 30 1 99 102
Van der Meulen S Bruinsma M Man as ‘aggregate of data’ AI Soc 2018 34 343 354 10.1007/s00146-018-0852-6
Wade S (2017) China’s Social Credit System: black mirror or red herring? China Digital Times 16 Feb. 2017. https://chinadigitaltimes.net/2017/02/qa-shazeda-ahmed-on-chinas-social-credit-system/. Accessed 23 May 2021
Wang A B (2017) Former Facebook VP says social media is destroying society with ‘dopamine-driven feedback loops’. The Washington Post 12 December 2017. https://www.washingtonpost.com/news/the-switch/wp/2017/12/12/former-facebook-vp-says-social-media-is-destroying-society-with-dopamine-driven-feedback-loops/?noredirect=on&utm_term=.eed38ca274ac. Accessed 29 Aug 2021
Warschauer M Demystifying the digital divide Sci Am 2003 289 2 42 47 10.1038/scientificamerican0803-42
Wegner DM Ward AF How Google is changing your brain Sci Am 2013 309 6 58 61 10.1038/scientificamerican1213-58
Wilcox W B, Sturgeon S (2018) Too much Netflix, not enough chill: why young Americans are having less sex. Politico Magazine 08 Feb. 2018. https://www.politico.com/magazine/story/2018/02/08/why-young-americans-having-less-sex-216953. Accessed 24 May 2021
Winseck D The geopolitical economy of the global Internet infrastructure J Inf Policy 2017 7 228 267 10.5325/jinfopoli.7.2017.0228
World Bank (2022) 70% of 10-year-olds now in learning poverty, unable to read and understand a simple text. The World Bank. https://www.worldbank.org/en/news/press-release/2022/06/23/70-of-10-year-olds-now-in-learning-poverty-unable-to-read-and-understand-a-simple-text. Accessed 16 July 2022
Wright R A short history of progress 2004 New York Carroll & Graf Publishers
Wright LE White CD Human biology in the classic Maya collapse: evidence from paleopathology and paleodiet J World Prehist 1996 10 2 147 198 10.1007/BF02221075
YouTube (2017) An update on our commitment to fight terror content. YouTube.com. https://youtube.googleblog.com/2017/08/an-update-on-our-commitment-to-fight.html. Accessed 13 Sept 2021
Zhao S Elesh D The second digital divide: unequal access to social capital in the online world Int Rev Mod Sociol 2007 33 2 171 192
| 36465190 | PMC9707148 | NO-CC CODE | 2022-12-01 23:20:21 | no | AI Soc. 2022 Nov 28;:1-10 | utf-8 | AI Soc | 2,022 | 10.1007/s00146-022-01599-5 | oa_other |
==== Front
Curr Neurol Neurosci Rep
Curr Neurol Neurosci Rep
Current Neurology and Neuroscience Reports
1528-4042
1534-6293
Springer US New York
1247
10.1007/s11910-022-01247-x
Neurology of Systemic Diseases (J. Biller, Section Editor)
Neurological Complications Following COVID-19 Vaccination
Chatterjee Aparajita 1
Chakravarty Ambar [email protected]
2
1 grid.496593.4 Department of Neurology, Fortis Hospital, Kolkata, India
2 grid.416884.7 Department of Neurology, Vivekananda Institute of Medical Sciences, Kolkata, India
29 11 2022
114
16 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose of Review
A variety of neurological complications have been reported following the widespread use of the COVID-19 vaccines which may lead to vaccine hesitancy and serve as a major barrier to the public health aim of achieving protective herd immunity by vaccination. In this article, we review the available evidence regarding these neurological adverse events reported, to provide clarity regarding the same so that unfounded fears maybe put to rest.
Recent Findings
There is a greater than expected occurrence of severe neurological adverse events such as cortical sinus venous thrombosis, Bell’s palsy, transverse myelitis, and Guillain–Barré syndromes along with other common effects such as headaches following different kinds of COVID-19 vaccination. Precipitation of new onset demyelinating brain lesions with or without detection of specific antibodies and worsening of pre-existing neurological disorders (like epilepsy, multiple sclerosis) are also a matter of great concern though no conclusive evidence implicating the vaccines is available as of now.
Summary
The COVID-19 pandemic is far from being over. Till such time that a truly effective anti-viral drug is discovered, or an appropriate therapeutic strategy is developed, COVID-appropriate behavior and highly effective mass vaccination remain the only weapons in our armamentarium to fight this deadly disease. As often occurs with most therapeutic means for the treatment and prevention of any disease, vaccination against COVID-19 has its hazards. These range from the most trivial ones like fever, local pain and myalgias to several potentially serious cardiac and neurological complications. The latter group includes conditions like cerebral venous thrombosis (curiously often with thrombocytopenia), transverse myelitis and acute inflammatory demyelinating polyneuropathy amongst others. Fortunately, the number of reported patients with any of these serious complications is far too low for the total number of people vaccinated. Hence, the current evidence suggests that the benefits of vaccination far outweigh the risk of these events in majority of the patients. As of now, available evidence also does not recommend withholding vaccination in patients with pre-existing neurological disorders like epilepsy and MS, though adenoviral vaccines should be avoided in those with history of thrombotic events.
Keywords
COVID-19 vaccination
Neurological complications
Cortical sinus venous thrombosis
Guillain–Barré syndrome
Transverse myelitis (TM)
Multiple sclerosis (MS)
==== Body
pmcIntroduction
Immunization has long been the foundation of public health policy and has rendered various deadly diseases which once felled millions into the “rare,” “eradicated,” and “easily treatable” categories [1]. The World Health Organization (WHO) estimates that 2–3 million lives are saved each year by the current national immunization programs [2]. When humanity was first faced with the SARS-CoV-2 virus in the winter of 2019, there followed an initial period of “trial and error” method of treating the virus and its myriad manifestations. Work also progressed at breakneck speed for the identification of the genomic sequence of the virus and then to develop a vaccine against it. Scientists and researchers were successful like never before in coming up with several effective vaccines against the deadly virus in record time, leading to an upsurge of hope around the world that this devastating pandemic would finally be over. However, like every other human creation, these vaccines proved to be a double-edged sword. Soon after the introduction of the widespread vaccine drives, several complications (some substantiated, some not) began to be reported. The neurological system seemed to be one of the most affected by these vaccines and is the subject of our discourse here [3, 4].
Principles of Vaccination
The fundamental concept of vaccination rests on the principle of “immune memory” [5]. A vaccine can be defined as a biological/biochemical product that can be administered to “safely” produce an immune response that confers “protection” against infection and/or disease on subsequent exposure to a pathogen. The term “protection” refers to a state in which an individual will not develop a disease after being challenged with a pathogen [5]. A vaccine contains “antigens” that are either derived from the pathogen or produced synthetically to represent components of the same. These antigens are capable of inducing an immune response in the host that the immune system will then remember, and therefore protect the host against subsequent exposures to these pathogens. This adaptive immune response is mediated by B cells which produce antibodies (humoral immunity) and by T cells (cellular immunity). All vaccines in routine use confer protection through the induction of antibodies [5]. These antigens are usually various protein components of the pathogen, but rarely polysaccharide vaccines have also been used against bacterial infections. Efficacy of a vaccine is measured in terms of the protection conferred by it and can be estimated in clinical trials endpoints such as prevention of infection, a reduction in disease severity or a decreased rate of hospitalization and /or death.
COVID Vaccines and Their Mechanisms of Action
COVID-19 vaccines prime the immune system to create antibodies against SARS-CoV-2. The vaccines use a structure similar to spike (S) protein, which is present on the surface of SARS-CoV-2 and is used for viral endocytosis to the host cells. The COVID-19 vaccines include messenger RNA (mRNA), vector, protein subunit, and inactivated/weakened whole virus vaccines [6•, 7•, 8•, 9•, 10•, 11•, 12•, 13•]. Engineered mRNA COVID-19 vaccines namely Pfizer-BioNTech (BNT162b2) and the Moderna (mRNA-1273) give the cells instructions to make S protein which then leads to the generation of protective antibodies. After delivering instructions, the injected mRNA breaks down and does not enter the nucleus of the host cells [6•, 7•]. In vector COVID-19 vaccines which include Ad26.COV2.S [8•], Sputnik V (rAd26-S and rAd5-S) [9•], and ChAdOx1 nCoV-19 SARS-COV-2 (AstraZeneca/Covishield) [10•], genetic material is inserted in a viral vector. The vector delivers the genetic material to the host cells that make copies of the S protein on their surfaces. The immune system then responds by creating antibodies and defensive mononuclear white blood cells like lymphocytes and plasma cells. Protein subunit COVID-19 vaccines such as Novavax (NVX-CoV2373) include only harmless S protein which stimulates the immune system [11•]. Inactivated or weakened COVID-19 vaccines do not cause disease, but still, stimulate the immune system, such as the Covaxin developed by Bharat Biotech and the Indian Council of Medical Research, and the CoronaVac and Sinopharm developed in China [12•, 13•, 14•]. The very recently introduced bivalent COVID-19 vaccines include a component of the original virus strain to provide broad protection against COVID-19 and a component of the omicron variant to provide better protection against COVID-19 caused by the omicron variant. The FDA authorized bivalent formulations of the Moderna and Pfizer-BioNTech COVID-19 vaccines for use as a single booster dose at least 2 months after completing primary or booster vaccination in subjects above the age of 12 years.
Non-neurological Complications of COVID-19 Vaccines
The most common non-neurological adverse events following COVID-19 vaccination include fever, chills and rigor, fatigue, muscle and joint pains, nausea, diarrhea, and rashes. Local site reactions including pain and tenderness, local warmth, pruritus, bruising, and axillary lymphadenopathy have also been commonly reported. Local and systemic allergic reactions may occur [15••, 16]. Pillay and colleagues conducted a systematic review of myocarditis and pericarditis after COVID-19 mRNA vaccination. The relative incidence of myocarditis was highest among young males between 12 and 29 years of age after a second dose, and with Moderna’s mRNA vaccine than after Pfizer-BioNTech’s vaccine. The risk of myocarditis/pericarditis is lower when the second dose is administered after 30 days of the first dose. Finally, the study concluded that most patients had a good outcome and mortality was low after these cardiac complications. The occurrence of myocarditis overall is rare after vaccination [17•].
Neurological Complications of COVID-19 Vaccines
Neurological adverse events following COVID-19 vaccination are generally mild and transient [18••]. The most commonly reported side effects are headache, fatigue, muscle, and joint pains. As with any other injectable vaccine, local injection site reactions such as swelling, redness, and pain are also well known. These mild neurological effects are seen with all kinds of COVID-19 vaccines.
Incidence of Neurological Adverse Events
In a Mexican study, 6536 adverse events were recorded after Pfizer-BioNTech vaccination among 704,003 subjects. Sixty-five percent of these had at least one neurologic complication, 99% were mild and transient. The most commonly reported events included headache, transient sensory symptoms, and generalized weakness. The only severe adverse events recorded were seizures in 7, Guillain-Barré syndrome in 3 patients, and transverse myelitis in 2 patients [19••]. Similar rates were seen in a South Korean study by Kim et al. looking at adverse effects after the first dose of ChAdOx1 nCoV-19 and BNT162b2 vaccines [20]. In another study conducted amongst healthcare workers in the UK after two doses of the BNT162b2 mRNA vaccine, the most common adverse effects were myalgia, headache, chills, fatigue, and fever. These complications were higher after the second dose than with the first dose [21].
Classification of Neurological Adverse Effects of COVID Vaccines
There are two ways in which one may classify the neurological side effects of these COVID vaccines; one is by classifying these as those affecting the central nervous system and those affecting the peripheral nervous system. Another way of classifying these maybe those that are considered serious/severe and those which are minor. The former classification scheme is depicted in Fig. 1. Below, we elaborate on individual neurological complications reported after vaccination.Fig. 1 COVID 19 vaccine related neurological adverse events
Neurological Adverse Events
Headache is the most frequent mild neurological complication reported and occurs with all of the approved COVID vaccines. This usually starts within the first few hours after the vaccination and in most cases, resolves spontaneously within the next 48 h. The location is usually bifrontal or temporal, dull aching in nature with varying intensity. Frequent accompaniments include fever, chills, myalgia, and fatigue [22]. However, there is another type of headache which is far more severe and needs to be distinguished from the first type. This occurs about a week after the vaccine and is secondary to more ominous neurological complications like cortical venous thrombosis, intracerebral, or subarachnoid hemorrhage [23•].
Severe Neurological Adverse Events
Severe or serious adverse reaction following immunization is defined as a post-vaccination event that is either life-threatening, requires hospitalization, or results in severe disability. The WHO lists Guillain-Barré syndrome, seizures, syncope, encephalitis, Bell’s palsy, and strokes as serious neurologic adverse events [24]. These may be the source of vaccine hesitancy, and it is important to report and analyze these events to better understand whether a clear-cut relationship between the two actually exists.
Cerebral Vascular Events
The whole range of cerebral vascular events from arterial strokes and venous thromboses to intracerebral and subarachnoid hemorrhages have been reported, predominantly with the adenoviral vector vaccines, often associated with severe vaccine-induced thrombotic thrombocytopenia (VITT), which manifests 5 to 30 days after the vaccination. The clinical picture is similar to that of heparin-induced thrombocytopenia [25••].
Cerebral Venous Sinus Thrombosis
Cerebral venous sinus thrombosis (CSVT) is the most feared COVID vaccine-associated neurological complication. This should be suspected in all patients who develop a persistent headache post-vaccination, which is unresponsive to analgesics and often associated with a dip in sensorium and/or seizures and focal deficits. As mentioned earlier, CSVT is commonly associated with VITT.
Mechanism
VITT is caused by antibodies that recognize platelet factor 4 (PF4) bound to platelets. The antibodies are immunoglobulin G (IgG) molecules that activate platelets via low-affinity platelet FcγIIa receptors causing platelet aggregation and activation, leading to hypercoagulability, but with decreased platelet count. A state of thrombotic thrombocytopenia ensues resulting in venous thrombosis with marked elevation of D-dimer due to increased generation of fibrin degradation products. Microscopic findings show inflammatory vascular thrombotic occlusions occurring in the vessels of multiple body organs [26••]. The vector-based vaccines contain genetic material of SARS-COV-2 that is capable of encoding the spike glycoprotein. Possibly leaked genetic material binds to platelet factor 4 that subsequently leads to the formation of autoantibodies [27••, 28].
Post-vaccination CSVT has a propensity for younger females [23•, 25••, 26••, 27••, 28–34, 35••, 36••, 37, 38]. The commonly implicated vaccines in CSVT have been the Astra Zeneca and Johnson and Johnson vaccines [26••]. Scully et al. published a case series of 21 patients who had venous sinus thrombosis with documented thrombocytopenia, which developed 6 to 24 days after the first dose of the viral vector-based vaccines [27••]. Presence of autoantibodies against platelet factor 4 and elevated D-dimer levels was also demonstrated.
Tiede reported five cases of prothrombotic immune thrombocytopenia after vaccination with viral vector-based vaccine (Vaxzevria) in Germany which clinically manifested as CSVT, splanchnic vein thrombosis, arterial cerebral thromboembolism, and thrombotic microangiopathy within 2 weeks of vaccination [31].
Pottegård et al. assessed the rates of thrombotic events in the first 28 days after vaccination with the Oxford-AstraZeneca vaccine ChAdOx1-S in Denmark in 148,792 people and 132,472 people in Norway and compared them with rates observed in the general populations to determine the excess risk, if any. Authors estimated an increased rate for venous thromboembolism corresponding to 11 excess events per 100,000 vaccinations with 2.5 times excess cerebral venous thrombosis events per 100,000 vaccinations [35••].
Krzywicki et al. analyzed data of 213 cases with post-vaccination CSVT, of which 187 occurred after adenoviral vector vaccines and 26 after mRNA vaccines; thrombocytopenia was only reported with adenoviral vaccines. CSVT after adenoviral vaccines also carried poorer prognosis; 38% patients in adenoviral vaccine group died, while 20% succumbed in the mRNA vaccine group [36••]. Perry et al. analyzed 95 cases of vaccine associated CSVT, 70 of which were associated with VITT [37]. Patients with VITT were younger, had more thrombosed intracranial veins and poorer outcomes. Use of non-heparin anticoagulants and IVIg was associated with better outcomes [38].
When diagnosing CSVT related to COVID-19 vaccination, a temporal relationship must be established (within 30 days of vaccination) and other prothrombotic states must be excluded. However, whether the vaccines have a role in precipitating the thrombosis in susceptible individuals with pre-existing prothrombic states is not established [39•] (Fig. 2). Such patients should also be checked for fresh COVID-19 infection by RT-PCR as CSVT may also occur in infected cases.Fig. 2 A, B MR venography in a 60-year-old lady with prior history of migraine, presenting with a three week history of increasing headaches starting about a fortnight after her second dose of AstraZeneca (COVISHIELD vaccine), showing segmental partial occlusion of both transverse sinuses. She did not have any thrombocytopenia or raised D-dimer but had a low protein S level suggesting a pre-existing prothrombotic state. She was started on apixaban 2.5 mg twice daily. Her headaches were relieved. Repeat venography done over 6 months later revealed recanalization of the thrombosed segments (C). Her protein S level was persistently low. She was advised to continue apixaban indefinitely. She had no problem after her third booster dose of the same vaccine (adapted from [39•], with permission)
National Institute for Health and Care Excellence (NICE) guidelines recommend treatment of patients with VITT and thrombosis with intravenous immunoglobulin, at a dose of 1 g/kg which may be repeated if there is further clinical deterioration. In patients with insufficient response, methylprednisolone 1 g intravenously for 3 days or dexamethasone 20 to 40 mg for 4 days can be used [40••]. Heparin should be avoided, and alternative anticoagulants like argatroban, bivalirudin, fondaparinux, rivaroxaban, or apixaban should be used [36••, 37, 38]. Patients with very low platelet count should be treated either with argatroban alone or a combination of argatroban and platelet transfusion [41].
Arterial Ischemic Stroke
Ischemic stroke following COVID-19 vaccination are also reported though rarely, usually seen in the context of VITT [42•]. Pottegard et al. described an increased rate of venous thromboembolic events but not of arterial events among recipients of ChAdOx1 nCoV-19 [35••]. Blauenfeldt et al. described a female who presented with bilateral adrenal necrosis, 7 days after receiving the adenoviral (ChAdOx1) vaccine. She subsequently developed a malignant right cerebral infarction, secondary to occlusion of the right internal carotid artery leading to her death. She had thrombocytopenia, elevated D-dimer, and PF-4 antibodies in her blood [43].
De Michele et al. described two fatal cases with VITT-associated malignant MCA infarcts one of which was treated with mechanical thrombectomy, with additional portal and pulmonary vein thrombosis [44]. Al-Mayhani et al. described three cases of VITT presenting with arterial strokes [45]. Authors opined that young patients with arterial stroke after receiving the COVID-19 vaccine should always be evaluated for vaccine-induced thrombotic thrombocytopenia.
Question may be raised whether vaccine-eligible subjects need prior screening for any coagulopathy before vaccination which would be very expensive. Duration of oral anticoagulation therapy may be 6 months for subjects without pre-existing prothrombotic states, in whom treatment should be continued lifelong.
Stroke Mimics and COVID-19 Vaccination
A recent report from Thailand highlights a large number of cases with a distinctive focal neurological syndrome among those receiving the Chinese vaccine CoronaVac but not with the AstraZeneca vaccine [46•]. This develops within the first few days of the first dose of vaccination and consists of transient hemisensory or hemimotor disturbances, at times associated with visual phenomenon developing in the corresponding hemifield of vision. Hemicranial or holocranial headaches are often present. Diffusion-weighted MRI brain and MR angiography were normal in all patients. SPECT studies however demonstrated hypoperfusion in the contralateral hemisphere during the acute phase followed by hyperperfusion with clearing of symptoms. This has been likened to the phenomenon of “cortical spreading depression” occurring in subjects with migraine with aura. Postulations regarding the triggers for this include precipitation by vaccination-related stress and an immune response to the vaccine.
ICH
Both intracerebral hemorrhage (ICH) and subarachnoid hemorrhage (SAH) can occur after COVID-19 vaccination, which can be primary or secondary to venous thrombosis [26••, 27••, 47–49]. While ICH after COVID-19 vaccination can occur in the context of VITT, de Melo Silva et al. described a primary hemorrhagic stroke following ChAdOx1nCoV-19 vaccination in a patient without thrombocytopenia, or coagulopathy [48]. Finsterer et al. suggested that the second dose of vaccination may be followed by ICH even when the first dose was uneventful [49]. Most of these cases were reported following ChAdOx1 nCoV-19 vaccine and in people 30 to 57 years of age, 5 to 12 days after the vaccination.
SCLS
This is a potentially life-threatening immune disorder characterized by transient, recurrent episodes of vascular endothelial hyperpermeability triggered by viral upper respiratory infections, and maybe seen after both adenoviral and mRNA COVID vaccines. Hypotension, hemo-concentration, hypoalbuminemia, and anasarca are prominent features. Treatment is aimed at correcting hypovolemia and avoiding end-organ damage; prophylactic monthly administration of immunoglobulins can prevent further episodes. Adenoviral vector vaccines against SARS-CoV-2 are not advised for patients with a history of systemic capillary leak syndrome (SCLS). Monoclonal gammopathy of uncertain significance is observed in 68–85% of patients with SCLS [50].
Encephalopathy
Acute encephalopathy clinically manifests either with delirium, decreased consciousness, and a fluctuating sensorium with impaired attention. Reports of acute encephalopathy have been reported following mRNA vaccines, especially the Moderna vaccine [51, 52•, 53]. Elderly patients are affected and present with decreased attention and confusion with diffuse slowing on electroencephalogram (EEG) and negative neuroimaging. EEG may also show triphasic waves or nonconvulsive status. Workup for other causes of delirium is negative. Steroid-responsive encephalopathy was also reported in a young male following vaccination by Ali-Mashdali with elevated CSF protein levels [53].
ADEM
This is an acute inflammatory demyelinating disorder of the central nervous system which is associated with recent infections and vaccinations. Case reports of acute disseminated encephalomyelitis (ADEM) have been reported post-COVID vaccination with SinoVac and the Astra-Zeneca vaccines [54••, 55, 56]. Clinical manifestations include altered sensorium, seizures and focal neurological deficits. MRI showed multiple T2/FLAIR hyperintense lesions involving the periventricular and juxtacortical regions. Treatment is in the form of high-dose intravenous steroids and/or intravenous immunoglobulins. Most patients recovered well, though mortality had been reported.
Post‑vaccinal Encephalitis
Zuhorn et al. reported a series 3 patients, who presented with post-vaccinal encephalitis, occurring 7 to 11 days after administration of adenoviral ChAdOx1 nCov-19 vaccine. All patients fulfilled the diagnostic criteria for possible autoimmune encephalitis and presented with cognitive decline, seizures, and gait dysfunction; neuroimaging was non-contributory. CSF pleocytosis and good response to corticosteroids was noted in all [57••]. Presentations of brainstem encephalitis with diplopia and opsoclonus-myoclonus syndromes after COVID vaccination have also been reported, all with good response to steroid therapy [58].
Acute necrotizing encephalopathy (ANE) is a rare neurological disorder arising from a para- or post-infectious “cytokine storm.” Bensaidane et al. reported the case of a 56-year-old male who developed ANE 2 days following the first dose of ChAdOx1 nCoV-19 vaccination. MRI brain showed hyperintensities involving bilateral thalami, with diffusion restriction. The patient was treated with high-dose corticosteroids [59].
Transverse Myelitis
Acute transverse myelitis is a rare, acquired inflammatory spinal cord disorder manifesting with rapid onset weakness, sensory loss and bowel and bladder dysfunction. MRI demonstrates T2/FLAIR hyperintensity which is often holocord and maybe longitudinally extensive. Autoimmunity via mechanism of molecular mimicry is implicated. Adenoviral-based vaccines are more frequently associated with this presentation, though even inactivated virus vaccine and mRNA-based vaccines have been implicated [60, 61••, 62–65]. Both short and long segment myelitis have been reported after COVID vaccination; outcomes are favorable with high dose intravenous steroids.
During the phase III trial of Oxford/AstraZeneca vaccine, 2 patients developed transverse myelitis; one 14 days after the booster shot which was considered to be idiopathic. The second case was reported 68 days post-vaccination in a patient who was previously diagnosed with multiple sclerosis and was considered to be not associated with vaccination [63, 64]. Possibly, antigens present in the COVID-19 vaccine, or its adenovirus adjuvant induce an immunological reaction in the spinal cord leading to myelopathy. The occurrence of 3 reported acute transverse myelitis adverse events amongst 11,636 participants in the vaccine trials was considered high and a cause of concern [65]. Overall, the incidence of transverse myelitis after COVID vaccination remains higher than is expected in the general population and prompt diagnosis and treatment is needed to ensure good outcomes.
COVID-19 Vaccination and Pre-existing or New-Onset CNS Demyelinating Disorders
Quintanilla-Bordás et al. reported a case series of four patients that received mRNA COVID-19 vaccination while already suffering from symptoms related to CNS demyelination due to yet undiagnosed MS [66•]. All of these patients presented with exacerbation of ongoing symptoms after vaccination with MRI features suggestive of highly active MS, fulfilling McDonald 2017 criteria; high serum light-neurofilament levels and oligoclonal bands restricted to the CSF. It is recommended that patients experiencing acute neurological symptoms should seek medical attention, especially before vaccination.
On the other hand, Epstein et al. in a multicenter US study including 1164 participants, found no evidence of neuroimmunological disease worsening after vaccination [67••]. Dinoto et al. in another multicenter Italian study assessed the frequency of relapses after SARS-CoV-2 vaccination (Pfizer and Moderna) in patients with NMOSD and MOGAD. The frequency of relapses within one month of vaccination was 4% (1/26) in the AQP4-IgG + NMOSD group and 0 in the MOGAD group [68]. In these patients, the potential benefits of vaccination outweighed the risk of relapses. The jury is still out about the safety, at least of the mRNA vaccines, for patients with active relapsing–remitting MS, NMOSD and MOGAD. The need of the hour is large-scale real-world evidence for further validation.
Neuroleptic Malignant Syndrome
This is a life-threatening complication of many antipsychotic medications presenting with hyperpyrexia, altered sensorium, muscle rigidity, and autonomic dysfunction with markedly elevated creatine phosphokinase levels and subsequent renal dysfunction due to myoglobinuria. Various case reports have recorded the occurrence of this syndrome in elderly patients on antipsychotics who had recently received the COVID-19 vaccine, the real offender remaining uncertain [69, 70].
Seizures
There have been case reports of new-onset seizures in patients after COVID vaccination, where neuroimaging revealed no acute changes, no pre-existing brain lesions and workup for other causes of acute symptomatic seizures was negative. Seizure semiology may range from focal seizures with impaired awareness to bilateral tonic–clonic seizures and status epilepticus [71, 72, 73•]. A more important question is whether COVID-19 vaccination is safe in subjects with pre-existing epilepsy. Various studies have demonstrated that there is no significant effect of COVID vaccines on seizure frequency in patients with epilepsy. Such studies demonstrate that COVID-19 vaccines have a good safety and tolerability profile in the short term, in patients with epilepsy though there is high levels of vaccine hesitancy amongst these patients [74••, 75•, 76–78]. A UK survey noted that in most people with Dravet syndrome, SARS-CoV-2 vaccine does not appear to be associated with an increase in the frequency or duration of seizures, even in those who develop fever post-vaccination [79•]. Similarly, the safety of a Chinese virus inactivated vaccine (Sirolimus) in children with tuberous sclerosis (TS) who were being treated with the mTOR inhibitor rapamycin has also been demonstrated [80].
Bell’s Palsy
Several case reports have demonstrated the relationship between COVID-19 vaccination and Bell’s palsy, especially with the mRNA vaccines [81–84, 85••]. An article published in the Lancet analyzed the combined phase 3 data of Pfizer and Moderna trials and noted that the rate of Bell’s palsy was 3.5 to 7 times higher than expected in the general population [85••]. Facial palsy in most cases recovered with or without treatment. This being such a self-limited side effect, considerations about an elevated risk for Bell’s palsy should not lead to any vaccine hesitancy.
Olfactory Dysfunction
Olfactory dysfunction is the most frequent neurological complication of COVID-19 infection. Lechien et al. reported 6 cases of post-COVID vaccine olfactory and gustatory disorders in patients with negative nasal swab, 4 after the AstraZeneca vaccine and 2 after Pfizer vaccine, lasting from 4 to 42 days with no long-term side effects [86]. Keir reported a case in which the subject felt as if he was “smelling smoke” after the Pfizer vaccine; MRI brain showed olfactory bulbs and bilateral olfactory tract enhancement—the exact pathogenesis remaining unknown [87].
Abducens Nerve Palsy
Multiple case reports have reported a temporal relationship between COVID vaccination and isolated sixth cranial nerve palsies, with normal neuroimaging where no other causative factors were found, likely secondary to immune-mediated localized demyelination or vasculitis [88].
Ocular Manifestations
A variety of ocular manifestations have been reported post-vaccination, including lid edema, superior ophthalmic vein thrombosis, Tolosa-Hunt Syndrome, uveitis, choroiditis, retinal necrosis, macular retinopathy, central retinal vein occlusion, and optic neuritis (ON). Again, the incidence of these adverse events was so low that no definitive causal relationship could be attributed to the vaccines [89–91]. A multinational review of ON after vaccination found that most cases occurred after AstraZeneca vaccination with one-fourth showing myelin oligodendrogliocyte (MOG) antibody positivity, and visual outcomes were good, with or without treatment. The MOG antibody-positive patients showed contrast enhancement in the whole length of the optic nerve. This study also found that the incidence of post-vaccination ON was lower than the incidence of ON otherwise [92••].
Otologic Manifestations
A variety of otologic manifestations have been reported following COVID-19 vaccination, including new-onset, or worsening of pre-existing hearing loss, vertigo, and tinnitus. However, most authors concluded that there were no significant increases in otologic complications after COVID-19 vaccination as compared with the incidence in the general population [93].
GBS
Guillain‑Barré syndrome (GBS) is a rare but serious post-infectious immune-mediated polyradiculoneuropathy which manifests as a lower motor neuron type of areflexic sensorimotor non-length dependent quadriparesis. Significant medical literature exists which has established the link between COVID infection and GBS. Similarly, all kinds of COVID-19 vaccines have also been found to be associated with Guillain-Barré syndrome, though the incidence is higher with adenovirus-based vaccines [94, 95•, 96–98]. Post-vaccination GBS generally affects older adults within 2 weeks of vaccine administration. Clinical presentation is indistinguishable from GBS secondary to other causes. Nerve conduction studies show a demyelinating sensorimotor polyradiculoneuropathy pattern, and CSF examination shows albumino-cytological dissociation. Response to immunotherapy is generally good. The proposed pathogenesis is via autoantibody-mediated immunological damage of peripheral nerves and radicles through the mechanism of molecular mimicry between structural components of peripheral nerves and the vaccine components.
Small Fiber Neuropathy
A couple of case reports have demonstrated that small fiber neuropathy may be a rare complication of COVID-19 vaccination. Patients present with a severe burning sensation involving the extremities with normal electrodiagnostic studies [99].
Parsonage‑Turner Syndrome
Parsonage-Turner syndrome or neuralgic amyotrophy usually manifests clinically with acute onset unilateral shoulder pain with progressive involvement of the brachial plexus and upper limb weakness. Some case reports have documented the occurrence of Parsonage-Turner syndrome following COVID-19 vaccination [100].
Herpes Zoster
McMahon et al. reviewed cutaneous reactions to mRNA COVID-19 vaccines in 414 patients, of which 5 (1.9%) developed herpes zoster [101]. These vaccines possibly induce dysregulation of T-cell function, thereby allowing latent herpes zoster virus to reactivate.
Myositis and Rhabdomyolysis
Cases of fatal myositis with rhabdomyolysis and acute kidney injury necessitating renal replacement therapy have been reported post-COVID vaccination [102•, 103]. Kimura and colleagues described a patient who developed recurrent, relapsing muscle weakness after the Pfizer vaccine with MRI demonstrating myoedema.
Worsening of Pre-existing Neurological Disorders
Exacerbations of previous neurological diseases have been documented with COVID vaccinations. The cases of demyelinating disorders and epilepsy had already been discussed. Other disorders showing transient worsening following vaccination include myasthenia gravis, dementia and movement disorders such as Parkinsonism where temporary worsening of motor symptoms and severe dyskinesias have been reported. Even though medical literature is far from conclusive in these cases because of the low numbers, it is evident that the risks of COVID-19 infection far outweigh the transient exacerbation of these pre-existing disorders [104••, 105, 106, 107••, 108•].
Functional Neurological Disorders
Functional neurological disorders (FNDs) are those which produce varied neurological manifestations without any demonstrable abnormalities on extensive workup and are often triggered by physical or emotional stress. A variety of these FNDs have come to forefront after the COVID-19 vaccine drive; non-epileptic seizures characterized by paroxysms of bizarre hyperkinetic movements with no electrographic correlate, while others presented as variable limb weakness mimicking a cerebral vascular event [108•, 109–111, 112••, 113]. Ercoli and colleagues described a middle-aged man who immediately reported bilateral facial weakness which resolved in an hour after vaccination. The same gentleman after the second dose of vaccine, complained of respiratory distress and swollen tongue which resolved after receiving corticosteroids; however, he developed new symptoms in the form of right hemiparesis, and then facial sensory loss. A detailed workup of this patient failed to demonstrate any abnormality to explain the aforementioned varied manifestations and a diagnosis of functional neurological disorder was arrived at [113].
Concluding Remarks
Within a little over 2.5 years of the emergence of the novel SARS-CoV-2 virus infection that caused the worst pandemic the world had ever seen, the causative pathogen was determined, vaccine targets identified, mass scale vaccine manufacturing developed, large-scale phase 1 through phase 3 trials were conducted at multinational levels, data have been reported and finally millions of the world population have been vaccinated at least with one or two doses. This process demonstrates what is possible in the context of motivated collaboration among key sectors of society, including medical personnel, research workers, government, industry regulators, and the general public. While lessons learned from this endeavor should allow us to be better prepared for the next pandemic pathogen, a very large section of the world population, especially in the developing world, remain far from receiving the desired protection. Risks associated with emergence of newer variants, which might be more transmissible or even more virulent than the delta variant, loom large in the minds of clinicians and scientists who cannot be complacent. Viruses spreading rampantly would invariably mutate; whether such mutations would lead to a “gain of function” or a “loss of function” remains unpredictable. Development of more effective vaccines is the need of the day and the risks from vaccination, neurological or otherwise, would never match the havoc caused by the virus [114]. True, complications like CVST, ANE, GBS, and ADEM are potentially fatal ones; but almost all of these are treatable ones if detected early and therapeutic results are commonly predictable. This is very much unlike cases with severe COVID-19 infection. Furthermore, most severe neurological adverse events have been reported either in isolated case reports or small cases series. A causal association of these events is controversial; larger collaborative prospective studies are needed to establish causality. The overwhelming evidence as of now supports widescale vaccination as the risks of COVID-19 infection far outweigh those which may occur secondary to vaccination, and hence should not be the cause for vaccine hesitancy.
Author Contribution
Both authors contributed equally to the preparation of the manuscript.
Declarations
Conflict of Interest
Aparajita Chatterjee and Ambar Chakravarty each declare no potential conflicts of interest.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
This article is part of Topical Collection on Neurology of Systemic Diseases
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
1. World Health Organization. Global vaccine action plan 2011–2020. WHO. https://www.who.int/immunization/global_vaccine_action_plan/GVAP_doc_2011_2020/en/ (2013). Accessed Oct 2022.
2. World Health Organization. Child mortality and causes of death. WHO. https://www.who.int/gho/child_health/mortality/mortality_under_five_text/en/ (2020). Accessed Oct 2022.
3. Graham BS Rapid COVID-19 vaccine development Science 2020 368 6494 945 946 10.1126/science.abb8923 32385100
4. Forni G Mantovani A COVID-19 Commission of Accademia Nazionale dei Lincei, Rome COVID-19 vaccines: where we stand and challenges ahead Cell Death Differ 2021 28 2 626 639 10.1038/s41418-020-00720-9 33479399
5. Pollard AJ Bijker EM A guide to vaccinology: from basic principles to new developments Nat Rev Immunol 2021 21 83 100 10.1038/s41577-020-00479-7 33353987
6. Polack FP Thomas SJ Kitchin N Absalon J Gurtman A Lockhart S Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine N Engl J Med 2020 383 2603 2615 10.1056/NEJMoa2034577 33301246
7. • Jackson LA, Anderson EJ, Rouphael NG, Roberts PC, Makhene M, Coler RN, et al. An mRNA vaccine against SARS-CoV-2— preliminary report. N Engl J Med Overseas Ed. 2020;383:1920–31. 10.1056/NEJMoa2022483. This article discusses the mechanism of action and rate of adverse events of the Moderna vaccine against COVID19.
8. Sadoff J Gray G Vandebosch A Cárdenas V Shukarev G Grinsztejn B Safety and efficacy of single-dose Ad26.COV2.S vaccine against Covid-19 N Engl J Med 2021 384 2187 201 10.1056/NEJMoa2101544 33882225
9. • Logunov DY, Dolzhikova IV, Zubkova OV, Tukhvatullin AI, Shcheblyakov DV, Dzharullaeva AS, et al. Safety and immunogenicity of an rAd26 and rAd5 vector-based heterologous prime-boost COVID-19 vaccine in two formulations: two open, non-randomised phase 1/2 studies from Russia. Lancet. 2020;396:887–97. 10.1016/S0140-6736(20)31866-3. Article discussing the principle of the Sputnik V vaccine along with phase III trial data regarding efficacy and tolerability.
10. Voysey M Clemens SAC Madhi SA Weckx LY Folegatti PM Aley PK Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomized controlled trials in Brazil, South Africa, and the UK Lancet 2021 397 99 111 10.1016/S0140-6736(20)32661-1 33306989
11. Heath PT Galiza EP Baxter DN Boffito M Browne D Burns F Safety and efficacy of NVX-CoV2373 Covid-19 vaccine N Engl J Med 2021 385 13 1172 1183 10.1056/NEJMoa2107659 34192426
12. Ella R Reddy S Blackwelder W Potdar V Yadav P Sarangi V Efficacy, safety, and lot-to-lot immunogenicity of an inactivated SARS-CoV-2 vaccine (BBV152): interim results of a randomised, double-blind, controlled, phase 3 trial Lancet 2021 398 10317 2173 2184 10.1016/S0140-6736(21)02000-6 34774196
13. Saeed BQ Al-Shahrabi R Alhaj SS Alkokhardi ZM Adrees AO Side effects and perceptions following Sinopharm COVID-19 vaccination Int J Infect Dis 2021 111 219 226 10.1016/j.ijid.2021.08.013 34384899
14. Wu Z Hu Y Xu M Chen Z Yang W Jiang Z Safety, tolerability, and immunogenicity of an inactivated SARS-CoV-2 vaccine (CoronaVac) in healthy adults aged 60 years and older: a randomised, double-blind, placebo-controlled, phase 1/2 clinical trial Lancet Infect Dis 2021 21 6 803 812 10.1016/S1473-3099(20)30987-7 33548194
15. Kaur RJ Dutta S Bhardwaj P Charan J Dhingra S Mitra P Adverse events reported from COVID-19 vaccine trials: a systematic review Indian J Clin Biochem 2021 36 4 427 439 10.1007/s12291-021-00968-z 33814753
16. Ganesan S Al Ketbi LMB Al Kaabi N Al Mansoori M Al Maskari NN Al Shamsi MS Vaccine side effects following COVID-19 vaccination among the residents of the UAE-an observational study Front Public Health 2022 10 876336 10.3389/fpubh.2022.876336 35602146
17. Pillay J Gaudet L Wingert A Bialy L Mackie AS Paterson DI Incidence, risk factors, natural history, and hypothesised mechanisms of myocarditis and pericarditis following covid-19 vaccination: living evidence syntheses and review BMJ 2022 378 e069445 10.1136/bmj-2021-069445 35830976
18. Garg RK Paliwal VK Spectrum of neurological complications following COVID-19 vaccination Neurol Sci 2022 43 1 3 40 10.1007/s10072-021-05662-9 34719776
19. •• García-Grimshaw M, Ceballos-Liceaga SE, Hernández-Vanegas LE, Núñez I, Hernández-Valdivia N, Carrillo-García DA, et al. Neurologic adverse events among 704,003 first-dose recipients of the BNT162b2 mRNA COVID-19 vaccine in Mexico: a nationwide descriptive study. Clin Immunol. 2021;229:108786. 10.1016/j.clim.2021.108786. The article reports on the incidence of neurological adverse events post-vaccination with the first dose of Pfizer-BioNTech vaccine in a Mexican cohort.
20. Kim SH Wi YM Yun SY Ryu JS Shin JM Lee EH Adverse events in healthcare workers after the first dose of ChAdOx1 nCoV-19 or BNT162b2 mRNA COVID-19 vaccination: a single center experience J Korean Med Sci 2021 36 14 e107 10.3346/jkms.2021.36.e107 33847085
21. Lee YW Lim SY Lee JH Lim JS Kim M Kwon S Adverse reactions of the second dose of the BNT162b2 mRNA COVID-19 vaccine in healthcare workers in Korea J Korean Med Sci 2021 36 21 e153 10.3346/jkms.2021.36.e153 34060261
22. Göbel CH Heinze A Karstedt S Morscheck M Tashiro L Cirkel A Headache attributed to vaccination against COVID-19 (coronavirus SARS-CoV-2) with the ChAdOx1 nCoV-19 (AZD1222) vaccine: a multicenter observational cohort study Pain Ther 2021 10 2 1309 1330 10.1007/s40122-021-00296-3 34313952
23. García-Azorín D Do TP Gantenbein AR Hansen JM Souza MNP Obermann M Delayed headache after COVID-19 vaccination: a red flag for vaccine induced cerebral venous thrombosis J Headache Pain 2021 22 1 108 10.1186/s10194-021-01324-5 34535076
24. World Health Organization. COVID-19 vaccines: safety surveillance manual. World Health Organization. 2020. https://apps.who.int/iris/handle/10665/338400. Accessed Oct 2022.
25. Iba T Levy JH Warkentin TE Recognizing vaccine-induced immune thrombotic thrombocytopenia Crit Care Med 2022 50 1 e80 e86 10.1097/CCM.0000000000005211 34259661
26. Greinacher A Thiele T Warkentin TE Weisser K Kyrle PA Eichinger S Thrombotic thrombocytopenia after ChAdOx1 nCov-19 vaccination N Engl J Med 2021 10.1056/NEJMoa2104840 34731557
27. Scully M Singh D Lown R Poles A Solomon T Levi M Pathologic antibodies to platelet factor 4 after ChAdOx1 nCoV-19 vaccination N Engl J Med 2021 384 23 2202 2211 10.1056/NEJMoa2105385 33861525
28. Bersinger S Lagarde K Marlu R Pernod G Payen JF Using nonheparin anticoagulant to treat a near-fatal case with multiple venous thrombotic lesions during ChAdOx1 nCoV-19 vaccination-related vaccine-induced immune thrombotic thrombocytopenia Crit Care Med 2021 49 9 e870 e873 10.1097/CCM.0000000000005105 34049307
29. Graf T, Thiele T, Klingebiel R, Greinacher A, Schäbitz WR, Greeve I. Immediate high-dose intravenous immunoglobulins followed by direct thrombin-inhibitor treatment is crucial for survival in Sars-Covid-19-adenoviral vector vaccine-induced immune thrombotic thrombocytopenia VITT with cerebral sinus nvenous and portal vein thrombosis. J Neurol. 2021:20211–3. 10.1007/s00415-021-10599-2
30. George G, Friedman KD, Curtis BR, Lind SE. Successful treatment of thrombotic thrombocytopenia with cerebral sinus venous thrombosis following Ad26.COV2.S vaccination. Am J Hematol. 2021. 10.1002/ajh.26237.
31. Tiede A Sachs UJ Czwalinna A Werwitzke S Bikker R Krauss JK Prothrombotic immune thrombocytopenia after COVID-19 vaccination Blood 2021 138 4 350 353 10.1182/blood.2021011958 34323939
32. Schulz JB Berlit P Diener HC Gerloff C Greinacher A Klein C COVID-19 Vaccine-Associated Cerebral Venous Thrombosis in Germany Ann Neurol 2021 90 4 627 639 10.1002/ana.26172 34288044
33. Gattringer T, Gressenberger P, Gary T, Wölfler A, Kneihsl M, Raggam RB. Successful management of vaccine-induced immune thrombotic thrombocytopenia-related cerebral sinus venous thrombosis after ChAdOx1 nCov-19 vaccination. Stroke Vasc Neurol.svn-2021–001142. 2021. 10.1136/svn-2021-001142
34. Clark RT Johnson L Billotti J Foulds G Ketels T Heard K Early Outcomes of Bivalirudin Therapy for Thrombotic Thrombocytopenia and Cerebral Venous Sinus Thrombosis After Ad26.COV2.S Vaccination Ann Emerg Med. 2021 78 4 511 514 10.1016/j.annemergmed.2021.04.035 34226070
35. Pottegård A Lund LC Karlstad Ø Dahl J Andersen M Hallas J Arterial events, venous thromboembolism, thrombocytopenia, and bleeding after vaccination with Oxford-AstraZeneca ChAdOx1-S in Denmark and Norway: population based cohort study BMJ 2021 373 n1114 10.1136/bmj.n1114 33952445
36. Krzywicka K Heldner MR Sánchez van Kammen M van Haaps T Hiltunen S Silvis SM Post-SARS-CoV-2-vaccination cerebral venous sinus thrombosis: an analysis of cases notified to the European Medicines Agency Eur J Neurol 2021 28 11 3656 3662 10.1111/ene.15029 34293217
37. McKeigue PM Burgul R Bishop J Robertson C McMenamin J O'Leary M Association of cerebral venous thrombosis with recent COVID-19 vaccination: case-crossover study using ascertainment through neuroimaging in Scotland BMC Infect Dis 2021 21 1 1275 10.1186/s12879-021-06960-5 34930153
38. Perry RJ Tamborska A Singh B Craven B Marigold R Arthur-Farraj P Cerebral venous thrombosis after vaccination against COVID-19 in the UK: a multicentre cohort study Lancet 2021 398 10306 1147 1156 10.1016/S0140-6736(21)01608-1 34370972
39. • Chakravarty A. Between the devil and the deep blue sea? Case 3.14. In: Jaypee CA, editor. Neurology & Internal Medicine – A Case Based Study. New Delhi, London: Brothers Publications; 2021, p. 195–99. Report of a case of cerebral venous sinus thrombosis in a lady following Covisheild COVID 19 vaccination . The patient had an underlying Protein S deficiency and hence had been put on oral anticoagulation for indefinite period.
40. •• Covid-19 rapid guideline: Vaccine-induced immune thrombocytopenia and thrombosis (VITT) [Internet]. National Center for Biotechnology Information. U.S. National Library of Medicine; [cited 2022Nov6]. Available from: https://pubmed.ncbi.nlm.nih.gov/34406720/. Accessed Nov 2022. NICE guidelines for the diagnosis and management of VITT.
41. Bersinger S Lagarde K Marlu R Pernod G Payen JF Using nonheparin anticoagulant to treat a near-fatal case with multiple venous thrombotic lesions during ChAdOx1 nCoV-19 vaccination-related vaccine-induced immune thrombotic thrombocytopenia Crit Care Med 2021 49 9 e870 e873 10.1097/CCM.0000000000005105 34049307
42. Kakovan M Ghorbani Shirkouhi S Zarei M Andalib S Stroke Associated with COVID-19 Vaccines J Stroke Cerebrovasc Dis 2022 31 6 106440 10.1016/j.jstrokecerebrovasdis.2022.106440 35339857
43. Blauenfeldt RA Kristensen SR Ernstsen SL Kristensen CCH Simonsen CZ Hvas AM Thrombocytopenia with acute ischemic stroke and bleeding in a patient newly vaccinated with an adenoviral vector-based COVID-19 vaccine J Thromb Haemost 2021 19 7 1771 1775 10.1111/jth.15347 33877737
44. De Michele M Iacobucci M Chistolini A Nicolini E Pulcinelli F Cerbelli B Malignant cerebral infarction after ChAdOx1 nCov-19 vaccination: a catastrophic variant of vaccine-induced immune thrombotic thrombocytopenia Nat Commun 2021 12 1 4663 10.1038/s41467-021-25010-x 34341358
45. Al-Mayhani T Saber S Stubbs MJ Losseff NA Perry RJ Simister RJ Ischaemic stroke as a presenting feature of ChAdOx1 nCoV-19 vaccine-induced immune thrombotic thrombocytopenia J Neurol Neurosurg Psychiatry 2021 92 11 1247 1248 10.1136/jnnp-2021-326984 34035134
46 Suwanwela NC Kijpaisalratana N Tepmongkol S Prolonged migraine aura resembling ischemic stroke following CoronaVac vaccination: an extended case series J Headache Pain. 2022 23 1 13 10.1186/s10194-022-01385-0 35062869
47. Bjørnstad-Tuveng TH, Rudjord A, Anker P. Fatal cerebral haemorrhage after COVID-19 vaccine. Fatal hjerneblødning etter covid-19-vaksine. Tidsskr Nor Laegeforen. 2021;141. 10.4045/tidsskr.21.0312
48. de Melo Silva ML Lopes DP Large hemorrhagic stroke after ChAdOx1 nCoV-19 vaccination: a case report Acta Neurol Scand 2021 10.1111/ane.13505 34273119
49. Finsterer J Korn M Aphasia seven days after second dose of an mRNA-based SARS-CoV-2 vaccine Brain Hemorrhages 2021 2 165 167 10.1016/j.hest.2021.06.001 34192245
50. Robichaud J Côté C Côté F Systemic capillary leak syndrome after ChAdOx1 nCOV-19 (Oxford-AstraZeneca) vaccination CMAJ 2021 193 34 E1341 E1344 10.1503/cmaj.211212 34362727
51. Rosso M Anziska Y Levine SR Acute transient encephalopathy after Moderna COVID-19 vaccine Case Rep Neurol 2022 14 2 231 236 10.1159/000523769.PMID:35702446;PMCID:PMC9149540 35702446
52. Liu BD Ugolini C Jha P Two cases of post-Moderna COVID-19 vaccine encephalopathy associated with nonconvulsive status epilepticus Cureus 2021 13 7 e16172 10.7759/cureus.16172 34367780
53. Al-Mashdali AF Ata YM Sadik N Post-COVID-19 vaccine acute hyperactive encephalopathy with dramatic response to methylprednisolone: a case report Ann Med Surg (Lond) 2021 69 102803 10.1016/j.amsu.2021.102803 34512961
54. Permezel F Borojevic B Lau S de Boer HH Acute disseminated encephalomyelitis (ADEM) following recent Oxford/AstraZeneca COVID-19 vaccination Forensic Sci Med Pathol 2022 18 1 74 79 10.1007/s12024-021-00440-7 34735684
55. OzgenKenangil G Ari BC Guler C Demir MK Acute disseminated encephalomyelitis-like presentation after an inactivated coronavirus vaccine [published correction appears in Acta Neurol Belg Acta Neurol Belg 2021 121 4 1089 1091 10.1007/s13760-021-01699-x 34018145
56. Cao L Ren L Acute disseminated encephalomyelitis after severe acute respiratory syndrome coronavirus 2 vaccination: a case report Acta Neurol Belg 2022 122 3 793 795 10.1007/s13760-021-01608-2 33527327
57. Zuhorn F Graf T Klingebiel R Schäbitz WR Rogalewski A Postvaccinal Encephalitis after ChAdOx1 nCov-19 Ann Neurol 2021 90 3 506 511 10.1002/ana.26182 34324214
58. Kobayashi Y Karasawa S Ohashi N Yamamoto K A case of encephalitis following COVID-19 vaccine J Infect Chemother 2022 28 7 975 977 10.1016/j.jiac.2022.02.009 35190257
59 Bensaidane MR Picher-Martel V Émond F De Serres G Dupré N Beauchemin P Case report: Acute necrotizing encephalopathy following COVID-19 vaccine Front Neurol 2022 13 872734 10.3389/fneur.2022.872734 35572945
60. Malhotra SH, Gupta P, Prabhu V, Garg RK, Dandu H, Agarwal V. COVID-19 vaccination-associated myelitis. QJM. 2021;hcab069. 10.1093/qjmed/hcab069
61. Tan WY Yusof Khan AHK MohdYaakob MN Abdul Rashid AM Loh WC Baharin J Longitudinal extensive transverse myelitis following ChAdOx1 nCOV-19 vaccine: a case report BMC Neurol. 2021 21 1 395 10.1186/s12883-021-02427-x 34641797
62. Pagenkopf C Südmeyer M A case of longitudinally extensive transverse myelitis following vaccination against Covid-19 J Neuroimmunol 2021 358 577606 10.1016/j.jneuroim.2021.577606 34182207
63 Chagla Z In adults, the Oxford/AstraZeneca vaccine had 70% efficacy against COVID-19 >14 d after the 2nd dose Ann Intern Med 2021 174 3 JC29 10.7326/ACPJ202103160-029 33646835
64. Knoll MD Wonodi C Oxford-AstraZeneca COVID-19 vaccine efficacy Lancet 2021 397 10269 72 74 10.1016/S0140-6736(20)32623-4 33306990
65. Román GC Gracia F Torres A Palacios A Gracia K Harris D Acute transverse myelitis (ATM): clinical review of 43 patients with COVID-19-associated ATM and 3 post-vaccination ATM serious adverse events with the ChAdOx1 nCoV-19 vaccine (AZD1222) Front Immunol 2021 26 12 653786 10.3389/fimmu.2021.653786.PMID:33981305;PMCID:PMC8107358
66. Quintanilla-Bordás C Gascón-Gimenez F Alcalá C Payá M Mallada J Silla R Case report: Exacerbation of relapses following mRNA COVID-19 vaccination in multiple sclerosis: a case series Front Neurol 2022 13 897275 10.3389/fneur.2022.897275 35572939
67. Epstein S Xia Z Lee AJ Dahl M Edwards K Levit E Vaccination against SARS-CoV-2 in neuroinflammatory disease: early safety/tolerability data Mult Scler Relat Disord 2022 57 103433 10.1016/j.msard.2021.103433 34923427
68. Dinoto A Gastaldi M Iorio R Marini S Damato V Farina A Safety profile of SARS-CoV-2 vaccination in patients with antibody-mediated CNS disorders Mult Scler Relat Disord 2022 63 103827 10.1016/j.msard.2022.103827 35508101
69. Alfishawy M Bitar Z Elgazzar A Elzoueiry M Neuroleptic malignant syndrome following COVID-19 vaccination Am J Emerg Med. 2021 49 408 409 10.1016/j.ajem.2021.02.011 33642127
70. Nagamine T Neuroleptic malignant syndrome associated with COVID-19 vaccination CJEM 2022 24 3 349 350 10.1007/s43678-021-00254-0 34919205
71. Ghosh R Dubey S Roy D Mandal A Naga D Benito-León J Focal onset non-motor seizure following COVID-19 vaccination: a mere coincidence? Diabetes Metab Syndr 2021 15 3 1023 1024 10.1016/j.dsx.2021.05.003 34000712
72. Makhlouf AT Van Alphen MU Manzano GS Freudenreich O A seizure after COVID-19 vaccination in a patient on clozapine J Clin Psychopharmacol 2021 41 6 689 690 10.1097/JCP.0000000000001488 34735102
73 Werner J Brandi G Jelcic I Galovic M New-onset refractory status epilepticus due to autoimmune encephalitis after vaccination against SARS-CoV-2: first case report Front Neurol. 2022 13 946644 10.3389/fneur.2022.946644 36051224
74. Martinez-Fernandez I Sanchez-Larsen A Gonzalez-Villar E Martínez-Martín Á von Quednow E Del Valle-Pérez JA Observational retrospective analysis of vaccination against SARS-CoV-2 and seizures: VACCI-COVID registry Epilepsy Behav 2022 134 108808 10.1016/j.yebeh.2022.10880 35809503
75. Özdemir HN Dere B Gökçay F Gökçay A Are COVID-19 vaccines safe for people with epilepsy? A cross-sectional study Neurol Sci 2022 43 6 3489 3496 10.1007/s10072-022-05956-6 35224704
76. von Wrede R Pukropski J Moskau-Hartmann S Surges R Baumgartner T COVID-19 vaccination in patients with epilepsy: First experiences in a German tertiary epilepsy center Epilepsy Behav 2021 122 108160 10.1016/j.yebeh.2021.108160 34166950
77. Massoud F Ahmad SF Hassan AM Alexander KJ Al-Hashel J Arabi M Safety and tolerability of the novel 2019 coronavirus disease (COVID-19) vaccines among people with epilepsy (PwE): a cross-sectional study Seizure 2021 92 2 9 10.1016/j.seizure.2021.08.001 34391030
78 Li N Chu C Lin W A survey of hesitancy and response to the COVID-19 vaccine among patients with epilepsy in northeast China Front Neurol 2021 2 778618 10.3389/fneur.2021.778618
79. Clayton LM Balestrini S Cross JH Wilson G Eldred C Evans H The impact of SARS-COV-2 vaccination in Dravet syndrome: a UK survey Epilepsy Behav 2021 124 108258 10.1016/j.yebeh.2021.108258 34536735
80. Lu Q Wang YY Wang QH Tang LN Yang XY Dun S Safety of inactivated COVID-19 vaccine in tuberous sclerosis complex patients with epilepsy treated with rapamycin Seizure 2022 99 71 74 10.1016/j.seizure.2022.05.010 35605444
81. Shemer A Pras E Hecht I Peripheral facial nerve palsy following BNT162b2 (COVID-19) vaccination Isr Med Assoc J 2021 23 3 143 144 33734623
82. Burrows A Bartholomew T Rudd J Walker D Sequential contralateral facial nerve palsies following COVID-19 vaccination first and second doses BMJ Case Rep 2021 14 7 e243829 10.1136/bcr-2021-243829 34281950
83. Obermann M Krasniqi M Ewers N Fayad J Haeberle U Bell’s palsy following COVID-19 vaccination with high CSF antibody response Neurol Sci 2021 42 11 4397 4399 10.1007/s10072-021-05496-5 34322761
84. Shemer A Pras E Einan-Lifshitz A Dubinsky-Pertzov B Hecht I Association of COVID-19 vaccination and facial nerve palsy: a case-control study JAMA Otolaryngol Head Neck Surg 2021 147 8 739 743 10.1001/jamaoto.2021.1259 34165512
85. Ozonoff A Nanishi E Levy O Bell’s palsy and SARSCoV-2 vaccines Lancet Infect Dis 2021 21 4 450 452 10.1016/S1473-3099(21)00076-1 33639103
86. Lechien JR, Diallo AO, Dachy B, Le Bon SD, Maniaci A, Vaira LA, Saussez S (2021) COVID-19: Post-vaccine smell and taste disorders: report of 6 cases. Ear Nose Throat J Sep 1:1455613211033125. 10.1177/01455613211033125. Epub ahead of print. PMID: 34467793.
87. Keir G Maria NI Kirsch CFE Unique imaging findings of neurologic phantosmia following Pfizer-BioNtech COVID-19 vaccination: a case report Top Magn Reson Imaging 2021 30 3 133 137 10.1097/RMR.0000000000000287 34096896
88. Reyes-Capo DP Stevens SM Cavuoto KM Acute abducens nerve palsy following COVID-19 vaccination J AAPOS 2021 25 5 302 303 10.1016/j.jaapos.2021.05.003 34044114
89. Sen M Honavar SG After the storm: ophthalmic manifestations of COVID-19 vaccines Indian J Ophthalmol 2021 69 12 3398 3420 10.4103/ijo.IJO_2824_21 34826968
90. Chuang TY Burda K Teklemariam E Athar K Tolosa-Hunt syndrome presenting after COVID-19 vaccination Cureus 2021 13 7 e16791 10.7759/cureus.16791 34513398
91. Bialasiewicz AA Farah-Diab MS Mebarki HT Central retinal vein occlusion occurring immediately after 2nd dose of mRNA SARS-CoV-2 vaccine Int Ophthalmol 2021 41 3889 3892 10.1007/s10792-021-01971-2 34426861
92. Alvarez LM NingNeo Y Davagnanam I Ashenhurst M Acheson J Abdel-Hay A Post vaccination optic neuritis: Observations from the SARS-CoV-2 pandemic SSRN Electron J 2021 10.2139/ssrn.3889990
93. Formeister EJ Wu MJ Chari DA Meek R 3rd Rauch SD Remenschneider AK Assessment of sudden sensorineural hearing loss after COVID-19 vaccination JAMA Otolaryngol Head Neck Surg 2022 148 4 307 315 10.1001/jamaoto.2021.4414 35201274
94 Waheed S Bayas A Hindi F Rizvi Z Espinosa PS Neurological Complications of COVID-19: Guillain-Barre syndrome following Pfizer COVID-19 vaccine Cureus 2021 13 2 e13426 10.7759/cureus.13426 33758714
95. • Márquez Loza AM, Holroyd KB, Johnson SA, Pilgrim DM, Amato AA. Guillain-Barré syndrome in the placebo and active arms of a COVID-19 vaccine clinical trial: temporal associations do not imply causality. Neurology. 2021. 10.1212/WNL.0000000000011881. Study of risk of GBS post-COVID vaccination demonstrating the absence of sufficient evidence to implicate vaccines as the cause.
96. Kohli S, Varshney M, Mangla S, Jaiswal B, Chhabra PH. Guillain-Barré syndrome after COVID-19 vaccine: should we assume a causal link? International Journal of Medical and Pharmaceutical Case Reports. 2021;20–4. 10.9734/ijmpcr/2021/v14i130124.
97. Bonifacio GB Patel D Cook S Purcaru E Couzins M Domjan J Bilateral facial weakness with paraesthesia variant of Guillain-Barré syndrome following Vaxzevria COVID-19 vaccine J Neurol Neurosurg Psychiatry 2021 93 3 341 342 10.1136/jnnp-2021-327027 34261746
98. McKean N Chircop C Guillain-Barré syndrome after COVID-19 vaccination BMJ Case Rep 2021 14 7 e244125 10.1136/bcr-2021-244125 34330729
99. Waheed W Carey ME Tandan SR Tandan R Post COVID-19 vaccine small fiber neuropathy Muscle Nerve 2021 64 1 E1 E2 10.1002/mus.27251 33851437
100. Mahajan S Zhang F Mahajan A Zimnowodzki S Parsonage Turner syndrome after COVID-19 vaccination Muscle Nerve 2021 64 1 E3 E4 10.1002/mus.27255 33890680
101. McMahon DE Amerson E Rosenbach M Lipoff JB Moustafa D Tyagi A Cutaneous reactions reported after Moderna and Pfizer COVID-19 vaccination: A registry-based study of 414 cases J Am Acad Dermatol 2021 85 1 46 55 10.1016/j.jaad.2021.03.092 33838206
102. • Nassar M, Chung H, Dhayaparan Y, Nyein A, Acevedo BJ, Chicos C, Zheng D, Barras M, Mohamed M, Alfishawy M, Nso N, Rizzo V, Kimball E. COVID-19 vaccine induced rhabdomyolysis: Case report with literature review. Diabetes Metab Syndr. 2021;15(4):102170. 10.1016/j.dsx.2021.06.007. Case report of possible vaccine-induced rhabdomyolysis with good outcome following prompt treatment.
103. Kimura M Niwa JI Doyu M Recurring weakness in rhabdomyolysis following Pfizer-BioNTech coronavirus disease 2019 mRNA vaccination Vaccines (Basel) 2022 10 6 935 10.3390/vaccines10060935 35746543
104. •• Sansone G, Bonifati DM. Vaccines and myasthenia gravis: a comprehensive review and retrospective study of SARS-CoV-2 vaccination in a large cohort of myasthenic patients. J Neurol. 2022;269:3965–81. 10.1007/s00415-022-11140-9. This review article examines the available data regarding safety of COVID vaccines in myasthenia patients.
105. Cosentino C Torres L Vélez M SARS-CoV-2 vaccines and motor symptoms in Parkinson’s disease Mov Disord 2022 37 1 233 10.1002/mds.28851 34775612
106. Erro R Buonomo AR Barone P Pellecchia MT Severe dyskinesia after administration of SARS-CoV2 mRNA vaccine in Parkinson’s disease Mov Disord 2021 36 10 2219 10.1002/mds.28772 34368991
107. Imbalzano G Ledda C Artusi CA Romagnolo A Montanaro E Rizzone MG SARS-CoV-2 vaccination, Parkinson’s disease, and other movement disorders: case series and short literature review Neurol Sci 2022 43 5165 5168 10.1007/s10072-022-06182-w 35666352
108. Naharci MI Tasci I Delirium in a patient with Alzheimer’s dementia following COVID-19 vaccination Psychogeriatrics 2021 21 5 846 847 10.1111/psyg.12747 34245485
109. Ng JH Chaudhuri KR Tan EK Functional neurological disorders and COVID-19 vaccination Ann Neurol 2021 90 2 328 10.1002/ana.26160 34231246
110. Takahashi O, Sakakibara R, Sawai S, Ogata T. Functional neurological disorders after COVID-19 vaccination: Case series and literature review. Psychiatry Clin Neurosci. 2022;25. 10.1111/pcn.13453.
111. Kim DD Kung CS Perez DL Helping the public understand adverse events associated with covid-19 vaccinations JAMA Neurol 2021 78 7 789 10.1001/jamaneurol.2021.1042 33835153
112. •• Butler M, Coebergh J, Safavi F, Carson A, Hallett M, Michael B, et al. 2021 Functional neurological disorder after SARS-CoV-2 vaccines: two case reports and discussion of potential public health implications. J Neuropsychiatry Clin Neurosci. 10.1176/appi.neuropsych.21050116. Reports of patients developing functional neurological disorder following COVID-vaccination and implications.
113. Ercoli T Lutzoni L Orofino G Muroni A Defazio G Functional neurological disorder after COVID-19 vaccination Neurol Sci 2021 42 1 2 10.1007/s10072-021-05504-8
114. Goss AL Samudralwar RD Das RR Nath A ANA investigates: neurological complications of COVID-19 vaccines Ann Neurol 2021 89 5 856 857 10.1002/ana.26065 33710649
| 36445631 | PMC9707152 | NO-CC CODE | 2022-12-01 23:20:21 | no | Curr Neurol Neurosci Rep. 2022 Nov 29;:1-14 | utf-8 | Curr Neurol Neurosci Rep | 2,022 | 10.1007/s11910-022-01247-x | oa_other |
==== Front
Mol Cell Biochem
Mol Cell Biochem
Molecular and Cellular Biochemistry
0300-8177
1573-4919
Springer US New York
36441353
4615
10.1007/s11010-022-04615-w
Article
Advances in ligase-based nucleic acid amplification technology for detecting gene mutations: a review
Li Ying
Wang Xiangjun
Wang Minghui
Liu Moyi
Wang Helin
Xia Wei
Liu Limei [email protected]
grid.411601.3 0000 0004 1798 0308 Department of Experiment Center, School of Medical Technology, Beihua University, No. 3999 Binjiang East Road, Fengman District, Jilin City, Jilin Province China
28 11 2022
111
7 7 2022
14 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Gene mutation has been a concern for researchers because it results in genetic variations with base changes in molecular structure. Researchers continue to explore methods to detect gene mutations, which may help in disease diagnosis, medication guidance, and so on. Currently, the detection methods, such as whole-genome sequencing and polymerase chain reaction, have some limitations in terms of cost and sensitivity. Ligase (an enzyme) can recognize base mismatch as a commonly used tool in genetic engineering. Therefore, the ligase-related nucleic acid amplification technology for detecting gene mutations has become a research hotspot. In this study, the main techniques explored for detecting gene mutations included the ligase detection reaction, ligase chain reaction, rolling circle amplification reaction, enzyme-assisted polymerase chain reaction, and loop-mediated isothermal amplification reaction. This review aimed to analyze the aforementioned techniques and mainly present their advantages and disadvantages, sensitivity, specificity, cost, detection time, applications, and so on. The findings may help develop sufficient grounds for further studies on detecting gene mutations.
Keywords
Amplification technique
Genetic mutations
Ligase
Nucleic acid
National Science Funding of China;Science and Technology Development Program of Jilin Province;Changchun Chenyu Biomedical Technology Cooperation Project;No.81201354;No.20200403118SF;No.2021-10 No.81201354;No.20200403118SF;No.2021-10 No.81201354;No.20200403118SF;No.2021-10 No.81201354;No.20200403118SF;No.2021-10 No.81201354;No.20200403118SF;No.2021-10 No.81201354;No.20200403118SF;No.2021-10 Li Ying Wang Xiangjun Wang Minghui Liu Moyi Wang Helin Xia Wei Beihua University Postgraduate Innovation Program Project2022023 2022023 2022023 2022023 2022023 2022023 Li Ying Wang Xiangjun Wang Minghui Liu Moyi Wang Helin Xia Wei
==== Body
pmcIntroduction
Gene mutation is a common phenomenon in biological growth and evolution. Base mismatch in DNA molecules easily causes gene mutations, especially point mutations, which are often an important cause of hereditary and genetic predisposition diseases [1]. In recent years, the detection and analysis of gene mutations have become a topic of increasing interest, such as pathogenic gene screening and prenatal diagnosis, which can provide the basis for early diagnosis and precise treatment of diseases [2]. At present, many methods are in place for detecting gene mutations, such as whole-genome sequencing, but they still have certain limitations in terms of experimental cost, sensitivity, and accuracy. Ligase is an enzyme that can catalyze the formation of chemical bonds between the head and tail of two molecules or a molecule and can be used as a tool in DNA replication and repair. Ligase recognition ability can effectively identify mismatched base pairs [3–5]. Therefore, the application of ligase for detecting gene mutations is of increasing interest and a research hotspot. This review mainly focused on the latest research progress in ligase and several related nucleic acid amplification technologies and also on their advantages, disadvantages, and applications.
Types and modes of action of ligases
Two types of ligases have been found to catalyze nucleic acid ligations: DNA ligase and RNA ligase. DNA ligase forms a phosphodiester bond by connecting the adjacent 5’-terminal phosphate group with the 3’-terminal hydroxyl group on the catalytic chain so that the single-stranded nick can be repaired, which is known as “the needle and thread of genes” [6]. Ligase can be divided into room-temperature and thermophilic DNA ligases based on the thermal stability of DNA ligase. The room-temperature DNA ligases include SplintR Ligase, T3 DNA ligase, and T4 DNA ligase. They are suitable for the reaction below 37 °C, and the reaction is fast but has low specificity. The thermophilic DNA ligases include Thermus aquaticus ligase (commonly known as HiFi, and so on, with high fidelity), 9°N ligase, Ampligase, Thermotoga maritima ligase, Pfu ligase, and so forth. The reaction temperature of the thermophilic DNA ligases is higher, and they are not inactivated even at 90 °C [7, 8]. RNA ligase can catalyze the intermolecular or intramolecular connection of DNA or RNA and can also be used for the specific modification of transfer RNA. The optimum reaction temperature is 37 °C, and T4 RNA ligase is common. T4 RNA ligase 2 is commonly used and specifically catalyzes linear intermolecular ligation and intramolecular circular ligation with high efficiency, low background, and good stability [9].
Ligase-related nucleic acid amplification techniques
Ligase chain reaction
Since its invention by Backman in 1995, ligase chain reaction (LCR) has been one of the techniques for detecting point mutations in target gene sequences [10]. LCR is an exponential amplification technology based on DNA ligase. The principle of LCR technology is shown by a schematic in Fig. 1. The LCR needs to be carried out in a thermal cycler. Two pairs of complementary oligonucleotide probes are designed and hybridized with the perfectly matched target sequence. The 5’-terminal phosphate group of one probe is combined with the 5’-terminal phosphate group of another probe by the action of DNA ligase. The 3’-terminal hydroxyl group of the other probe is connected, and the new fragment obtained is denatured again and used as a template to continue the reaction. The other two probes are hybridized into the second template strand in a similar manner. After cycling, a large number of ligation products are formed to achieve the purpose of exponential amplification. Therefore, this method is more sensitive than ligase detection reaction (LDR) [11]. The adjacent probes cannot be connected if a base mutation occurs at the ligation site, and no further amplifications occur. Therefore, another advantage of LCR is that it has a stronger ability to identify base mutations while ensuring the same amplification efficiency as polymerase chain reaction (PCR). Hence, LCR is currently the best method for detecting point mutations in known sequences. However, it requires the use of a thermal cycler, and the biggest drawback is the possibility of target-independent connections that increase the background signal.Fig. 1 Schematic diagram depicting the principle of LCR technology
LCR is widely used in detecting gene mutations. Malik used this technology to screen the hepatitis B virus G1896A mutation in patients with chronic liver disease negative for hepatitis B surface antigen in northern India and found that the G1896A mutation caused one-third of the pronuclear stop-codon mutations. Experiments showed that this method had lower cost and higher detection efficiency compared with sequencing. However, the results obtained were reliable and consistent with the sequencing results [12, 13]. The product of LCR can be detected by electrophoresis, autoradiography, and so on, but the process is complicated. With the gradual deepening of nucleic acid research, various combined application methods are being used for detecting gene mutations based on LCR, such as combined fluorescence, electrochemistry, single quantum dots, cationic polymers, and nanoparticles.
Combining LCR with fluorescent biosensors
Feng Rui et al. developed a new method combining LCR and flow cytometry microspheres to detect the tumor-associated Janus kinase 2 (JAK2) gene V617F mutation site. They labeled the LCR products with biotin and fluorophores, fixed them on the microspheres, and quantitatively detected them via flow imaging. The results showed that the fluorescence gradually increased with the increase in the mutant DNA concentration [14]. This method could detect mutant DNA in a linear range of 10−15 to 10−11 mol/L, and the detection limit was as low as 10−15 mol/L. The process was simple and involved one-tube typing. The quantitative imaging results were intuitive and clear, but the sensitivity needed improvement.
Combining LCR with electrochemical biosensor
Hu et al. designed an LCR method combined with a ferrocene (Fc)-labeled electrochemical sensor for the ultrasensitive detection of the mutation site of the gene epidermal growth factor receptor (EGFR) T790M. They hybridized the capture probe on the nanotube-modified electrode using the Fc-labeled LCR product to measure the current, which could be identified by comparing the difference in electrical signals. The method had good specificity and high precision. The linear dynamic range of the detected mutant target sequence concentration was 10−18 to 10−11 mol/L, and the detection limit was 10−19 mol/L. Moreover, the product did not need separation and purification [15]. However, this method also had drawbacks. For example, the modification of carboxyl multi-wall carbon nanotube was cumbersome. Chen et al. transformed the thiol- or biotin-labeled short double-stranded DNA into a recombined complete double-stranded DNA via LCR exponential amplification in the presence of the target sequence, that is, “probe lengthening” and then immobilized it on the surface of the gold electrode to measure the electrical signal. The results showed distinct differences in DNA electrical signals between perfectly paired and single-base mismatches, allowing easy identification of genetic mutations. The linear dynamic range of the mutant DNA concentration measured using this method was 10−16 to 10−11 mol/L, and the sensitivity was 1000 times higher than that of gel electrophoresis. It also had the characteristics of strong specificity and high resolution and could be used for genomic DNA mutations in clinical serum samples [16]. Liu et al. tested cytochrome P450 family 2 subfamily C member 19 (CYP2C19) gene mutation site in whole blood with high sensitivity. Experiments showed that when the concentration of the mutant target sequence was as low as 10−16 mol/L, the method still had good specificity for the presence of more than 104 times the wild-type gene [17].
Combining LCR with DNA melting temperature
In 2020, Hu et al. developed a marker-free and economical method to analyze the mutation sites of the Kirsten rat sarcoma viral oncogene homolog (KRAS) gene by analyzing the DNA melting temperature based on LCR. Compared with the original target DNA, the double-stranded DNA bases produced by LCR and the melting temperature significantly increased. They accurately screened out site-specific mutations from guanine to adenine, thymine, or cytosine in the KRAS gene, and the linear dynamic range of mutant DNA concentration was 0 to 10−7 mol/L. Experiments showed that this method was highly sensitive, had good selectivity and reliability, did not need purification products and expensive and sophisticated instruments, and avoided cumbersome chemically modified probes [18]. At the same time, it could also amplify the signal. This method might be used for the early diagnosis of tumors.
Combining LCR with argonaute pyrococcus
Argonaute Pyrococcus (Pyrococcus furiosus Argonaute, PfAgo) coupled with LCR was a new method proposed by Wang et al. in 2021 that could rapidly distinguish between the simulated wild-type new coronavirus and the single-base mutation of the spike D614G gene. PfAgo is a thermophilic DNA-guided programmed enzyme that specifically cleaves phosphorylated target DNAs longer than 14 bases [19]. Wang et al. added the LCR product to a solution containing PfAgo protein and a fluorophore. When the target gene was present, PfAgo could cleave the phosphorylated DNA and release the fluorophore for detection. Experiments showed that the detection limit of this method was 10−17 mol/L, the linear dynamic range of the mutant concentration was 10−17 to 10−11 mol/L, the time used was less than that for PCR, and the sensitivity and specificity were better [20]. This method provided new ideas for detecting novel coronaviruses and their more infectious mutated genes. However, this method also had some shortcomings. For example, the LCR and PfAgo cleavage assays could not be carried out in the same tube, which needs to be improved in future.
Ligase detection reaction
LDR is a method that uses DNA ligase to repeat thermal cycling for linear amplification. The principle of the LDR technique is shown in Fig. 2. This reaction requires the design of a pair of specific oligonucleotide probes upstream and downstream of the mutation site. When one of the probes is fully complementary to the target gene, the 3’ end of the probe forms a gapped duplex with the 5’ end of the adjacent probe, the gap is connected by ligase, and the amplified product is detected using capillary electrophoresis after multiple thermal cycles [21]. LDR overcomes the shortcomings of traditional LCR. It only uses a pair of oligonucleotides to reduce the background signal of false ligation. The biggest advantage is that it can quickly screen mutant genes at multiple sites and is simpler than quantitative PCR, but the signal amplification is lesser than that of LCR.Fig. 2 Schematic diagram depicting LDR technology principles
Many researchers combined PCR with LDR for gene identification. For example, Ruiz et al. applied this method to establish a specific, rapid, and practical liquid biopsy strategy [22]. Zhang Xinya et al. used this method to detect the D614G mutation in the fragmented coronavirus S gene down to a 40-nt fragment length, which was a shorter template length than that required for the probe method of fluorescent quantitative PCR and had an advantage in detecting fragmented templates [23]. Zhang et al. used this method to integrally detect the mutation site of the EGFR T790M gene. The results showed that the linear dynamic range of the concentration of the mutant target sequence was 10−16 to 10−12 mol/L, and the detection limit could reach 10−17 mol/L [24]. Li et al. used an improved multiplex LDR (iMLDR), that is, multiplex PCR products as templates for ligation, and found that the lncRNA nuclear enriched abundant transcript1 (NEAT1) rs3825071 locus variant was significantly associated with sputum smear positivity in patients with pulmonary tuberculosis [25]. The applications of LDR in recent years are presented in Table 1.Table 1 Current applications of LDR
Format Target Disease Reference
LDR–fluorescence–magnetic nanospheres SLC30A8 gene Type 2 diabetes [26]
LDR–PCR BRAF gene Colon adenocarcinoma [22]
CYP2C19 gene Screening before clopidogrel treatment [27]
T790M gene Non-small-cell lung cancer [24]
CASP7 gene Noise-induced hearing loss [28]
PD-L1 3’-UTR Epithelial ovarian cancer [29]
ALDH2 gene Coronary artery disease [30]
MDM4 gene Esophageal squamous cell carcinoma [31]
Calgranulin gene Ischemic stroke [32]
ATG5 gene Thyroid-related eye disease [33]
IL-27 gene Aplastic anemia [34]
ABCA1 gene Hypertension [35]
CTLA-4 gene Epithelial ovarian cancer [36]
MIF-AS1 gene Stomach cancer [37]
Novel coronavirus S gene D614G Novel coronavirus pneumonia [23]
iMLDR ncRNA NEAT1gene Tuberculosis [25]
AHR signaling pathway gene Rheumatoid arthritis [38]
Rolling circle amplification technology
Rolling circle amplification (RCA) is an isothermal amplification reaction in which ligase participates in specific ligation. The key to this reaction is the construction of a single-stranded DNA circle. The principle of probe looping involving DNA ligase is schematically shown in Fig. 3. When the target gene and the probe are completely matched, the 3’-terminal hydroxyl group and the 5’-terminal phosphate group of the probe are connected to form a circular closed loop under the action of ligase and then the subsequent amplification reaction is carried out. The reaction can achieve the effect of amplifying the signal hundreds of times. For a situation where the bases cannot be matched, the probe cannot be closed into a loop and signal amplification cannot be achieved [39, 40]. The RCA reaction that ligase participates in improves the specificity and sensitivity of the detection, requires less experimental temperature conditions, and does not require a thermal cycler.Fig. 3 Schematic diagram of the principle of probe looping involving DNA ligase
Qian et al. reported a method of RCA combined with fluorescence detection to identify mutations simply and efficiently. A large number of pyrophosphate molecules could be generated using this method during the signal amplification process, and the fluorescence intensity could be determined to detect gene mutations. This method could measure the concentration of mutant DNA up to 10−13 mol/L and specifically distinguish the difference in the base site of the target gene. This method was less expensive than sequencing and provided a new option for detecting gene mutations [41]. Similarly, Kim et al. also described a method for detecting low-abundance EGFR exon 19-del mutant genomic DNA: generation of long segments containing G-quadruplex structures in RCA by thioflavin T detection based on the fluorescence intensity of single-stranded DNA. This method detected as low as 0.01% of mutant genes in pooled normal plasma [42]. Chung et al. used RCA combined with surface-enhanced Raman scattering to detect multiple point mutations in the KRAS gene. When the linear probe hybridizes with a matched target gene, the probe forms a specific loop structure. Experiments showed that the scattering intensity increased proportionally with the increase in the target mutant gene concentration in the range of 10−11 to 10−8 mol/L. This method turns on Raman scattering at the same time as the ligation reaction, shortens the detection time, and the clever design of the probe avoids false-positive results caused by self-hybridization. This method can be considered as an alternative method to PCR for detecting gene mutations under restricted conditions [43].
Ligase-assisted PCR
PCR generally uses DNA polymerase to achieve amplification through continuous cycles of denaturation, annealing, and extension. However, this method has disadvantages, such as being prone to false results and a long detection time. Considering these shortcomings, it was proposed to use ligase-assisted PCR with better recognition ability, as shown in Fig. 4. When the target gene is completely matched and complementary to the two oligonucleotide probes, under the action of DNA ligase, probes 1 and 2 connect the gap to form a long DNA chain and then carry out subsequent PCR amplification. However, for a base mutation in the target gene, the two probes cannot be connected, indicating the detection of the base mutation of the target gene. Li Bo developed a simple method, using T4 DNA ligase-assisted PCR, to detect mutations in various pathogenic genes such as tumor protein 53 (TP53) and serine/threonine kinase 11 (STK11) related to lung squamous cell carcinoma. This method could detect 0.1% of mutant genes and was expected to be a powerful tool for detecting gene mutations [44]. However, this method had disadvantages in terms of low ligation efficiency and time consumption. Although these defects were corrected by optimizing the experimental conditions, the activity of T4 DNA ligase was weakened, thereby inhibiting the ligation reaction. Therefore, this method needs further improvement.Fig. 4 Schematic diagram depicting the ligation reaction principle of the DNA ligase-assisted PCR
Ligase-assisted loop-mediated isothermal amplification technology
Loop-mediated isothermal amplification (LAMP) can amplify a small amount of template multiplicity in a short period under constant-temperature conditions and its sensitivity is higher than that of traditional PCR and the determination is faster [45]. However, the ability of LAMP to identify base mismatches is not adequate. Therefore, it has been proposed to use high-specificity recognition ligase to assist in detecting gene mutations, thus improving the accuracy and lowering the cost of sequencing. The schematic diagram depicting the principle of DNA ligase-assisted LAMP ligation reaction is shown in Fig. 5. In the ligation reaction, a pair of stem-loop probes 1 and 2 are designed: probe 1 is completely complementary to the gene to be tested (perfect match/mismatch) and probe 2 is only complementary to the perfectly matched target gene. In the catalysis of ligase, probes 1 and 2 can be connected to form a dumbbell-shaped product, which served as an initial template for the LAMP reaction. When the target sequence is mutated, the bases cannot be matched and the ligation will not occur, thus distinguishing the mutation of the gene sites. For example, Wang et al. designed two stem-loop probes complementary to mutant DNA to carry out ligase reaction to initiate LAMP and quickly identified the mutation site via a fluorescence curve. The results showed that the fluorescence signal increased with an increase in the concentration of mutant DNA, and the mutant DNA with a concentration as low as 10−17 mol/L could be accurately detected. The method was simple and universal [46]. Zhang et al. also used this method to screen for breast cancer nucleic acid markers with high sensitivity [47]. Sun et al. reported a novel technique based on artificial mismatch ligation probes combined with ligase-assisted LAMP amplification to detect mutation sites in the exons of the p53 gene. They designed the position of mismatch introduction at the third position of the 3’ end of the probe ligation range. They generated three overhanging bases, effectively avoiding the false-positive signal caused by the mismatch target as a template ligation. Their proposed method could clearly measure the target DNA concentration as low as 10−17 mol/L with ultra-high specificity, and the detection results were completely consistent with the sequencing results. Therefore, this method might be used in clinical and medical diagnosis [48]. Recently, Choi et al. developed a new double-site ligation-assisted LAMP (dLig-LAMP) method for bedside detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA. They designed three DNA oligonucleotide templates that bound perfectly to the target sequence and used the SplintR ligase double-site ligation to form complementary DNA, which was then further amplified. The ligation is impossible for a mismatch in the sequence to be tested or if one of the templates is not bound. This method is more specific in its selection compared with the reverse transcription LAMP, allowing selection at the ligation binding step without being performed in reverse transcription, thus overcoming problems, such as primer mismatches and false results. Studies showed that this method could achieve a detection limit of 10−15 mol/L in 1 h and also had the advantage in terms of high sensitivity and clear clinical selectivity [49]. This method may improve reverse transcription-related selectivity and bedside RNA detection.Fig. 5 Schematic illustration of the principle of DNA ligase-assisted LAMP ligation reaction
Conclusions and future outlook
In the past, most people used sequencing or PCR to detect gene mutations. Certain new developments have been made in ligase-based nucleic acid amplification technology in recent years. The comparative analysis of these methods is presented in Table 2. In the past 5 years, many applications of LDR–PCR and LCR combined with electrochemical sensors have emerged, especially in the screening of genetic diseases and tumors and significant achievements have been made in guiding clinical medication. At present, SARS-CoV-2 mutations still occur frequently. Therefore, sensitive, specific, and field-applicable diversified detection techniques need to be developed. dLig-LAMP and LDR–PCR techniques in this study may provide new alternative ideas for SARS-CoV-2 detection. Unlike LCR, LDR, and PCR requiring a thermal cycler, RCA and LAMP need low experimental temperature. Moreover, they also exhibit improved sensitivity and specificity and hence may be considered for potential use in detecting gene mutations. However, this method can still be improved a lot in terms of cost and detection time.Table 2 Comparison of ligase-related nucleic acid amplification techniques for detecting gene mutations
Method Applications (detection of diseases) Sensitivity (mol/L) Linear dynamic range (mol/L) Specificity (%) Cost Time (min)
LDR Novel coronavirus pneumonia 10−12 10−12 to 10−11 0.01 Usually relatively cheap 160
Non-small-cell lung cancer 10−17 10−16 to 10−12 0.1 Relatively inexpensive with real-time PCR systems 50
LCR Hematological tumor 10−15 10−15 to 10−11 0.1 Cost-effective LCR amplification with multiplex detection 90
Non-small-cell lung cancer 10−19 10−18 to 10−11 0.01 Electrochemical probe using Fc-labeled DNA double strand, no purification required, but more expensive than conventional LCR 118
Abnormal clopidogrel drug metabolism 10−16 10−14 to 10−10 0.01 Miniaturized, easy to operate, and relatively inexpensive instrument 110
Novel coronavirus pneumonia 10−17 10−17 to 10−11 Not reported Relatively inexpensive compared with CRISPR, which does not require the use of costly single-guide RNA 70
Tumors 10−10 0–10−7 5 More convenient and economical than traditional PCR techniques or nanoparticle-based methods 33
RCA Tumors 10−11 10−11 to 10−8 1 Direct connection of nanoparticles for SERS amplification without further signal amplification; cost-effective with inexpensive instrumentation 60
Ligase-assisted PCR Squamous lung cancer Not reported Not reported 0.1 Relatively cheap with real-time PCR systems 110
Ligase-assisted LAMP Breast cancer 10−16 10−14 to 10−11 Not reported Lower cost than ribonucleotide-modified DNA probes 60
Tumors 10−17 10−17 to 10−11 0.01 Modified probe using simple manual mismatch, slightly more expensive than traditional LAMP 66
Novel coronavirus pneumonia 10−15 10−18 to 10−8 Not reported Use of colorimetric and fluorescent detection is not expensive, and costs are relatively low 60
In conclusion, the techniques described in this study reduced the sample requirements and could not only detect DNA in serum samples but also directly detect genomic DNA mutations in whole blood. Compared with PCR, most of these techniques improved in terms of detection specificity. Compared with sequencing, these techniques improved in terms of reduction in the detection cost and were economically competitive. In future, it may be possible to replace sequencing and widely use these techniques in routine and large-scale rapid genetic mutation screening. Therefore, it is believed that the method of detecting gene mutations based on ligase nucleic acid amplification technology will be continuously innovated and may help in disease diagnosis and designing a precise treatment regime.
Acknowledgements
The authors thank the Department of Experiment Center, School of Medical Technology, Beihua University for guidance and advice.
Author contributions
All authors contributed to the work conception and design. Data collection and analysis were performed by YL, XJW, MHW, MYL, and HLW. The manuscript was written by YL and reviewed by LML and WX. All authors read and approved the final manuscript.
Funding
This study was supported by the National Science Funding of China (Grant No. 81201354), the Science and Technology Development Program of Jilin Province (Grant No. 20200403118SF), the Changchun Chenyu Biomedical Technology Cooperation Project (Grant No. 2021-10), and the Beihua University Postgraduate Innovation Program Project (Grant No. 2022023).
Data availability
Not applicable.
Declarations
Conflict of interest
The authors declare no conflict of interest.
Ethical approval
This work does not include any experiments conducted on human or animal participants by any of the authors.
Informed consent
Not applicable.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Wang W Wang X Yang K Fan Y Association of BCL2 polymorphisms and the IL19 single nucleotide polymorphism rs2243188 with systemic lupus erythematosus J Int Med Res 2021 49 5 3000605211019187 10.1177/03000605211019187 34044633
2. Chen Y Xie Y Jiang Y Luo Q Shi L Zeng S Zhuang J Lyu G The genetic etiology diagnosis of fetal growth restriction using single-nucleotide polymorphism-based chromosomal microarray analysis Front Pediatr 2021 9 743639 10.3389/fped.2021.743639 34722424
3. Li T, Zou H, Zhang J, Ding H, Li C, Chen X, Li Y, Feng W, Kageyama K (2022) High-efficiency and high-fidelity ssDNA circularisation via the pairing of five 3'-terminal bases to assist LR-LAMP for the genotyping of single-nucleotide polymorphisms. Analyst 147(18):3993–3999. 10.1039/d2an01042a
4. Potapov V Ong JL Langhorst BW Bilotti K Cahoon D Canton B Knight TF Evans TC Jr Lohman GJS A single-molecule sequencing assay for the comprehensive profiling of T4 DNA ligase fidelity and bias during DNA end-joining Nucleic Acids Res 2018 46 13 e79 10.1093/nar/gky303 29741723
5. Yan X, Zhang J, Jiang Q, Jiao D, Cheng Y (2022) Integration of the ligase chain reaction with the CRISPR-Cas12a system for homogeneous, ultrasensitive, and visual detection of microRNA. Anal Chem 94(9):4119–4125. 10.1021/acs.analchem.2c00294
6. Azuara-Liceaga E Betanzos A Cardona-Felix CS Castañeda-Ortiz EJ Cárdenas H Cárdenas-Guerra RE Pastor-Palacios G García-Rivera G Hernández-Álvarez D Trasviña-Arenas CH Diaz-Quezada C Orozco E Brieba LG The sole DNA ligase in entamoeba histolytica is a high-fidelity DNA ligase involved in DNA damage repair Front Cell Infect Microbiol 2018 8 214 10.3389/fcimb.2018.00214 30050869
7. Lei Y Washington J Hili R Efficiency and fidelity of T3 DNA ligase in ligase-catalysed oligonucleotide polymerisations Org Biomol Chem 2019 17 7 1962 1965 10.1039/c8ob01958d 30357247
8. Bilotti K Potapov V Pryor JM Duckworth AT Keck JL Lohman GJS Mismatch discrimination and sequence bias during end-joining by DNA ligases Nucleic Acids Res 2022 50 8 4647 4658 10.1093/nar/gkac241 35438779
9. Chen H Cheng K Liu X An R Komiyama M Liang X Preferential production of RNA rings by T4 RNA ligase 2 without any splint through rational design of precursor strand Nucleic Acids Res 2020 48 9 e54 10.1093/nar/gkaa181 32232357
10. Shimer GH Jr Backman KC Ligase chain reaction Methods Mol Biol 1995 46 269 278 10.1385/0-89603-297-3:269 7550715
11. Zhong GX Ye CL Wei HX Yang LY Wei QX Liu ZJ Fu LX Lin XH Chen JY Ultrasensitive detection of RNA with single-base resolution by coupling electrochemical sensing strategy with chimeric DNA probe-aided ligase chain reaction Anal Chem 2021 93 2 911 919 10.1021/acs.analchem.0c03563 33284015
12. Malik A Kumar D Khan AA Khan AA Chaudhary AA Husain SA Kar P Hepatitis B virus precore G1896A mutation in chronic liver disease patients with HBeAg negative serology from north India Saudi J Biol Sci 2018 25 7 1257 1262 10.1016/j.sjbs.2016.05.004 30505167
13. Malik A Aldakheel F Rabbani S Alshehri M Chaudhary AA Alkholief M Alshamsan A Based quick detection of hotspot G1896A mutation in patients with different spectrum of hepatitis B J Infect Public Health 2021 14 5 651 654 10.1016/j.jiph.2021.01.013 33857724
14. Feng R (2021) Detection of gene mutations and 5-hydroxymethylcytosine by ligase chain reaction combined with flow technique (Master's thesis, Hebei University). 10.27103/d.cnki.ghebu.2021.001481
15. Hu F Zhang W Meng W Ma Y Zhang X Xu Y Wang P Gu Y Ferrocene-labeled and purification-free electrochemical biosensor based on ligase chain reaction for ultrasensitive single nucleotide polymorphism detection Anal Chim Acta 2020 1109 9 18 10.1016/j.aca.2020.02.062 32252909
16. Chen JY Liu ZJ Wang XW Ye CL Zheng YJ Peng HP Zhong GX Liu AL Chen W Lin XH Ultrasensitive electrochemical biosensor developed by probe lengthening for detection of genomic DNA in human serum Anal Chem 2019 91 7 4552 4558 10.1021/acs.analchem.8b05692 30838849
17. Liu ZJ Yang LY Wei QX Ye CL Xu XW Zhong GX Zheng YJ Chen JY Lin XH Liu AL A novel ligase chain reaction-based electrochemical biosensing strategy for highly sensitive point mutation detection from human whole blood Talanta 2020 216 120966 10.1016/j.talanta.2020.120966 32456905
18. Xue C Yu X Hu S Luo M Shen Z Yuan P Wu ZS Biocomputing label-free security system based on homogenous ligation chain reaction-induced dramatic change in melting temperature for screening single nucleotide polymorphisms Talanta 2020 218 121141 10.1016/j.talanta.2020.121141 32797898
19. He R Wang L Wang F Li W Liu Y Li A Wang Y Mao W Zhai C Ma L Pyrococcus furiosus argonaute-mediated nucleic acid detection Chem Commun (Camb) 2019 55 88 13219 13222 10.1039/c9cc07339f 31589231
20. Wang L He R Lv B Yu X Liu Y Yang J Li W Wang Y Zhang H Yan G Mao W Liu L Wang F Ma L Pyrococcus furiosus argonaute coupled with modified ligase chain reaction for detection of SARS-CoV-2 and HPV Talanta 2021 227 122154 10.1016/j.talanta.2021.122154 33714462
21. Deng H Wang J Kong X Zhang H Wang T Tian W Yi T Wang L Associations of BAFF rs2893321 polymorphisms with myasthenia gravis susceptibility BMC Med Genet 2019 20 1 168 10.1186/s12881-019-0906-8 31666013
22. Ruiz C Huang J Giardina SF Feinberg PB Mirza AH Bacolod MD Soper SA Barany F Single-molecule detection of cancer mutations using a novel PCR-LDR-qPCR assay Hum Mutat 2020 41 5 1051 1068 10.1002/humu.23987 31950578
23. Xinya Z Zhang Jian Lu Chen XY Zhidan L Detection of fragmented new coronavirus S gene D614G mutation by ligase detection reaction J Biol 2022 39 03 107 110 10.3969/j.issn
24. Zhang W Liu K Zhang P Cheng W Zhang Y Li L Yu Z Chen M Chen L Li L Zhang X All-in-one approaches for rapid and highly specific quantifcation of single nucleotide polymorphisms based on ligase detection reaction using molecular beacons as turn-on probes Talanta 2021 224 121717 10.1016/j.talanta.2020.121717 33378999
25. Li HM Wang LJ Tang F Pan HF Zhang TP Association between genetic polymorphisms of lncRNA NEAT1 and pulmonary tuberculosis risk, clinical manifestations in a Chinese population Infect Drug Resist 2022 12 15 2481 2489 10.2147/IDR.S354863
26. Chen Y Zhao Y Li YB Wang YJ Zhang GZ Detection of SNPs of T2DM susceptibility genes by a ligase detection reaction-fluorescent nanosphere technique Anal Biochem 2018 540–541 38 44 10.1016/j.ab.2017.11.003 29128291
27. Zhang J Zhang H Li K Shi M Development of a polymerase chain reaction/ligase detection reaction assay for detection of CYP2C19 polymorphisms Genet Test Mol Biomarkers 2018 22 1 62 73 10.1089/gtmb.2017.0086 29345984
28. Ruan Y Zhang J Mai S Zeng W Huang L Gu C Liu K Ma Y Wang Z Role of CASP7 polymorphisms in noise-induced hearing loss risk in Han Chinese population Sci Rep 2021 11 1 1803 10.1038/s41598-021-81391-5 33469117
29. Sun H Li Y Si W Hua T Chen J Kang S Genetic variation of PD-L1 gene affects its expression and is related to clinical outcome in epithelial ovarian cancer Front Oncol 2022 26 12 763134 10.3389/fonc.2022.763134
30. Ye CY Xin JR Li Z Yin XY Guo SL Li JM Zhao TY Wang L Yang L ALDH2, ADCY3 and BCMO1 polymorphisms and lifestyle-induced traits are jointly associated with CAD risk in Chinese Han people Gene 2022 10 807 145948 10.1016/j.gene.2021.145948
31. Zhou R Li Y Wang N Niu C Huang X Cao S Huo X MDM4 polymorphisms associated with the risk but not the prognosis of esophageal cancer in Cixian high-incidence region from northern China Asia Pac J Clin Oncol 2022 10.1111/ajco.13746 36101936
32. Chen L Chen X Wang Y Li S Huang S Wu Z He J Chen S Deng F Zhu P Zhong W Zhao B Ma G Li Y Polymorphisms of calgranulin genes and ischemic stroke in a Chinese population J Inflamm Res 2022 9 15 3355 3368 10.2147/JIR.S360775
33. Wang W Yu ZY Song RH He ST Shi LF Zhang JA Polymorphisms of ATG5 gene are associated with autoimmune thyroid diseases, especially thyroid eye disease J Immunol Res 2022 26 2022 3881417 10.1155/2022/3881417
34. Yu W Yang W Interlukin-27 rs153109 polymorphism confers the susceptibility and prognosis of aplastic anemia in Chinese population Int J Lab Hematol 2022 44 1 150 156 10.1111/ijlh.13700 34528397
35. Ren Y Tong E Di C Zhang Y Xu L Tan X Yang L Association between ABCA1 gene polymorphisms and the risk of hypertension in the Chinese Han population Front Public Health 2022 20 10 878610 10.3389/fpubh.2022.878610
36. Chen J Kang S Wu J Zhao J Si W Sun H Li Y CTLA-4 polymorphism contributes to the genetic susceptibility of epithelial ovarian cancer J Obstet Gynaecol Res 2022 48 5 1240 1247 10.1111/jog.15186 35150042
37. Ni P Wang G Wang Y Liu K Chen W Xiao J Fan H Ma X Li Z Shen K Xu Z Yang L Correlation of MIF-AS1 polymorphisms with the risk and prognosis of gastric cancer Pathol Res Pract 2022 233 153850 10.1016/j.prp.2022.153850 35367937
38. Zhang TP Li R Li HM Xiang N Tan Z Wang GS Li XM The contribution of genetic variation and aberrant methylation of aryl hydrocarbon receptor signaling pathway genes to rheumatoid arthritis Front Immunol 2022 2 13 823863 10.3389/fimmu.2022.823863
39. Woo CH Jang S Shin G Jung GY Lee JW Sensitive fluorescence detection of SARS-CoV-2 RNA in clinical samples via one-pot isothermal ligation and transcription Nat Biomed Eng 2020 4 12 1168 1179 10.1038/s41551-020-00617-5 32948855
40. Jonchhe S Selvam S Karna D Mandal S Wales-McGrath B Mao H Ensemble sensing using single-molecule DNA copolymers Anal Chem 2020 92 19 13126 13133 10.1021/acs.analchem.0c02196 32829637
41. Ma Q Gao Z A simple and ultrasensitive fluorescence assay for single-nucleotide polymorphism Anal Bioanal Chem 2018 410 13 3093 3100 10.1007/s00216-018-0874-4 29644378
42. Kim DM Zhang S Kim M Kim DE Fluorometric detection of low-abundance EGFR exon 19 deletion mutation using tandem gene amplification J Microbiol Biotechnol 2020 30 5 662 667 10.4014/jmb.2004.04010 32482931
43. Chung CH Kim JH One-step isothermal detection of multiple KRAS mutations by forming SNP specific hairpins on a gold nanoshell Analyst 2018 143 15 3544 3548 10.1039/c8an00525g 29687792
44. Li B (2021) T4 DNA ligase-assisted single base mutation detection by PCR (Master’s thesis, Inner Mongolia University). 10.27224/d.cnki.gnmdu.2021.000336
45. Liu Z Guo C Zhang Y Zhao L Hao Z Rapid and sensitive detection of the colistin resistance gene mcr-3 by loop-mediated isothermal amplification and visual inspection Microb Drug Resist 2021 27 10 1328 1335 10.1089/mdr.2020.0129 34264742
46. Wang H Wang H Sun Y Liu X Liu Y Wang C Zhang P Li Z A general strategy for highly sensitive analysis of genetic biomarkers at single-base resolution with ligase-based isothermallyexponential amplification Talanta 2020 212 120754 10.1016/j.talanta.2020.120754 32113533
47. Zhang M, Wang H, Wang H, Wang F, Li Z (2021) CRISPR/Cas12a-Assisted ligation-initiated loop-mediated isothermal amplification (CAL-LAMP) for highly specific detection of microRNAs. Anal Chem 93(22):7942–7948. 10.1021/acs.analchem.1c00686
48. Sun Y Han B Sun F Ultra-specific genotyping of single nucleotide variants by ligase-based loop-mediated isothermal amplification coupled with a modified ligation probe RSC Adv 2021 11 28 17058 17063 10.1039/d1ra00851j 35479710
49. Choi MH Lee J Seo YJ Dual-site ligation-assisted loop-mediated isothermal amplification (dLig-LAMP) for colorimetric and point-of-care determination of real SARS-CoV-2 Mikrochim Acta 2022 189 5 176 10.1007/s00604-022-05293-7 35381892
| 36441353 | PMC9707155 | NO-CC CODE | 2022-12-01 23:20:21 | no | Mol Cell Biochem. 2022 Nov 28;:1-11 | utf-8 | Mol Cell Biochem | 2,022 | 10.1007/s11010-022-04615-w | oa_other |
==== Front
Surg Endosc
Surg Endosc
Surgical Endoscopy
0930-2794
1432-2218
Springer US New York
36443563
9763
10.1007/s00464-022-09763-0
Dynamic Manuscript
Reperfused human cadaver as a new simulation model for colonoscopy: a pilot study
Rohr A. 1
Perrenot C. 23
Pitta A. 1
Celerier I. 2
Labrousse M. 24
Renard Y. 23
Cadiot G. 12
http://orcid.org/0000-0001-7574-9364
Brugel M [email protected]
1
1 grid.139510.f 0000 0004 0472 3476 Gastroenterology and Digestive Oncology Department, Centre Hospitalier Universitaire de Reims, Hospital Robert Debré, Rue du Général Koenig, 51100 Reims, France
2 grid.11667.37 0000 0004 1937 0618 Anatomy Department, Université de Reims Champagne Ardenne, Reims, France
3 grid.139510.f 0000 0004 0472 3476 Digestive and Endocrine Surgery, Centre Hospitalier Universitaire de Reims, Reims, France
4 grid.139510.f 0000 0004 0472 3476 Department of Otorhinolaryngology, Head and Neck Surgery, Centre Hospitalier Universitaire de Reims, Reims, France
28 11 2022
19
2 9 2022
6 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Background
The reperfused human cadaver is a validated simulator for surgery. We aimed to use it as a colonoscopy simulator.
Methods
Novices, intermediates and skilled participants in gastrointestinal endoscopy were included. They performed one colonoscopy on a reperfused human cadaver and reaching rates, time, and length needed to reach anatomical landmarks were reported for construct validity analysis. We also assessed our model realism (SRS survey), educational content (CVS survey) and task load (NASA-TLX index). Score items were collected and defined as “favorable” when items were rated ≥ 5/7 with an inter-quartile range (IQR) overlapping four, and “very favorable” when rated ≥ 5/7 with an IQR excluding four (neutral).
Primary endpoints were the rectosigmoid junction (RSJ) reaching rate and the descending colon (DC) reaching time. Secondary objectives were SRS, CVS and NASA-TLX questionnaire results.
Results
A total of 11 skilled participants, 5 intermediates and 8 novices were included. Skilled participants reached RSJ more often than novice and intermediary groups, respectively, 100%, 80% and 75% without differing significantly. They reached DC more frequently (100% for skilled, 80% for intermediates and 50% for novices, p = 0.018). The median time to reach RSJ (59, 272 and 686 s for skilled, intermediates and novices group, respectively) and DC (90, 534 and 1360 s for skilled, intermediates and novices) was significantly shorter for skilled participants (both p < .01).
Nineteen out of the 22 items composing the realism survey obtained “very favorable” and “favorable” scores. Educational content was designated as “very favorable”. Mental, physical, and technical demands were gradually higher the lower the initial level of experience.
Conclusions
Reperfused human cadaver model has the potential to be valid simulation tool for diagnostic colonoscopy training.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00464-022-09763-0.
Keywords
Endoscopy, gastrointestinal
Simulation training
Education, medical
Learning curve
==== Body
pmcDiagnostic and therapeutic colonoscopy is a cornerstone in the gastroenterologist’s (GE) set of skills. Several years of practice are needed to perform complete, safe and quick procedures. Learning curves have enabled the estimation of a minimal competency threshold recognizing the ability to perform a procedure correctly [1–4]. Simulation-based training has been widely developed and helps trainees to achieve curriculum fulfilment. It improves endoscopy teaching and trainees’ endoscopic performance, especially in the early stages [5–8]. Novices can acquire skills in handling, progression, anatomy and more difficult techniques to optimize their valuable time in a caring environment without compromising patients’ safety and comfort [9, 10].
Validated simulators in gastrointestinal endoscopy (GIE) consist of mechanical, animal-based, or virtual reality (VR) models. They all have their advantages and limitations in terms of realism, force feedback, educational feedback or ethical concerns [11, 12]. Unfortunately, their high cost and low accessibility hamper residents’ and fellows’ learning skills. There is a need for an affordable and accessible alternative.
The Human-cadaveric model is a historical simulation tool for surgical training programs due to its high anatomical fidelity and great educational value [13, 14]. Data about the human-cadaveric model in GIE training is inexistent regarding lower gastrointestinal endoscopy. The main objective was to evaluate the feasibility of a human-cadaveric model as a colonoscopy training model. Secondary objectives were to assess realism, effectiveness as a learning tool and task load.
Material and methods
We conducted a monocentric prospective feasibility and validity study. Every simulation was performed at the medical school’s Anatomy Department (Université de Reims Champagne Ardenne) from April to July 2021. The Anatomy Department provides anatomy lessons, dissection lectures, surgical and emergency training on cadavers to its students. Cadavers can be used either fresh or defrosted. All of them were tested negative for COVID-19 (RT-PCR method) and respected the core ethical practice in human dissection (International Federation of Associations of Anatomists, IFAA, 2012). cadavers were individuals who had bequeathed their bodies to scientific research. Anonymity is ensured by the administrative process. We had no access to cadavers’ medical or surgical records. We selected experts in GIE (AR, GC, MB), medical education (AR, IC, CP, GC, MB) and simulation-based training (CP). Regular informal discussions allowed us to conceive a step-by-step reproducible protocol of cadaver preparation and bowel cleansing.
Reperfused human cadaver colonoscopy simulator (RHCCS)
Cadavers were slowly defrosted at 15 °C 3–5 days before use and were connected to a Pulse for Practice® device (P4P®, Simedys® 2020, Poitiers, France). P4P® system is a specific device allowing pulsatile arterial and venous revascularization by connecting eight catheters infusing 37 °C-heated fake blood in both femoral and common carotid arteries (input) and both femoral and internal jugular veins (output) (Photo 1A). It aims to recolor, reheat organs and tissues and restore an intravascular pressure with a pump to simulate blood circulation [15, 16]. Many parameters are available to enhance realism such as blood pressure in mmHg, percentage of diastole duration, pulse per minute and continuous or pulsating mode. The cadavers in this study could have been used in previous medical or surgical training sessions as long as these excluded digestive surgery exercises modifying colonic anatomy. Cadavers were not embalmed or formalin-fixed, and could be used for 2 days in a row. Each cadaver was used either until excessive damage, such as intestinal perforation, compromised the simulation experience, or until the end of the session.
Bowel preparation
A digital rectal examination was performed to confirm anus presence and extract hard stools from the rectum. A rectal probe or rectally inserted Flexi-Seal® (ConvaTec, Reading, United Kingdom) (Photo 1B) device was used to send cool water into the colon lumen for at least 3 min until clean water emerged to achieve complete colonic defrosting and stool evacuation. After satisfactory stool extraction, a first colonoscopy was performed by one of the investigators (AR, MB) to check colonic anatomy, improve mucosal cleaning using the endoscopic flushing pump and determine the highest reachable point by trainees. Every cadaver with a major colonic resection or at least, left colectomy, was not used.
We used the Olympus® EVIS EXERA III Video Colonoscope (CF-H190L/I) (Olympus Corporation, Tokyo, Japan) (Photo 1C), and the entire set of endoscopy equipment was for sole use of the anatomy department.
Study protocol
All participants were selected among volunteers and spontaneous offers by caregivers (medical students, residents, and medical staff) at the Reims University Medical Center. Each screened participant underwent a basic demographic self-questionnaire (sex, age, dominant hand, job), signed an agreement-of-participation form, and was asked about their manual abilities and endoscopic experience (see Supplementary material). Based on their endoscopy experience, participants were divided into three distinct groups:—novices, who had never performed GIE,—intermediates who had already performed one to ten colonoscopies alone and—skilled participants, who had already performed > 10 complete colonoscopies in complete autonomy.
Several sessions were scheduled. Participants from different group-levels were assigned to a session and started the colonoscopy according to their order of arrival. All included participants received a short 5-min tutorial briefing about gastrointestinal endoscopes and how to use them.(i) We used objective performance parameters such as time and length to different anatomical landmarks (rectosigmoid junction (RSJ), descending colon (DC), splenic flexure, hepatic flexure, and caecum), including total time of procedure and maximal point of insertion. A maximum of 25 min was allowed for each participant to reach the left colon. Investigators performed a complete supervision including live feedback and a debriefing session an open evaluation of each participant during the colonoscopy.
Each participant’s experience was debriefed, and a multidimensional assessment of participants’ experience was performed at the end of each procedure.
(ii) Skilled participants completed a simulation realism survey (SRS) to evaluate the similarity between the human-cadaver model and human-living endoscopy (face validity). Detailed SRS is available in Supplementary material—Figure S1. Each of these parameters was measured using a Likert scale (1 = strongly disagree, 4 = neutral, 7 = strongly agree). Technical difficulty fidelity (resemblance of technical difficulty to real-life conditions) was also measured with a Likert Scale (1 = easier, 4 = neutral, 7 = harder than human living colonoscopy). Realism was defined as favorable when items were rated over or equal to 5 with an inter-quartile range (IQR) overlapping 4, and very favorable when rated over or equal to 5 with an inter-quartile range (IQR) excluding 4 (neutral). Technical difficulty fidelity was defined as ideal when equalling 4 or overlapping.
(iii) Skilled participants also assessed the educational quality of this model (content validity) using a content validity survey (CVS) designed by Jones et al. [17], based on the GAGES evaluation for endoscopic proficiency [18]. The CVS form is available as supplementary material—Figure S2. It is composed of seven items on a scale from one to seven (1 = strongly disagree, 4 = neutral, 7 = strongly agree). The definition of favorable and very favorable items results was the same as for the SRS.
(iv) Finally, we used the multidimensional NASA Task Load Index (NASA TLX) to assess perceived mental and physical load (Supplementary material—Figure S3). We added one item (mental disturbance) to the original questionnaire (mental demand, physical demand, temporal demand, performance, technical issue, and frustration) to encompass this dimension. Each item had a scale score from 1 (very low) to 20 (very high), except for the item “performance” where 1 is for “perfect” and 20 for “failure”.
The main objective was to evaluate the feasibility of using a human-cadaveric model as a colonoscopy training model (construct validity). Primary endpoints were RSJ reaching rate and DC reaching time between groups. Secondary objectives were to assess realism (face validity), effectiveness as a learning tool (content validity) and task load using SRS, CVS and NASA-TLX questionnaires, respectively. Every participant signed a written consent before inclusion. No institutional review board approval was needed as this was a non-interventional monocentric study.
Data analysis
Quantitative data were described with their median and IQR (Q1–Q3), and compared with the non-parametric Kruskal–Wallis test. Qualitative data were described using frequencies and percentages and compared with the Chi-square test or Fisher’s exact test when appropriate. No minimum number of participants was required for this study as its main purpose was exploratory. When RSJ was not reached, inserted colonoscope length was not considered and a standard time of 25 min was assigned to the participant. Statistical significance was defined as a p-value < 0.05 for all tests. All statistical analyses were performed using R (R Development Core Team, 2005). All authors and co-authors had access to the study data, reviewed and approved the final manuscript.
Ethical concern
Cadavers had been donated to the Department of Anatomy, Faculty of Medicine and University Hospital, Champagne Ardenne university, for anatomical education and research purposes. As described by the French laws, all the ethical rules concerning work on cadaveric material in our institution (Law 94‐653 of July 29, 1994 relative to the respect of the human body) were followed for this study. The staff supervising cadaver manipulations was trained and ensured the respect of ethical principles.
Results
A total of 31 potential participants volunteered or were invited. Among them, seven refused to take part in the study (four in our medical staff, two third-year gastroenterology residents, and one GIE nurse). Eleven skilled participants (median age of 29, IQR: 27–39), five intermediates (median age of 26, IQR: 25–27), and eight novices (median age of 24, IQR: 23.8–25.5) were included. Included participants’ characteristics are presented in Table 1 and supplementary material—Table S1A and B. One of the novices included was a pneumologist resident who had already performed bronchoscopies on a living patient. A total of six cadavers were used to perform 24 colonoscopies.Table 1 Sociodemographic characteristics of the included population
Variable Skilled (n = 11) Intermediate (n = 5) Novice (n = 8) Total (n = 24) P
Gender, n (%) Woman 6 (54.5) 1 (20.0) 5 (62.5) 12 (50.0) 0.303
Age, median (IQR) Years 29 (27–39) 26 (25–27) 24 (23.8–25.5) 26.5 (24.8–29)
Professional status, n (%) Medical doctor 6 (54.5) 0 (0.0) 0 (0.0) 6 (25.0) < .01
Resident 5 (45.5) 5 (100.0) 2 (25.0) 12 (50.0)
Medical student 0 (0.0) 0 (0.0) 6 (75.0) 6 (25.0)
Medical specialty, n (%) GE 11 (100.0) 5 (100.0) 0 (0.0) 16 (66.6) < .01
Radiotherapy 0 (0.0) 0 (0.0) 1 (12.5) 1 (4.2)
Pneumology 0 (0.0) 0 (0.0) 1 (12.5) 1 (4.2)
Medical student 0 (0.0) 0 (0.0) 6 (75.0) 6 (25.0)
n number of people, IQR Interquartile range
Skilled participants were composed of five GE residents and six members of staff with different levels of experiences. Five participants in the intermediary and skilled group had performed simulation-based endoscopy on human cadavers prior to this study.
There was no difference between groups in terms of their dominant hand or practice of activities requiring dexterity (musical instrument or video games).
A video of a simulated colonoscopy is available (Video 1).(i) Construct validity
Skilled participants reached RSJ more often than novice and intermediary groups, respectively, 100%, 80% and 75% without significantly differing (Table 2). They also reached other anatomic landmarks more frequently: DC (p = 0.018) and splenic flexure (p < 0.01). The time to reach RSJ and DC was significantly shorter for skilled participants (p < 0.01 and p < 0.01, respectively) (Fig. 1). Furthermore, reaching rates and speed gradually increased according to experience.
No difference was observed in inserted scope length to reach different anatomical landmarks. Colonic perforation occurred twice: once by a novice and once by an expert who mentioned that she had not noticed the typical pre-perforation “red flags”.
(ii) Face validity
Simulator fidelity assessment results by skilled participants are shown in Supplementary Material—Table S2. Level of realism was rated as very favorable in “endoscopic view” (5.0, IQR: 4.5–6.0), “stool look” (5.0, IQR: 4.2–6.0), “colon length” (6.0, IQR: 5.8–7.0), “haustra/folds” (6.0, IQR: 4.5–7.0), “sigmoid angle” (5.0, IQR: 5.0–6.0), “splenic flexure” (5.0, IQR: 5.0–6.0), “hepatic flexure” (5.5, IQR: 5.0–6.0), “resistance to scope advancement” (6.0, IQR: 4.5–6.5), “paradoxical scope motion” (5.0, IQR: 4.5–7.0), “location of paradoxical motion encountered” (6.0, IQR: 5.0–7.0), and “frequency and degree of looping occurrence” (5.0, IQR: 5.0–6.0) items. Seven other items were rated as favorable: “mucosal realism”, “mucosal vascularization”, “response to loop reduction techniques”, “response to abdominal compressive maneuvers”, “occurrence of abdominal distension after excessive insufflation”, “exsufflation quality assessment at the end of the procedure”, and “overall simulation fidelity”. Technical difficulty was comparable to that of a normal colonoscopy (5.0, IQR 4,0 to 5,0).
(iii) Content validity
Content validity results are presented in Supplementary Material—Table S3. Items “This simulator closely replicates the necessary skills for a real colonoscopy”, “This simulator would be effective to teach residents/fellows scope navigation”, “This simulator would be effective to teach residents/fellows basic maneuvers such as insufflation, suction and retroflexion”, “This simulator would be effective to teach residents/fellows to keep a clear endoscopic field”, and “This simulator would be effective to teach residents/fellows instrumentation (biopsy and snare polypectomy)” scored > 5, without overlapping the neutral score of 4.
(iv) Workload study
Modified NASA-TLX survey showed that mental, physical, and technical demands were gradually higher the lower the level was (Supplementary material—Table S4). Novices were significantly less disturbed working on human cadavers than intermediaries or skilled participants (p = 0.029).
Table 2 Reaching rates, time and length of the inserted endoscope for each anatomical landmark per group
Skilled (n = 11) Intermediate (n = 5) Novice (n = 8) P (n = 24)
Rectosigmoid junction reached, n (%) 11 (100.0) 4 (80.0) 6 (75.0) 0.196
Time to rectosigmoid junction (seconds) 59.0 (20.0–76.0) 272.0 (150.0–396.0) 686.0 (476.8–1015.5) < .01
Length to rectosigmoid junction* (centimeters) 20.0 (20.0–25.0) 20.0 (20.0–20.0) 22.5 (20.0–25.0) 0.329
Descending colon reached, n (%) 11 (100.0) 4 (80.0) 4 (50.0) 0.018
Time to descending colon (seconds) 90.0 (61.0–133.0) 534.0 (303.0–1318.0) 1360.0 (847.2–1500.0) < .01
Length to descending colon* (centimeters) 30.0 (30.0–40.0) 35.0 (30.0–41.2) 30.0 (30.0–32.5) 0.766
Left angle reached, n (%) 9 (81.8) 1 (20.0) 1 (12.5) < .01
Time to left angle* (seconds) 185.0 (123.0–223.0) 760.0 (760.0–760.0) 909.0 (909.0–909.0) 0.103
Length to left angle* (centimeters) 60.0 (45.0–60.0) 40.0 (40.0–40.0) 30.0 (30.0–30.0) 0.282
Right angle reached, n (%) 5 (45.5) 1 (20.0) 1 (12.5) 0.379
Time to right angle* (seconds) 196.0 (172.0–230.0) 959.0 (959.0–959.0) 1329.0 (1329.0–1329.0) 0.145
Length to right angle* (centimeters) 80.0 (80.0–85.0) 80.0 (80.0–80.0) 75.0 (75.0–75.0) 0.575
Time (seconds) and length of inserted endoscope (centimeters) are expressed as median (IQR)
n number of people, IQR Interquartile range
*All participants who did not reach this point were not analyzed
Fig. 1 A Time to reach the rectosigmoid junction per group; B Time to reach the descending colon per group; C Total colonoscopy time per group. Participants who did not reach the rectosigmoid junction and descending colon are represented by orange points. Participants who reached the rectosigmoid junction and descending colon are represented by blue points
Photo 1 A Pulse for Practice® device; B FlexiSeal device used for bowel cleansing; C Endoscopic devices used to perform the colonoscopy on cadavers
Discussion
After bowel preparation, re-perfused cadavers offer a reliable simulation model for colonoscopy. In this study, construct, face and content validation were provided to define the RHCCS as a new tool for colonoscopy training.
Traditionally, endoscopy training is based on an apprenticeship model with a student under the direct supervision of an expert, performing endoscopic procedures on a living patient. The European Society of Gastrointestinal Endoscopy and American Society for Gastrointestinal Endoscopy (ESGE and ASGE) developed a list of core technical, non-technical, and cognitive skills to acquire for colonoscopy [4, 19]. ESGE suggests that personal experience of at least 300 upper GIE and/or 300 lower GIE is required to achieve general competence [4]. Training on living patients brings with it ethical concerns as it includes risks of bleeding (2,4/1000) and intestinal perforation (0,3/1000) [20]. Therefore, simulation-based training allows fellows to perform procedures in a safe and stressless environment for patients and for themselves. GIE simulators include mechanical trainers, animal-based models, or virtual reality (VR) tools. Nevertheless, prohibitive costs, restricted accessibility and sometimes low educational quality hinder the integration of simulation-based training into the GIE curriculum [21, 22].
The human cadaver is the historical and gold-standard validated training tool for surgical residents. It provides a high level of realism in terms of technical gesture and feel of human tissue and it mimics the operating conditions perfectly (body positions on the operating table, operating fields, anatomy…) [13, 14]. It increases residents’ confidence and may lead to earlier operative autonomy [23]. Moreover, each cadaver can be reused for different surgical training workshops. Despite its validity in the surgical simulation domain, the human cadaver model has never been validated in GIE. An Indian team have described the use of cadavers to perform upper GIE simulation and endoscopic techniques without face, content and construct validation [18, 24]. This study confirmed that cadavers are usable for upper GIE simulation. However, no colonoscopy has ever been performed previously. With the RHCCS, we reported a reliable tool for colonoscopy simulation on cadavers for the first time.
In France, endoscopic simulation training is based on animal ex-vivo (79%) and VR models (71%) used in 14 workshops around the country [21]. Unfortunately, french residents do not benefit from training with these simulators as they are only exposed to them once or twice during their 5-year programme [21]. In the USA, simulation-based training is represented by computerized (61.5%), mechanical (30.8%) and animal-explanted (7.7%) tools in fewer than half of all training centers [22]. Moreover, a computerized simulator costs between $50,000 and $140,0008. Composite animal simulators may seem cheaper (e.g., $2,200–$2,600 for EASIE-R3® simulator, Endosim, Büchenbach, Germany), but require the purchase of the main unit and single-use prepared porcine organs (from $125 for porcine stomach specimen to $350 for complete porcine explant specimen in Endosim). In French medical schools, cadavers are donations. The cost of a cadaver preparation could be estimated around 160€ (including 70€ of infusion, 50€ of fake blood, 40€ of protection equipment). However, a P4P device costs around 80 000 € but can be used for all simulations (including surgical simulations). The contribution of the P4P device in this setting remains unclear and also needs to be validated. However, this added equipment to the model could play a role in the recoloration of the mucosae. The human cadaver model can be mutualized: it can become a shared simulation device between medical and surgical specialties lowering its economic burden and enhancing the teaching of multidisciplinarity. To be further implemented if validated, every academic center with a cadaver dissection theater would need to be trained to cadaveric bowel cleansing and taking into account the ethical dimension. In our study, we have also shown that colonoscopy on a human cadaver is possible with a simple and thorough bowel preparation in advance. This reproducible bowel cleansing protocol allowed fellows to perform colonoscopies with high fidelity without either disrupting the educational purpose or increasing its cost.
The cadaver model was described as very favorable in terms of realism for 11 items, and favorable for eight items over 22. The RHCCS model can accurately replicate the visual and mechanical properties of a colonoscopy on a living patient (e.g., occurrence of paradoxical motion, resistance to scope advancement and intestinal loop). Realism is a fundamental aspect of a validated simulation training tool. In surgery, realism is an essential tool allowing comparison between different laparoscopic training systems [25]. In GIE, several studies have investigated the realism of VR, mechanical and ex-vivo colonoscopy simulators [26]. Mechanical simulators reproduced satisfying mechanical properties and realistic force feedback [11]. Nevertheless, visual realism is not achieved with a score of only 32,69/100 for the Koken Colonoscopy Training Model Type 1-B device, and a score of only 21,77/100 for the Kyoto Kagaku Colonoscope Training Model [27]. These scores were also low for VR devices such as the Endoscopy AccuTouch System, Immersion Medical Corporation, and the Gi Mentor II, Simbionix Corporation, with 50.64/100 and 42.86/100, respectively. Finally, the visual realism of the ex-vivo colonoscopy bovine model obtained a score of 6.0/7 [22]. However, as these studies used different scores with heterogeneous low-numbered populations, no face-to-face comparison seems appropriate. A face-to-face comparison with other simulators is necessary to better investigate the input of this model.
The RHCCS can differentiate between participants based on their endoscopic level thus ensuring a high construct validity. Even if no significant statistical difference in RSJ reaching rate was shown between the three groups, we can observe a gradual increase in reaching rates and speed, according to experience. Construct validity of the RHCCS may be an opportunity for certification in GIE before practicing on a patient.
Moreover, a good endoscopic simulator should be effective for teaching colonoscopy competencies (evidence of content validity). Skilled participants have recognized the educational quality of this model, based on the GAGES criteria for endoscopic proficiency [18]. All items obtained a very favorable score. The model’s educational value needs to be validated in a larger multicentric cohort. The RHCCS is a valid tool regarding content validity.
Finally, the mental, physical and technical load survey based on the modified NASA Task Load Index is a good measurement of each participant’s level. As expected, the lower the level of skills, the more demanding the task. This assessment, although not usually used in validation studies of endoscopic simulation models, highlights the stress that endoscopy can generate for inexperienced students. This confirms that the RHCCS offers realistic conditions for GIE trainees. In the future, comparison with the modified NASA-TLX score in real-life colonoscopies on a living patient could be interesting. With reference to the additional question included in this study, the sight of a cadaver does not seem to be a problem for those who accepted to be included, even for younger or inexperienced participants.
Multimodal evaluation is the mainstay of simulation model validation. It measures the model’s ability to reflect the reality of a simulated situation. In our RHCCS model, we successively showed high visual, mechanical and task load fidelity, capacity to distinguish operators based on their experience, and educational purpose.
Some limitations in our study can be pointed out, however. First, albeit its prospective format, it is a monocentric study with a limited number of participants which results in low statistical power. Secondly, no minimum number of participants was required for this study as the main purpose was exploratory and mainly qualitative. Nevertheless, as it is considered a pilot study, the number of included participants seems acceptable. The limited number of participants per group limits the statistical analysis in its interpretation. The definition of our skilled group could be criticized as we considered participants with > 10 autonomous colonoscopies as being “skilled”. This definition is not in line with the definition of other authors who set the thresholds at > 200 or sometimes > 1000 [28, 29]. Our main objective was designed for initial endoscopic education and essential skill acquisition in a mistake-tolerant environment. For this reason, and to lower ambiguity, we chose to call this group “skilled participants” instead of “experts”. Also, only one colonoscopy per participant may be insufficient to conclude on the validity of the model and more powerful longitudinal studies are required.
We can also notice that one participant in the novice group had a small amount of experience in bronchoscopy, which could have enhanced the novice group’s performance score. Despite this, the skilled group performed significantly faster than the other two groups. The gap between the skilled participants and the other two groups suggests that this model could be a useful tool to make the intermediary progress towards expertise.
We only included participants who had accepted to perform a colonoscopy on a human cadaver. We obtained seven refusals (five in our medical staff, one in intermediaries and one in novices). The acceptability of human cadavers as a simulation model has to be investigated because of personal or religious convictions. Nevertheless, this kind of workshop is commonplace in the medical curriculum (e.g., suture or dissection classes) and the majority of medical students are voluntary and agree to train on human cadavers. Additional measures could be applied to lower the emotional burden of this training (e.g., covering bodies with surgical drapes).
This model could also lack sufficient feedback. Contrary to VR devices, the human cadaveric model does not provide objective quality measurement tools such as patient’s pain level, percentage of mucosa visualized, or amount of air insufflated, which have significant educational value. Supervision by skilled endoscopists seems mandatory to guarantee the educational value of the model. To provide a better level of content validity, future research should compare the ability to perform high-quality GIE depending on the learning device used.
Also, according to skilled participants, the colonic mucosa lacked realism in terms of mucosal color despite reperfusion with a P4P® device. We suggested that colonic mucosal rendering could be improved with the use of a fresh cadaver rather than an defrosted cadaver, and higher blood pressure from the mechanical pump of the P4P® device.
Two intestinal perforations occurred during our study: one by a novice, and one by a skilled participant. Perforation on a human cadaveric model may thus occur more frequently as it seems to show fewer pre-perforation red flags than on a living body (e.g., whitening mucosa, vanishing capillaries).
The RHCCS is a satisfying model in terms of realism and educational value, which can be improved over time. In the future, this model could be used as a way to teach the skill of therapeutic colonoscopy (e.g., snare polypectomy, endoscopic clipping, electrocautery, submucosal injection).
This feasibility study on the reperfused human cadaver model for colonoscopy simulation demonstrated construct, face and content validities. This study calls for more robust studies to definitely validate the use of our human-cadaver simulation model for diagnostic colonoscopy training and evaluation.
Supplementary Information
Below is the link to the electronic supplementary material.Table 1A. Sociodemographic characteristics of the included population (extra)Abbreviations: n: number of participants, IQR: Interquartile range, h: hours
Table 1B. Previous experience of the included populationAbbreviations: n: number of participants, IQR: Interquartile range, GE: gastroenterology, h: hours, VR: virtual reality** Apprehension is assessed using a Likert scale from 1: strongly disagree to 7: strongly agree.
Table 2. Simulation realism survey (SRS) median item score for skilled participants. Scores are expressed as median (IQR). Abbreviations: IQR: Interquartile rangeEach of these parameters was measured using a Likert scale (1= strongly disagree, 4= neutral, 7= strongly agree). Technical difficulty fidelity was also measured with a Likert Scale (1= easier, 4= neutral, 7= harder than human living colonoscopy). Realism was defined as favorable when items were rated over or equal to 5 with an inter-quartile range (IQR) overlapping 4, and very favorable when rated over or equal to 5 with an inter-quartile range (IQR) excluding 4 (neutral). Technical difficulty fidelity was defined as ideal when equalling 4 or overlapping.
Table 3. Content validity survey (CVS) median item score for skilled participants. Scores are expressed as median (IQR). Abbreviations: IQR: Interquartile range. Each of these parameters was measured using a Likert scale (1= strongly disagree, 4= neutral, 7= strongly agree). Item was defined as favorable when rated over or equal to 5 with an inter-quartile range (IQR) overlapping 4, and very favorable when rated over or equal to 5 with an inter-quartile range (IQR) excluding 4 (neutral).
Table 4. NASA Task Load Index modified median item score per group Scores are expressed as median (IQR). Each item had a scale score from 1 (very low) to 20 (very high), except for the item “performance” where 1 is for “perfect” and 20 for “failure”.
Supplementary file1 (DOCX 1479 KB)
Fig. 1 Simulation Realism Survey form
Fig. 2 Content Validity Survey form
Fig. 3 NASA Task Load Index Modified form
Fig. 4 Bowel cleansing protocol
Supplementary file2 (MP4 63454 KB)
Acknowledgements
The investigators would like to thank the Université de Reims Champagne Ardenne for its intellectual support and the Anatomy Laboratory staff for their kindness and availability. The authors would also like to thank Daniela Pellot of the SERRA at Reims Faculty of Medicine for assistance with English language editing. All of the authors would like to sincerely thank the donors and their families.
Authors contributions
AR: Conceptualization, Validation, Investigation, Resources, Data Curation, Writing—Original Draft, Writing—Review & Editing, Visualization, Project administration; CP: Conceptualization, Methodology, Validation, Writing—Review & Editing; AP: Validation, Writing—Review & Editing; IC: Validation, Writing—Review & Editing; ML: Writing—Review & Editing, YR: Conceptualization, Writing—Review & Editing; GC: Conceptualization, Validation, Resources, Writing—Review & Editing; MB: Conceptualization, Validation, Formal Analysis, Investigation, Resources, Data Curation, Writing—Original Draft, Writing—Review & Editing, Visualization, Supervision, Project administration.
Funding
None.
Declarations
Disclosures
Cyril Perrenot received honoraria as a consultant for Medtronic, BD and Hartmann SA unrelated to this work. All other authors (Amélie Rohr, Anaïs Pitta, Isabelle Celerier, Marc Labrousse, Yohann Renard, Guillaume Cadiot, Mathias Brugel) have no conflict of interest or financial ties to disclose.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Ward ST Hancox A Mohammed MA The learning curve to achieve satisfactory completion rates in upper GI endoscopy: an analysis of a national training database Gut 2017 66 1022 1033 10.1136/gutjnl-2015-310443 26976733
2. Ward ST Mohammed MA Walt R An analysis of the learning curve to achieve competency at colonoscopy using the JETS database Gut 2014 63 1746 1754 10.1136/gutjnl-2013-305973 24470280
3. Sedlack RE Coyle WJ Sedlack RE Assessment of competency in endoscopy: establishing and validating generalizable competency benchmarks for colonoscopy Gastrointest Endosc 2016 83 516 523.e1 10.1016/j.gie.2015.04.041 26077455
4. Dekker E Houwen BBSL Puig I Curriculum for optical diagnosis training in Europe: European Society of Gastrointestinal Endoscopy (ESGE) position statement Endoscopy 2020 52 899 923 10.1055/a-1231-5123 32882737
5. Singh S Sedlack RE Cook DA Effects of simulation-based training in gastrointestinal endoscopy: a systematic review and meta-analysis Clin Gastroenterol Hepatol 2014 12 1611 1623.e4 10.1016/j.cgh.2014.01.037 24509241
6. Khan R Plahouras J Johnston BC Virtual reality simulation training in endoscopy: a Cochrane review and meta-analysis Endoscopy 2019 51 653 664 10.1055/a-0894-4400 31071757
7. Ekkelenkamp VE Koch AD de Man RA Kuipers EJ Training and competence assessment in GI endoscopy: a systematic review Gut 2016 65 607 615 10.1136/gutjnl-2014-307173 25636697
8. Technology Committee ASGE Goodman AJ Melson J Aslanian HR Bhutani MS Krishnan K Lichtenstein DR Navaneethan U Pannala R Parsi MA Schulman AR Sethi A Sullivan SA Thosani N Trikudanathan G Trindade AJ Watson RR Maple JT Endoscopic simulators Gastrointest Endosc 2019 90 1 12 10.1016/j.gie.2018.10.037 31122746
9. Sedlack RE Kolars JC Alexander JA Computer simulation training enhances patient comfort during endoscopy Clin Gastroenterol Hepatol 2004 2 348 352 10.1016/S1542-3565(04)00067-9 15067632
10. Ahlberg G Hultcrantz R Jaramillo E Virtual reality colonoscopy simulation: a compulsory practice for the future colonoscopist? Endoscopy 2005 37 1198 1204 10.1055/s-2005-921049 16329017
11. Finocchiaro M Cortegoso Valdivia P Hernansanz A Training simulators for gastrointestinal endoscopy: current and future perspectives Cancers 2021 13 1427 10.3390/cancers13061427 33804773
12. King N Kunac A Merchant AM A review of endoscopic simulation: current evidence on simulators and curricula J Surg Educ 2016 73 12 23 10.1016/j.jsurg.2015.09.001 26699281
13. Ahmed K Aydin A Dasgupta P A novel cadaveric simulation program in urology J Surg Educ 2015 72 556 565 10.1016/j.jsurg.2015.01.005 25683152
14. Wyles SM Miskovic D Ni Z Analysis of laboratory-based laparoscopic colorectal surgery workshops within the English national training programme Surg Endosc 2011 25 1559 1566 10.1007/s00464-010-1434-y 21058021
15. Danion J Breque C Oriot D La technologie SimLife® dans la formation du chirurgien J Chir Viscérale 2020 157 S119 S124 10.1016/j.jchirv.2020.02.005
16. Faure JP Breque C Danion J SIM Life: a new surgical simulation device using a human perfused cadaver Surg Radiol Anat 2017 39 211 217 10.1007/s00276-016-1715-9 27315801
17. Jones MW Deere MJ Harris JR Fabrication of an inexpensive but effective colonoscopic simulator JSLS 2017 21 e2017.00002 10.4293/JSLS.2017.00002 29353990
18. Gages: a valid measurment tool for technical skills in flexible endoscopy—SAGES abstract archives. SAGES. Im Internet: https://www.sages.org/meetings/annual-meeting/abstracts-archive/gages-a-valid-measurment-tool-for-technical-skills-in-flexible-endoscopy/; Stand: 10.10.2021
19. Walsh C Umar S Ghassemi S Colonoscopy core curriculum Gastrointest Endosc 2020 10.1016/j.gie.2020.06.054 32854938
20. Reumkens A Rondagh EJA Bakker CM Post-colonoscopy complications: a systematic review, time trends, and meta-analysis of population-based studies Am J Gastroenterol 2016 111 1092 1101 10.1038/ajg.2016.234 27296945
21. Camus M, Jacques J, Mouel JL et al (2019) L’offre de formation initiale en simulation en endoscopie digestive en France. In Endoscopy. Georg Thieme Verlag KG, p. 000535
22. Jirapinyo P Thompson CC Current status of endoscopic simulation in gastroenterology fellowship training programs Surg Endosc 2015 29 1913 1919 10.1007/s00464-014-3884-0 25303910
23. Kim SC Fisher JG Delman KA Cadaver-based simulation increases resident confidence, initial exposure to fundamental techniques, and may augment operative autonomy J Surg Educ 2016 73 e33 e41 10.1016/j.jsurg.2016.06.014 27488813
24. Balekuduru A Appaji A Therapeutic endoscopic procedures on a human cadaver—a pilot feasibility study J Dig Endosc 2021 12 036 042 10.1055/s-0041-1728223
25. Liu W Zheng X Wu R Novel laparoscopic training system with continuously perfused ex-vivo porcine liver for hepatobiliary surgery Surg Endosc 2018 32 743 750 10.1007/s00464-017-5731-6 28733731
26. Ansell J Mason J Warren N Systematic review of validity testing in colonoscopy simulation Surg Endosc 2012 26 3040 3052 10.1007/s00464-012-2332-2 22648104
27. Hill A Horswill MS Plooy AM Watson MO Karamatic R Basit TA Wallis GM Riek S Burgess-Limerick R Hewett DG Assessing the realism of colonoscopy simulation: the development of an instrument and systematic comparison of 4 simulators Gastrointest Endosc 2012 75 631 640 10.1016/j.gie.2011.10.030 22341108
28. Koch AD Haringsma J Schoon EJ A second-generation virtual reality simulator for colonoscopy: validation and initial experience Endoscopy 2008 40 735 738 10.1055/s-2008-1077508 18698536
29. Moorthy K Munz Y Orchard TR Gould S Rockall T Darzi A An innovative method for the assessment of skills in lower gastrointestinal endoscopy Surg Endosc 2004 18 1613 1619 10.1007/s00464-004-9002-y 15931472
| 36443563 | PMC9707156 | NO-CC CODE | 2022-12-01 23:20:21 | no | Surg Endosc. 2022 Nov 28;:1-9 | utf-8 | Surg Endosc | 2,022 | 10.1007/s00464-022-09763-0 | oa_other |
==== Front
Curr Allergy Asthma Rep
Curr Allergy Asthma Rep
Current Allergy and Asthma Reports
1529-7322
1534-6315
Springer US New York
36445652
1052
10.1007/s11882-022-01052-z
Anaphylaxis and Drug Allergy (M Castells, Section Editor)
Utilizing Biologics in Drug Desensitization
http://orcid.org/0000-0001-5018-095X
Yang Barbara C. [email protected]
12
http://orcid.org/0000-0001-6451-0163
Castells Mariana C. 1
1 grid.38142.3c 000000041936754X Division of Allergy and Clinical Immunology, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Hale Building for Transformative Medicine, 60 Fenwood Road, Room 5002-B, Boston, MA 02115 USA
2 grid.510013.6 0000 0004 6004 4363 Ribon Therapeutics, 35 Cambridgepark Drive Suite 300, Cambridge, MA 02140 USA
29 11 2022
111
3 10 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose of Review
The purpose of this literature review was to review the latest advancements with biologics in rapid drug desensitization. Our methodology was to highlight both desensitization to biologics themselves and the use of biologics in desensitization to both biologic and nonbiologic drugs.
Recent Findings
Biologics are a vast category of drugs that include monoclonal antibodies, nanobodies, modern vaccinations, and even hormones. Desensitization to biologics can be safely performed through standardized procedure. Biomarkers are used both in vitro and in vivo to help identify and classify hypersensitivity reactions. Hypersensitivity reactions to the mRNA vaccinations against SARS-CoV-2 present their own unique challenges to management. There are specific excipients in monoclonal antibodies that are thought to be responsible for many of their hypersensitivity reactions. Certain biologics can even be used to assist in desensitization to other drugs.
Summary
Rapid drug desensitization is a standardized procedure that may be able to help many patients who have experienced hypersensitivity reactions to biologics and would best be treated with them to continue to receive them. Biologic drugs have opened a new era in medicine for the prevention and treatment of infectious diseases, cancer, and inflammatory diseases. Hypersensitivity reactions to biologics are quite common. This literature review presents the latest advancements in our understanding of hypersensitivity reactions to biologics, how rapid drug desensitization can be used to continue therapy despite history of hypersensitivity, and how biologics themselves can be used to aid in desensitization itself.
Keywords
Desensitization
Hypersensitivity
Monoclonal antibodies
Biologics
Nanobodies
COVID-19 vaccine
==== Body
pmcIntroduction
Biologics are drugs comprised of organic molecules that only living systems can produce and always have either a gene or a protein as their therapeutic target [1]. They comprise a diverse class of medications that includes monoclonal antibodies, nanobodies, modern vaccinations, and hormones. Monoclonal antibodies were once categorized and named based on the degree of immunogenicity of their Fab fragments (e.g., having the suffix “-mab” with infixes of “-o-” for murine, “-xi-” for chimeric, “-zu-” for humanized, “-u-” for fully human, or the suffix “-cept” for receptor fusion) but those made after November 2021 are now categorized and named differently—with suffixes “-tug” for unmodified immunoglobulins, “-bart” for monospecific antibodies with artificially engineered constant domains, “-mig” for multispecific antibodies, and “-ment” for fragments without an Fc domain [2–4] (Fig. 1). Nanobodies are specifically just the antigen-binding fragment of what could be recognized as a monoclonal antibody with no Fc portion or corresponding Fab fragment, making them the smallest known natural molecules that can bind a target protein epitope [5]. There have been numerous advances in the therapy of specific cancers and neurodegenerative diseases with the use of nanobodies without a single reported case of hypersensitivity to date. Modern vaccinations that use an mRNA platform, first used clinically with vaccinations against SARS-CoV-2, are by their very design biologic drugs and have shown tremendous efficacy [6–8]. Exogenous hormones used for the management of various endocrine diseases are also of biologic origin.Fig. 1 International Nonproprietary Names (INN) nomenclature for monoclonal antibodies. A The former classification (before November 2021) first required all monoclonal antibodies (MAbs) to carry the suffix -mab unless they were formed by the fusion of a receptor ligand and an Fc segment, where it would carry the suffix -cept. An infix would be included in the middle of the term (-o-, -xi-, -zu-, or -u-) based on how the MAb was produced (murine, chimeric, humanized, or human). A chimeric antibody has foreign amino acids making up the entire V heavy and light chains and is linked to heavy and light C regions of human origin. A humanized antibody has segments of foreign amino acids interspersed with V segments of human amino acids and the V heavy and light domains are linked to heavy and light C regions of human origin. Other infixes exist in the naming convention but have never been used (-e- for hamster, -i- for primate, etc.). B The current classification applies to all MAbs made after November 2021 and classifies them by their complementarity-determining regions (CDRs) and C regions. MAbs with unmodified C regions and identical sets of CDRs that recognize the same epitope of their target have the suffix -tug. This applies no matter the source of the rest of the MAb, even if chimeric or humanized. Full-sized MAbs with engineered amino acid changes in C regions (even the hinge region) and identical sets of CDRs recognizing the same epitope have the suffix -bart. Immunoglobulins that are bispecific and multispecific (having two different sets of CDRs) regardless of shape have the suffix -mig. These MAbs need not be full-sized immunoglobulins; they only need to have CDRs for two different epitopes, including for example BiTEs (bispecific T-engaging antibodies) and BiKEs (bispecific killer cell engagers) shown here, which are sets of V light and heavy chain CDRs directly linked together without an Fc region. Lastly, monospecific fragments of a full-sized immunoglobulin that are lacking a partial or entire C region have the suffix -ment. Examples shown here are a Fab fragment (fragment antigen-binding), a Fab(‘ab’)2 fragment (immunoglobulin cleaved by pepsin below the hinge region), and a scFv (single-chain variable fragment). The infixes used in the previous classification are no longer a part of the new classification. C Shown for comparison are a nanobody (a.k.a. single domain antibody) and a nanobody linked to a single-chain variable fragment. Abbr: V, variable; C, constant (created, in part, with BioRender.com)
Desensitization to Biologics
A major concern of the use of biologics in everyday clinical practice is that, when compared to other drugs, biologics have a relatively greater rate of hypersensitivity reactions, with rates reported upwards of a quarter of all patients, depending on the drug [9]. Even the first exposure to a biologic can lead to a hypersensitivity reaction [10]. There are seven types of reactions defined by their clinical presentation (phenotype), pathophysiology (endotype), and biomarkers: infusion reaction, type I IgE/non-IgE, cytokine release, mixed, type II, type III, and type IV (Fig. 2) [11••]. Rapid drug desensitization is a standardized procedure where successively increasing doses of the causative medication are rapidly administered that induce specific tolerance, allowing for continued administration of the medication [12, 13]. To determine eligibility for desensitization, patients are risk stratified by the severity of their reaction and by the results of skin testing (Fig. 3) [14, 15]. Sometimes, drug provocation testing is performed to further determine those who would most benefit from desensitization [16, 17]. Patients are premedicated (Fig. 4) and then reintroduced to the medication with rapidly increasing doses [11••]. Reactions that may occur during desensitization are managed as they arise (Fig. 5). Resulting desensitization to the medication is only maintained through continually repeated exposure to the specific drug. The patient will become sensitized again if there is any prolonged lack of exposure to the drug. With rapid drug desensitization, most hypersensitivity reactions to biologics can be safely overcome [18–23].Fig. 2 The seven types of drug reactions and whether rapid drug desensitization is indicated (created, in part, with BioRender.com. From Yang BC, Castells MC. The Who, What, Where, When, Why, and How of Drug Desensitization. Immunology and Allergy Clinics of North America. 2022. https://doi.org/10.1016/j.iac.2021.12.004; with permission)
Fig. 3 Algorithm for risk stratification for intravenous rapid drug desensitization. SCARs, severe cutaneous adverse reactions (toxic epidermal necrolysis, Stevens-Johnson syndrome, drug eruption with eosinophilia and systemic symptoms, acute generalized exanthematous pustulosis) (from Yang BC, Castells MC. Rituximab hypersensitivity and desensitization: a personalized approach to treat cancer and connective tissue diseases. Ann Allergy Asthma Immunol. 2019;123(1):11–5. https://doi.org/10.1016/j.anai.2019.03.008; with permission)
Fig. 4 Premedication for rapid drug desensitization is personalized for each patient and is based on the symptoms of the original reaction (from Yang BC, Castells M. Diagnosis and Treatment of Drug Hypersensitivity Reactions to Biologicals: Medical Algorithm. Allergy. 2020. https://doi.org/10.1111/all.14432; with permission)
Fig. 5 Algorithm for the management of breakthrough reactions during intravenous rapid drug desensitization (from Yang BC, Castells MC. Rituximab hypersensitivity and desensitization: a personalized approach to treat cancer and connective tissue diseases. Ann Allergy Asthma Immunol. 2019;123(1):11–5. https://doi.org/10.1016/j.anai.2019.03.008; with permission)
Advancements in Biomarkers
Serum tryptase is a helpful biomarker to be measured during a hypersensitivity reaction and helps to identify the type of reaction [24]. Given its short half-life, it should be initially measured between 30 and 120 min of the start of the hypersensitivity reaction. Baseline tryptase should also be measured 2 to 3 weeks after the event [11••]. An elevation of 2 + (1.2 × the baseline tryptase level) would be considered elevated. The acute serum tryptase level is difficult to interpret without the baseline serum tryptase measurement. When elevated, it is highly suggestive of a type I reaction, possibly IgE-mediated, as it is released when there is systemic mast cell degranulation [25]. In addition, the measurement of serum tryptase 24 h prior to surgery and again during episodes of possible anaphylaxis as manifested by hypotension and/or cutaneous, gastrointestinal, and respiratory changes during surgery has been found to be useful in identifying perioperative hypersensitivity and should be accompanied by allergist referral for further workup [26].
Hereditary alpha tryptasemia (HAT) is an autosomal dominant genetic disorder from having extra copies of the TPSAB1 allele and leads to a far increased risk of anaphylaxis [27]. In a study at our institution of 101 patients with genetically confirmed HAT, 58 of those patients (57.4%) had experienced at least one episode of anaphylaxis [28•]. Baseline serum tryptase levels of these patients averaged 17.2 ng/mL and ranged from 6.2 to 51.3 ng/mL. Data from this study suggests that any patient with even a single tryptase measurement of > 8.0 ng/mL has a significant possibility of having hereditary alpha tryptasemia and should be genotyped.
Cetuximab is a monoclonal antibody against EGFR with uses in the management of metastatic colorectal cancer and head and neck squamous cell cancers. The first biomarker of potential hypersensitivity is the presence of IgE against alpha-1,3-galactose (a.k.a. alpha-Gal) in patients who are to undergo treatment with cetuximab [29]. It was discovered that those who lived in the southeastern USA and Stockholm area of Sweden were more likely to carry such specific IgE. It was found in those who had hypersensitivity to consuming beef in these areas, the cross-reactivity in sensitization presumed to be from tick bites. Prior to testing for alpha-Gal, 80% of patients would experience hypersensitivity reaction to cetuximab, 3% of them severe [30]. It is now standard of care to test for IgE to alpha-Gal before beginning treatment with cetuximab to prevent this type of hypersensitivity reaction.
The activation of basophils also occurs in type I hypersensitivity reactions when IgE is crosslinked. There are two markers of basophil activation: CD203c and CD63 [31]. CD203c has a very low expression in resting cells but is upregulated after basophil activation. CD63 is normally expressed on the inner side of secretory granules and become exposed to the cytoplasmic membrane upon degranulation. The basophil activation test (BAT) is an in vitro assay that measures the degree of basophil activation by measuring CD63 and/or CD203c by flow cytometry in the hopes of detecting type I hypersensitivity. In a case series of 18 patients receiving rituximab, 5 patients who experienced hypersensitivity reaction had significantly elevated CD63 + basophils compared to the others [32]. Gradual decrease of BAT positivity to adalimumab has been demonstrated in 4 patients with each successive injection of the drug during rapid desensitization, showing a higher activation threshold in basophils because of desensitization [33].
BAT may be positive even when skin testing may be negative. In one case study, two patients with localized hypersensitivity reaction etanercept had negative skin tests but positive BAT and underwent successful rapid desensitization [34]. Another case study shows a patient with severe anaphylaxis to pertuzumab with negative skin prick and intradermal tests with a positive BAT for both markers who then underwent successful rapid desensitization [35]. This shows that BAT may be a useful adjunct to skin testing in the detection of type I hypersensitivity.
There are currently limitations to the use of BAT. Although its specificity is high, sensitivity ranges from 42 to 70%; a positive test is suggestive of type I hypersensitivity, but a negative test cannot yet rule it out [31]. BAT requires dedicated laboratory expertise and cannot detect non-IgE-mediated hypersensitivity reactions.
Another in vitro test that has the potential to identify multiple types of hypersensitivity reactions is the lymphocyte transformation test (LTT). Since T cells are involved in all types of hypersensitivity reactions, the increased activity of drug-specific T cells can be a sign of drug-specific hypersensitivity. The LTT measures the proliferation of T cells in response to a drug in vitro [36]. For example, in a patient with delayed cutaneous hypersensitivity reaction (type IV) to atezolizumab, both skin prick and intradermal tests were negative [37]. Drug provocation test reproduced the same delayed symptoms and LTT was positive; rapid desensitization was thereafter performed successfully. The LTT is a very technically demanding test and not commercially available as a ready-to-use assy. It is important to remember that a negative test cannot rule out hypersensitivity.
The clinical phenotype of a cytokine release hypersensitivity reaction (fever, chills, pain, nausea, vomiting, headache) is driven by IL-6. In a study from our institution, 21 patients who experienced a combined 38 breakthrough reactions during rapid desensitization to either chemotherapy or monoclonal antibodies had their serum IL-6 levels drawn at the time of their reactions [38]. In those 29 reactions where IL-6 levels were elevated, their complaints were primarily fever and neuromuscular in nature. A comparable rise in serum tryptase was not seen. IL-6 is a useful biomarker for the detection of cytokine release reactions. The fact that these reactions occurred during rapid desensitization also suggests that desensitization itself can control some but not all pathways of hypersensitivity reactions.
Vaccines
When the mRNA vaccinations against SARS-CoV-2 were given emergency use authorization by the FDA in December 2020, surveillance data was carefully collected by the CDC about adverse events [39]. Keeping in mind that this data is self-reported and thus may be over-estimated, anaphylaxis from the vaccine is rare, occurring in 2.8 to 5 per million people [40]. Thirty-one percent of those with anaphylaxis to the vaccine had a prior history of anaphylaxis. Sixty percent of those who reacted had previous history of allergic reactions (e.g., food allergy, drug allergy, Hymenoptera allergy, reaction to allergen immunotherapy, and/or carry an epinephrine autoinjector) [41]. Over 90% of patients who reported a reaction to the vaccination are female. To put these numbers into perspective, when compared to common vaccines available for other diseases ranging from 0.5 to 1 in 1,000,000, the rate of anaphylaxis to the mRNA vaccinations is thus approximated to be 10 × higher [42]. This was somewhat unexpected as there were no cases of anaphylaxis in the large phase 3 clinical trials for either mRNA vaccination, but patients who had a history of severe allergy to any component of the vaccines were excluded [7, 8]. The reported reactions are mostly vasovagal in nature (i.e., diaphoresis, bradycardia, nausea, vomiting) rather than anaphylaxis (i.e., urticaria, angioedema, hypotension) [43].
The exact cause of hypersensitivity reaction in the mRNA vaccinations is not yet fully known. It has been conjectured that “a certain portion” of patients with anaphylaxis to these vaccines may have underlying undiagnosed hereditary alpha tryptasemia [44]. These vaccinations have none of the excipients that are found to be culprits of anaphylaxis in other vaccines, such as gelatin, latex, egg protein, or surfactant (polysorbate 80) [45•]. The mRNA vaccinations package the mRNA in lipid nanoparticles as naked mRNA is directly immunogenic and would be broken down by the host immune system before it reaches its destination. The negative charge of the mRNA is counteracted by the positive charge of the lipid nanoparticle “sheath” that envelops it at low pH. There is some evidence in vitro that mast cells may possibly take up the sheathed vaccine particles and that there may be disruption of the phagosome caused by the lipid nanoparticle that leads to mast cell degranulation [46]. It is also not known if the lipid nanoparticle sheath can lead to direct mast cell degranulation by activation of other receptors, such as MRGPRX2 [47]. Even though the exact lipid component of the Pfizer-BioNTech vaccine differs from that of the Moderna vaccine, they are both made more water soluble with polyethylene glycol (PEG 2000). If there is existence of anti-PEG IgG or IgM, this may cause direct compliment activation, leading to anaphylaxis [48]. However, this very same compliment activation may lead to greater efficacy of the vaccine by increasing dendritic cell uptake of the vaccine particles and increased spike protein expression and more efficient antigen presentation to T cells [45•]. No other vaccination has PEG as an excipient, so there is no prior data available for comparison from other vaccinations. As PEG is a likely source of anaphylaxis, it is not yet proven, and more investigation is necessary.
In adults, the COVID-19 mRNA vaccinations require 2 primary doses (or 3 primary doses in specific at-risk populations) as well as 1 to 2 boosters based on risk to complete the series [49]. If one experiences a reaction on the first dose, the risk of reaction increases drastically on any subsequent dose. Assuming PEG is the most likely culprit excipient in the mRNA vaccines, then skin testing can be performed to determine sensitivity to PEG or even to related polysorbate in those who experienced hypersensitivity reaction, although the clinical utility of such testing remains under debate as it has not been determined if the mechanism of reactions are IgE-mediated [50]. In a case series of six patients who had an episode of anaphylaxis to their first dose of mRNA COVID-19 vaccine, two of the six patients were positive on skin prick test for PEG or polysorbate [51]. All had successful desensitization to the Moderna COVID-19 vaccination. In another study, 159 patients with first-dose reactions (19 of whom had anaphylaxis) were given premedication with antihistamines and given the second dose and tolerated it with either no reaction or, at most, a mild reaction that was self-limited [52]. Skin testing to PEG was performed in 80 patients that were considered to have the highest risk, none of whom were positive. Given this study, subsequent doses of vaccination should not be avoided; it may be safely given with premedication and desensitization may not be required as the mechanism of reaction is still not known.
Immune Checkpoint Inhibitors
Newer monoclonal antibody therapies are available, used in the management of multiple types of cancer, collectively called immune checkpoint inhibitors (CPIs). They directly target T cell ligands, preventing T cell silencing. These include monoclonal antibodies against programmed cell death protein 1 (PD-1), its ligand (PD-L1), cutaneous T lymphocyte antigen 4 (CTLA-4 a.k.a. CD152), and lymphocyte activation gene 3 (LAG-3) [53]. Between 50 and 75% of all patients treated with CPIs experience an adverse event [54]. Hypersensitivity is mostly reported as skin reactions, ranging from drug rashes to severe cutaneous adverse reactions (e.g., toxic epidermal necrolysis or Stevens-Johnson syndrome); the degree of type I hypersensitivity is not known. However, multiple case studies have reported success with rapid desensitization to CPIs [55].
Excipients in Biologics
Polysorbate (PS-20 and PS-80) is a surfactant that is commonly used as an FDA-approved excipient in many monoclonal antibody drugs (as well as being the active ingredient in many lubricant eye drops). High-dose systemic administration of polysorbates leads to hypotension and tachycardia [56]. In studies both in animals and in humans, blood was incubated with polysorbate demonstrating in vitro generation of anaphylatoxins C3a and C5a, suggesting that the immunogenicity of these excipients is complement-driven [57, 58]. In one case study where two patients were treated with omalizumab, one patient experienced ocular angioedema from lubricant eye drops; the other patient had a positive intradermal test to polysorbate [59]. In another study, a patient had anaphylaxis after their first injection of omalizumab, and when skin testing was performed to other monoclonal antibodies, reaction to certolizumab was positive (the only common excipient being PEG/macrogol in certolizumab, which is cross-reactive with polysorbate) [60]. In another patient being treated for psoriasis, adalimumab caused generalized wheals and pruritus and switching to ustekinumab caused severe urticaria [61]. In this patient, skin prick test to polysorbate was positive; the only common excipient between the two monoclonal antibodies was polysorbate. Another patient on infliximab therapy experienced breathing difficulty and facial flushing on their 4th infliximab injection, switched then experienced urticaria on their 9th adalimumab injection, and even switched to ustekinumab only to experience urticaria and breathing difficulties on their 3rd injection [62]. Skin prick test was positive to the only common excipient among all three medications: polysorbate. When evaluating patients who have experienced hypersensitivity reactions to monoclonal antibodies, skin prick testing to polysorbate may yield useful in identifying type I hypersensitivity and risk stratification for rapid desensitization.
Nanobodies
Nanobodies have several possible clinical applications, ranging from their use in molecular imaging (nuclear and optical imaging) and therapy (drug delivery systems, agonist, and antagonist capabilities) [63]. To date, there are no known hypersensitivity reactions to nanobodies, and the expected potential immunogenicity is extremely low as there is an absence of any significant activation of dendritic cells [64].
Hormones
Diabetes mellitus is a disorder of insulin: caused either by the lack of production of enough insulin or peripheral tissue resistance to insulin. Essential to the overall management of diabetes is administration of exogenous insulin. The incidence of hypersensitivity reactions to insulin has vastly decreased over the past 100 years as it was purified from animal origin to what we now use as human insulin, the current incidence being < 1 to 2.4% [65]. The most common excipient that contributes to insulin allergy is metacresol, a preservative used in every commercial preparation of insulin [66]. Other possible excipients that could cause reactions include protamine, zinc, or even glycerol [67]. Those who experience immediate local hypersensitivity to multiple preparations of subcutaneous insulin commonly experience the same such reactions when exposed to an insulin “saline” training pen that includes all the excipients without the insulin itself [68]. Rapid desensitization to insulin with premedication and administration of progressively increasing doses of the desired type of insulin has been successfully performed [67, 69]. Even a patient who presented with diabetic ketoacidosis requiring urgent hemodialysis to correct severe metabolic acidosis experienced a successful rapid desensitization to insulin, transitioning from intravenous insulin to subcutaneous insulin over a course of days [70].
The mainstay of treatment for hypothyroidism is replacement thyroid hormone in the form of levothyroxine. With its predictable pharmacodynamics, it is the drug of choice over liothyronine or desiccated thyroid extracts [71]. For those who experience hypersensitivity reaction to levothyroxine, since the first-published oral rapid desensitization in 2013 [72], it has been repeatedly replicated successfully by others [73, 74].
Progesterone hypersensitivity has also been reported when used for in vitro fertilization (IVF) when women are exposed to high doses of progesterone [75]. Symptoms can extend from a rash to anaphylaxis. In this case series of 6 patients who reacted to progesterone for IVF, they all underwent rapid drug desensitization with vaginal suppositories to achieve successful tolerance of the high doses required for IVF.
Desensitization Using Biologics
When a patient’s primary drug hypersensitivity reaction presents as having a high probability of being IgE-mediated (i.e., immediate reaction phenotype, positive skin testing, measured serum anti-drug IgE), premedication that directly controls IgE levels before rapid desensitization may be useful. First, there is the biologic omalizumab, a humanized IgG1 monoclonal antibody against free IgE at the site of the FcεRI receptor, preventing IgE crosslinking and mast cell degranulation [76]. It is currently FDA approved for both the management of poorly controlled severe allergic asthma, chronic idiopathic urticaria, and nasal polyps [77]. It has also been successfully used as an adjunctive treatment in patients with mastocytosis to dampen the effects of venom immunotherapy [76, 78–80]. Research into its utility as premedication in aspirin desensitization has yielded conflicting results. In a series of 275 patients who underwent successful standardized rapid desensitization to aspirin, 8 of 9 patients who were on omalizumab during desensitization reacted, suggesting that omalizumab does not prevent reactions during desensitization specifically to aspirin [81]. In another study, 5 of 7 patients on omalizumab undergoing rapid desensitization to aspirin did not experience respiratory symptoms during the desensitization; the other two experienced aspirin-provoked bronchospasm [82]. In a case report of a patient with insulin allergy who failed traditional rapid desensitization in the past, a combination of biologics was given: first rituximab to reduce IgE production from active plasma cells, then omalizumab to prevent IgE crosslinking [83]. The patient remained asymptomatic while also receiving mycophenolate and only 2 mg of prednisone daily. In a patient receiving chemotherapy with FOLFIRINOX (folinic acid, 5-fluorouracil, irinotecan, and oxaliplatin) who had an episode of anaphylaxis during third desensitization, there were no further episodes of anaphylaxis during 13 additional desensitizations after starting the patient on omalizumab [84]. Omalizumab has even been used successfully for desensitization to rituximab in children [85].
Ligelizumab is a monoclonal antibody against IgE with a manifold greater affinity for IgE in vitro and greater potency in vivo than omalizumab that is currently in phase III clinical trials [86–92]. Their latest published information, however, does not show superiority in efficacy for the indication of chronic idiopathic urticaria when compared to omalizumab, even though the studies met their clinical endpoints. Quilizumab is a humanized IgG1 monoclonal antibody that binds membrane IgE, not soluble IgE, and phase II trials performed thus far do not show significant clinical improvements in chronic idiopathic urticaria [93, 94]. Lastly, MEDI4212 is a monoclonal antibody against IgE that just published its phase I study, showing increased rapid decrease of total IgE but no sustained decrease of total IgE, suggesting limited potential over omalizumab [95]. Further investigation is necessary to determine if any of these drugs can be used in premedication to control reactions during rapid desensitization.
Conclusion
Biologics, whether chimeric, humanized, and human antibodies, targeted mRNA, or nanobodies have opened a new era in medicine for the prevention and treatment of infectious diseases, cancer, and inflammatory diseases. While their safety has been largely proven, hypersensitivity reactions are rare, but remain the most important factor preventing the use of first-line therapies for reactive and allergic patients. Understanding the underlying mechanisms of hypersensitivity reactions requires the wide utilization of biomarkers such as skin testing, currently cost limited, and others such as tryptase and now IL-6, which permit risk stratification of patients and their qualification for desensitization. Desensitization is primarily limited to type I, including anaphylaxis, and type IV reactions and is a highly effective and safe procedure for re-exposure to biologics in severely allergic patients. While desensitization requires a multidisciplinary approach between pharmacy, nursing, specialists, and allergists, its implementation should be the standard of care to provide all patients their best medication.
Author Contribution
All authors whose names appear on this submission made substantial contributions to the conception and design of the work, drafted the work, revised it critically for important intellectual content, approve this version to be published, and agree to be accountable for all aspects of the work.
Data Availability
Data sharing is not applicable to this article as no new data were created or analyzed in this article.
Declarations
Conflict of Interest
Dr. Castells is the Anaphylaxis and Drug Allergy Section Editor for the Journal. Dr. Yang is currently employed at Ribon Therapeutics. The authors did not receive support from any organization for the submitted work. The authors have no financial or proprietary interests in any material discussed in this article.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
This article is part of the Topical Collection on Anaphylaxis and Drug Allergy
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
1. Morrow T Felcone LH Defining the difference: what makes biologics unique Biotechnol Healthc 2004 1 4 24 29 23393437
2. Balocco R, De Sousa Guimaraes Koch S, Thorpe R, Weisser K, Malan S. New INN nomenclature for monoclonal antibodies. Lancet. 2022;399(10319):24. 10.1016/S0140-6736(21)02732-X.
3. WHO. New INN nomenclature scheme for monoclonal antibodies. Geneva: World Heal Org. 2022.
4. WHO. Nomenclature for monoclonal antibodies. Geneva: World Heal Org. 2009.
5. Awad RM Meeus F Ceuppens H Ertveldt T Hanssens H Lecocq Q Emerging applications of nanobodies in cancer therapy Int Rev Cell Mol Biol 2022 369 143 199 10.1016/bs.ircmb.2022.03.010 35777863
6. Xu S, Yang K, Li R, Zhang L. mRNA Vaccine era-mechanisms, drug platform and clinical prospection. Int J Mol Sci. 2020;21(18). 10.3390/ijms21186582.
7. Polack FP Thomas SJ Kitchin N Absalon J Gurtman A Lockhart S Safety and Efficacy of the BNT162b2 mRNA Covid-19 vaccine N Engl J Med 2020 383 27 2603 2615 10.1056/NEJMoa2034577 33301246
8. Baden LR El Sahly HM Essink B Kotloff K Frey S Novak R Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine N Engl J Med 2021 384 5 403 416 10.1056/NEJMoa2035389 33378609
9. Maggi E Vultaggio A Matucci A Acute infusion reactions induced by monoclonal antibody therapy Expert Rev Clin Immunol 2011 7 1 55 63 10.1586/eci.10.90 21162650
10. Caiado J Castells MC Drug desensitizations for chemotherapy: safety and efficacy in preventing anaphylaxis Curr Allergy Asthma Rep 2021 21 6 37 10.1007/s11882-021-01014-x 34232411
11. •• Yang BC, Castells MC. The who, what, where, when, why, and how of drug desensitization. immunology and allergy clinics of North America. 2022. 10.1016/j.iac.2021.12.004. Thorough guide on the pathophysiologic mechanisms of rapid drug desensitization and how to perform desensitization.
12. Yang BC Castells MC Rituximab hypersensitivity and desensitization: a personalized approach to treat cancer and connective tissue diseases Ann Allergy Asthma Immunol 2019 123 1 11 15 10.1016/j.anai.2019.03.008 30885801
13. Yang BC Castells M Diagnosis and treatment of drug hypersensitivity reactions to biologicals: medical algorithm Allergy 2020 10.1111/all.14432 33319354
14. Brown SG Clinical features and severity grading of anaphylaxis J Allergy Clin Immunol 2004 114 2 371 376 10.1016/j.jaci.2004.04.029 15316518
15. Picard M, Pur L, Caiado J, Giavina-Bianchi P, Galvao VR, Berlin ST, et al. Risk stratification and skin testing to guide re-exposure in taxane-induced hypersensitivity reactions. J Allergy Clin Immunol. 2016;137(4):1154–64 e12. 10.1016/j.jaci.2015.10.039.
16. Madrigal-Burgaleta R Bernal-Rubio L Berges-Gimeno MP Carpio-Escalona LV Gehlhaar P Alvarez-Cuesta E A large single-hospital experience using drug provocation testing and rapid drug desensitization in hypersensitivity to antineoplastic and biological agents J Allergy Clin Immunol Pract 2019 7 2 618 632 10.1016/j.jaip.2018.07.031 30098410
17. Madrigal-Burgaleta R Vazquez-Revuelta P Marti-Garrido J Lleonart R Ali FR Alvarez-Cuesta E Importance of diagnostics prior to desensitization in new drug hypersensitivity: chemotherapeutics and biologicals Current Treatment Options in Allergy 2020 7 1 1 13 10.1007/s40521-020-00238-y
18. Castells M Drug hypersensitivity and anaphylaxis in cancer and chronic inflammatory diseases: the role of desensitizations Front Immunol 2017 8 1472 10.3389/fimmu.2017.01472 29163536
19. Castells MC Tennant NM Sloane DE Hsu FI Barrett NA Hong DI Hypersensitivity reactions to chemotherapy: outcomes and safety of rapid desensitization in 413 cases J Allergy Clin Immunol 2008 122 3 574 580 10.1016/j.jaci.2008.02.044 18502492
20. Hsu Blatman KS Castells MC Desensitizations for chemotherapy and monoclonal antibodies: indications and outcomes Curr Allergy Asthma Rep 2014 14 8 453 10.1007/s11882-014-0453-5 24994467
21. Hong DI, Dioun AF. Indications, protocols, and outcomes of drug desensitizations for chemotherapy and monoclonal antibodies in adults and children. J Allergy Clin Immunol Pract. 2014;2(1):13–9; quiz 20. 10.1016/j.jaip.2013.11.007.
22. Brennan PJ Rodriguez Bouza T Hsu FI Sloane DE Castells MC Hypersensitivity reactions to mAbs: 105 desensitizations in 23 patients, from evaluation to treatment J Allergy Clin Immunol 2009 124 6 1259 1266 10.1016/j.jaci.2009.09.009 19910036
23. Castells M Diagnosis and management of anaphylaxis in precision medicine J Allergy Clin Immunol 2017 140 2 321 333 10.1016/j.jaci.2017.06.012 28780940
24. Picard M Galvao VR Current knowledge and management of hypersensitivity reactions to monoclonal antibodies J Allergy Clin Immunol Pract 2017 5 3 600 609 10.1016/j.jaip.2016.12.001 28110056
25. Isabwe GAC, Garcia Neuer M, de Las Vecillas Sanchez L, Lynch DM, Marquis K, Castells M. Hypersensitivity reactions to therapeutic monoclonal antibodies: phenotypes and endotypes. J Allergy Clin Immunol. 2018;142(1):159–70 e2. 10.1016/j.jaci.2018.02.018.
26. Vitte J Sabato V Tacquard C Garvey LH Michel M Mertes PM Use and interpretation of acute and baseline tryptase in perioperative hypersensitivity and anaphylaxis J Allergy Clin Immunol Pract 2021 9 8 2994 3005 10.1016/j.jaip.2021.03.011 33746087
27. Lyons JJ Yu X Hughes JD Le QT Jamil A Bai Y Elevated basal serum tryptase identifies a multisystem disorder associated with increased TPSAB1 copy number Nat Genet 2016 48 12 1564 1569 10.1038/ng.3696 27749843
28. Giannetti MP Weller E Bormans C Novak P Hamilton MJ Castells M Hereditary alpha-tryptasemia in 101 patients with mast cell activation-related symptomatology including anaphylaxis Ann Allergy Asthma Immunol 2021 126 6 655 660 10.1016/j.anai.2021.01.016 33465452
29. Chung CH Mirakhur B Chan E Le QT Berlin J Morse M Cetuximab-induced anaphylaxis and IgE specific for galactose-alpha-1,3-galactose N Engl J Med 2008 358 11 1109 1117 10.1056/NEJMoa074943 18337601
30. Chinuki Y Morita E Alpha-Gal-containing biologics and anaphylaxis Allergol Int 2019 68 3 296 300 10.1016/j.alit.2019.04.001 31053502
31. Campos L Galvao VR Kalil J Castells M Giavina-Bianchi P BAT in the diagnosis of drug allergy: a novel tool in clinical daily practice? Curr Allergy Asthma Rep 2019 19 4 20 10.1007/s11882-019-0852-8 30859323
32. Piva E Chieco-Bianchi F Krajcar V Aversa S Plebani M Adverse reactions in patients with B-cell lymphomas during combined treatment with rituximab: in vitro evaluation of rituximab hypersensitivity by basophil activation test Am J Hematol 2012 87 11 E130 E131 10.1002/ajh.23329 22987289
33. Thevenot J, Ferrier le Bouedec MC, Buisson A, Bommelaer G, D'Incan M, Rouzaire P. Rapid desensitization to adalimumab is associated with decreased basophil sensitivity. J Investig Allergol Clin Immunol. 2019;29(2):141–3. 10.18176/jiaci.0350.
34. de la Varga MR Gutierrez Fernandez D Foncubierta Fernandez A Andres Garcia JA Medina VF Rapid subcutaneous desensitization for treatment of hypersensitivity reactions to etanercept in two patients with positive basophil activation test Allergol Int 2017 66 2 357 359 10.1016/j.alit.2016.09.002 27780662
35. Gonzalez-de-Olano D Morgado JM Juarez-Guerrero R Sanchez-Munoz L Letellez-Fernandez J Malon-Gimenez D Positive basophil activation test following anaphylaxis to pertuzumab and successful treatment with rapid desensitization J Allergy Clin Immunol Pract 2016 4 2 338 340 10.1016/j.jaip.2015.10.007 26725157
36. Pichler WJ Tilch J The lymphocyte transformation test in the diagnosis of drug hypersensitivity Allergy 2004 59 8 809 820 10.1111/j.1398-9995.2004.00547.x 15230812
37. Giraldo-Tugores M Fernandez-Lozano C Carron-Herrero A Gajate P Martinez-Botas J Pueyo-Lopez C Successful rapid desensitization to atezolizumab in delayed hypersensitivity confirmed with lymphocyte transformation test J Allergy Clin Immunol Pract 2022 10.1016/j.jaip.2021.12.041 35074597
38. Jakubovic BD Sanchez-Sanchez S Hamadi S Lynch DM Castells M Interleukin-6: a novel biomarker for monoclonal antibody and chemotherapy-associated hypersensitivity confirms a cytokine release syndrome phenotype-endotype association Allergy 2020 10.1111/all.14644 33119137
39. Shimabukuro TT Nguyen M Martin D DeStefano F Safety monitoring in the Vaccine Adverse Event Reporting System (VAERS) Vaccine 2015 33 36 4398 4405 10.1016/j.vaccine.2015.07.035 26209838
40. Blumenthal KG Robinson LB Camargo CA Jr Shenoy ES Banerji A Landman AB Acute allergic reactions to mRNA COVID-19 vaccines JAMA 2021 325 15 1562 1565 10.1001/jama.2021.3976 33683290
41. Tanno LK Castells M Caminati M Senna G Demoly P Anaphylaxis and coronavirus disease 2019 vaccine: a danger relationship? Curr Opin Allergy Clin Immunol 2021 21 5 411 417 10.1097/ACI.0000000000000778 34334663
42. Dreskin SC Halsey NA Kelso JM Wood RA Hummell DS Edwards KM International Consensus (ICON): allergic reactions to vaccines World Allergy Organ J 2016 9 1 32 10.1186/s40413-016-0120-5 27679682
43. Castells MC Phillips EJ Maintaining safety with SARS-CoV-2 vaccines N Engl J Med 2021 384 7 643 649 10.1056/NEJMra2035343 33378605
44. Castells M, Demoly P, Tanno LK. [Anaphylaxis and COVID-19 vaccines]. Rev Fr Allergol (2009). 2021;61(8):8S30–8S5. 10.1016/S1877-0320(21)00439-5.
45. • Risma KA, Edwards KM, Hummell DS, Little FF, Norton AE, Stallings A, et al. Potential mechanisms of anaphylaxis to COVID-19 mRNA vaccines. J Allergy Clin Immunol. 2021;147(6):2075–82 e2. 10.1016/j.jaci.2021.04.002. Good review of theoretical mechanisms of COVID-19 vaccine hypersensitivity.
46. Duguay BA Huang KW Kulka M Lipofection of plasmid DNA into human mast cell lines using lipid nanoparticles generated by microfluidic mixing J Leukoc Biol 2018 104 3 587 596 10.1002/JLB.3TA0517-192R 29668121
47. McNeil BD Pundir P Meeker S Han L Undem BJ Kulka M Identification of a mast-cell-specific receptor crucial for pseudo-allergic drug reactions Nature 2015 519 7542 237 241 10.1038/nature14022 25517090
48. Yang Q Lai SK Anti-PEG immunity: emergence, characteristics, and unaddressed questions Wiley Interdiscip Rev Nanomed Nanobiotechnol 2015 7 5 655 677 10.1002/wnan.1339 25707913
49. Interim Clinical Considerations for Use of COVID-19 Vaccines Currently Approved or Authorized in the United States. 2020. https://www.cdc.gov/vaccines/covid-19/clinical-considerations/interim-considerations-us.html . Accessed July 20, 2022.
50. Greenhawt M Shaker M Golden DBK PEG/polysorbate skin testing has no utility in the assessment of suspected allergic reactions to SARS-CoV-2 vaccines J Allergy Clin Immunol Pract 2021 9 9 3321 3322 10.1016/j.jaip.2021.06.025 34393075
51. AlMuhizi F Ton-Leclerc S Fein M Tsoukas C Garvey LH Lee D Successful desensitization to mRNA COVID-19 vaccine in a case series of patients with a history of anaphylaxis to the first vaccine dose Front Allergy 2022 3 825164 10.3389/falgy.2022.825164 35386647
52. Krantz MS Kwah JH Stone CA Jr Phillips EJ Ortega G Banerji A Safety evaluation of the second dose of messenger RNA COVID-19 vaccines in patients with immediate reactions to the first dose JAMA Intern Med 2021 181 11 1530 1533 10.1001/jamainternmed.2021.3779 34309623
53. Seidel JA Otsuka A Kabashima K Anti-PD-1 and anti-CTLA-4 therapies in cancer: mechanisms of action, efficacy, and limitations Front Oncol 2018 8 86 10.3389/fonc.2018.00086 29644214
54. Ibraheim H, Perucha E, Powell N. Pathology of immune-mediated tissue lesions following treatment with immune checkpoint inhibitors. Rheumatology (Oxford). 2019;58(Suppl 7):vii17-vii28. 10.1093/rheumatology/kez465.
55. Labella M Castells M Hypersensitivity reactions and anaphylaxis to checkpoint inhibitor-monoclonal antibodies and desensitization Ann Allergy Asthma Immunol 2021 126 6 623 629 10.1016/j.anai.2021.03.008 33781937
56. Singh SK Mahler HC Hartman C Stark CA Are injection site reactions in monoclonal antibody therapies caused by polysorbate excipient degradants? J Pharm Sci 2018 107 11 2735 2741 10.1016/j.xphs.2018.07.016 30055223
57. Weiszhar Z Czucz J Revesz C Rosivall L Szebeni J Rozsnyay Z Complement activation by polyethoxylated pharmaceutical surfactants: Cremophor-EL, Tween-80 and Tween-20 Eur J Pharm Sci 2012 45 4 492 498 10.1016/j.ejps.2011.09.016 21963457
58. Qiu S Liu Z Hou L Li Y Wang J Wang H Complement activation associated with polysorbate 80 in beagle dogs Int Immunopharmacol 2013 15 1 144 149 10.1016/j.intimp.2012.10.021 23159336
59. Price KS Hamilton RG Anaphylactoid reactions in two patients after omalizumab administration after successful long-term therapy Allergy Asthma Proc 2007 28 3 313 319 10.2500/aap.2007.28.3003 17619560
60. Perino E Freymond N Devouassoux G Nicolas JF Berard F Xolair-induced recurrent anaphylaxis through sensitization to the excipient polysorbate Ann Allergy Asthma Immunol 2018 120 6 664 666 10.1016/j.anai.2018.02.018 29481891
61. Perez-Perez L Garcia-Gavin J Pineiro B Zulaica A Biologic-induced urticaria due to polysorbate 80: usefulness of prick test Br J Dermatol 2011 164 5 1119 1120 10.1111/j.1365-2133.2011.10220.x 21219296
62. Kato M Oiso N Uchida S Yanagihara S Sano H Tohda Y Biologic-induced urticaria due to polysorbate 20 J Dermatol 2019 46 7 e230 e232 10.1111/1346-8138.14808 30758866
63. Oliveira S, Heukers R, Sornkom J, Kok RJ, van Bergen En Henegouwen PM. Targeting tumors with nanobodies for cancer imaging and therapy. J Control Release. 2013;172(3):607–17. 10.1016/j.jconrel.2013.08.298.
64. Ackaert C Smiejkowska N Xavier C Sterckx YGJ Denies S Stijlemans B Immunogenicity risk profile of nanobodies Front Immunol 2021 12 632687 10.3389/fimmu.2021.632687 33767701
65. Schernthaner G Immunogenicity and allergenic potential of animal and human insulins Diabetes Care 1993 16 Suppl 3 155 165 10.2337/diacare.16.3.155 8299472
66. Kim D Baraniuk J Delayed-type hypersensitivity reaction to the meta-cresol component of insulin Ann Allergy Asthma Immunol 2007 99 2 194 195 10.1016/S1081-1206(10)60645-X 17718109
67. Elfekih H Hadjkacem F Elleuch M Ghorbel D Charfi N Mnif F Successful treatment of insulin allergy with desensitization therapy: a case report and literature review Iran J Allergy Asthma Immunol 2019 18 5 572 583 10.18502/ijaai.v18i5.1927 32245301
68. Heinzerling L Raile K Rochlitz H Zuberbier T Worm M Insulin allergy: clinical manifestations and management strategies Allergy 2008 63 2 148 155 10.1111/j.1398-9995.2007.01567.x 18186805
69. Wheeler BJ Taylor BJ Successful management of allergy to the insulin excipient metacresol in a child with type 1 diabetes: a case report J Med Case Rep 2012 6 263 10.1186/1752-1947-6-263 22937994
70. Shuster S Borici-Mazi R Awad S Houlden RL Rapid desensitization with intravenous insulin in a patient with diabetic ketoacidosis and insulin allergy AACE Clin Case Rep 2020 6 4 e147 e150 10.4158/ACCR-2019-0562 32671215
71. Jonklaas J Bianco AC Bauer AJ Burman KD Cappola AR Celi FS Guidelines for the treatment of hypothyroidism: prepared by the American Thyroid Association Task Force on thyroid hormone replacement Thyroid 2014 24 12 1670 1751 10.1089/thy.2014.0028 25266247
72. Fevzi D Mustafa G Ozgur K Cetin T Abdullah B Sait Y Successful oral desensitization to levothyroxine Ann Allergy Asthma Immunol 2013 111 2 146 147 10.1016/j.anai.2013.06.003 23886239
73. Bahloul N Ben Mahmoud L Ghozzi H Hadjkacem F Zeghal K Abid M Hypersensitivity to levothyroxine: a case report of a successful oral desensitization Therapie 2018 73 4 349 350 10.1016/j.therap.2017.10.009 29373130
74. Jiang CY Shen BS Yun TZ Singarayar C Hui FS Successful oral levothyroxine desensitization in a patient with severe hypothyroidism post radioactive iodine therapy: a case report J ASEAN Fed Endocr Soc 2021 36 2 213 215 10.15605/jafes.036.02.10 34966208
75. Prieto-Garcia A, Sloane DE, Gargiulo AR, Feldweg AM, Castells M. Autoimmune progesterone dermatitis: clinical presentation and management with progesterone desensitization for successful in vitro fertilization. Fertil Steril. 2011;95(3):1121 e9–13. 10.1016/j.fertnstert.2010.10.038.
76. Carter MC Metcalfe DD Matito A Escribano L Butterfield JH Schwartz LB Adverse reactions to drugs and biologics in patients with clonal mast cell disorders: a work group report of the Mast Cells Disorder Committee, American Academy of Allergy, Asthma & Immunology J Allergy Clin Immunol 2019 143 3 880 893 10.1016/j.jaci.2018.10.063 30528617
77. Novartis/Genentech: Xolair Prescribing Information (Package Insert). https://www.gene.com/download/pdf/xolair_prescribing.pdf . Accessed.
78. Kontou-Fili K High omalizumab dose controls recurrent reactions to venom immunotherapy in indolent systemic mastocytosis Allergy 2008 63 3 376 378 10.1111/j.1398-9995.2007.01604.x 18269681
79. Kontou-Fili K Filis CI Prolonged high-dose omalizumab is required to control reactions to venom immunotherapy in mastocytosis Allergy 2009 64 9 1384 1385 10.1111/j.1398-9995.2009.02045.x 19432937
80. Kontou-Fili K Filis CI Voulgari C Panayiotidis PG Omalizumab monotherapy for bee sting and unprovoked "anaphylaxis" in a patient with systemic mastocytosis and undetectable specific IgE Ann Allergy Asthma Immunol 2010 104 6 537 539 10.1016/j.anai.2010.04.011 20568389
81. Waldram J Walters K Simon R Woessner K Waalen J White A Safety and outcomes of aspirin desensitization for aspirin-exacerbated respiratory disease: a single-center study J Allergy Clin Immunol 2018 141 1 250 256 10.1016/j.jaci.2017.05.006 28550988
82. Lang DM Aronica MA Maierson ES Wang XF Vasas DC Hazen SL Omalizumab can inhibit respiratory reaction during aspirin desensitization Ann Allergy Asthma Immunol 2018 121 1 98 104 10.1016/j.anai.2018.05.007 29777744
83. Yong PF Malik R Arif S Peakman M Amiel S Ibrahim MA Rituximab and omalizumab in severe, refractory insulin allergy N Engl J Med 2009 360 10 1045 1047 10.1056/NEJMc0808282 19264698
84. Prieto-Garcia A Noguerado B Rojas P Torrado I Rodriguez-Fernandez A Tornero P Unexpected anaphylaxis after completing a desensitization protocol to oxaliplatin: successful adjuvant use of omalizumab J Investig Allergol Clin Immunol 2019 29 1 53 55 10.18176/jiaci.0326 30785102
85. Gemici Karaaslan HB Karabag Yilmaz E Gulmez R Canpolat N Kiykim A Cokugras HC Omalizumab may facilitate drug desensitization in patients failing standard protocols Pediatr Allergy Immunol 2022 33 5 e13783 10.1111/pai.13783 35616895
86. Arm JP Bottoli I Skerjanec A Floch D Groenewegen A Maahs S Pharmacokinetics, pharmacodynamics and safety of QGE031 (ligelizumab), a novel high-affinity anti-IgE antibody, in atopic subjects Clin Exp Allergy 2014 44 11 1371 1385 10.1111/cea.12400 25200415
87. Gauvreau GM Arm JP Boulet LP Leigh R Cockcroft DW Davis BE Efficacy and safety of multiple doses of QGE031 (ligelizumab) versus omalizumab and placebo in inhibiting allergen-induced early asthmatic responses J Allergy Clin Immunol 2016 138 4 1051 1059 10.1016/j.jaci.2016.02.027 27185571
88. Maurer M Gimenez-Arnau AM Sussman G Metz M Baker DR Bauer A Ligelizumab for Chronic Spontaneous Urticaria N Engl J Med 2019 381 14 1321 1332 10.1056/NEJMoa1900408 31577874
89. Gasser P Tarchevskaya SS Guntern P Brigger D Ruppli R Zbaren N The mechanistic and functional profile of the therapeutic anti-IgE antibody ligelizumab differs from omalizumab Nat Commun 2020 11 1 165 10.1038/s41467-019-13815-w 31913280
90. Jensen RK Jabs F Miehe M Molgaard B Pfutzner W Mobs C Structure of intact IgE and the mechanism of ligelizumab revealed by electron microscopy Allergy 2020 75 8 1956 1965 10.1111/all.14222 32037590
91. Maurer M Gimenez-Arnau A Bernstein JA Chu CY Danilycheva I Hide M Sustained safety and efficacy of ligelizumab in patients with chronic spontaneous urticaria: A one-year extension study Allergy 2021 10.1111/all.15175 34862605
92. Turk M Yilmaz I High-Dose Omalizumab versus Ligelizumab for the Treatment of Chronic Spontaneous Urticaria: Do Not We Need a Head-To-Head Comparison? Int Arch Allergy Immunol 2021 182 5 461 462 10.1159/000512373 33321510
93. Harris JM Cabanski CR Scheerens H Samineni D Bradley MS Cochran C A randomized trial of quilizumab in adults with refractory chronic spontaneous urticaria J Allergy Clin Immunol 2016 138 6 1730 1732 10.1016/j.jaci.2016.06.023 27567329
94. Harris JM Maciuca R Bradley MS Cabanski CR Scheerens H Lim J A randomized trial of the efficacy and safety of quilizumab in adults with inadequately controlled allergic asthma Respir Res 2016 17 29 10.1186/s12931-016-0347-2 26993628
95. Sheldon E Schwickart M Li J Kim K Crouch S Parveen S Pharmacokinetics, pharmacodynamics, and safety of MEDI4212, an anti-IgE monoclonal antibody, in subjects with atopy: a phase I study Adv Ther 2016 33 2 225 251 10.1007/s12325-016-0287-8 26843086
| 36445652 | PMC9707161 | NO-CC CODE | 2022-12-01 23:21:38 | no | Curr Allergy Asthma Rep. 2022 Nov 29;:1-11 | utf-8 | Curr Allergy Asthma Rep | 2,022 | 10.1007/s11882-022-01052-z | oa_other |
==== Front
Appl Intell (Dordr)
Appl Intell (Dordr)
Applied Intelligence (Dordrecht, Netherlands)
0924-669X
1573-7497
Springer US New York
4288
10.1007/s10489-022-04288-4
Article
Semi-supervised adversarial discriminative domain adaptation
Nguyen Thai-Vu [email protected]
12
Nguyen Anh [email protected]
3
Le Nghia [email protected]
24
http://orcid.org/0000-0002-4306-6945
Le Bac [email protected]
12
1 grid.454160.2 0000 0004 0642 8526 Faculty of Information Technology, University of Science, Ho Chi Minh City, Vietnam
2 grid.444808.4 0000 0001 2037 434X Vietnam National University, Ho Chi Minh City, Vietnam
3 grid.10025.36 0000 0004 1936 8470 Department of Computer Science, University of Liverpool, London, UK
4 grid.444808.4 0000 0001 2037 434X University of Information Technology, Ho Chi Minh City, Vietnam
29 11 2022
114
20 10 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Domain adaptation is a potential method to train a powerful deep neural network across various datasets. More precisely, domain adaptation methods train the model on training data and test that model on a completely separate dataset. The adversarial-based adaptation method became popular among other domain adaptation methods. Relying on the idea of GAN, the adversarial-based domain adaptation tries to minimize the distribution between the training and testing dataset based on the adversarial learning process. We observe that the semi-supervised learning approach can combine with the adversarial-based method to solve the domain adaptation problem. In this paper, we propose an improved adversarial domain adaptation method called Semi-Supervised Adversarial Discriminative Domain Adaptation (SADDA), which can outperform other prior domain adaptation methods. We also show that SADDA has a wide range of applications and illustrate the promise of our method for image classification and sentiment classification problems.
Keywords
Domain adaptation
Semi-supervised domain adaptation
Semi-supervised adversarial discriminative domain adaptation
==== Body
pmcIntroduction
Over the past few years, deep neural networks have achieved significant achievements in many applications. One of the major limitations of deep neural networks is the dataset bias or domain shift problems [1]. These phenomena occur when the model obtains good results on the training dataset; however, showing poor performance on a testing dataset or a real-world sample.
As shown in Fig. 1, because of numerous reasons (illumination, image quality, background), there is always a different distribution between two datasets, which is the main factor reducing the performance of deep neural networks. Even though various research has proved that deep neural networks can learn transferable feature representation over different datasets [2, 3], Donahue et al. [4] showed that domain shift still influences the accuracy of the deep neural network when testing these networks in a different dataset. Fig. 1 Examples of images from different datasets. (a) Some digit images from MNIST [5], USPS [6], MNIST-M [7], and SVHN [8] datasets. (b) Some object images from the “bird” category in CALTECH [9], LABELME [10], PASCAL [11], and SUN [12] datasets
The solution for the aforementioned problems is domain adaptation techniques [13, 14]. The main idea of domain adaptation techniques is to learn how a deep neural network can map the source domain and target domain into a common feature space, which minimize the negative influence of domain shift or dataset bias.
The adversarial-based adaptation method [15, 16] has become a well-known technique among other domain adaptation methods. Adversarial adaptation includes two networks - an encoder and a discriminator, trained simultaneously with conflicting objectives. The encoder is trained to encode images from the original domain (source domain) and new domain (target domain) such that it puzzles the discriminator. In contrast, the discriminator tries to distinguish between the source and target domain. Recently, Adversarial Discriminative Domain Adaptation (ADDA) by Tzeng et al. [16] has shown that adversarial adaptation can handle dataset bias and domain shift problems. From there, we extend the ADDA method to the semi-supervised learning context by obliging the discriminator network to predict class labels.
Semi-supervised learning [17] is an approach that builds a predictive model with a small labeled dataset and a large unlabeled dataset. The model must learn from the small labeled dataset and somehow exploit the larger unlabeled dataset to classify new samples. In the context of unsupervised domain adaptation tasks, the semi-supervised learning approach needs to take advantage of the labeled source dataset to map to the unlabeled target dataset, thereby correctly classifying the labels of the target dataset. The Semi-Supervised GAN [18] is designed to handle the semi-supervised learning tasks and inspired us to develop our model.
In this paper, we present a novel method called Semi-supervised Adversarial Discriminative Domain Adaptation (SADDA), where the discriminator is a multi-class classifier. Instead of only distinguishing between source images and target images (method like ADDA [16]), the discriminator learns to distinguish N + 1 classes, where N is the number of classes in the classification task, and the last one uses to distinguish between the source dataset or the target dataset. The discriminator focuses not only on the domain label between two datasets but also on the labeled images from the source dataset, which improves the generalization ability of the discriminator and the encoder as well as the classification accuracy.
To validate the effectiveness of our methodology, we experiment with domain adaptation tasks on digit datasets, including MNIST [5], USPS [6], MNIST-M [7], and SVHN [8]. In addition, we also prove the robustness ability of the SADDA method by using t-SNE visualization of the digit datasets, the SADDA method keeps the t-SNE clusters as tight as possible and maximizes the separation between two clusters. We also test its potential with a more sophisticated dataset, by object recognition task with CALTECH [9], LABELME [10], PASCAL [11], and SUN [12] datasets. In addition, we evaluate our method for the natural language processing task, with three text datasets including Women’s E-Commerce Clothing Reviews [19], Coronavirus tweets NLP - Text Classification [20], and Trip Advisor Hotel Reviews [21]. The Python code of the SADDA method for object recognition tasks can be downloaded at https://github.com/NguyenThaiVu/SADDA.
Our contributions can be summarized as follows: We propose a new Semi-supervised Adversarial Discriminative Domain Adaptation method (SADDA) for addressing the unsupervised domain adaptation task.
We illustrate that SADDA improves digit classification tasks and achieves competitive performance with other adversarial adaptation methods.
We also demonstrate that the SADDA method can apply to multiple applications, including object recognition and natural language processing tasks.
Related work
Domain adaptation
is an active research field, which can handle numerous problems such as imbalanced data [22], dataset bias [23], and domain shift [24]. Recent research has focused on domain adaptation from a labeled source dataset to an unlabeled target dataset, also known as unsupervised domain adaptation [25, 26]. The principle technique is minimizing the distinction between the source and target distribution [27]. Some popular approaches are Maximum Mean Discrepancy (MMD) [1], deep reconstruction classification network (DRCN) [28] or Autoencoder-based domain adaptation [29].
Adversarial-based domain adaptation
With the rise of generative adversarial networks [15], the adversarial-based made huge advancements in the domain adaptation task [30–32]. Adversarial-based techniques try to achieve domain adaptation by using domain discriminators, which increases domain confusion through an adversarial process. A popular adversarial-based domain adaptation method is the Adversarial Discriminative Domain Adaptation (ADDA) by Tzeng et al. [16]. ADDA approach aims to diminish the distance between the source encoder and target encoder distributions through the domain-adversarial process. However, this method only distinguishes between the source and target domain. Instead, our SADDA method not only predicts whether the source domain or the target domain, but also classifies the label of the source dataset. More concretely, we force the adversarial-based method to the semi-supervised context. We will show that this creation can produce a more efficient classification model.
Combining adversarial-based domain adaptation with other auxiliary tasks
Recently, some works have focused on combining auxiliary tasks for adversarial-based adaptation to exploit more information [33, 34]. Xavier and Bengio introduce Stacked Denoising Autoencoders [35, 36], reconstructing the merging data from numerous domains with the same network, such that the representations can be symbolized by both the source and target domain. Deep reconstruction classification network (DRCN) [28] attempts to solve two sub-problem at the same time: classification of the source data, and reconstruction of the unlabeled target data. However, these auxiliary tasks are not towards the same goal. We observe that during the adversarial process, we can classify the source or target dataset and predict the label of the source dataset simultaneously. That allows us to re-use the same output layers in the discriminator model as well as forces two discriminator models towards the same goal (Section 3.2 for more details). In addition, we also demonstrate that our SADDA method not only applies to computer vision tasks but also the natural language processing task.
Proposed method
Semi-supervised adversarial discriminative domain adaptation
In this section, we describe in detail our Semi-supervised Adversarial Discriminative Domain Adaptation (SADDA) method. An overview of our method can be found in Fig. 2. Fig. 2 An overview of the SADDA. Firstly, training the source encoder (Ms) and the classification (Cs) using the source labeled images (Xs, Ys). Secondly, training a target encoder (Mt) through the domain adversarial process. Finally, in the testing phase, concatenate the target encoder (Mt) and the classification (Cs) to create the complete model, which will predict the label of the target dataset precisely
In the unsupervised domain adaptation task, we already have source images Xs and source labels Ys come from the source domain distribution ps(x,y). Besides that, a target dataset Xt comes from a target distribution pt(x,y), where the label of the target dataset is non-exist. We desire to learn a target encoder Mt and classifier Ct, which can accurately predict the target image’s label. In an adversarial-based adaptation approach, we aim to diminish the distance between the source mapping distribution (Ms(Xs)) and target mapping distributions (Mt(Xt)). As a result, we can straightly apply the source classifier Cs to classify the target images, in other words, C = Ct = Cs. The summary process of SADDA includes three steps: pre-training, training target encoder, and testing.
Pre-training
In the pre-training phase, training source encoder (Ms) and source classifier (Cs), by using the source labeled images (Xs,Ys). This step is a standard supervised classification task, a common form can be denoted as:
1 argminMs,CsLcls(Xs,Ys)=−E(x,y)∼(Xs,Ys)∑n=1NynlogCs(Ms(xn))
where Lcls is a supervised classification loss (categorical crossentropy loss), and N is the number of classes.
Training target encoder
In the training target encoder phase, we first present a training discriminator process and then present a procedure for training the target encoder.
Firstly, training the discriminator (D) in two modes, each giving a corresponding output. (1) Supervised mode, where the supervised discriminator (Dsup) predicts N labels from the original classification task. (2) Unsupervised mode, where the unsupervised discriminator (Dunsup) classifies between Xs and Xt. Discriminator correlates with unconstrained optimization:
2 argminDsupLcls(Xs,Ys)=−E(x,y)∼(Xs,Ys)∑n=1NynlogDsup(Ms(xn))
3 argminDunsupLadvD(Xs,Xt,Ms,Mt)=−Exs∼XslogDunsup(Ms(xs))−Ext∼Xtlog(1−Dunsup(Mt(xt)))
In (2), Lcls is a supervised classification loss corresponding to predicting N labels from the original classification task in the source dataset (Xs), which will update the parameter in Dsup. In (3), LadvD is an adversarial loss for unsupervised discriminator Dunsup, which trains (Dunsup) to maximize the probability of predicting the correct label from the source dataset or target dataset. One thing to notice is that the unsupervised discriminator uses a custom activation function (5), which returns a probability to determine whether a source image or target image (Section 3.2 for more details).
Secondly, training the target encoder Mt with the standard loss function and inverted labels [15]. This implies that the unsupervised discriminator Dunsup is fooled by the target encoder Mt, in other words, Dunsup is unable to determine between Xs and Xt. The feedback from the unsupervised discriminator Dunsup allows the Mt to learn how to produce a more authentic encoder. The loss for LadvM can be denoted:
4 argminMtLadvM(Xs,Xt,D)=−Ext∼XtlogDunsup(Mt(xt))
Testing
In thetesting phase, we concatenate the target encoder Mt and the source classifier Cs to predict the label of target images Xt.
The discriminator model
In this section, we describe the detail of the discriminator model and provide some arguments to prove the effectiveness of the discriminator model in the SADDA method.
In the training target encoder step, the discriminator model is trained to predict N + 1 classes, where N is the number of classes in the original classification task (supervised mode) and the final class label predicts whether the sample comes from the source dataset or target dataset (unsupervised mode). The supervised discriminator and the unsupervised discriminator have different output layers but have the same feature extraction layers - via backpropagation when we train the network, updating the weights in one model will impact the other model as well. Table 1 Experimental compute on custom activation - the output of unsupervised discriminator model
Output probabilities (prior softmax) Custom activation Entropy
[9.0, 1.0, 1.0] 0.9999 Low
[5.0, 1.0, 1.0] 0.9935 Low
[-5.0, -5.0, -5.0] 0.0198 High
The supervised discriminator model produces N output classes (with a softmax activation function). The unsupervised discriminator is defined such that it grabs the output layer of the supervised mode prior softmax activation and computes a normalized sum of the exponential outputs (custom activation). When training the unsupervised discriminator, the source sample will have a class label of 1.0, while the target sample will be labeled as 0.0. The explicit formula of custom activation [37] is: 5 D(x)=Z(x)Z(x)+1
where 6 Z(x)=∑n=1Nexp[ln(x)]
The experiment of (5) is described in Table 1, and the outputs are between 0.0 and 1.0. If the probability of output value prior softmax activation is a large number (meaning: low entropy) then the custom activation output value is close to 1.0. In contrast, if the output probability is a small value (meaning: high entropy), then the custom activation output value is close to 0.0. Implied that the discriminator is encouraged to output a confidence class prediction for the source sample, while it predicts a small probability for the target sample. That is an elegant method allowing re-use of the same feature extraction layers for both the supervised discriminator and the unsupervised discriminator.
It is reasonable that learning the well supervised discriminator will improve the unsupervised discriminator. Moreover, training the discriminator in unsupervised mode allows the model to learn useful feature extraction capabilities from huge unlabeled datasets. As a sequence, improving the supervised discriminator will improve the unsupervised discriminator and vice versa. Improving the discriminator will enhance the target encoder [18]. In total, this is one kind of advantage circle, in which three elements (unsupervised discriminator, supervised discriminator, and target encoder) iteratively make each other better.
Guideline for stable SADDA
In general, training SADDA is an extremely hard process, there are two losses we need to optimize: the loss for the discriminator and the loss for the target encoder. For that reason, the loss landscape of SADDA is fluctuating and dynamic (detail in Section 4.4). When implementing and training the SADDA, we find that is a tough process. To overcome the limitation of the adversarial process, we present a full architecture of SADDA. This designed architecture increases training stability and prevents non-convergence. In this section, we present the key ideas in designing the model for the image classification and the sentiment classification task. Readers can see section A for more details about our designed architecture.
Image classification
The design of SADDA is inspired by Deep Convolutional GAN (DCGAN) architecture [38]. The summary architecture of the SADDA method for digit recognition is shown in Fig. 5. On the one hand, the encoder is used to capture the content in the image, increasing the number of filters while decreasing the spatial dimension by the convolutional layer. On the other hand, the discriminator is symmetric expansion with the encoder by using fractionally-strided convolutions (transpose convolution).
Moreover, our recommendation for efficient training SADDA in the image classification task: Replace any pooling layers with convolution layers (or transposed convolution) with strides larger than 1.
Remove fully connected layers in both encoder and discriminator (except the last fully connected layers, which are used for prediction).
Use ReLU activation [39] in the encoder and LeakyReLU activation [39] (with alpha= 0.2) in the discriminator.
Sentiment classification
The design of the SADDA method for sentiment classification is inspired by the architecture called Autoencoders LSTM [40, 41]. The summary architecture of the SADDA method for sentiment classification is demonstrated in Fig. 7. In general, the architecture of the model used in the sentiment classification task has many similarities with the architecture used in image classification. Firstly, we remove fully connected layers in both the encoder and the discriminator. Instead, we use the Long Short Term Memory [42] (LSTM) to handle sequences of text data. Secondly, the network is organized into an architecture called the Encoder-Decoder LSTM, with the Encoder LSTM being the encoder block and the Decoder LSTM being the discriminator block respectively. The Encoder-Decoder LSTM was built for the NLP task where it illustrated state-of-the-art performance, such as machine translation [43]. From the empirical, we find that the Encoder-Decoder is suitable for the unsupervised domain adaptation task.
Experiments
In this section, we evaluate our SADDA method for unsupervised domain adaptation tasks in three scenarios: digit recognition, object recognition, and sentiment classification.
In the experiments, we focus on probing how the SADDA method improves the unsupervised domain adaptation task. For this purpose, we only choose shallow architecture rather than a deep network. We leave the sophisticated design for a future job.
Digit recognition
Datasets and domain adaptation scenarios
We evaluate SADDA on various unsupervised domain adaptation experiments, examining the following popular used digits datasets and settings (the visualization is in Fig. 1):
MNIST ⇔ USPS: MNIST [5] includes 28x28 pixels, which are grayscale images of digit numbers. USPS [6] is a digit dataset, which contains 9298 grayscale images. The image is 16x16 pixels. In this experiment, we follow the evaluation protocol of [44].
MNIST → MNIST-M: MNIST-M [7] is made by merging MNIST digits with the patches arbitrarily extracted from color images of BSDS500 [45]. In this experiment, we set the input size is 28x28x3 pixels, and we follow the evaluation protocol of [44]
SVHN → MNIST: The Street View House Number (SVHN) [8] is a digit dataset, which contains 600000 32×32 RGB images. In this experiment, we convert the SVHN dataset to grayscale images and resize the MNIST images into 32x32 grayscale images. We use the evaluation protocol of [28].
Implementation details
The SADDA model is trained with different learning rates in different phases. In the pre-training phases, this is a standard classification task, we use a learning rate is 0.001 in our experiment. In the training target encoder phases, we suggested a learning rate of 0.0002 as well as an Adam optimizer [46] and setting the β1 equal to 0.5 to help stabilize training. In the LeakyReLU activation, we set α = 0.2 in the whole model.
In this experiment, the encoder consists of four convolutional layers with 4 x 4 kernel size, 2 x 2 strides, same padding, and ReLU activation. The number of filters for four convolution layers are 32, 64, 128, and 256, respectively. The target encoder has the same architecture as the source encoder, and the source encoder is used as an initialization for the target encoder.
The classifier takes the outputs of the encoder as input. Next, a fully connected layer with 100 feature channels, followed by ReLU activation. Finally, the fully connected layer with ten feature channels and the softmax activation.
In the training target encoder phases, the outputs of the encoder serve as the input of the discriminator. The discriminator consists of four Transpose Convolutional layers with 4 x 4 kernels size, 2 x 2 strides, the same padding, and the Leaky ReLU activation (alpha = 0.2). The number of kernels for four Transpose Convolutional is 256, 128, 64, and 32, respectively. We illustrate the overall architecture in Fig. 5.
Results on digit datasets
In this experiment, we compare our SADDA method against multiple state-of-the-art unsupervised domain adaptation methods.
Experimental results are shown in Table 2. In the real-world dataset SVHN → MNIST, the SADDA model showed approximately 26% improvement over the Source only model, 10% more than the ADDA method [16]. In addition, in the first two experiments (MNIST → USPS and USPS → MNIST), the SADDA method achieves extremely high accuracy, which results in 98.1% and 97.8% respectively. However, SADDA has a little lower accuracy than other methods like SBADA-GAN [47], SHOT [48], and DFA-MCD [49] in some experiments. Table 2 Experimental results on unsupervised domain adaptation on digit datasets
Method mnist→usps usps→mnist mnist→mnist-m svhn→mnist
Source only 78.9 57.1 ± 1.7 63.6 60.1 ± 1.1
DANN [7] 85.1 73.0 ± 2.0 77.4 73.9
DRCN [28] 91.8 ± 0.1 73.7 ± 0.1 - 82.0 ± 0.2
ADDA [16] 89.4 ± 0.2 90.1 ± 0.8 - 76.0 ± 1.8
SBADA-GAN [47] 97.6 95.0 99.4 76.1
SHOT [48] 98.0 98.4 - 98.9
DFA-MCD [49] 98.6 96.6 - 98.9
SADDA (our) 98.1 97.8 78.2 86.5
The results are not re-implement, instead, we select based on the available result in the previous publication (some experimental results have the standard deviation because that publication has the standard deviation while others do not.)
Although, our method did not achieve the highest accuracy in any of the experiments. Our method still has competitive accuracy when compared with the state-of-the-art methods in the last two years (SHOT [48], DFA-MCD [49]). For example, our SADDA method compared with the DFA-MCD method, in the MNIST → USPS experiment, we have lower accuracy (98.1% versus 98.6%); however, in the USPS → MNIST experiment, we achieved a higher accuracy (SADDA achieved 97.8% compared to 96.6%).
For further insight into the SADDA model effect on the digit classification tasks, we use t-SNE [50] to visualize the 2D point of the last encoder layer of SADDA (as described in Fig. 3). Ten labels are from 0 to 9 corresponding, and 100 samples per label. The domain invariance is determined by the degree of overlap between features. Regarding the Source only model, the distribution and the density are messy. In contrast, the SADDA method splits different labels into different regions, and the overlap is more prominent. Fig. 3 t-SNE embedding of digit classification, using (2 x 2 x 256) dimensional representation, with Source only (on the left) and SADDA (on the right) on the target dataset. Note that SADDA minimizes intra-class distance and maximizes inter-class distance
Object recognition
Datasets and preprocessing
In this subsection, we present the experiments for evaluating the SADDA method. The experiment is performed on the VLCS [23] dataset, including PASCAL VOC2007 (V) [11], LABELME (L) [10], CALTECH (C) [9], and SUN (S) [12] datasets. Each dataset contains five categories: bird, car, chair, dog, and person. Since the number of images per class is not equal, we use data augmentation techniques to balance the number of images. In this experiment, we use the Albumentations library [51] to increase to 5000 images per class. We divide the dataset into a training set (60%), validation set (20%), and test set (20%).
The detailed architecture is shown in Fig. 6. The rest of the other installation (optimization algorithm, learning rate) is the same as Section 4.1.2.
In the experiments, one dataset is used as the source domain and the rest is used as the target domain, resulting in four different cases (Table 3). In addition, we do not compare our SADDA model with other domain adaptation methods due to different setups. Table 3 The accuracy (%) on the VLCS dataset
LABELME CALTECH SUN
Source only 33.26 45.10 33.78
SADDA 37.73 55.30 36.21
Train on target 86.52 99.27 88.26
(a)Source domain: PASCAL
PASCAL CALTECH SUN
Source only 28.73 32.75 27.71
SADDA 32.71 39.36 31.01
Train on target 75.28 99.27 88.26
(b)Source domain: LABELME
PASCAL LABELME SUN
Source only 26.71 28.27 31.62
SADDA 31.82 31.39 37.67
Train on target 75.28 86.52 88.26
(c)Source domain: CALTECH
PASCAL LABELME CALTECH
Source only 29.72 25.87 39.66
SADDA 33.04 27.35 41.52
Train on target 75.28 86.52 99.27
(d)Source domain: SUN
Results
The results on the VLCS are shown in Table 3. Source only is a model that only trains on the source dataset without using any domain adaptation methods. Overall, the accuracy when applying the SADDA method overcomes the Source only model in all cases. In some specific cases like PASCAL → CALTECH, the classification accuracy goes from 45.10 to 55.30 (improving approximately 10%). In case LABELME → CALTECH, the accuracy grows from 32.75% to 39.36%.
Examining the results in Table 3, the Source only model has low accuracy, which reveals that the domain shift is quite large. In other words, the Source only model does not learn any knowledge about the source dataset to predict the target dataset. In contrast, the SADDA model learns a more useful feature representation, leading to higher accuracy when performing a prediction on the target dataset.
Although the SADDA method has certain improvements compared to the Source only method, the accuracy of the SADDA method is still low and not ideal. For further comparison, we also test the hypothesis situation where the target labels are present (the train on target model). There is still a big gap between the accuracy of the SADDA method and the train on target method.
Sentiment classification
Datasets and preprocessing
In this subsection, we evaluate the SADDA method for the sentiment classification task. We use three sentimental datasets, including Women’s E-Commerce Clothing Reviews [19], Coronavirus tweets NLP - Text Classification [20], and Trip Advisor Hotel Reviews [21]:
Women’s E-Commerce Clothing Reviews
[19] This is real commercial data, where the reviews are written by customers. In this task, we only use two features called Review Text (the raw text review) and Rating (the positive integer for the product, provided by the customer from 1 Worst to 5 Best). Regarding the Rating, we relabel into the sets {positive, neutral, negative} with the following rule: if a review is greater than 3, it is considered a positive comment; if a review is equal to 3, it is considered a neutral comment; if a review is less than 3, it is considered a negative comment.
Coronavirus tweets NLP - Text Classification
[20] The tweets were downloaded from Twitter and tagged manually. Although there are four columns in total, we only use the Original Tweet feature and Label in our experiment. In the case of Label, the original label includes Extremely Negative, Negative, Neutral, Positive, and Extremely Positive. However, we convert to the sets {positive, neutral, negative} respectively.
Trip Advisor Hotel Reviews
[21] This contains reviews crawled from the travel company called Tripadvisor. The dataset contains two features, including Review Text and Rating (the positive integer from 1 Worst to 5 Best). Regarding Rating, we process the same as the case Women’s E-Commerce Clothing Reviews above.
In all three datasets, we perform the following text preprocessing steps: removing the punctuation, URL, hashtags, mentions, and stop words (with the support of the NLTK [52] library). We limit the input sentence to a max length equal to 50. The GloVe [53] word embedding is applied to map the word in the text review to the vector space.
Because the number of samples per class is not balanced, we use the data augmentation techniques to balance the number of samples - with the support of the TextAugment [54] library. Particularly, we perform data augmentation such that each class has up to 20 000 samples. The dataset is divided into a training set (60%), validation set (20%), and test set (20%).
Experiments and results
For this experiment, our architecture is illustrated in Fig. 5. Elements in that architecture such as the LSTM layer and the Repeat Vector layer are implemented by the TensorFlow library with default settings. The optimization algorithms and learning rates are set up as in Section 4.1.2. In addition, we do not attempt to fine-tune the architecture and leave it for future work.
The results of our experiment are provided in Table 4. Compared with the Source only model, the SADDA method shows a little improvement in the accuracy of this sentiment analysis task. For a certain experiment, like T → W, the classification accuracy goes from 41.07% to 49.28%. However, not all experiments improve, such as experiment T → C, the accuracy even dropped a bit (from 38.46% down to 38.05%). Additionally, a comparison with the “Train on target” model exposes that the SADDA model is far from the ideal model. We hope that is the motivation for future development. Table 4 The accuracy (%) of unsupervised domain adaptation on the sentiment classification task
W → C W → T C → W C → T T → W T → C
Source only 37.97 50.11 45.91 49.93 41.07 38.46
SADDA 43.88 55.54 48.02 56.10 49.28 38.05
Train on target 79.01 96.10 93.02 96.10 93.02 79.01
In the table, there are three datasets: Women’s E-Commerce Clothing Reviews (W) [19], Coronavirus tweets (C) [20], Trip Advisor Hotel Reviews (T) [21]
Challenge and convergence analysis
In this subsection, we will discuss the challenge of training a stable SADDA model and how to trigger the early stopping of the training progress to achieve the convergent state.
In the SADDA model, we have three stages: Pre-training, Training target encoder, and Testing. The difficulty comes from the Training target encoder process. The reason is that both the discriminator and target encoder are trained simultaneously in that procedure, which can lead to updating the parameter of one model will reduce the performance of the other model. More concretely, there are two loss functions we need to optimize: the discriminator loss and the adversarial loss. The discriminator loss (3) is the loss when the unsupervised discriminator predicts the source or target samples. The adversarial loss (4) is the loss of the target encoder when training with the inverted labels.
When training the target encoder, we do not try to find a minimum value for either discriminator loss or adversarial loss. Instead, we are looking for a Nash equilibrium state for both losses [37]. In practice, we observe the discriminator loss and adversarial loss after each epoch, when both loss values no longer change, we trigger an early stopping. Early stopping is the technique to stop the training process at a certain point before the model overfits the training dataset and has poor performance on the test set. In that case, we consider that our model has converged. Keep in mind that the loss of 0.0 in the discriminator loss or the adversarial loss during the training process is a failure mode.
Regarding Fig. 4, the convergence point is in the epoch 20. At that point, we will stop the training process because the discriminator loss and the adversarial loss are saturated at around 0.33 and 2.30 respectively. In that experiment (and other object recognition tasks), it took around 1 hour on a single Tesla T4 GPU to complete the training procedure. Fig. 4 The discriminator loss and adversarial loss in the LABEL → CALTECH experiment
Conclusion
We proposed a more stable and high-accuracy architecture for training adversarial-based domain adaptation methods. The key idea of this approach is to train discriminators in two modes: supervised mode and unsupervised mode. Moreover, utilize this to create a more efficient target encoder, which will help improve the classification accuracy.
While the SADDA method has demonstrated an improvement in many tasks like image classification or sentiment classification, there are still open challenges. Particularly, the SADDA model in object recognition and sentiment classification is far from the desired accuracy model. We hope that the intuition of this research will facilitate further advances in domain adaptation tasks.
Appendix
In this appendix section, we present in detail the design architecture of our SADDA method in three experiments, including digit recognition, object recognition, and sentiment classification. Fig. 5 The overview of the SADDA method for the digit recognition task. We found that Global Average Pooling (GAP) [55] increased model stability and reduce the number of parameter
Fig. 6 The overview of the SADDA method for the object recognition task on the VLCS [23] dataset. With the input image’s shape is 64 x 64 x 3
Fig. 7 The overview of the SADDA method for the sentiment classification task. The input sentence has a max length equal to 50. In the design above, to prevent overfitting, the LSTM layer is always followed by the dropout layer with 0.2 rates. The numbers under the particular layer are the output shape of that layer
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Gretton A, Smola A, Huang J, Schmittfull M, Borgwardt K, Schölkopf B (2009) Covariate shift and local learning by distribution matching, pp 131–160. Cambridge, MA USA: MIT Press
2. Long M, Cao Y, Wang J, Jordan MI (2015) Learning transferable features with deep adaptation networks. arXiv:1502.02791
3. Nguyen A, Nguyen N, Tran K, Tjiputra E, Tran QD (2020) Autonomous navigation in complex environments with deep multimodal fusion network. In: 2020 IEEE/RSJ international conference on intelligent robots and systems (IROS)
4. Donahue J, Jia Y, Vinyals O, Hoffman J, Zhang N, Tzeng E, Darrell T (2013) Decaf: A deep convolutional activation feature for generic visual recognition
5. Lecun Y Bottou L Bengio Y Haffner P Gradient-based learning applied to document recognition Proc IEEE 1998 86 11 2278 2324 10.1109/5.726791
6. Hull JJ A database for handwritten text recognition research IEEE Trans Pattern Anal Mach Intell 1994 16 5 550 554 10.1109/34.291440
7. Ganin Y, Ustinova E, Ajakan H, Germain P, Larochelle H, Laviolette F, Marchand M, Lempitsky V (2015) Domain-adversarial training of neural networks
8. Netzer Y, Wang T, Coates A, Bissacco A, Wu B, Ng AY (2011) Reading digits in natural images with unsupervised feature learning. In: NIPS Workshop on deep learning and unsupervised feature learning 2011
9. Fei-Fei L Fergus R Perona P One-shot learning of object categories IEEE Trans Pattern Anal Mach Intell 2006 28 4 594 611 10.1109/TPAMI.2006.79 16566508
10. Russell BC Torralba A Murphy KP Freeman WT Labelme: a database and web-based tool for image annotation Int J Comput Vis 2007 77 157 173 10.1007/s11263-007-0090-8
11. Everingham M, Van Gool L, Williams CKI, Winn J, Zisserman A (2022) The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Results. http://www.pascal-network.org/challenges/VOC/voc2007/workshop/index.html
12. Choi MJ, Lim JJ, Torralba A, Willsky AS (2010) Exploiting hierarchical context on a large database of object categories. In: 2010 IEEE Computer society conference on computer vision and pattern recognition, pp 129–136
13. Gheisari M Baghshah MS Unsupervised domain adaptation via representation learning and adaptive classifier learning Neurocomput 2015 165 300 311 10.1016/j.neucom.2015.03.020
14. Pan SJ Tsang IW Kwok JT Yang Q Domain adaptation via transfer component analysis IEEE Trans Neural Netw 2011 22 2 199 210 10.1109/TNN.2010.2091281 21095864
15. Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. In: Advances in neural information processing systems (Z. Ghahramani, M. Welling, C. Cortes, N. Lawrence, and K. Q. Weinberger, eds.), vol 27, Curran Associates., Inc.
16. Tzeng E, Hoffman J, Saenko K, Darrell T (2017) Adversarial discriminative domain adaptation. In: 2017 IEEE conference on computer vision and pattern recognition (CVPR), pp 2962–2971
17. Reddy Y Pulabaigari V E. B Semi-supervised learning: a brief review Int J Eng Technol 2018 7 81, 02
18. Odena A (2016) Semi-supervised learning with generative adversarial networks
19. Nicapotato (2018) Women’s e-commerce clothing reviews
20. Miglani A (2020) Coronavirus tweets nlp - text classification
21. Alam MH Ryu W-J Lee S Joint multi-grain topic sentiment: modeling semantic aspects for online reviews Inf Sci 2016 339 206 223 10.1016/j.ins.2016.01.013
22. Krawczyk B Learning from imbalanced data: open challenges and future directions Prog Artif Intell 2016 5 221 232 10.1007/s13748-016-0094-0
23. Torralba A, Efros AA (2011) Unbiased look at dataset bias, CVPR 2011, pp 1521–1528
24. Chi W, Dagnino G, Kwok TM, Nguyen A, Kundrat D, Abdelaziz ME, Riga C, Bicknell C, Yang G-Z (2020) Collaborative robot-assisted endovascular catheterization with generative adversarial imitation learning. In: 2020 IEEE International conference on robotics and automation (ICRA), pp 2414–2420 IEEE
25. Kouw WM Loog M A review of domain adaptation without target labels IEEE Trans Pattern Anal Mach Intell 2021 43 766 785 10.1109/TPAMI.2019.2945942 31603771
26. Margolis A (2011) A literature review of domain adaptation with unlabeled data, Rapport Technique, University of Washington, p 01
27. Wang M Deng W Deep visual domain adaptation: a survey Neurocomputing 2018 312 135 153 10.1016/j.neucom.2018.05.083
28. Ghifary M, Kleijn WB, Zhang M, Balduzzi D, Li W (2016) Deep reconstruction-classification networks for unsupervised domain adaptation
29. Deng J Zhang Z Eyben F Schuller B Autoencoder-based unsupervised domain adaptation for speech emotion recognition IEEE Signal Process Lett 2014 21 9 1068 1072 10.1109/LSP.2014.2324759
30. Long M, Zhu H, Wang J, Jordan MI (2016) Deep transfer learning with joint adaptation networks
31. Long M, Cao Z, Wang J, Jordan MI (2017) Conditional adversarial domain adaptation
32. Saito K, Watanabe K, Ushiku Y, Harada T (2018) Maximum classifier discrepancy for unsupervised domain adaptation
33. Ghifary M, Kleijn WB, Zhang M, Balduzzi D (2015) Domain generalization for object recognition with multi-task autoencoders
34. Bousmalis K, Trigeorgis G, Silberman N, Krishnan D, Erhan D (2016) Domain separation networks
35. Glorot X, Bordes A, Bengio Y (2011) Domain adaptation for large-scale sentiment classification:, A deep learning approach
36. Vincent P Larochelle H Lajoie I Bengio Y Manzagol P-A Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion J Mach Learn Res 2010 11 3371 3408
37. Salimans T, Goodfellow I, Zaremba W, Cheung V, Radford A, Chen X (2016) Improved techniques for training gans
38. Radford A, Metz L, Chintala S (2015) Unsupervised representation learning with deep convolutional generative adversarial networks
39. Maas AL, Hannun AY, Ng AY (2013) Rectifier nonlinearities improve neural network acoustic models. In: ICML Workshop on deep learning for audio, speech and language processing
40. Srivastava N, Mansimov E, Salakhutdinov R (2015) Unsupervised learning of video representations using lstms
41. Brownlee J (2020) A gentle introduction to lstm autoencoders
42. Hochreiter S Schmidhuber J Long short-term memory Neural Comput 1997 12 1735 80 10.1162/neco.1997.9.8.1735
43. Cho K, van Merriënboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, Bengio Y (2014) Learning phrase representations using RNN encoder–decoder for statistical machine translation. In: Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), (Doha, Qatar), pp 1724–1734, Association for Computational Linguistics
44. Bousmalis K, Silberman N, Dohan D, Erhan D, Krishnan D (2016) Unsupervised pixel-level domain adaptation with generative adversarial networks
45. Arbeláez P Maire M Fowlkes C Malik J Contour detection and hierarchical image segmentation IEEE Trans Pattern Anal Mach Intell 2011 33 5 898 916 10.1109/TPAMI.2010.161 20733228
46. Kingma DP, Ba J (2014) Adam: A method for stochastic optimization
47. Russo P, Carlucci FM, Tommasi T, Caputo B (2017) From source to target and back: symmetric bi-directional adaptive gan
48. Liang J, Hu D, Feng J (2020) Do we really need to access the source data? source hypothesis transfer for unsupervised domain adaptation. In: international conference on machine learning (ICML), pp 6028–6039
49. Wang J, Chen J, Lin J, Sigal L, de Silva CW (2021) Discriminative feature alignment: Improving transferability of unsupervised domain adaptation by gaussian-guided latent alignment. Pattern Recognition, p 107943
50. van der Maaten L Hinton G Visualizing data using t-sne J Mach Learn Res 2008 9 86 2579 2605
51. Buslaev A, Iglovikov VI, Khvedchenya E, Parinov A, Druzhinin M, Kalinin AA (2020) Albumentations: fast and flexible image augmentations, information, vol 11, no. 2
52. Loper E, Bird S (2002) Nltk: the natural language toolkit, CoRR, vol. cs.CL/0205028, 07
53. Pennington J, Socher R, Manning C (2014) GLoVe: Global vectors for word representation. In: Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), (Doha, Qatar), pp 1532–1543, Association for Computational Linguistics
54. Marivate V, Sefara T (2020) Improving short text classification through global augmentation methods
55. Lin M, Chen Q, Yan S (2013) Network in network
| 36466775 | PMC9707164 | NO-CC CODE | 2022-12-01 23:20:21 | no | Appl Intell (Dordr). 2022 Nov 29;:1-14 | utf-8 | Appl Intell (Dordr) | 2,022 | 10.1007/s10489-022-04288-4 | oa_other |
==== Front
Curr Oncol Rep
Curr Oncol Rep
Current Oncology Reports
1523-3790
1534-6269
Springer US New York
36445569
1347
10.1007/s11912-022-01347-4
Evolving Therapies (RM Bukowski, Section Editor)
Development of Effective Siglec-9 Antibodies Against Cancer
Wang Jun Hui Shawn 1
Jiang Nan 2
Jain Amit 2
http://orcid.org/0000-0003-4344-6822
Lim Jackwee [email protected]
1
1 grid.430276.4 0000 0004 0387 2429 Singapore Immunology Network, A*STAR, 8a Biomedical Grove, Singapore, 138648 Singapore
2 grid.410724.4 0000 0004 0620 9745 Division of Medical Oncology, National Cancer Centre Singapore, Singapore, 169610 Singapore
29 11 2022
19
27 9 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose of Review
This study aims to review state-of-the-art advances in Siglec-9-directed antibodies and to highlight specific aspects of Siglec-9 antibodies that are suitable to mount anti-tumor immunity.
Recent Findings
Controversies surrounding studies on Siglec-9 antibodies can confound future studies. In this review, we have highlighted some controversies, explained the distinction between Siglec-9 agonistic and antagonistic (endocytic) antibodies, and discussed their suitability in sustaining anti-tumor immunity.
Summary
Siglec-9 is an immune checkpoint target and an immunoinhibitory receptor that can engage either sialic acid ligands or agonistic antibodies. Through Siglec-9 sialic acid interactions, activated immunoreceptor tyrosine-based inhibitory signaling of the immune cells can lead to unfavorable immunosuppression. To overcome tumor-related immunosuppression, different types of Siglec-9 antibody blockade need to be developed. However, whether a Siglec-9-directed antibody is agonistic or antagonistic is probably affinity-dependent and not epitope-dependent. Additionally, unlike immune-modulatory antibodies such as agonistic antibodies (OX40, CD28, ICOS, and 4-1BB) or Fc-inert antibodies (PD1 and PD-L1) directed against cancer cells, the nature of antagonistic Siglec-9 antibodies is more suitable to enhance anti-tumor immunity and will be discussed.
Keywords
Siglec-9 antibody
Antagonism
Agonism
Affinity
Sialic acid
==== Body
pmcIntroduction
First discovered in 2000, Siglec-9 (CD329) is a member of the CD33-related sialic acid-binding immunoglobulin-like lectins with 84% sequence homology to Siglec-7 [1, 2]. Like most other immunoinhibitory Siglec family members, its cytoplasmic tail contains immunoreceptor tyrosine-based inhibitory (ITIM) and ITIM-like motifs. The classical mechanism of inhibitory Siglec-9 signaling upon binding to their sialic acid ligand, would result in phosphorylation of tyrosine residues in the ITIM domain by SRC kinase and the recruitment of inhibitory tyrosine phosphatase Src homology region 2 domain-containing phosphatase (SHP) SHP1 and/or SHP2 [3]. These lead to a range of cell-type-specific inhibitory functional signalings, such as suppressing neutrophil activation, CD8+ T cell, and NK cell effector functions [4, 5, 6••]. Furthermore, its basal expression is found in healthy human immune cell types such as monocytes, macrophages, neutrophils, dendritic cells, and CD16Bright/CD56−/dim NK cells [2, 4, 7••]. of note, Siglec-9 is also weakly expressed on CD16dim/−/CD56Bright NK, B cells, tumor-infiltrating CD4+ T cells, and PD-1+/CD8+ T cells in humans but not found in NK cells and T cells in mice (Siglec-E, an ortholog of human Siglec-9) [4, 8–11].
Siglec-9 is an emerging immune checkpoint therapeutic target, and blockade appears to boost anti-cancer immunity [12]. However, the success of this targeted approach depends not only on the specific monoclonal antibodies but also on their ability to induce a proper immune function. Here, we will highlight the consistencies and attempt to resolve opposing studies that have yielded conflicting results. This review covers how different sialic acid and non-sialic acid blocking Siglec-9 antibodies can be agonistic or antagonistic (endocytic), leading to varying levels of intended anti-tumor immunity (Table 1), and provide a rationale when designing Siglec-9 antibodies.Table 1 An overview of Siglec-9-directed antibodies and their functions
Effector function Anti-Siglec-9 clone Observation Reference
Blocking, agonist 191240 Downregulated NF-κB activity in HEKBlue hTLR4 cells [30]
Inhibited MUC1 binding to Siglec-9 myeloid cells and reduces TAM polarization [13, 14•]
Inhibited T cell activation in a dose-dependent manner like clone E10-286 [9]
Competed with LGALS3BP, a Siglec-9 ligand involved in inhibiting neutrophil activation [16]
Failed to block Siglec-7 positive NK cells, resulting in cytotoxicity. Reduced cytotoxicity for Siglec-9 positive NK cells [55]
Activated Siglec-9 positive neutrophils resulting in tumor cell killing [11]
Inhibited MUC16 positive ovarian cancer cells binding to monocytes [10]
Blocking (undefined) Did not inhibit NK cell cytotoxicity against chronic myelogenous leukemia, K562 cells unlike clone E10-286. This finding may be refuted if the studied Siglec-9’s V-set was masked by sialic acids [4]
Reduced differentiation of monocytes to CD163+ macrophages in co-culture with tumor cell lines BxPC3 and PaTuS. Greater reduction was observed when used with anti-Siglec-7 [15•]
Blocking, partial agonist Loss of agonism with a monomeric 191240 Fab fragment [9]
Non-blocking, agonist E10-286 Inhibited Siglec-9 positive CD8+ T cells and IFNγ production. Triggers phosphorylation of Siglec-9 and SHP-1 [6••]
Reduced CD8+ T cell cytotoxicity against mastocytoma P815 cells [6••]
Inhibited T cell activation in a dose-dependent manner like clone 191240 [9]
Did not compete with LGALS3BP, a Siglec-9 ligand involved in inhibiting neutrophil activation [16]
Reduced activation of Siglec-9-positive neutrophils when compared to the clone 191240. This finding did not elaborate on binary sialic acid and clone E10-286 binding [11]
Inhibited NK cell cytotoxicity against chronic myelogenous leukemia, K562 cells [4]
Non-blocking, antagonist 2D4 Reduced Siglec-9 surface expression and exhibited anti-tumor responses when used in combination with pembrolizumab [7••]
Blocking (undefined) 68D4 and 224B1 Activated CD4+ and CD8+ T cells. Increased cytotoxicity of NK92 cell line against chronic myelogenous leukemia, K562 cells [43••]
Blocking (undefined) 8A1E9 Increased NK cell cytotoxicity against chronic myelogenous leukemia, K562 cells [20, 21••]
Blocking (undefined) mAbA Enhanced anti-tumor immunity in Siglec-7/-9 expressing mice, used in combination with an anti-Siglec-7 sialic acid blocking antibody [12]
Blocking (undefined) mAbA Increased NK cell cytotoxicity against colorectal adenocarcinoma, HT29 cells [18••]
Blocking (undefined) KALLI Siglec-9 mediated endocytosis of antibodies [37, 56]
Agonist (undefined) 1–3-A Suppressed induction of LPS-mediated COX2 mRNA production in histiocytic lymphoma, U937 cells [17]
More than Just Sialic Acid-Binding Epitopes on Siglec-9 Receptor
Studies on designing blocking antibodies that compete with the sialic acid-binding site may require an awareness of other possible cryptic binding sites. Hyper-sialylated cancer cells gain survival advantages from their ability to bind to Siglecs expressed on immune cells and manipulate the tumor microenvironment (TME) by inducing the formation of tumor-associated macrophages (TAMs) to promote their growth and metastasis while evading NK cell and T cell-mediated cytotoxicity [13, 14•, 15•]. Examples of glycoproteins in cancer that display an aberrantly sialylated glycan landscape capable of binding to Siglec-9 include mucin 1 (MUC1) and lectin galactoside-binding soluble 3 binding protein (LGALS3BP) [13, 16]. Earlier reports on Siglec-9 sialic acid ligands have shown that Siglec-9 displays preferential binding to a predominant sialic acid, Neu5Acα2-3, found on both sialyl 6-sulfo Lewisx and the cancer-associated sialyl Lewis x (sLewisx) [1, 17, 18••]. This is unlike its close homolog, Siglec-7, which preferably binds to Neu5Acα2-8Neu5Acα2-3 on gangliosides. This sialic acid ligand preference is attributed to key arginines in their binding grooves; Siglec-9 has one at R120, while Siglec-7 has two, R67 and R124 [1, 2, 19]. Furthermore, tryptophans W132 and W128 in Siglec-7 and Siglec-9, respectively, may engage alternate hydrophobic interaction with the glycerol moiety on the sialic acid [1, 18••]. Moreover, sialic acid in the 3rd position on the GlcNAc sugar is often poorly recognized by Siglec-9 receptor, as reported in a comprehensive binding study by Miyazaki et al. (2012) using glycan expressing colonic epithelial cells. In this study, Siglec-9 preferably binds sialic acids α2-3 linked Galβ1-4GlcNAc (sialyl LewisX) to sialic acids α2-3 linked Galβ1-3GlcNAc (sialyl LewisA/C, where LewisA carries an additional fucose sugar) [17]. A Newman projection shows that only the β1-4 linkage presents the sialic acid-bearing group in the 4th position on the GlcNAc or GalNAc sugar at a favorable steric-free position for Siglec-9 recognition (indicated as R’’) (Fig. 1A). Additionally, Siglec-9 prefers disialyl LewisA/C to sialyl LewisA/C ligand, which is again explained using the Newman projection, which exposes one sialic acid in the R’’’ group position. These suggest that terminal sialic acid ligands have to be exposed for Siglec-9 or Siglec family interaction. Indeed, two independent studies by Angata et al. (2000) and Beatson et al. (2016) also showed that Siglec-9 binds very weakly to sialic acid α2-3 linked Galβ1-3GalNAc (ST) [1, 13], with no observable Siglec-9 binding to both ST-coated glycan array and ST-coated polymers in vitro. Taken together, the 3rd position on the GlcNAc or GalNac sugar puts the sialic acid-bearing group in an unfavorable steric position (R’ group position) for Siglec-9 receptor recognition.Fig. 1 Siglec-9 and its partner sialic acids or directed antibodies. A Newman projection-based explanation for different sialic acid ligand binding preferences. Only the sialic acid (Neu5Ac, a predominant form)-bearing groups located at the 4th and 6th positions on GlcNAc/GalNAc (on R’’ and R’’’, respectively) remain exposed for Siglec-9 receptor recognition. In the steric conformations, the sialic acid in the 3rd position on GlcNAc/GalNAc (on R’) is in close proximity to the bulky O-linked protein. GalNAc in MUC1-ST is not shown. B The unsolved Siglec-9 receptor was modeled using AlphaFold, and the previously determined conformational epitopes were mapped onto it [7••, 53]. Full antagonistic antibodies that deplete surface Siglec-9 receptor density such as 5C6 and 2D4 have a binding affinity in low nM. Here, an inert antibody, clone 2D5, binds weakly and neither blocks sialic acid binding nor depletes surface Siglec-9 receptor density. It is still unknown about the binding affinity range for agonistic antibodies such as clones 191240 and E10-286. A structural interpretation shows that sialic acid-binding epitope plays no role in dictating the antagonism of an antibody. n.d. denotes not determined. C (Top) A mechanistic overview of Siglec-9 receptor oligomerization driven by (1) sialic acid supercluster or (2) FcγR-mediated agonistic antibody clusters. (Bottom) Diagram illustrating the competitive and non-competitive nature of agonistic antibodies (3) clone 191240 and (4) clone E10-286. The dose–response relationship of agonistic antibodies is bell-shaped [38, 54]. The optimum affinity range associated with agonism remains unknown. D A mechanistic overview of Siglec-9 endocytosis when strongly bound to antagonistic antibodies. This phenomenon may be harnessed by antibody conjugates to deliver cytotoxic drugs into the cancer cells. The dose–response relationship of antagonistic antibodies is sigmoidal [38]. The optimum affinity range associated with antagonism remains unknown. Figures 1C and D were made in BioRender
Besides sialic acid binding, Beatson et al. continued to show that Siglec-9 expressing myeloid cells can bind strongly to MUC1-ST, suggesting boosted interaction involving Siglec-9, sialic acid, and MUC1 protein. This binding was abolished in asialoMUC1 (MUC1-T) [13]; this underscored a cryptic protein–protein interaction between the MUC1 protein backbone and Siglec-9 receptors, as alternative epitopes for Siglec-9 antibodies.
Mapping Conformational Epitopes of Siglec-9 Antibodies
Identifying a monoclonal antibody’s epitope is integral for characterizing binding or blocking specificities. However, the choice of linear peptides for epitope mapping may misinform conformational epitopes. For example, Choi et al. have employed Siglec-9-derived linear peptides to report clone 191240 binding to the membrane-proximal C2-set and later showed their most potent clone 8A1E9 binding at the N-terminal region [20, 21••]. Instead, using CLIPS cyclization, Rosenthal et al. showed that the conformational epitopes for clone 191240 are at the N-terminal region. Unlike Choi et al.’s study, Rosenthal et al.’s report is more consistent with the sialic acid blocking nature of clone 191240 at the N-terminal ligand binding site on Siglec-9 receptor [7••]. Alternatively, soluble Siglec-9 protein can be produced with several point mutations to get a more accurate description of antibody recognition. Nonetheless, Choi et al. have reported a potent Siglec-9 antibody that targets the N-terminal region with enhanced anti-tumor activity in vitro.
Agonistic Siglec-9-Directed Antibodies for Anti-inflammatory Therapies
Agonistic Siglec-9 antibodies mimic natural sialic acid ligands, which induce SHP-1 phosphorylation and trigger inhibitory cell signaling. Siglec-9 cross-linking due to sialic acid clusters can deplete neutrophils, a normal mechanism in Siglecs on leukocytes to downregulate inflammation [22, 23]. Notably, FcγR expressed on antigen-presenting cells can crosslink agonist antibodies bound to receptors and result in receptor cross-linking or oligomerization. Receptor oligomerization is a prerequisite for activating downstream signaling in tumor necrosis factor receptor superfamily, but Siglec receptor oligomerization still remains poorly understood [24]; CD22 is reported to be a multimer; Siglec-5, Siglec-8, and Siglec-11 are thought to be disulfide-linked dimers [25–28]. Siglec-7 and Siglec-9 exist as monomers [29, 30]. Nonetheless, soluble sialoside analog pS9-sol is unable to induce Siglec-9 receptor oligomerization. On the other hand, membrane-tethered pS9-L provided evidence of induced Siglec-9 receptor clustering [30]. This indirectly suggests that Siglec-9 antibody (FcγR mediated) multimerization may trigger Siglec-9 receptor clustering [31].
Here, two popular human Siglec-9 antibody clones, sialic acid blocking clone 191240 and non-sialic acid blocking clone E10-286 (using Siaα2-3Galβ1-4GlcNAc on glycan array), have been described as agonistic antibodies [32]. In this review, we have mapped their binding epitopes onto the Siglec-9 receptor based on a competitive antibody study (Fig. 1B) [7••]. The clone E10-286 was found to be more agonistic than clone 191240, resulting in a stronger inhibition of Siglec-9 expressing NK cell cytotoxicity [4]. This concurs with the possibility that the Siglec-9 receptor can engage both sialic acid and clone E10-286 as binary ligands (Fig. 1C). On the other hand, clone 191240 would compete for the same sialic acid-binding cleft, thus likely resulting in a weaker agonistic effect.
Agonism is not limited to intact antibodies because antibody fragment (hS9-Fab03) has also been shown to be specific to Siglec-9 and exhibit immunoinhibitory effects in PBMC- or THP-1-differentiated macrophages [33]. However, in their study, hS9-Fab03 generated in E. coli BL21 cells as agonists of toll-like receptors may carry endotoxins [34].The authors did not perform endotoxin removal on the hS9-Fab03 in their LPS-stimulated studies. Thus, it remains debatable about the true avidity of the antibody fragment. Next, Siglec-9 can be an autoantigen, whereby agonistic anti-Siglec-9 autoantibodies are found in IVIg preparations for treating autoimmune diseases, which explain the observed neutropenia among patients [35, 36]. In comparison, synthetic Siglec-9 agonists or glycopolymers have been used to curb neutrophil hyperinflammation in COVID-19 by favoring apoptosis and not NETosis [23].
There is another debatable finding associated with macrophages. Laubli et al. have used LS180 tumor co-cultures and observed a marginal increase in the number of tumor-associated CD11b+CD206+ M2 macrophages using agonistic Siglec-9 antibody clone 191,240 [11]. On the other hand, Beatson et al. first observed a significant increase in the number of PD-L1+CD206+ M2 macrophages in the presence of MUC1-ST. Second, Beatson et al. further showed that MUC1-ST-induced TAM-like macrophage formation can be blocked using the same clone 191240, which contradicts Laubli et al. [13]. More studies are needed to validate both opposing studies on the propensity of M2 macrophage polarization. In this unique situation, Siglec-9 agonism via SHP-1 is deemed as the therapeutic way to treat both cancer and infectious diseases via reducing the TAM-like macrophage population [23].
Designing Antagonistic (Endocytic) Siglec-9-Directed Antibodies for Cancer Therapies
Many studies have observed reduced tumor growth in Siglec-9 knockout experiments in vivo and in vitro, which are the Rosetta stones to develop Siglec-9 antibodies to treat cancer today. In oncology, epithelial tumor cells overexpress heavily glycosylated mucins that bind to Siglec-9 [10, 13]. Additionally, Siglec-9 is overexpressed on primary acute myeloid leukemia cells but absent in progenitor cells in patient bone marrow samples [37]. These suggest that blocking Siglec-9 sialic acid interactions can become potential therapeutic strategies. Interestingly, a naturally occurring Siglec-9 K131Q (A391C) polymorphism (rs16988910) with diminished sialic acid binding was associated with improved early overall survival in non-small cell lung cancer (NSCLC) patients in only the first 2 years but not in subsequent years [11]. This finding motivated others to develop antagonistic Siglec-9 antibodies that can either reduce Siglec-9 expression on the cell surface and/or inhibit Siglec-9 ligand interactions and reduce tumor growth and volume [12]. This is unlike the development of agonistic antibodies (OX40, CD28, ICOS, 4-1BB) in clinical development against cancer, which has been nicely reviewed elsewhere [38].
Today, there is still no known clinical trial using antagonistic Siglec-9 antibodies to treat cancer. The idea of antagonistic Siglec-9 antibodies that serves to potentiate immune reactivity towards cancer is not new. The mechanistic nature of antagonistic antibodies in depleting surface Siglec-9 receptors is via endocytosis in a tyrosine-ITIM dependent manner [37]. However, an empirical approach is still needed to characterize the antibody using cell-based or in vivo models [39–41]. Unlike “stepping on the gas” OX40-targeted agonistic antibodies which can activate OX40 on activated T cells to potentiate T cell receptor (TCR) signaling and depletes the number of FoxP3+ regulatory T (Treg) cells via antibody-dependent cellular cytotoxicity (ADCC), Siglec agonism suppresses immune reactivity, which is more appropriate for anti-inflammatory therapies [23, 42]. Ideally, the design of Siglec-9 antibodies as cancer therapeutics should not activate downstream ITIM inhibitory signaling pathways, a brake in immune responses.
The design of agonistic or antagonistic antibodies is not epitope-specific. For example, both clones E10-286 (an agonist) and 2D4 (an antagonist) can target the same epitopes found away from the sialic acid-binding region on the Siglec-9 receptor. As an antagonist, clone 2D4 has earlier been shown to downregulate surface Siglec-9 expression on monocytes, dendritic cells, and macrophages by up to 80% for 21 days [7••]. The depletion of Siglec-9 surface expression would favorably overcome immunosuppression due to limited immunoinhibitory Siglec-sialic acid interactions. Indeed, in another study, Choi et al. have reported a sialic acid blocking Siglec-9 antibody that reduced the tumor burden of humanized immunodeficient NOD scid gamma mice inoculated with SKOV3 ovarian cancer cells. They also found that sialic acid blocking Siglec-9 antibodies in combination with a MUC1 vaccine increased CD4+ and CD8+ MUC1-specific, IFN-γ-expressing T cells and NK cell CD56 expression and cell count when compared to mice that were untreated or only vaccinated [20, 21••]. Of note, in a separate study, reduced melanoma tumor volume was observed in humanized mice when clone 2D4 was used in combinatorial treatment with Keytruda®, an anti-PD-1 immune checkpoint inhibitor compared to 2D4 untreated mice [7••].
Although the features of antagonistic antibodies are not well-defined, the general rule is one with the highest binding strength. In some cases, the native sialic acid ligand binds weakly and may be out-competed by the higher affinity antibody. In other cases, the antagonistic antibody does not compete with sialic acid ligand. For instance, both antagonistic antibody clones 5C6 (sialic acid blocking) and 2D4 (non-sialic acid blocking) bind with low nanomolar affinity, which falls in the range of high-affinity antibodies (10−9 M) but not very high-affinity antibodies (10−12 M) [7••]. In comparison, an inert antibody, a weaker affinity binding clone 2D5, can bind to macrophages (93.2%) but not strongly to monocytes (9.1%) and dendritic cells (34.2%), when compared to clone 2D4 (macrophages, 99.1%; monocytes, 95%; dendritic cells, 99.3%) [7••]. Notably, the clone 2D5 also does not cause reduced surface Siglec-9 receptor expression like clones 5C6 and 2D4 (Fig. 1B). In a separate study, weaker affinity clone 5C6 and negative control 53C3 did not induce any stimulatory effects on effector CD4+ and CD8+ T cells, as suggested by the absence of CD69 and CD25 [43••]. For this reason, more studies are needed to evaluate the optimum binding affinities of Siglec-9-directed antibodies to be full antagonist and neither inert nor agonistic.
Finally, increasing evidence also highlights that FcγR clustering via the Fc region can enhance agonistic antibody-mediated receptor clustering, ADCC, and antibody-dependent cellular phagocytosis (ADCP) [44]. However, for antagonistic antibodies, the endocytic event leading to Siglec-9 downregulation is independent of FcγRs. Therefore, silencing or removal of the Fc regions to minimize interactions with FcγRs might be a key consideration in the design of future antagonistic antibodies. For example, Fc-inert antibodies against PD-1 and PD-L1 have shown clinical efficacy against melanoma, cutaneous squamous cell carcinoma, NSCLC, and urothelial carcinoma as FDA-approved therapeutics [45–48]. We have included well-defined Fc mutations to guide the design of such Fc-inert Siglec-9-directed antibodies in Table 2.Table 2 Known FcR mutations to either enhance or abolish FcR binding. Siglec-9 downregulation is independent of FcγRs. Therefore, silencing or removal of the Fc region to minimize interactions with FcγRs is a key consideration in the design of antagonistic antibodies and vice versa. Of note, FcγRs-independent receptor cross-linking/agonism has also been reported due to hexameric antibody formation in solution due to mutations made on the Fc region [44, 59]. #The authors misannotated L235A as F235A
Effector function Mutations Fc isotype Reference
Silence binding to Fcγ receptors E233P, L234V, L235A, G236 del., P238A, D265A, N297A, A327Q/G, and P329A IgG1 [7••]
V234A, G237A, P238S, H268A, V309L, A330S, and P331S IgG2
E233P, F234V, L235A#, G237A, N297A/Q, and E318A IgG4
Silence binding to Fcγ receptors L234A, L235A, G237A, P238S, H268A, A330S, and P331S IgG1 [57]
Enhance binding to FcγRIIB and FcγRIIA (Arg 131 allotype) S267E and L328F IgG1 [44]
Enhance binding to FcγRIIB E233D, G237D, P238D, H268D, P271G, and A330R IgG1 [58]
Formation of hexamers in solution E345R, E430G, and S440Y IgG1 [59]
Other Forms of Siglec-9 Antibodies as Anti-Cancer Agents
As of writing, no Siglec-9 antibody conjugate is in clinical trials for treating cancer. In the case of acute myeloid leukemia (AML), surface Siglec-9 is expressed to a similar magnitude as CD33 in 7 out of 21 AML cases [37]. Antagonistic antibodies can undergo endocytosis; thus, antibody–drug conjugate design can take advantage of the constitutive recycling of Siglec-9 as seen in certain malignancies such as AML (Fig. 1D). Similar strategies to deplete Siglec-expressing tumors have also been employed, for example, gemtuzumab ozogamicin is a calicheamicin-conjugated humanized anti-CD33 mAb (IgG4) used for delivering cytotoxic payloads into Siglec-3 expressing target AML cells and is administered at a lower dosage of 3 mg/m2 to reduce the chance of human fatalities [49]. In another approach, an inert antibody fused to sialidase, anti-HER2 sialidase, cleaves off sialic acids recognized by Siglec-7/-9 and has resulted in NK-mediated killing of HER2-positive breast tumor cells [50]. However, many factors such as affinity, epitope specificity, and receptor occupancy still remain unclear for inert Siglec-9-directed antibodies, which bind but do not activate ITIM inhibitory signaling.
Conclusion
Siglec-9 is emerging as an immune checkpoint target for cancer immunotherapy [51•]. In this review, we have highlighted some controversies and certain differentiating features of a binding sialic acid ligand and an antagonistic Siglec-9 antibody to mount anti-tumor immunity. These features include exposed sialic acid in glycan-bearing groups and reduce surface Siglec-9 receptor density and endocytic and higher affinity. It is also clear that agonistic Siglec-9 antibodies are more appropriate for anti-inflammatory therapies such as depletion of neutrophils associated with severe COVID-19 [52]. On the other hand, little is known about the binding affinity range distinguishing agonistic, inert, and antagonistic Siglec-9-directed antibodies. Nonetheless, high (nM) affinity antibodies have been associated with antagonism, and in this case, antagonistic Siglec-9 antibodies can piggyback on the endocytosing Siglec-9 receptors to either reduce surface Siglec-9 receptor density or deliver cytotoxic drugs into the cancer cells, in an epitope non-specific manner.
The challenges of identifying antagonistic Siglec-9 antibodies still remain largely empirical. There is no biophysical guideline that can reliably predict antagonism. In this review, we have observed high-affinity antibodies (KD ~ low 10–9 M) as likely antagonistic for Siglec-9 antibodies that may reduce surface Siglec-9 receptor expression and inhibit a “brake” in immunosuppression in cancer. But more studies are still needed to correlate higher binding affinities and Siglec-9 antibody antagonism. Also, whether higher affinity sialic acid ligands can synergize with antagonism either alone or as a binary antibody-sialic acid conjugate instead of sialidase conjugate. Presently, only full functional characterization can confirm that the antibody functions as an antagonist—a cancer cell killing approach. In this regard, we have included known characteristics of an antibody that may help guide the design of a new generation of Siglec-9-directed antibodies.
Data Availability
All data generated or analysed during the current review are available from the corresponding author on reasonable request.
Declarations
Conflict of Interest
Jun Hui Shawn Wang, Nan Jiang, and Amit Jain declare no conflict of interest. Jackwee Lim has received funding from the A*STAR Career Development Fund 2022 to study the Siglec family members.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
This article is part of the Topical collection on Evolving Therapies
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
1. Angata T Varki A Cloning, characterization, and phylogenetic analysis of Siglec-9, a new member of the CD33-related group of Siglecs: evidence for co-evolution with sialic acid synthesis pathways J Biol Chem 2000 275 22127 22135 10.1074/jbc.M002775200 10801860
2. Zhang JQ Nicoll G Jones C Crocker PR Siglec-9, a novel sialic acid binding member of the immunoglobulin superfamily expressed broadly on human blood leukocytes J Biol Chem 2000 275 22121 22126 10.1074/jbc.M002788200 10801862
3. Avril T Floyd H Lopez F Vivier E Crocker PR The membrane-proximal immunoreceptor tyrosine-based inhibitory motif is critical for the inhibitory signaling mediated by Siglecs-7 and -9, CD33-related Siglecs expressed on human monocytes and NK cells J Immunol Am Assoc Immunol 2004 173 6841 9
4. Jandus C Boligan KF Chijioke O Liu H Dahlhaus M Démoulins T Interactions between Siglec-7/9 receptors and ligands influence NK cell-dependent tumor immunosurveillance J Clin Inv. Am Soc Clin Inv 2014 124 1810 20 10.1172/JCI65899
5. Lizcano A Secundino I Dohrmann S Corriden R Rohena C Diaz S Erythrocyte sialoglycoproteins engage Siglec-9 on neutrophils to suppress activation Blood Am Soc Hematol 2017 129 3100 3110
6. •• Haas Q, Boligan KF, Jandus C, Schneider C, Simillion C, Stanczak MA, et al. Siglec-9 regulates an effector memory CD8+ T-cell subset that congregates in the melanoma tumor microenvironment. Cancer Immunol Res. American Association for Cancer Research Inc. 2019;7:707–18. This study highlights a combinatorial approach using both anti-Siglec-9 and anti-PD-1 to enhance CD8 + T cell cytotoxicity.
7. •• Rosenthal Arnon, Monroe Kate, Lee Seung-Joo. Anti-siglec-9 antibodies and methods of use thereof. Patent US20210095023A1. 2021. This study describes the use of CLIPS technology to map conformational epitopes recognized by anti-Siglec-9 clones 2D4, 2D5, 191240, E10–286, and 5C6.
8. Gonzalez-Gil A, Schnaar RL. Siglec ligands. Cells. 2021;10:1260.
9. Stanczak MA Siddiqui SS Trefny MP Thommen DS Boligan KF von Gunten S Self-associated molecular patterns mediate cancer immune evasion by engaging Siglecs on T cells J Clin Inv Am Soc Clin Inv 2018 128 4912 23 10.1172/JCI120612
10. Belisle JA, Horibata S, Jennifer GA, et al. Identification of Siglec-9 as the receptor for MUC16 on human NK cells, B cells, and monocytes. Mol Cancer. 2010;9:118.
11. Läubli H, Pearce OMT, Schwarz F, Siddiqui SS, Deng L, Stanczak MA, et al. Engagement of myelomonocytic Siglecs by tumor-associated ligands modulates the innate immune response to cancer. Proc Natl Acad Sci U S A. National Academy of Sciences. 2014;111:14211–6.
12. Ibarlucea-Benitez I, Weitzenfeld P, Smith P, Ravetch J v. Siglecs-7/9 function as inhibitory immune checkpoints in vivo and can be targeted to enhance therapeutic antitumor immunity. Proc Natl Acad Sci U S A [Internet]. 2021;118.
13. Beatson R Tajadura-Ortega V Achkova D Picco G Tsourouktsoglou TD Klausing S The mucin MUC1 modulates the tumor immunological microenvironment through engagement of the lectin Siglec-9 Nat Immunol Nature Publishing Group 2016 17 1273 1281 10.1038/ni.3552
14. • Beatson R, Graham R, Grundland Freile F, et al. Cancer-associated hypersialylated MUC1 drives the differentiation of human monocytes into macrophages with a pathogenic phenotype. Commun Biol. 2020;3(1):644. This study describes Siglec-9 signaling pathways that drive the formation of TAMs in breast cancer patients.
15. • Rodriguez E, Boelaars K, Brown K, et al. Sialic acids in pancreatic cancer cells drive tumour associated macrophage differentiation via the Siglec receptors Siglec-7 and Siglec-9. Nat Commun. 2021;12(1):1270. This study describes Siglec-9 signaling pathways that drive the formation of TAMs in pancreatic ductal adenocarcinoma.
16. Läubli H, Alisson-Silva F, Stanczak MA, Siddiqui SS, Deng L, Verhagen A, et al. Lectin galactoside-binding soluble 3 binding protein (LGALS3BP) is a tumor-associated immunomodulatory ligand for CD33-related siglecs. J Biol Chem Am Soc Biochem Mol Biol Inc. 2014;289:33481–91.
17. Miyazaki K Sakuma K Kawamura YI Izawa M Ohmori K Mitsuki M Colonic epithelial cells express specific ligands for mucosal macrophage immunosuppressive receptors Siglec-7 and -9 J Immunol Am Assoc Immunolog. 2012 188 4690 700
18. •• Cornen Stéphanie, Rossi Benjamin, Wagtmann Nicolai, Gauthier Laurent. Siglec-9-neutralizing antibodies. Patent US20200369765A1. 2020. This study describes the Siglec-9 blocking antibodies in reducing tumor size in colorectal adenocarcinoma, HT29 cells.
19. Yamakawa N, Yasuda Y, Yoshimura A, et al. Discovery of a new sialic acid binding region that regulates Siglec-7. Sci Rep. 2020;10(1):8647.
20. Choi H, Ho M, Adeniji OS, Giron L, Bordoloi D, Kulkarni AJ, Puchalt AP, Abdel-Mohsen M, Muthumani K. Development of Siglec-9 blocking antibody to enhance anti-tumor immunity. Front Oncol. 2021;11:778989.
21. •• Kar Muthumani, Mohamed Abdel-Mohsen, David Weiner, Shyam Somasundaram. Monoclonal antibodies against Siglec-9 and use thereof for immunotherapy. Patent WO2021247821. 2021. This study describes the Siglec-9 blocking antibodies in promoting NK cell cytotoxicity against chronic myelogenous leukemia, K562 cells.
22. Yu H Gonzalez-Gil A Wei Y Fernandes SM Porell RN Vajn K Siglec-8 and Siglec-9 binding specificities and endogenous airway ligand distributions and properties Glycobiology Oxford Univ Press 2017 27 657 668 10.1093/glycob/cwx026
23. Delaveris CS Wilk AJ Riley NM Stark JC Yang SS Rogers AJ Synthetic Siglec-9 agonists inhibit neutrophil activation associated with COVID-19 ACS Cent Sci Am Chem Soc 2021 7 650 657 10.1021/acscentsci.0c01669
24. Kucka K, Wajant H. Receptor oligomerization and its relevance for signaling by receptors of the tumor necrosis factor receptor superfamily. Front Cell Dev Biol. 2021;8:615141.
25. Han S Collins BE Bengtson P Paulson JC Homomultimeric complexes of CD22 in B cells revealed by protein-glycan cross-linking Nat Chem Biol 2005 1 93 97 10.1038/nchembio713 16408005
26. Cornish AL Freeman S Forbes G Characterization of siglec-5, a novel glycoprotein expressed on myeloid cells related to CD33 Blood. 1998 92 6 2123 2132 10.1182/blood.V92.6.2123 9731071
27. Floyd H Ni J Cornish AL Zeng Z Liu D Carter KC Siglec-8 A novel eosinophil-specific member of the immunoglobulin superfamily J Biol Chem 2000 275 861 6 10.1074/jbc.275.2.861 10625619
28. Angata T Kerr SC Greaves DR Varki NM Crocker PR Varki A Cloning and characterization of human Siglec-11: a recently evolved signaling molecule that can interact with SHP-1 and SHP-2 and is expressed by tissue macrophages, including brain microglia J Biol Chem 2002 277 24466 24474 10.1074/jbc.M202833200 11986327
29. Nicoll G Ni J Liu D Klenerman P Munday J Dubock S Identification and characterization of a novel siglec, siglec-7, expressed by human natural killer cells and monocytes J Biol Chem 1999 274 34089 34095 10.1074/jbc.274.48.34089 10567377
30. Delaveris CS, Chiu SH, Riley NM, Bertozzi CR. Modulation of immune cell reactivity with CIS binding Siglec agonists. Proc Natl Acad Sci U S A. 2021;118:e2012408118.
31. Wilson NS Yang B Yang A Loeser S Marsters S Lawrence D An Fcγ receptor-dependent mechanism drives antibody-mediated target-receptor signaling in cancer cells Cancer Cell Cell Press 2011 19 101 113 10.1016/j.ccr.2010.11.012
32. Carlin AF, Uchiyama S, Chang YC, Lewis AL, Nizet V, Varki A. Molecular mimicry of host sialylated glycans allows a bacterial pathogen to engage neutrophil Siglec-9 and dampen the innate immune response. Blood. 2009;113(14):3333–6.
33. Chu S, Zhu X, You N, Zhang W, Zheng F, Cai B, Zhou T, Wang Y, Sun Q, Yang Z, Zhang X, Wang C, Nie S, Zhu J, Wang M. The fab fragment of a human Anti-Siglec-9 monoclonal antibody suppresses LPS-induced inflammatory responses in human macrophages. Front Immunol. 2016;7:649.
34. Mamat U, Wilke K, Bramhill D, et al. Detoxifying Escherichia coli for endotoxin-free production of recombinant proteins. Microb Cell Fact. 2015;14:57.
35. von Gunten S Schaub A Vogel M Stadler BM Miescher S Simon HU Immunologic and functional evidence for anti-Siglec-9 autoantibodies in intravenous immunoglobulin preparations Blood 2006 108 4255 4259 10.1182/blood-2006-05-021568 16902148
36. von Gunten S Yousefi S Seitz M Jakob SM Schaffner T Seger R Siglec-9 transduces apoptotic and nonapoptotic death signals into neutrophils depending on the proinflammatory cytokine environment Blood 2005 106 1423 1431 10.1182/blood-2004-10-4112 15827126
37. Biedermann B Gil D Bowen DT Crocker PR Analysis of the CD33-related siglec family reveals that Siglec-9 is an endocytic receptor expressed on subsets of acute myeloid leukemia cells and absent from normal hematopoietic progenitors Leuk Res 2007 31 211 220 10.1016/j.leukres.2006.05.026 16828866
38. Mayes PA, Hance KW, Hoos A. The promise and challenges of immune agonist antibody development in cancer. Nat Rev Drug Discov. 2018;17(7):509–27.
39. Weiss A Manger B Imboden J Synergy between the T3/antigen receptor complex and Tp44 in the activation of human T cells J Immunol. 1986 137 3 819 825 3088111
40. Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S. Stimulation of CD25(+)CD4(+) regulatory T cells through GITR breaks immunological self-tolerance. Nat Immunol. 2002;3(2):135–42.
41. Kjaergaard J Tanaka J Kim JA Rothchild K Weinberg A Shu S Therapeutic efficacy of OX-40 receptor antibody depends on tumor immunogenicity and anatomic site of tumor growth Cancer Res. 2000 60 19 5514 5521 11034096
42. Bulliard Y Jolicoeur R Zhang J Dranoff G Wilson NS Brogdon JL OX40 engagement depletes intratumoral Tregs via activating FcγRs, leading to antitumor efficacy Immunol Cell Biol Nature Publishing Group 2014 92 475 480 10.1038/icb.2014.26
43. •• Heinz Laubli, Simone Schmitt, Christoph Esslinger. Anti-Siglec-9 antibody molecules. Patent WO2021094545. 2021. This study describes Siglec-9 blocking antibody which activates CD4 + and CD8 + T cells, as well as increased cytotoxicity of NK92 cell line against chronic myelogenous leukemia, K562 cells. Additionally, they showed that weak binding anti-Siglec-9 clone 5C6 demonstrated no functional activity.
44. Zhang D, Goldberg MV, Chiu ML. Fc engineering approaches to enhance the agonism and effector functions of an Anti-OX40 antibody. J Biol Chem Am Soc Biochem Mol Biol Inc. 2016;291:27134–46.
45. Vaddepally RK, Kharel P, Pandey R, Garje R, Chandra AB. Review of indications of FDA approved immune checkpoint inhibitors per NCCN guidelines with the level of evidence. Cancers (Basel). 2020;12(3):738.
46. Ahmed SR, Petersen E, Patel R, Migden MR. Cemiplimab-rwlc as first and only treatment for advanced cutaneous squamous cell carcinoma. Expert Rev Clin Pharmacol. 2019;12(10):947–51.
47. Barone A, Hazarika M, Theoret MR, et al. FDA approval summary: pembrolizumab for the treatment of patients with unresectable or metastatic melanoma. Clin Cancer Res. 2017;23(19):5661–5.
48. Zhang Q, Huo GW, Zhang HZ, Song Y. Efficacy of pembrolizumab for advanced/metastatic melanoma: a meta-analysis. Open Med (Wars). 2020;15(1):447–56.
49. Norsworthy KJ Ko C-W Lee JE Liu J John CS Przepiorka D FDA approval summary: Mylotarg for treatment of patients with relapsed or refractory CD33-positive acute myeloid leukemia Oncologist Oxford Univ Press (OUP) 2018 23 1103 1108
50. Xiao H, Woods EC, Vukojicic P, Bertozzi CR. Precision glycocalyx editing as a strategy for cancer immunotherapy. Proc Natl Acad Sci U S A. Natl Acad Sci. 2016;113:10304–9.
51. • Wu Y, Huang W, Xie Y, Wang C, Luo N, Chen Y, Wang L, Cheng Z, Gao Z, Liu S. Siglec-9, a putative immune checkpoint marker for cancer progression across multiple cancer types. Front Mol Biosci. 2022;9:743515. This study highlights the broad application of Siglec-9 as an immune checkpoint target across various cancers.
52. Lim J, Puan KJ, Wang LW, et al. Data-driven analysis of COVID-19 reveals persistent immune abnormalities in convalescent severe individuals. Front Immunol. 2021;12:710217.
53. Jumper J Evans R Pritzel A Green T Figurnov M Ronneberger O Highly accurate protein structure prediction with AlphaFold Nature Nature Research 2021 596 583 589 10.1038/s41586-021-03819-2
54. Chodorge M Züger S Stirnimann C Briand C Jermutus L Grütter MG A series of Fas receptor agonist antibodies that demonstrate an inverse correlation between affinity and potency Cell Death Differ 2012 19 1187 1195 10.1038/cdd.2011.208 22261618
55. Hudak JE Canham SM Bertozzi CR Glycocalyx engineering reveals a Siglec-based mechanism for NK cell immunoevasion Nat Chem Biol Nature Publishing Group 2014 10 69 75 10.1038/nchembio.1388
56. Khatua B Bhattacharya K Mandal C Sialoglycoproteins adsorbed by Pseudomonas aeruginosa facilitate their survival by impeding neutrophil extracellular trap through siglec-9 J Leukoc Biol Wiley 2012 91 641 655 10.1189/jlb.0511260
57. Rodrigues E, Jung J, Park H, et al. A versatile soluble siglec scaffold for sensitive and quantitative detection of glycan ligands. Nat Commun. 2020;11:5091.
58. Mimoto F Katada H Kadono S Igawa T Kuramochi T Muraoka M Engineered antibody Fc variant with selectively enhanced FcγRIIb binding over both FcγRIIaR131 and FcγRIIaH131 Protein Eng Des Sel 2013 26 589 598 10.1093/protein/gzt022 23744091
59. Diebolder CA, Beurskens FJ, de Jong RN, Koning RI, Strumane K, Lindorfer MA, et al. Complement is activated by IgG hexamers assembled at the cell surface. Science (1979). Am Assoc Adv Sci. 2014;343:1260–3.
| 36445569 | PMC9707166 | NO-CC CODE | 2022-12-01 23:21:38 | no | Curr Oncol Rep. 2022 Nov 29;:1-9 | utf-8 | Curr Oncol Rep | 2,022 | 10.1007/s11912-022-01347-4 | oa_other |
==== Front
Obes Surg
Obes Surg
Obesity Surgery
0960-8923
1708-0428
Springer US New York
36441488
6365
10.1007/s11695-022-06365-6
Original Contributions
Long-term Reported Outcomes Following Primary Laparoscopic Sleeve Gastrectomy
http://orcid.org/0000-0002-6658-8981
Sakran Nasser [email protected]
123
Soifer Kim [email protected]
12
Hod Keren [email protected]
4
Sherf-Dagan Shiri [email protected]
56
Soued Sharon [email protected]
12
Kessler Yafit [email protected]
15
Adelson Dana [email protected]
1
Biton Reut [email protected]
1
Buchwald J. N. [email protected]
7
Goitein David [email protected]
189
Raziel Asnat [email protected]
1
1 grid.414003.2 0000 0004 0644 9941 Assia Medical Group, Assuta Medical Center, Tel Aviv, Israel
2 grid.22098.31 0000 0004 1937 0503 The Azrieli Faculty of Medicine Safed, Bar-Ilan University, Ramat Gan, Israel
3 grid.414321.1 0000 0004 0371 9846 Department of General Surgery, Holy Family Hospital, P.O. Box 8, Nazareth, Israel
4 grid.414003.2 0000 0004 0644 9941 Department of Academy and Research, Assuta Medical Center, Tel-Aviv, Israel
5 grid.411434.7 0000 0000 9824 6981 Department of Nutrition Sciences, School of Health Sciences, Ariel University, Ariel, Israel
6 grid.414003.2 0000 0004 0644 9941 Department of Nutrition, Assuta Medical Center, Tel Aviv, Israel
7 Division of Scientific Research Writing, Medwrite Medical Communications, Maiden Rock, WI USA
8 grid.12136.37 0000 0004 1937 0546 Sackler Faculty of Medicine, Tel-Aviv University, Tel Aviv, Israel
9 grid.413795.d 0000 0001 2107 2845 Department of Surgery C, Sheba Medical Center, Ramat Gan, Israel
28 11 2022
112
7 5 2022
30 10 2022
18 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Background
Long-term laparoscopic sleeve gastrectomy (LSG) outcomes in patients with obesity are scarce. We aimed to examine the outcomes and subjective experience of patients who underwent primary LSG with long-term follow-up.
Methods
The study is a retrospective observational analysis of patients who underwent primary LSG in a single center with 5–15 years of follow-up. Patients’ hospital chart data supplemented by a detailed follow-up online questionnaire and telephone interview were evaluated.
Results
The study sample included 578 patients (67.0% female) with 8.8 ± 2.5 years of mean follow-up, with a response rate to the survey of 82.8%. Mean baseline age and body mass index (BMI) were 41.9 ± 10.6 years and 42.5 ± 5.5 kg/m2, respectively. BMI at nadir was 27.5 ± 4.9 kg/m2, corresponding to a mean excess weight loss (EWL) of 86.9 ± 22.8%. Proportion of patients with weight regain, defined as nadir ≥ 50.0% EWL, but at follow-up < 50.0% EWL, was 34.6% (n = 200) and the mean weight regain from nadir was 13.3 ± 11.1 kg. BMI and EWL at follow-up were 32.6 ± 6.4 kg/m2 and 58.9 ± 30.1%, respectively. The main reasons for weight regain given by patients included “not following guidelines,” “lack of exercise,” “subjective impression of being able to ingest larger quantities of food in a meal,” and “not meeting with the dietitian.” Resolution of obesity-related conditions at follow-up was reported for hypertension (51.7%), dyslipidemia (58.1%) and type 2 diabetes (72.2%). The majority of patients (62.3%) reported satisfaction with LSG.
Conclusions
In the long term, primary LSG was associated with satisfactory weight and health outcomes. However, weight regain was notable.
Graphical Abstract
Keywords
Bariatric surgery
Sleeve gastrectomy
Obesity
Long-term follow-up
Weight regain
==== Body
pmcIntroduction
Laparoscopic sleeve gastrectomy (LSG) is the most frequently performed bariatric/metabolic surgery (BMS) procedure worldwide largely due to its straightforward operative technique, minimal alteration of anatomy, absence of gastrointestinal anastomoses and foreign material, successful short- and mid-term weight loss, low morbidity, and capacity for conversion to alternate BMS procedures [1–4]. Still, a paucity of long-term LSG studies with ≥ 7 years of mean follow-up has restricted the assessment of its sustained effectiveness and safety [5]. We herein present our long-term primary LSG results analyzing weight outcomes, obesity-associated disease changes, and patient satisfaction.
Patients and Methods
Analysis of a detailed online questionnaire addressing long-term LSG follow-up supplemented by hospital chart data was conducted. Inclusion criteria included patients aged 18 ≤ years old who underwent primary LSG in our center and were followed for at least 5 years after surgery. Exclusion criteria included patients who underwent another MBS after the LSG. All patients met the standard inclusion criteria for BMS (i.e., body mass index [BMI] ≥ 40.0 or ≥ 35.0 kg/m2 with weight-associated medical conditions) [6]. LSG technique has been previously described in detail [7]. Venous thromboembolic prophylaxis was initiated within 12 h of surgery and continued for 2 weeks [8]. A bariatric dietitian provided detailed dietary guidance during recovery [9]. Patients were instructed to attend routine follow-up meetings with the bariatric multidisciplinary team.
An online questionnaire containing open-ended and multiple-choice questions was administered electronically by email or SMS text and verified by a telephone interview, when needed. The questionnaire (Appendix) included information regarding weight outcomes, coexisting obesity-related conditions (hypertension, hyperlipidemia, type 2 diabetes mellitus [T2DM], obstructive sleep apnea), patient satisfaction, and reasons for weight regain. Follow-up data were obtained from the online questionnaire.
Percent excess weight loss (%EWL − ([initial weight − follow-up weight] / [initial weight − ideal body weight]) × 100) and percentage of total body weight loss (%TWL) (([initial weight − follow-up weight]/[initial weight]) × 100) were calculated. Weight regain was reported as the (1) proportion of patients that initially achieved ≥ 50% EWL at nadir, but then then fell to < 50% EWL at long-term follow-up; (2) weight gain in kilograms post nadir; (3) percentage of nadir weight; and (4) percentage of maximum weight lost. Information regarding post-LSG long-term weight loss and weight regain is presented for the whole population and for two subgroups of patients with follow-up of 5–10 years and ≥ 10 years. Patient privacy throughout the study was ensured through a password-protected de-identified participant code. The study protocol was approved by the institutional review board (# 2,012,021-ASMC).
Statistical Analysis
Analyses were performed using Microsoft Excel 2016 (Microsoft Corp. Redmond, WA) and the SPSS statistical package (IBM Corp. Released 2019. IBM SPSS Statistics for Windows, version 28.0. Armonk, NY: IBM Corp). Continuous variables were presented as means ± standard deviations (SD), while dichotomous and categorical variables were presented as percentages and frequencies. Measures of change from baseline were analyzed with the dependent-samples t-test or the Wilcoxon test, as appropriate. In addition, comparisons of baseline characteristics between subgroups were calculated by Mann–Whitney U test and chi-square tests, as appropriate. To assess consistency between self-reported coexisting obesity-related conditions (i.e., diabetes, hypertension, hyperlipidemia, and obstructive sleep apnea) with diagnoses obtained from electronic medical records, Cohen’s kappa was calculated for baseline data. The level of agreement was classified as poor (≤ 0.01), slight (0.01–0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), and almost perfect (0.81–1.00) [10–12]. All statistical tests were two-tailed and statistical significance was set at p < 0.05.
Results
Patient Baseline Characteristics
Of the 698 patients who were reached to participate in the study and met the inclusion criteria, 578 gave their consent and responded to the survey (response rate of 82.8%). Furthermore, 33 patients were excluded from the study analysis due to conversion to another BMS. Most of these patients converted to one anastomosis gastric bypass (OAGB) (51.5%), Roux-en-Y gastric bypass (RYGB) (30.3%), and additional LSG (18.2%), while the leading reported reasons for additional MBS were weight regain (69.7%), and reflux, frequent vomiting, and eating difficulties (45.5%).
Overall, the study sample included 578 (67.0% female) patients who responded to the questionnaires, and underwent primary LSG between May 1, 2006, and December 31, 2016, in our center. Their mean pre-surgery BMI and age were 42.5 ± 5.5 kg/m2 and 41.9 ± 10.6 years, respectively. The most prevalent weight-related medical illnesses pre-surgery were dyslipidemia 43.9% (n = 254), followed by hypertension 27.7% (n = 160), T2DM 25.3% (n = 146), and obstructive sleep apnea 14.2% (n = 82). The most common additional procedure performed during primary LSG was a laparoscopic cholecystectomy (7.1%, n = 41), followed by hiatal hernia repair (6.6%, n = 38) (Table 1). The mean follow-up period was 8.8 ± 2.5 years (range: 5–14.4 years). Patients who responded to the survey had a similar pre-surgical age, BMI, and gender distribution in comparison to non-respondents (41.9 ± 10.6 vs 43.3 ± 11.9 years, p = 0.235; 42.5 ± 5.5 vs 42.9 ± 5.6 kg/m2, p = 0.463; 67 vs 70% female, p = 0.593; respectively).Table 1 Baseline patient characteristics and additional procedures during primary LSG
Characteristics Total LSG population (n = 578)
Age (yrs), mean ± SD 41.9 ± 10.6
Initial weight (kg), mean ± SD 118.9 ± 19.6
Initial height (m), mean ± SD 1.67 ± 0.09
Initial body mass index (k/m2), mean ± SD 42.5 ± 5.5
Females, n (%) 387 (67.0)
Female to male ratio 2.0:1.0
Smoking status, n (%) 105 (18.2)
Marital status, n (%)
• Married 424 (73.4)
• Divorced 74 (12.8)
• Other 80 (13.8)
Prevalent associated medical conditions, n (%)
• Hyperlipidemia 254 (43.9)
• Hypertension 160 (27.7)
• Obstructive sleep apnea 82 (14.2)
• Type 2 diabetes 146 (25.3)
Additional laparoscopic procedures during LSG, n (%)
• Cholecystectomy 41 (7.1)
• Hiatal hernia repair 38 (6.6)
• Umbilical or ventral hernia repair w/ mesh 7 (1.2)
All data are based on electronic medical records
Weight Outcomes
Mean nadir weight at a mean 12.8 ± 13.4 months following surgery was 77.9 ± 15.5 kg. This weight loss equated to mean nadir TWL of 34.3 ± 8.1% and mean EWL of 86.9 ± 22.8%. Overall, only 5.5% (n = 32) of primary LSG patients in our sample had insufficient weight loss (i.e., EWL < 50.0%) at nadir. Mean weight at 8.8 ± 2.5 years of follow-up was 91.3 ± 20.4 kg for the whole study population. This corresponded to a mean TWL of 23.1 ± 11.4% and mean EWL of 58.9 ± 30.1%. Significant differences between patients with 5–10 and ≥ 10 years of follow-up were found for mean weight at present (89.5 ± 19.2 vs 95.4 ± 22.5 kg, p = 0.010), and corresponding mean TWL (24.1 ± 10.6 vs 20.9 ± 12.6%, p = 0.005), and mean EWL (61.7 ± 28.4 vs 52.8 ± 32.8, p = 0.002).
The proportion of patients who reported on any weight gained from nadir was 91.7% (n = 530). The mean time interval reported between LSG and weight regain in years was 3.4 ± 2.0, and was similar between patients with different time intervals of follow-up. Mean weight regain was 13.3 ± 11.1 kg for the whole study population, while less weight regain was reported in the group with 5–10 years of follow-up in comparison to the group with ≥ 10 years of follow-up (12.2 ± 9.5 kg vs 15.6 ± 13.7, p = 0.041).
Rate of weight regain, defined as nadir ≥ 50.0% EWL, but at follow-up < 50.0% EWL was 34.6% (n = 200), while this rate was lower in the group with 5–10 years of follow-up in comparison to the group with ≥ 10 years of follow-up (n = 125 (31.3%) vs n = 75 (42.1%), p = 0.007) (Table 2).Table 2 Post-LSG long-term weight loss and weight regain
Parameter Total LSG population (n = 578) 5–10 years of follow-up
(n = 400) ≥ 10 years of follow-up
(n = 178) P-value between 5–10 and ≥ 10 yrs follow-up
Short follow-up
Time to achieve nadir (lowest weight from LSG) (mo.), mean ± SD 12.8 ± 12.4 12.9 ± 12.0 12.4 ± 12.1 0.826
Nadir BMI (lowest BMI attained) (kg/m2), mean ± SD 27.5 ± 4.9 27.1 ± 4.6 28.4 ± 5.3 0.011
Nadir weight (kg), mean ± SD 77.9 ± 15.5 77.2 ± 15.2 79.7 ± 16.1 0.153
Nadir weight loss (kg), mean ± SD 40.9 ± 12.5 40.9 ± 12.1 40.8 ± 13.4 0.708
Nadir TWL (%), mean ± SD 34.3 ± 8.1 34.5 ± 8.1 33.7 ± 8.2 0.475
Nadir EWL (%) mean ± SD 86.9 ± 22.8 88.3 ± 23.1 83.7 ± 22.2 0.075
Insufficient weight loss, n (%) 32 (5.5) 21 (5.3) 11 (6.2) 0.696
Short follow-up successful patients, n (%)§§ 541 (94.4) 374 (93.5) 167 (93.8) 0.696
Short follow-up insufficient weight loss, n (%)§§§ 32 (5.5) 21 (5.3) 11 (6.2) 0.696
Long follow-up
Follow-up duration (yrs.), mean ± SD 8.8 ± 2.5 7.3 ± 1.9 11.9 ± 0.9 < 0.001
Current weight (kg), mean ± SD 91.3 ± 20.4 89.5 ± 19.2 95.4 ± 22.5 0.010
Current BMI (kg/m2), mean ± SD 32.6 ± 6.4 32.0 ± 5.9 34.1 ± 7.1 0.002
Weight loss (kg), mean ± SD 27.5 ± 14.5 28.5 ± 13.5 25.2 ± 16.3 0.005
Current TWL (%), mean ± SD 23.1 ± 11.4 24.1 ± 10.6 20.9 ± 12.6 0.005
Current EWL (%), mean ± SD 58.9 ± 30.1 61.7 ± 28.4 52.8 ± 32.8 0.002
Weight regain
Proportion that gained any weight from nadir, n (%)* 530 (91.7) 369 (92.2) 161 (90.4) 0.308
Time after LSG to weight regain (yrs), mean ± SD† 3.4 ± 2.0 3.3 ± 1.8 3.8 ± 2.4 0.123
Weight regain post nadir (kg), mean ± SD†,‡‡ 13.3 ± 11.1 12.2 ± 9.5 15.6 ± 13.7 0.041
% weight regain (% nadir weight), mean ± SD†,‡ 17.3 ± 14.6 16.1 ± 12.7 19.8 ± 17.9 0.097
Proportion weight regain (≥ 15.0% of nadir weight)†,‡ 50.9 (270/530) 48.8 (180/369) 55.9 (90/161) 0.156
% weight regain (%maximum weight lost), mean ± SD†,§ 33.4 ± 26.1 30.8 ± 22.7 39.2 ± 31.7 0.017
Proportional distribution of weight regain (% of maximum weight lost), n (%)§
No regain* 45 (7.8) 27 (7.3) 18 (11.2) 0.308
< 25%† 213 (37.0) 156 (42.3) 57 (35.4) 0.150
≥ 25– < 50%† 181 (31.5) 133 (36.0) 48 (29.8) 0.165
≥ 50%† 136 (23.6) 80 (21.7) 56 (34.8) 0.002
Long follow-up successful patients, n (%)§§§§ 345 (59.9) 253 (63.3) 92 (51.7) 0.008
Long follow-up insufficient weight loss, n (%)§§§§§ 231 (40.0) 145 (36.3) 86 (48.3) 0.008
Proportion of successful patients at nadir that succeeded at long follow-up†† 341 (59.0) 249 (62.3) 92 (51.7) 0.007
Proportion of successful patients at nadir that failed at long follow-up††† 200 (34.6) 125 (31.3) 75 (42.1) 0.007
BMI, body mass index; TWL, total weight loss; EWL, excess weight loss
*Data were available for 575 patients
†Data were available for 530 patients
‡‡Weight regain post nadir (kg) was calculated based on the following equation: (current weight − nadir weight)
‡% weight regain of nadir weight was calculated based on the following equation: ([current weight − nadir weight] / [nadir weight]) × 100
§% weight regain was of maximum weight lost was calculated based on the following equation: ([current weight − nadir weight] / (baseline weight − nadir weight]) × 100
Proportional distribution of weight regain (< 25%, ≥ 25– < 50%, ≥ 50%) was available for 530 patients since those who did not regain weight (n = 45) were excluded from these analyses
§§Short follow-up successful patients, n (%) = patients with Nadir %EWL ≥ 50%, n (%), data were available for 573 patients
§§§Short follow-up insufficient weight loss, n (%) = patients with Nadir %EWL < 50%, n (%), data were available for 573 patients
§§§§Long follow-up successful patients, n (%) = patients with current weight %EWL ≥ 50%, n (%). Data were available for 576 patients
§§§§§Long follow-up insufficient weight loss, n (%) = patients with current weight %EWL < 50%, data were available for 576 patients
††Proportion of successful patients at nadir that succeeded at long follow-up = [number of patients with nadir EWL ≥ 50% that succeeded also at long follow-up as their current EWL ≥ 50% / number of patients with nadir EWL > 50%] × 100
†††Proportion of successful patients at nadir that failed at long follow-up = [number of patients with nadir EWL ≥ 50% that failed at long follow-up as their current EWL < 50% / number of patients with nadir EWL > 50%] × 100
Baseline characteristics of patients whose weight loss was successful at nadir (EWL ≥ 50%) but failed at long-term follow-up (EWL < 50%) were similar, except for higher initial weight (121.8 ± 21.0 kg) and BMI (43.6 ± 5.2 kg/m2) and lower proportion of current smoking (12.0%, n = 24) as compared to patients that succeeded both at nadir and at long-term follow-up (115.9 ± 17.6 kg, 41.3 ± 4.9 kg/m2 and 24.0% (n = 82), P = 0.002, p < 0.001, and P = 0.001, respectively).
Primary LSG patient-reported reasons for weight regain are presented in Table 3. Over half of all patients (56.8%, n = 302) with weight regain reported that they “did not follow the MBS guidelines” recommended by their health professionals after LSG. The second most frequently reported reason for regain was “lack of exercise” (51.9%, n = 276), followed by “patients’ subjective impression of being able to ingest larger quantities of food in a meal” (41.1%, n = 218), and “not meeting with a dietitian” (38.6%, n = 205). Also, weight regain due to “Covid-19 consequences” was mentioned in textual responses by some patients (1.7%, n = 9). Despite the fact that 91.7% (n = 530) reported at least some weight regain, the majority (62.3%, n = 360) were “sufficiently satisfied” with the procedure and “would choose LSG again.” Interestingly, similarity between patients with different follow-up time intervals was found with respect to of the reasons reported for weight regain (Table 3).Table 3 Patient-reported reasons for weight regain
Reasons for weight regain*† Total LSG population
(n = 578)
n (%) LSG population with 5–10 years of follow-up
(n = 400)
n (%) LSG population with ≥ 10 years of follow-up
(n = 178)
n (%) P-value between 5–10 and ≥ 10 years of follow-up
• Did not follow guidelines 302 (56.8) 199 (49.8) 103 (57.9) 0.123
• Lack of exercise 276 (51.9) 203 (50.7) 73 (41.0) 0.018
• Sleeve enlargement¥ 218 (41.1) 138 (34.5) 80 (44.9) 0.032
• Did not meet with dietitian 205 (38.6) 152 (38.0) 53 (29.7) 0.038
• Difficulty controlling portion size, sweets, or alcoholic beverages 21 (3.9) 16 (4.0) 5 (2.8) 0.632
• Mental crisis, emotional eating 19 (3.5) 12 (3.0) 7 (3.9) 0.618
• COVID-19 consequences 9 (1.7) 7 (1.8) 2 (1.1) 0.727
• Consequences of other diseases 6 (1.1) 4 (1.0) 2 (1.1) > 0.999
• Antidepressants, steroids 7 (1.3) 5 (1.2) 2 (1.1) > 0.999
• Intentional weight gain 6 (1.1) 6 (1.5) 0 (0.0) 0.184
• Lack of support 5 (0.9) 2 (0.5) 3 (1.7) 0.179
• Hypothyroidism 3 (0.5) 1 (0.3) 2 (1.1) 0.232
• As a consequence of IVF treatments, pregnancies, births‡ 20/379 (5.3) 14 (3.5) 6 (3.4) > 0.999
• Smoking cessation 2 (0.4) 1 (0.3) 1 (0.6) 0.528
Patients sufficiently satisfied: would choose LSG again§ 360 (62.3) 250 (62.5) 110 (61.8) 0.926
COVID-19, coronavirus disease 2019; IVF, in vitro fertilization
*Data available only for patients who regain weight (n = 530)
†Each patient had the option of providing more than one answer
¥Patient’s subjective impression of being able to ingest larger quantities of food in a meal
‡Data available only for women who regain weight (n = 379)
§Data available for the whole study population (n = 578)
The questionnaire also offered possible reasons for choosing a bariatric procedure other than LSG in the future, such as “today there are more advanced procedures” (28.0%, n = 162, answered “yes” on this item) and “because I’m not satisfied with the LSG” (13.8%, n = 80, answered “yes” on this item).
Hospitalizations, Procedures, and Clinical Follow-up
Nine percent of patients (n = 52) visited the emergency room, and 12.3% (n = 71) were hospitalized at some point after primary LSG, not necessarily related to their surgery (Table 4). The majority of stated emergency room visits were related to BMS (78.8%, n = 41), with gallbladder stones reported as the most frequent cause (39%, n = 16), followed by abdominal pain, dehydration and vomiting (26.8%, n = 11), leak or necrosis (9.8%, n = 4), stenosis or stricture of the sleeve (7.3%, n = 3), bleeding (2.4%, n = 1), incisional hernia (2.4%, n = 1), iron deficiency anemia (2.4%, n = 1), and kidney stone (2.4%, n = 1).Table 4 Post-LSG clinical follow-up and surgical procedures
Post-primary LSG procedures Total LSG population
(n = 578)
n (%)
Readmission rate 52 (9.0)
Hospitalizations following LSG 71 (12.3)
Additional non-bariatric surgery 230 (39.8)
Upper endoscopy 179 (31.0)
Plastic surgery 56 (9.7)
Cholecystectomy 54 (9.3)
Gynecologic surgery 39 (6.7)
Orthopedic surgery 36 (6.2)
Hiatal/umbilical/ventral hernia repair 23 (4.0)
Dietitian follow-up
• Not at all 0 (0.0)
• Only in the 1–2 years after the LSG 260 (45.0)
• Intermittently during the years 84 (14.5)
• Routine follow-up 40 (6.9)
Blood tests performed
• Did not perform blood tests at all since LSG 8 (1.4)
• Only in the first 1–2 years after the LSG 37 (6.4)
• Intermittently over the years 214 (37.0)
• Routine follow-up 319 (55.2)
Additional non-bariatric surgery after LSG was reported by 39.8% of patients (n = 230). Upper endoscopy was the most common invasive procedure reported (31.0%, n = 179), followed by plastic surgeries (9.7%, n = 56), cholecystectomy (9.3%, n = 54), gynecologic procedures (6.7%, n = 39), orthopedic surgeries (6.2%, n = 36), and hiatal, umbilical, and ventral hernia repair (4.0%, n = 23). About half (47.2%) of the patients reported that they suffered from heartburn or took any drugs to treat heartburn. Only a minority of patients reported routine contact with the dietitian throughout long-term follow-up (6.9%, n = 40), and 55.2% (n = 319) reported undergoing routine blood tests after LSG (Table 4).
Change in Associated Medical Problems
Significant reductions from baseline in the proportion of LSG patients suffering from hypertension, dyslipidemia, and T2DM were observed (p < 0.001 for all). Rates of no change, improvement, and resolution for selected associated medical problems are depicted in Fig. 1.Fig. 1 Resolution of associated medical conditions following laparoscopic sleeve gastrectomy
Discussion
Long-term LSG Outcomes
The predominance of the LSG among other BMS procedures arose before its rates of long-term weight loss and revision were known. Until recently, ≥ 7-year mean LSG outcomes were scarce. In 2018, a meta-analysis by Clapp and colleagues [5] reported outcomes synthesized from prior studies incorporating 652 patients who completed ≥ 7 years of mean follow-up. From 2018 through 2021, several definitively long-term LSG outcome reports in patients with obesity were published (8 observational studies, 1 randomized controlled trial) [13–21]. Our current investigation studied outcomes in 578 LSG patients; to our knowledge, this is the largest single series of LSG outcomes analyzed at ≥ 8.8 years. Herein, we relate the key study findings of the current study to the aforementioned recent reports.
Weight Outcomes
The meta-analysis of studies with > 7 years of mean follow-up by Clapp et al. found that 72.2% of patients attained ≥ 50.0% EWL at follow-up. Nine other long-term studies published later also demonstrated significant sustained weight loss at their respective follow-up points. Six reported a follow-up weight loss comparable to that of the current study [13–15, 19, 20], although three of these studies had 3–4-year longer follow-up durations [15, 19, 20]. In three of the later studies, follow-up weight loss was markedly less than in the current study, with ≤ 50.0% total EWL: Major et al. reported on a significant 16.3 BMI-point reduction at nadir, but at 10-year follow-up EWL was 45.4% [21]; Gronroos et al.’s randomized controlled trial (RCT) reported a 47.0% EWL [17]; and Hauters et al., 42.0% EWL [18].
A study by Felsenreich and co-workers is notable for its achievement of a ≥ 15-year follow-up with 96.0% patient retention, nearly twice the follow-up duration of the current study and most of the other LSG studies [52]. The authors saw a stable 61.0% EWL at this extended endpoint, but at the cost of a high revision rate — almost half of the patients (49.1%) were converted to another procedure. The report of Fiorani et al. is also unique in the long-term LSG literature, in that, not only was there no weight regain, but weight loss continued beyond the 1-year nadir through 8 years of follow-up. This report, however, is limited by both the small initial sample size (44 patients undergoing LSG) and the high lost-to-follow-up rate of 75.0% in the LSG group [16].
Weight Regain
The cause of weight regain in LSG is multifactorial, influenced by individual physiology, dietary choices, activity level, culture, and genetics [22, 23]. In addition, assessment of BMS weight regain is not yet standardized, and there is wide variability in its reporting complicating outcome comparison [24–27]. Herein, we focused on only studies that had specifically defined and quantified weight regain.
Perhaps the most frequently used definition of weight regain in recent years is the proportion of patients that initially achieved ≥ 50.0% EWL but then regained weight to have < 50.0% EWL at follow-up. In a meta-analysis of long-term LSG data (≥ 7-year follow-up), Clapp and co-workers analyzed nine studies with respect to this measure, yielding a pooled weighted mean proportion of 27.8% (95% CI: 22.8, 32.7) with a range of 14.0–37.0% [5]. Our study had a similar sample size and follow-up period, and using this definition, weight regain was on the higher end, (34.6%) of the range reported by Clapp et al. This was because the current study’s subgroup of LSG patients with ≥ 10-year follow-up contained a significantly higher proportion of patients who regained weight (42.1%) relative to LSG patients with 5–10 years of follow-up (31.3%).
Perhaps in recognition of the arbitrary nature of the 50% EWL cutoff, other methods of assessing weight regain have been proposed (i.e., weight regain from nadir weight calculated as weight in kilograms, BMI, percentage of preoperative weight, percentage of nadir weight, and percentage of maximum weight lost). Using percentage of maximum weight lost, Capoccia et al. reported long-term weight regain results in 104 LSG patients with more than 5-year follow-up [14].
Weight regain ≥ 15.0– < 30.0% of maximum weight lost was considered mild, or > 30.0%, severe. By this classification, 25.0% and 22.0% experienced mild and severe weight regain, respectively. Using slightly different thresholds, our study reported 37.0% of patients as having experienced mild weight regain (< 25.0% of maximum weight lost), and 55.1% as having experienced significant weight regain (≥ 25.0% of maximum weight lost). The different thresholds confound the proportional comparison; however, the current study’s cohort had a mean percentage weight regain of maximum weight loss of 33.4%, slightly higher than the 31.5% reported by Capoccia et al. As should be expected, the current study’s subgroup of LSG patients with 5–10 years of follow-up, possibly a more apt comparator, had a slightly lower mean percentage weight regain of maximum weight lost than that reported by Capoccia et al. (30.8% vs 31.5%).
A final recent example of weight regain in LSG patients, calculated using yet another measure, is provided by Ben-Porat and colleagues. In their study of 212 patients, weight regain was defined as the proportion of patients regaining ≥ 15% of nadir weight [13]. According to this measure, exactly 50% of their patients were classified as experiencing weight regain. Similarly, using this definition, just over half (50.9%) of the current study’s entire cohort experienced significant weight regain (with 48.8% of LSG patients with 5–10-year follow-up experiencing weight regain vs 55.9% of patients with ≥ 10-year follow-up). In addition, weight regain from nadir weight calculated in kilograms in both studies was nearly equivalent (14.6 kg vs 13.3 kg). These comparative results parallel those of other long-term LSG patients over the long term, regardless of how they were measured and recorded (e.g., at clinical visit or by questionnaire) [28].
Resolution of Coexisting Conditions
In addition to significant and sustained weight loss, this study and most of the other LSG studies suggest that, at 7–15 years after the procedure, LSG provides meaningful resolution of obesity-related medical problems. Kraljevic et al. saw high rates of resolution in dyslipidemia (54.8%), hypertension (60.5%), and T2DM (61.0%) [20]. In the current study, good resolution of long-term coexisting conditions was found, including hypertension (51.7%), dyslipidemia (58.1%), and T2DM (72.2%). These rates were comparable to those reported by Capoccia et al. [14] and notably greater than those of Ben-Porat and Hauters, both including markedly smaller cohorts [13, 18].
In Felsenreich et al.’s study, although very few patients at baseline had any obesity-associated conditions, hypertension was resolved in 6/12 (50.0%) at ≥ 15 years of follow-up [15]. All patients in Ismail and co-worker’s cohort achieved complete resolution of hypertension, T2DM, and GERD between 6 months and 2 years but no long-term follow-up data was given [19]. Coexisting medical condition data provided in the included studies of the meta-analysis were too sparse for comparison, which is also true of other published work [16].
In Kraljevic et al., 52.0%, and in Hauters et al., 80.0% of the LSG cohort with baseline GERD experienced ongoing or worsened reflux at follow-up; of patients without preoperative reflux, 32.4% and 43.0%, respectively, developed de novo GERD in follow-up [18, 20]. While there are centers that have reported favorable mid-term LSG outcomes with respect to GERD [29, 30], the implications for complications due to both pre-existing and de novo GERD remain among the most important considerations for selection of this procedure. In the present study, about half of the patients reported that they suffered from heartburn or took any drugs to treat heartburn. However, this was reported subjectively with no available imaging test results to support it.
Patient Satisfaction and Reasons for Weight Regain
Several of the long-term LSG study comparators recorded quality of life (QoL) data using standardized assessment instruments [15, 16]. Others, as in the current study, assessed patient-reported characterization of the behaviors they believed influenced their weight regain through a questionnaire [13]. Study responses diverged in some of the parameters investigated, suggesting differing patient perspectives regarding whether maintaining a regimen of consulting with a dietitian or surgeon affected their LSG outcomes. The contrasting findings suggest the importance of further research to better clarify patient-reported experiences and behavioral choices in order to improve long-term BMS care.
Adherence to close follow-up of BMS patients is important but challenging beyond 2 years, where the rate is reported to be 40–62% [31–33]. We observed in the current study, as in most other BMS studies, that the rate of postoperative patient-reported dietitian interaction and routine blood testing was quite low. This finding confirmed the challenging nature of preserving interactive contact with patients over the long term. Designing studies that consider the input of patients regarding their weight regain over their years of long-term follow-up should provide insights that can be integrated into the longitudinal treatment regimen [34].
Limitations and Strengths
The major strengths of this study include the use of a large representative sample of LSG patients similar in baseline data composition to that of both a national and an international registry [35, 36]. However, there are some limitations to be acknowledged. First, data were collected mainly via a self-report questionnaire generated by the research team that has not been validated. Though it has become an acceptable method, initiated by the COVID-19 pandemic restrictions [37, 38], relying on patient-reported outcomes entails a potential for information, recall, and response bias. Nevertheless, Chao and colleagues observed in a study of 585 LSG and RYGB patients at 38 sites that patient-reported BMS outcomes had a high level of agreement with hospital records in early follow-up [39]. In this manner, to increase the validity of the data in the present study, some data were also collected by hospital charts and when needed data were verified by a telephone interview. Moreover, testing reliability of data collection is an important element of confidence in a study’s accuracy [11]. In our study, self-reported coexisting obesity-related conditions data at baseline presented fair to moderate agreement with the available data obtained from hospital records (Cohen’s kappa of 0.574, 0.559, 0.451, and 0.284 for reporting on T2DM, hypertension, obstructive sleep apnea, and dyslipidemia [p < 0.001], respectively). Nonetheless, reporting bias for follow-up data collection should be considered.
Second, the study lacked objective measurements such as a medication list, blood test results, and imaging test results, and it lacked a validated QoL instrument. Finally, study follow-up was performed during the COVID-19 pandemic which may have affected the study results. However, weight regain due to “Covid-19 consequences” was reported by only a few patients in the present study.
Conclusions
At long-term follow-up, primary LSG was associated with satisfactory weight and health outcomes. However, weight regain was notable. More efforts should be made to increase patients’ engagement in follow-up routines in the long-term following MBS.
Appendix
LSG Patient Experience and Satisfaction Questionnaire
Initials of your name and surname
Research code number
Age
Family status
- Married
- Single
- Divorced
- Widower
- Other5. Current weight (in kg)
6. Height (in cm)
7. Are you a smoker?
- Yes
- No
- I used to smoke in the past8. What is the minimal weight you achieved since the surgery? (in kg)
9. How long did it take to achieve the above weight? (in months)
10. What was the duration of the weight loss since the surgery? (in months)
11. Did you gain weight after the weight loss? (even if it’s few kilos)
- Yes
- No12. If yes, when did you started to gain weight? (in years)
13. If you gained weight since the surgery, what did you think was the main cause? (you can choose more than one of the options)
- I didn’t eat as recommended
- The sleeve enlarged
- I didn’t exercise
- I wasn’t followed by a dietitian
- Other14. Did you have to go to the ER in the post-operative period?
- Yes.
- No.15. If yes, explain what the reason was
16. Were you hospitalized in the post-operative period?
- Yes.
- No.17. If yes, explain what the reason was
18. Did you undergo another bariatric surgery before the sleeve gastrectomy?**
- Yes
- No19. If yes, which surgery did you undergo?**
- LSG – laparoscopic sleeve gastrectomy
- LAGB – Laparoscopic adjustable gastric banding
- SRVG—Silastic ring vertical gastroplasty20. If yes, how long before the sleeve gastrectomy? (in years)**
21. If yes, what was the reason to undergo the sleeve gastrectomy?**
- weight gain
- drug-resistant reflux and heartburn
- a complication such as a sleeve stenosis or a twisted sleeve
- vomiting and eating difficulties
- nutritional deficiencies
- other22. Did you undergo another surgery after the sleeve gastrectomy?**
23. If yes, how long after the sleeve gastrectomy? (in years)**
24. If yes, what surgery did you undergo?**
- sleeve reduction (an additional sleeve gastrectomy)
- bypass with a single anastomosis (mini-gastric bypass)
- bypass with two anastomoses (classic bypass)
- duodenal bypass with two anastomoses
- duodenal bypass with one anastomosis (SADI)
- addition of a gastric band
- other25. If yes, what was the reason to undergo another surgery?**
- weight gain.
- drug-resistant reflux and heartburn
- a complication such as a sleeve stenosis or a twisted sleeve
- vomiting and eating difficulties
- nutritional deficiencies
- other
Questions 26–29 are directed only to women
26. Did you get pregnant since the surgery?
- Yes
- No27. If yes, how many times have you gotten pregnant since the surgery?
28. Did you give birth after surgery?
- Yes
- No29. If yes, how many times did you give birth?
30. Did you suffer from gallstones before the surgery?
- Yes
- No
- I don’t know31. If yes, what is true for you?
- I suffered from gallstones before the surgery and I underwent a cholecystectomy when I underwent the sleeve gastrectomy
- I suffered from gallstones before the surgery and I underwent a cholecystectomy after the sleeve gastrectomy
- I suffered from gallstones before the surgery and I underwent a cholecystectomy before the sleeve gastrectomy32. Were you diagnosed with gallstones since the sleeve gastrectomy?
- Yes
- No
- I don’t know33. Did you undergo a cholecystectomy since the sleeve gastrectomy?
- Yes
- No34. Did you undergo any surgery since the sleeve gastrectomy?
- Yes
- No35. If yes, explain which surgery.
36. Do you suffer from heartburn or do you take any drugs to treat heartburn?
- Yes
- No37. Did you undergo a gastroscopy since the surgery?
- Yes
- No38. If yes, what were the findings?
39. Are you satisfied with the surgery? Grade from 1 to 10 (1 = not satisfied at all, 10 = very satisfied)
40. If you underwent another bariatric surgery, are you satisfied with the surgery? Grade from 1 to 10 (1 = not satisfied at all, 10 = very satisfied)**
41. If you could choose again the type of bariatric surgery, would you choose again a sleeve gastrectomy?
- Yes
- No42. If not, why? (you can choose more than one of the options)
- today there are more advanced procedures
- because I’m not satisfied from the surgery43. Did you suffer from diabetes type 2 before the surgery? (diagnosis by medical record/taking pills to treat diabetes)
- Yes
- No
- I don’t know44. If yes, what was the treatment? (you can choose more than one of the options)
- Diet
- Drugs
- Insulin
- Drugs and insulin
- Not treated
- I don’t know45. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement46. Did you suffer from hypertension before the surgery?
- Yes
- No
- I don’t know47. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement48. Did you suffer from sleep apnea or did you use a CPAP mask while asleep?
- Yes
- No
- I don’t know49. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement50. Did you suffer from hyperlipidemia before the surgery? (high cholesterol/triglycerides/LDL in your blood)
51. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement52. Were you followed by a dietitian after the surgery?
- No
- Only the first/second year after the surgery
- Intermittently during the years
- Regularly during the years, but not currently
- Regularly during the years to this day53. With which frequency did you undergo blood tests since the surgery?
-I didn’t undergo any blood test since the surgery
-I underwent blood tests only the first/second year after the surgery
-I underwent blood tests intermittently during the years
-I undergo blood tests regularly since the surgery
**These questions were asked to be marked only for patients who undergo another bariatric surgery in their past.
Acknowledgements
We thank TW McGlennon, Medwrite, Maiden Rock, WI, USA, for statistical analysis support.
Funding
Medwrite received a grant for assistance with manuscript development.
Declarations
Statement of Ethics
The study protocol and all procedures of the study were approved as compliant with the ethical standards of the Assuta Medical Center (IRB #43–20-ASMC).
Written Informed Consent
Written informed consent was obtained from all study participants.
Conflict of Interest
The authors declare no competing interests.
Human and Animal Rights
The study was performed in accord with the ethical standards of the 1964 Declaration of Helsinki and subsequent amendments.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
David Goitein and Asnat Raziel contributed equally to the study.
==== Refs
References
1. Angrisani L Santonicola A Iovino P Bariatric Surgery Survey 2018: similarities and disparities among the 5 IFSO chapters Obes Surg 2021 31 5 1937 1948 10.1007/s11695-020-05207-7 33432483
2. Angrisani L Santonicola A Iovino P IFSO Worldwide Survey 2016: primary, endoluminal, and revisional procedures Obes Surg 2018 10.1007/s11695-018-3450-2 30128644
3. English WJ DeMaria EJ Brethauer SA American Society for Metabolic and Bariatric Surgery estimation of metabolic and bariatric procedures performed in the United States in 2016 Surg Obes Relat Dis 2018 14 3 259 263 10.1016/j.soard.2017.12.013 29370995
4. Lo Menzo E Szomstein S Rosenthal RJ Changing trends in bariatric surgery Scand J Surg 2015 104 1 18 23 10.1177/1457496914552344 25269945
5. Clapp B Wynn M Martyn C Long term (7 or more years) outcomes of the sleeve gastrectomy: a meta-analysis Surg Obes Relat Dis 2018 14 6 741 747 10.1016/j.soard.2018.02.027 29625744
6. NIH conference Gastrointestinal surgery for severe obesity Consensus Development Conference Panel Ann Intern Med 1991 115 12 956 961 10.7326/0003-4819-115-12-956 1952493
7. Sakran N Raziel A Goitein O Laparoscopic sleeve gastrectomy for morbid obesity in 3003 patients: results at a high-volume bariatric center Obes Surg 2016 26 9 2045 2050 10.1007/s11695-016-2063-x 26757919
8. Telem DA Gould J Pesta C American Society for Metabolic and Bariatric Surgery: care pathway for laparoscopic sleeve gastrectomy Surg Obes Relat Dis 2017 13 5 742 749 10.1016/j.soard.2017.01.027 28416400
9. Sherf Dagan S Goldenshluger A Globus I Nutritional recommendations for adult bariatric surgery patients: clinical practice Adv Nutr 2017 8 2 382 394 10.3945/an.116.014258 28298280
10 Ranganathan P Pramesh CS Aggarwal R Common pitfalls in statistical analysis: measures of agreement Perspect Clin Res 2017 8 4 187 191 10.4103/picr.PICR_123_17 29109937
11 McHugh ML Interrater reliability: the kappa statistic Biochem Med (Zagreb) 2012 22 3 276 282 10.11613/BM.2012.031 23092060
12. Landis JR Koch GG The measurement of observer agreement for categorical data Biometrics 1977 33 1 159 174 10.2307/2529310 843571
13. Ben-Porat T Mashin L Kaluti D Weight loss outcomes and lifestyle patterns following sleeve gastrectomy: an 8-year retrospective study of 212 patients Obes Surg 2021 31 11 4836 4845 10.1007/s11695-021-05650-0 34403078
14 Capoccia D Coccia F Guarisco G Long-term metabolic effects of laparoscopic sleeve gastrectomy Obes Surg 2018 28 8 2289 2296 10.1007/s11695-018-3153-8 29497961
15. Felsenreich DM Artemiou E Steinlechner K Fifteen years after sleeve gastrectomy: weight loss, remission of associated medical problems, quality of life, and conversions to Roux-en-Y gastric bypass-long-term follow-up in a multicenter study Obes Surg 2021 31 8 3453 3461 10.1007/s11695-021-05475-x 34021882
16. Fiorani C Coles SR Kulendran M Long-term quality of life outcomes after laparoscopic sleeve gastrectomy and Roux-en-Y gastric bypass-a comparative study Obes Surg 2021 31 3 1376 1380 10.1007/s11695-020-05049-3 33064260
17. Gronroos S Helmio M Juuti A Effect of laparoscopic sleeve gastrectomy vs Roux-en-Y gastric bypass on weight loss and quality of life at 7 years in patients with morbid obesity: the SLEEVEPASS Randomized Clinical Trial JAMA Surg 2021 156 2 137 146 10.1001/jamasurg.2020.5666 33295955
18. Hauters P Dubart JW Desmet J Degolla R Roumain M Malvaux P Ten-year outcomes after primary vertical sleeve gastrectomy for morbid obesity: a monocentric cohort study Surg Endosc 2021 35 12 6466 6471 10.1007/s00464-020-08137-8 33140154
19. Ismail M Nagaraj D Rajagopal M Seven-year outcomes of laproscopic sleeve gastectomy in Indian patients with different classes of obesity Obes Surg 2019 29 1 191 196 10.1007/s11695-018-3506-3 30238216
20. Kraljevic M Cordasco V Schneider R Long-term effects of laparoscopic sleeve gastrectomy: what are the results beyond 10 years? Obes Surg 2021 31 8 3427 3433 10.1007/s11695-021-05437-3 33890225
21. Major P Stefura T Dziurowicz B Quality of life 10 years after bariatric surgery Obes Surg 2020 30 10 3675 3684 10.1007/s11695-020-04726-7 32535784
22. Athanasiadis DI Martin A Kapsampelis P Factors associated with weight regain post-bariatric surgery: a systematic review Surg Endosc 2021 35 8 4069 4084 10.1007/s00464-021-08329-w 33650001
23. El Ansari W Elhag W Weight regain and insufficient weight loss after bariatric surgery: definitions, prevalence, mechanisms, predictors, prevention and management strategies, and knowledge gaps-a scoping review Obes Surg 2021 31 4 1755 1766 10.1007/s11695-020-05160-5 33555451
24. Baig SJ Priya P Mahawar KK Shah S Indian Bariatric Surgery Outcome Reporting G Weight regain after bariatric surgery-a multicentre study of 9617 patients from Indian Bariatric Surgery Outcome Reporting Group Obes Surg 2019 29 5 1583 1592 10.1007/s11695-019-03734-6 30729366
25. Brethauer SA Kim J El Chaar M Standardized outcomes reporting in metabolic and bariatric surgery Obes Surg 2015 25 4 587 606 10.1007/s11695-015-1645-3 25802064
26. King WC Hinerman AS Belle SH Comparison of the performance of common measures of weight regain after bariatric surgery for association with clinical outcomes JAMA 2018 320 15 1560 1569 10.1001/jama.2018.14433 30326125
27. Voorwinde V Steenhuis IHM Janssen IMC Definitions of long-term weight regain and their associations with clinical outcomes Obes Surg 2020 30 2 527 536 10.1007/s11695-019-04210-x 31677016
28. Oved I Endevelt R Mardy-Tilbor L Health status, eating, and lifestyle habits in the long term following sleeve gastrectomy Obes Surg 2021 31 7 2979 2987 10.1007/s11695-021-05336-7 33829384
29. Garg H Aggarwal S Misra MC Mid to long term outcomes of Laparoscopic Sleeve Gastrectomy in Indian population: 3–7 year results - a retrospective cohort study Int J Surg 2017 48 201 209 10.1016/j.ijsu.2017.10.076 29122706
30. Melissas J Braghetto I Molina JC Gastroesophageal reflux disease and sleeve gastrectomy Obes Surg 2015 25 12 2430 2435 10.1007/s11695-015-1906-1 26428250
31. Owolabi EO Mac Quene T Louw J Telemedicine in surgical care in low- and middle-income countries: a scoping review World J Surg 2022 10.1007/s00268-022-06549-2 35428920
32. van Erkel FM Pet MJ Bossink EH Experiences of patients and health care professionals on the quality of telephone follow-up care during the COVID-19 pandemic: a large qualitative study in a multidisciplinary academic setting BMJ Open 2022 12 3 e058361 10.1136/bmjopen-2021-058361 35273062
33. Chao GF Bonham AJ Ross R Patient-reported comorbidity assessment after bariatric surgery: a potential tool to improve longitudinal follow-up Ann Surg 2021 10.1097/SLA.0000000000004841 35129487
34. Fried M Hainer V Basdevant A Interdisciplinary European guidelines for surgery for severe (morbid) obesity Obes Surg 2007 17 2 260 270 10.1007/s11695-007-9025-2 17476884
35. Israeli Society for Metabolic and Bariatric Surgery. National Center for Disease Control and Israeli Society for Metabolic and Bariatric Surgery Annual Summary 2020. https://www.health.gov.il/PublicationsFiles/Bariatric_2020.pdf
36. Brown W, Kow L, Shikora S, et al. Sixth IFSO Global Registry Report 2021. Henley-on-Thames: IFSO & Dendrite Clinical Systems Ltd; 2021. Available at: https://www.e-dendrite.com/IFSO6
37. Larjani S Spivak I Hao Guo M Preoperative predictors of adherence to multidisciplinary follow-up care postbariatric surgery Surg Obes Relat Dis 2016 12 2 350 356 10.1016/j.soard.2015.11.007 26806726
38. Wheeler E Prettyman A Lenhard MJ Adherence to outpatient program postoperative appointments after bariatric surgery Surg Obes Relat Dis 2008 4 4 515 520 10.1016/j.soard.2008.01.013 18586576
39. Aminian A Kashyap SR Wolski KE Patient-reported outcomes after metabolic surgery versus medical therapy for diabetes: insights from the STAMPEDE Randomized Trial Ann Surg 2021 274 3 524 532 10.1097/SLA.0000000000005003 34132694
| 36441488 | PMC9707167 | NO-CC CODE | 2022-12-01 23:20:21 | no | Obes Surg. 2022 Nov 28;:1-12 | utf-8 | Obes Surg | 2,022 | 10.1007/s11695-022-06365-6 | oa_other |
==== Front
Obes Surg
Obes Surg
Obesity Surgery
0960-8923
1708-0428
Springer US New York
36441488
6365
10.1007/s11695-022-06365-6
Original Contributions
Long-term Reported Outcomes Following Primary Laparoscopic Sleeve Gastrectomy
http://orcid.org/0000-0002-6658-8981
Sakran Nasser [email protected]
123
Soifer Kim [email protected]
12
Hod Keren [email protected]
4
Sherf-Dagan Shiri [email protected]
56
Soued Sharon [email protected]
12
Kessler Yafit [email protected]
15
Adelson Dana [email protected]
1
Biton Reut [email protected]
1
Buchwald J. N. [email protected]
7
Goitein David [email protected]
189
Raziel Asnat [email protected]
1
1 grid.414003.2 0000 0004 0644 9941 Assia Medical Group, Assuta Medical Center, Tel Aviv, Israel
2 grid.22098.31 0000 0004 1937 0503 The Azrieli Faculty of Medicine Safed, Bar-Ilan University, Ramat Gan, Israel
3 grid.414321.1 0000 0004 0371 9846 Department of General Surgery, Holy Family Hospital, P.O. Box 8, Nazareth, Israel
4 grid.414003.2 0000 0004 0644 9941 Department of Academy and Research, Assuta Medical Center, Tel-Aviv, Israel
5 grid.411434.7 0000 0000 9824 6981 Department of Nutrition Sciences, School of Health Sciences, Ariel University, Ariel, Israel
6 grid.414003.2 0000 0004 0644 9941 Department of Nutrition, Assuta Medical Center, Tel Aviv, Israel
7 Division of Scientific Research Writing, Medwrite Medical Communications, Maiden Rock, WI USA
8 grid.12136.37 0000 0004 1937 0546 Sackler Faculty of Medicine, Tel-Aviv University, Tel Aviv, Israel
9 grid.413795.d 0000 0001 2107 2845 Department of Surgery C, Sheba Medical Center, Ramat Gan, Israel
28 11 2022
112
7 5 2022
30 10 2022
18 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Background
Long-term laparoscopic sleeve gastrectomy (LSG) outcomes in patients with obesity are scarce. We aimed to examine the outcomes and subjective experience of patients who underwent primary LSG with long-term follow-up.
Methods
The study is a retrospective observational analysis of patients who underwent primary LSG in a single center with 5–15 years of follow-up. Patients’ hospital chart data supplemented by a detailed follow-up online questionnaire and telephone interview were evaluated.
Results
The study sample included 578 patients (67.0% female) with 8.8 ± 2.5 years of mean follow-up, with a response rate to the survey of 82.8%. Mean baseline age and body mass index (BMI) were 41.9 ± 10.6 years and 42.5 ± 5.5 kg/m2, respectively. BMI at nadir was 27.5 ± 4.9 kg/m2, corresponding to a mean excess weight loss (EWL) of 86.9 ± 22.8%. Proportion of patients with weight regain, defined as nadir ≥ 50.0% EWL, but at follow-up < 50.0% EWL, was 34.6% (n = 200) and the mean weight regain from nadir was 13.3 ± 11.1 kg. BMI and EWL at follow-up were 32.6 ± 6.4 kg/m2 and 58.9 ± 30.1%, respectively. The main reasons for weight regain given by patients included “not following guidelines,” “lack of exercise,” “subjective impression of being able to ingest larger quantities of food in a meal,” and “not meeting with the dietitian.” Resolution of obesity-related conditions at follow-up was reported for hypertension (51.7%), dyslipidemia (58.1%) and type 2 diabetes (72.2%). The majority of patients (62.3%) reported satisfaction with LSG.
Conclusions
In the long term, primary LSG was associated with satisfactory weight and health outcomes. However, weight regain was notable.
Graphical Abstract
Keywords
Bariatric surgery
Sleeve gastrectomy
Obesity
Long-term follow-up
Weight regain
==== Body
pmcIntroduction
Laparoscopic sleeve gastrectomy (LSG) is the most frequently performed bariatric/metabolic surgery (BMS) procedure worldwide largely due to its straightforward operative technique, minimal alteration of anatomy, absence of gastrointestinal anastomoses and foreign material, successful short- and mid-term weight loss, low morbidity, and capacity for conversion to alternate BMS procedures [1–4]. Still, a paucity of long-term LSG studies with ≥ 7 years of mean follow-up has restricted the assessment of its sustained effectiveness and safety [5]. We herein present our long-term primary LSG results analyzing weight outcomes, obesity-associated disease changes, and patient satisfaction.
Patients and Methods
Analysis of a detailed online questionnaire addressing long-term LSG follow-up supplemented by hospital chart data was conducted. Inclusion criteria included patients aged 18 ≤ years old who underwent primary LSG in our center and were followed for at least 5 years after surgery. Exclusion criteria included patients who underwent another MBS after the LSG. All patients met the standard inclusion criteria for BMS (i.e., body mass index [BMI] ≥ 40.0 or ≥ 35.0 kg/m2 with weight-associated medical conditions) [6]. LSG technique has been previously described in detail [7]. Venous thromboembolic prophylaxis was initiated within 12 h of surgery and continued for 2 weeks [8]. A bariatric dietitian provided detailed dietary guidance during recovery [9]. Patients were instructed to attend routine follow-up meetings with the bariatric multidisciplinary team.
An online questionnaire containing open-ended and multiple-choice questions was administered electronically by email or SMS text and verified by a telephone interview, when needed. The questionnaire (Appendix) included information regarding weight outcomes, coexisting obesity-related conditions (hypertension, hyperlipidemia, type 2 diabetes mellitus [T2DM], obstructive sleep apnea), patient satisfaction, and reasons for weight regain. Follow-up data were obtained from the online questionnaire.
Percent excess weight loss (%EWL − ([initial weight − follow-up weight] / [initial weight − ideal body weight]) × 100) and percentage of total body weight loss (%TWL) (([initial weight − follow-up weight]/[initial weight]) × 100) were calculated. Weight regain was reported as the (1) proportion of patients that initially achieved ≥ 50% EWL at nadir, but then then fell to < 50% EWL at long-term follow-up; (2) weight gain in kilograms post nadir; (3) percentage of nadir weight; and (4) percentage of maximum weight lost. Information regarding post-LSG long-term weight loss and weight regain is presented for the whole population and for two subgroups of patients with follow-up of 5–10 years and ≥ 10 years. Patient privacy throughout the study was ensured through a password-protected de-identified participant code. The study protocol was approved by the institutional review board (# 2,012,021-ASMC).
Statistical Analysis
Analyses were performed using Microsoft Excel 2016 (Microsoft Corp. Redmond, WA) and the SPSS statistical package (IBM Corp. Released 2019. IBM SPSS Statistics for Windows, version 28.0. Armonk, NY: IBM Corp). Continuous variables were presented as means ± standard deviations (SD), while dichotomous and categorical variables were presented as percentages and frequencies. Measures of change from baseline were analyzed with the dependent-samples t-test or the Wilcoxon test, as appropriate. In addition, comparisons of baseline characteristics between subgroups were calculated by Mann–Whitney U test and chi-square tests, as appropriate. To assess consistency between self-reported coexisting obesity-related conditions (i.e., diabetes, hypertension, hyperlipidemia, and obstructive sleep apnea) with diagnoses obtained from electronic medical records, Cohen’s kappa was calculated for baseline data. The level of agreement was classified as poor (≤ 0.01), slight (0.01–0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), and almost perfect (0.81–1.00) [10–12]. All statistical tests were two-tailed and statistical significance was set at p < 0.05.
Results
Patient Baseline Characteristics
Of the 698 patients who were reached to participate in the study and met the inclusion criteria, 578 gave their consent and responded to the survey (response rate of 82.8%). Furthermore, 33 patients were excluded from the study analysis due to conversion to another BMS. Most of these patients converted to one anastomosis gastric bypass (OAGB) (51.5%), Roux-en-Y gastric bypass (RYGB) (30.3%), and additional LSG (18.2%), while the leading reported reasons for additional MBS were weight regain (69.7%), and reflux, frequent vomiting, and eating difficulties (45.5%).
Overall, the study sample included 578 (67.0% female) patients who responded to the questionnaires, and underwent primary LSG between May 1, 2006, and December 31, 2016, in our center. Their mean pre-surgery BMI and age were 42.5 ± 5.5 kg/m2 and 41.9 ± 10.6 years, respectively. The most prevalent weight-related medical illnesses pre-surgery were dyslipidemia 43.9% (n = 254), followed by hypertension 27.7% (n = 160), T2DM 25.3% (n = 146), and obstructive sleep apnea 14.2% (n = 82). The most common additional procedure performed during primary LSG was a laparoscopic cholecystectomy (7.1%, n = 41), followed by hiatal hernia repair (6.6%, n = 38) (Table 1). The mean follow-up period was 8.8 ± 2.5 years (range: 5–14.4 years). Patients who responded to the survey had a similar pre-surgical age, BMI, and gender distribution in comparison to non-respondents (41.9 ± 10.6 vs 43.3 ± 11.9 years, p = 0.235; 42.5 ± 5.5 vs 42.9 ± 5.6 kg/m2, p = 0.463; 67 vs 70% female, p = 0.593; respectively).Table 1 Baseline patient characteristics and additional procedures during primary LSG
Characteristics Total LSG population (n = 578)
Age (yrs), mean ± SD 41.9 ± 10.6
Initial weight (kg), mean ± SD 118.9 ± 19.6
Initial height (m), mean ± SD 1.67 ± 0.09
Initial body mass index (k/m2), mean ± SD 42.5 ± 5.5
Females, n (%) 387 (67.0)
Female to male ratio 2.0:1.0
Smoking status, n (%) 105 (18.2)
Marital status, n (%)
• Married 424 (73.4)
• Divorced 74 (12.8)
• Other 80 (13.8)
Prevalent associated medical conditions, n (%)
• Hyperlipidemia 254 (43.9)
• Hypertension 160 (27.7)
• Obstructive sleep apnea 82 (14.2)
• Type 2 diabetes 146 (25.3)
Additional laparoscopic procedures during LSG, n (%)
• Cholecystectomy 41 (7.1)
• Hiatal hernia repair 38 (6.6)
• Umbilical or ventral hernia repair w/ mesh 7 (1.2)
All data are based on electronic medical records
Weight Outcomes
Mean nadir weight at a mean 12.8 ± 13.4 months following surgery was 77.9 ± 15.5 kg. This weight loss equated to mean nadir TWL of 34.3 ± 8.1% and mean EWL of 86.9 ± 22.8%. Overall, only 5.5% (n = 32) of primary LSG patients in our sample had insufficient weight loss (i.e., EWL < 50.0%) at nadir. Mean weight at 8.8 ± 2.5 years of follow-up was 91.3 ± 20.4 kg for the whole study population. This corresponded to a mean TWL of 23.1 ± 11.4% and mean EWL of 58.9 ± 30.1%. Significant differences between patients with 5–10 and ≥ 10 years of follow-up were found for mean weight at present (89.5 ± 19.2 vs 95.4 ± 22.5 kg, p = 0.010), and corresponding mean TWL (24.1 ± 10.6 vs 20.9 ± 12.6%, p = 0.005), and mean EWL (61.7 ± 28.4 vs 52.8 ± 32.8, p = 0.002).
The proportion of patients who reported on any weight gained from nadir was 91.7% (n = 530). The mean time interval reported between LSG and weight regain in years was 3.4 ± 2.0, and was similar between patients with different time intervals of follow-up. Mean weight regain was 13.3 ± 11.1 kg for the whole study population, while less weight regain was reported in the group with 5–10 years of follow-up in comparison to the group with ≥ 10 years of follow-up (12.2 ± 9.5 kg vs 15.6 ± 13.7, p = 0.041).
Rate of weight regain, defined as nadir ≥ 50.0% EWL, but at follow-up < 50.0% EWL was 34.6% (n = 200), while this rate was lower in the group with 5–10 years of follow-up in comparison to the group with ≥ 10 years of follow-up (n = 125 (31.3%) vs n = 75 (42.1%), p = 0.007) (Table 2).Table 2 Post-LSG long-term weight loss and weight regain
Parameter Total LSG population (n = 578) 5–10 years of follow-up
(n = 400) ≥ 10 years of follow-up
(n = 178) P-value between 5–10 and ≥ 10 yrs follow-up
Short follow-up
Time to achieve nadir (lowest weight from LSG) (mo.), mean ± SD 12.8 ± 12.4 12.9 ± 12.0 12.4 ± 12.1 0.826
Nadir BMI (lowest BMI attained) (kg/m2), mean ± SD 27.5 ± 4.9 27.1 ± 4.6 28.4 ± 5.3 0.011
Nadir weight (kg), mean ± SD 77.9 ± 15.5 77.2 ± 15.2 79.7 ± 16.1 0.153
Nadir weight loss (kg), mean ± SD 40.9 ± 12.5 40.9 ± 12.1 40.8 ± 13.4 0.708
Nadir TWL (%), mean ± SD 34.3 ± 8.1 34.5 ± 8.1 33.7 ± 8.2 0.475
Nadir EWL (%) mean ± SD 86.9 ± 22.8 88.3 ± 23.1 83.7 ± 22.2 0.075
Insufficient weight loss, n (%) 32 (5.5) 21 (5.3) 11 (6.2) 0.696
Short follow-up successful patients, n (%)§§ 541 (94.4) 374 (93.5) 167 (93.8) 0.696
Short follow-up insufficient weight loss, n (%)§§§ 32 (5.5) 21 (5.3) 11 (6.2) 0.696
Long follow-up
Follow-up duration (yrs.), mean ± SD 8.8 ± 2.5 7.3 ± 1.9 11.9 ± 0.9 < 0.001
Current weight (kg), mean ± SD 91.3 ± 20.4 89.5 ± 19.2 95.4 ± 22.5 0.010
Current BMI (kg/m2), mean ± SD 32.6 ± 6.4 32.0 ± 5.9 34.1 ± 7.1 0.002
Weight loss (kg), mean ± SD 27.5 ± 14.5 28.5 ± 13.5 25.2 ± 16.3 0.005
Current TWL (%), mean ± SD 23.1 ± 11.4 24.1 ± 10.6 20.9 ± 12.6 0.005
Current EWL (%), mean ± SD 58.9 ± 30.1 61.7 ± 28.4 52.8 ± 32.8 0.002
Weight regain
Proportion that gained any weight from nadir, n (%)* 530 (91.7) 369 (92.2) 161 (90.4) 0.308
Time after LSG to weight regain (yrs), mean ± SD† 3.4 ± 2.0 3.3 ± 1.8 3.8 ± 2.4 0.123
Weight regain post nadir (kg), mean ± SD†,‡‡ 13.3 ± 11.1 12.2 ± 9.5 15.6 ± 13.7 0.041
% weight regain (% nadir weight), mean ± SD†,‡ 17.3 ± 14.6 16.1 ± 12.7 19.8 ± 17.9 0.097
Proportion weight regain (≥ 15.0% of nadir weight)†,‡ 50.9 (270/530) 48.8 (180/369) 55.9 (90/161) 0.156
% weight regain (%maximum weight lost), mean ± SD†,§ 33.4 ± 26.1 30.8 ± 22.7 39.2 ± 31.7 0.017
Proportional distribution of weight regain (% of maximum weight lost), n (%)§
No regain* 45 (7.8) 27 (7.3) 18 (11.2) 0.308
< 25%† 213 (37.0) 156 (42.3) 57 (35.4) 0.150
≥ 25– < 50%† 181 (31.5) 133 (36.0) 48 (29.8) 0.165
≥ 50%† 136 (23.6) 80 (21.7) 56 (34.8) 0.002
Long follow-up successful patients, n (%)§§§§ 345 (59.9) 253 (63.3) 92 (51.7) 0.008
Long follow-up insufficient weight loss, n (%)§§§§§ 231 (40.0) 145 (36.3) 86 (48.3) 0.008
Proportion of successful patients at nadir that succeeded at long follow-up†† 341 (59.0) 249 (62.3) 92 (51.7) 0.007
Proportion of successful patients at nadir that failed at long follow-up††† 200 (34.6) 125 (31.3) 75 (42.1) 0.007
BMI, body mass index; TWL, total weight loss; EWL, excess weight loss
*Data were available for 575 patients
†Data were available for 530 patients
‡‡Weight regain post nadir (kg) was calculated based on the following equation: (current weight − nadir weight)
‡% weight regain of nadir weight was calculated based on the following equation: ([current weight − nadir weight] / [nadir weight]) × 100
§% weight regain was of maximum weight lost was calculated based on the following equation: ([current weight − nadir weight] / (baseline weight − nadir weight]) × 100
Proportional distribution of weight regain (< 25%, ≥ 25– < 50%, ≥ 50%) was available for 530 patients since those who did not regain weight (n = 45) were excluded from these analyses
§§Short follow-up successful patients, n (%) = patients with Nadir %EWL ≥ 50%, n (%), data were available for 573 patients
§§§Short follow-up insufficient weight loss, n (%) = patients with Nadir %EWL < 50%, n (%), data were available for 573 patients
§§§§Long follow-up successful patients, n (%) = patients with current weight %EWL ≥ 50%, n (%). Data were available for 576 patients
§§§§§Long follow-up insufficient weight loss, n (%) = patients with current weight %EWL < 50%, data were available for 576 patients
††Proportion of successful patients at nadir that succeeded at long follow-up = [number of patients with nadir EWL ≥ 50% that succeeded also at long follow-up as their current EWL ≥ 50% / number of patients with nadir EWL > 50%] × 100
†††Proportion of successful patients at nadir that failed at long follow-up = [number of patients with nadir EWL ≥ 50% that failed at long follow-up as their current EWL < 50% / number of patients with nadir EWL > 50%] × 100
Baseline characteristics of patients whose weight loss was successful at nadir (EWL ≥ 50%) but failed at long-term follow-up (EWL < 50%) were similar, except for higher initial weight (121.8 ± 21.0 kg) and BMI (43.6 ± 5.2 kg/m2) and lower proportion of current smoking (12.0%, n = 24) as compared to patients that succeeded both at nadir and at long-term follow-up (115.9 ± 17.6 kg, 41.3 ± 4.9 kg/m2 and 24.0% (n = 82), P = 0.002, p < 0.001, and P = 0.001, respectively).
Primary LSG patient-reported reasons for weight regain are presented in Table 3. Over half of all patients (56.8%, n = 302) with weight regain reported that they “did not follow the MBS guidelines” recommended by their health professionals after LSG. The second most frequently reported reason for regain was “lack of exercise” (51.9%, n = 276), followed by “patients’ subjective impression of being able to ingest larger quantities of food in a meal” (41.1%, n = 218), and “not meeting with a dietitian” (38.6%, n = 205). Also, weight regain due to “Covid-19 consequences” was mentioned in textual responses by some patients (1.7%, n = 9). Despite the fact that 91.7% (n = 530) reported at least some weight regain, the majority (62.3%, n = 360) were “sufficiently satisfied” with the procedure and “would choose LSG again.” Interestingly, similarity between patients with different follow-up time intervals was found with respect to of the reasons reported for weight regain (Table 3).Table 3 Patient-reported reasons for weight regain
Reasons for weight regain*† Total LSG population
(n = 578)
n (%) LSG population with 5–10 years of follow-up
(n = 400)
n (%) LSG population with ≥ 10 years of follow-up
(n = 178)
n (%) P-value between 5–10 and ≥ 10 years of follow-up
• Did not follow guidelines 302 (56.8) 199 (49.8) 103 (57.9) 0.123
• Lack of exercise 276 (51.9) 203 (50.7) 73 (41.0) 0.018
• Sleeve enlargement¥ 218 (41.1) 138 (34.5) 80 (44.9) 0.032
• Did not meet with dietitian 205 (38.6) 152 (38.0) 53 (29.7) 0.038
• Difficulty controlling portion size, sweets, or alcoholic beverages 21 (3.9) 16 (4.0) 5 (2.8) 0.632
• Mental crisis, emotional eating 19 (3.5) 12 (3.0) 7 (3.9) 0.618
• COVID-19 consequences 9 (1.7) 7 (1.8) 2 (1.1) 0.727
• Consequences of other diseases 6 (1.1) 4 (1.0) 2 (1.1) > 0.999
• Antidepressants, steroids 7 (1.3) 5 (1.2) 2 (1.1) > 0.999
• Intentional weight gain 6 (1.1) 6 (1.5) 0 (0.0) 0.184
• Lack of support 5 (0.9) 2 (0.5) 3 (1.7) 0.179
• Hypothyroidism 3 (0.5) 1 (0.3) 2 (1.1) 0.232
• As a consequence of IVF treatments, pregnancies, births‡ 20/379 (5.3) 14 (3.5) 6 (3.4) > 0.999
• Smoking cessation 2 (0.4) 1 (0.3) 1 (0.6) 0.528
Patients sufficiently satisfied: would choose LSG again§ 360 (62.3) 250 (62.5) 110 (61.8) 0.926
COVID-19, coronavirus disease 2019; IVF, in vitro fertilization
*Data available only for patients who regain weight (n = 530)
†Each patient had the option of providing more than one answer
¥Patient’s subjective impression of being able to ingest larger quantities of food in a meal
‡Data available only for women who regain weight (n = 379)
§Data available for the whole study population (n = 578)
The questionnaire also offered possible reasons for choosing a bariatric procedure other than LSG in the future, such as “today there are more advanced procedures” (28.0%, n = 162, answered “yes” on this item) and “because I’m not satisfied with the LSG” (13.8%, n = 80, answered “yes” on this item).
Hospitalizations, Procedures, and Clinical Follow-up
Nine percent of patients (n = 52) visited the emergency room, and 12.3% (n = 71) were hospitalized at some point after primary LSG, not necessarily related to their surgery (Table 4). The majority of stated emergency room visits were related to BMS (78.8%, n = 41), with gallbladder stones reported as the most frequent cause (39%, n = 16), followed by abdominal pain, dehydration and vomiting (26.8%, n = 11), leak or necrosis (9.8%, n = 4), stenosis or stricture of the sleeve (7.3%, n = 3), bleeding (2.4%, n = 1), incisional hernia (2.4%, n = 1), iron deficiency anemia (2.4%, n = 1), and kidney stone (2.4%, n = 1).Table 4 Post-LSG clinical follow-up and surgical procedures
Post-primary LSG procedures Total LSG population
(n = 578)
n (%)
Readmission rate 52 (9.0)
Hospitalizations following LSG 71 (12.3)
Additional non-bariatric surgery 230 (39.8)
Upper endoscopy 179 (31.0)
Plastic surgery 56 (9.7)
Cholecystectomy 54 (9.3)
Gynecologic surgery 39 (6.7)
Orthopedic surgery 36 (6.2)
Hiatal/umbilical/ventral hernia repair 23 (4.0)
Dietitian follow-up
• Not at all 0 (0.0)
• Only in the 1–2 years after the LSG 260 (45.0)
• Intermittently during the years 84 (14.5)
• Routine follow-up 40 (6.9)
Blood tests performed
• Did not perform blood tests at all since LSG 8 (1.4)
• Only in the first 1–2 years after the LSG 37 (6.4)
• Intermittently over the years 214 (37.0)
• Routine follow-up 319 (55.2)
Additional non-bariatric surgery after LSG was reported by 39.8% of patients (n = 230). Upper endoscopy was the most common invasive procedure reported (31.0%, n = 179), followed by plastic surgeries (9.7%, n = 56), cholecystectomy (9.3%, n = 54), gynecologic procedures (6.7%, n = 39), orthopedic surgeries (6.2%, n = 36), and hiatal, umbilical, and ventral hernia repair (4.0%, n = 23). About half (47.2%) of the patients reported that they suffered from heartburn or took any drugs to treat heartburn. Only a minority of patients reported routine contact with the dietitian throughout long-term follow-up (6.9%, n = 40), and 55.2% (n = 319) reported undergoing routine blood tests after LSG (Table 4).
Change in Associated Medical Problems
Significant reductions from baseline in the proportion of LSG patients suffering from hypertension, dyslipidemia, and T2DM were observed (p < 0.001 for all). Rates of no change, improvement, and resolution for selected associated medical problems are depicted in Fig. 1.Fig. 1 Resolution of associated medical conditions following laparoscopic sleeve gastrectomy
Discussion
Long-term LSG Outcomes
The predominance of the LSG among other BMS procedures arose before its rates of long-term weight loss and revision were known. Until recently, ≥ 7-year mean LSG outcomes were scarce. In 2018, a meta-analysis by Clapp and colleagues [5] reported outcomes synthesized from prior studies incorporating 652 patients who completed ≥ 7 years of mean follow-up. From 2018 through 2021, several definitively long-term LSG outcome reports in patients with obesity were published (8 observational studies, 1 randomized controlled trial) [13–21]. Our current investigation studied outcomes in 578 LSG patients; to our knowledge, this is the largest single series of LSG outcomes analyzed at ≥ 8.8 years. Herein, we relate the key study findings of the current study to the aforementioned recent reports.
Weight Outcomes
The meta-analysis of studies with > 7 years of mean follow-up by Clapp et al. found that 72.2% of patients attained ≥ 50.0% EWL at follow-up. Nine other long-term studies published later also demonstrated significant sustained weight loss at their respective follow-up points. Six reported a follow-up weight loss comparable to that of the current study [13–15, 19, 20], although three of these studies had 3–4-year longer follow-up durations [15, 19, 20]. In three of the later studies, follow-up weight loss was markedly less than in the current study, with ≤ 50.0% total EWL: Major et al. reported on a significant 16.3 BMI-point reduction at nadir, but at 10-year follow-up EWL was 45.4% [21]; Gronroos et al.’s randomized controlled trial (RCT) reported a 47.0% EWL [17]; and Hauters et al., 42.0% EWL [18].
A study by Felsenreich and co-workers is notable for its achievement of a ≥ 15-year follow-up with 96.0% patient retention, nearly twice the follow-up duration of the current study and most of the other LSG studies [52]. The authors saw a stable 61.0% EWL at this extended endpoint, but at the cost of a high revision rate — almost half of the patients (49.1%) were converted to another procedure. The report of Fiorani et al. is also unique in the long-term LSG literature, in that, not only was there no weight regain, but weight loss continued beyond the 1-year nadir through 8 years of follow-up. This report, however, is limited by both the small initial sample size (44 patients undergoing LSG) and the high lost-to-follow-up rate of 75.0% in the LSG group [16].
Weight Regain
The cause of weight regain in LSG is multifactorial, influenced by individual physiology, dietary choices, activity level, culture, and genetics [22, 23]. In addition, assessment of BMS weight regain is not yet standardized, and there is wide variability in its reporting complicating outcome comparison [24–27]. Herein, we focused on only studies that had specifically defined and quantified weight regain.
Perhaps the most frequently used definition of weight regain in recent years is the proportion of patients that initially achieved ≥ 50.0% EWL but then regained weight to have < 50.0% EWL at follow-up. In a meta-analysis of long-term LSG data (≥ 7-year follow-up), Clapp and co-workers analyzed nine studies with respect to this measure, yielding a pooled weighted mean proportion of 27.8% (95% CI: 22.8, 32.7) with a range of 14.0–37.0% [5]. Our study had a similar sample size and follow-up period, and using this definition, weight regain was on the higher end, (34.6%) of the range reported by Clapp et al. This was because the current study’s subgroup of LSG patients with ≥ 10-year follow-up contained a significantly higher proportion of patients who regained weight (42.1%) relative to LSG patients with 5–10 years of follow-up (31.3%).
Perhaps in recognition of the arbitrary nature of the 50% EWL cutoff, other methods of assessing weight regain have been proposed (i.e., weight regain from nadir weight calculated as weight in kilograms, BMI, percentage of preoperative weight, percentage of nadir weight, and percentage of maximum weight lost). Using percentage of maximum weight lost, Capoccia et al. reported long-term weight regain results in 104 LSG patients with more than 5-year follow-up [14].
Weight regain ≥ 15.0– < 30.0% of maximum weight lost was considered mild, or > 30.0%, severe. By this classification, 25.0% and 22.0% experienced mild and severe weight regain, respectively. Using slightly different thresholds, our study reported 37.0% of patients as having experienced mild weight regain (< 25.0% of maximum weight lost), and 55.1% as having experienced significant weight regain (≥ 25.0% of maximum weight lost). The different thresholds confound the proportional comparison; however, the current study’s cohort had a mean percentage weight regain of maximum weight loss of 33.4%, slightly higher than the 31.5% reported by Capoccia et al. As should be expected, the current study’s subgroup of LSG patients with 5–10 years of follow-up, possibly a more apt comparator, had a slightly lower mean percentage weight regain of maximum weight lost than that reported by Capoccia et al. (30.8% vs 31.5%).
A final recent example of weight regain in LSG patients, calculated using yet another measure, is provided by Ben-Porat and colleagues. In their study of 212 patients, weight regain was defined as the proportion of patients regaining ≥ 15% of nadir weight [13]. According to this measure, exactly 50% of their patients were classified as experiencing weight regain. Similarly, using this definition, just over half (50.9%) of the current study’s entire cohort experienced significant weight regain (with 48.8% of LSG patients with 5–10-year follow-up experiencing weight regain vs 55.9% of patients with ≥ 10-year follow-up). In addition, weight regain from nadir weight calculated in kilograms in both studies was nearly equivalent (14.6 kg vs 13.3 kg). These comparative results parallel those of other long-term LSG patients over the long term, regardless of how they were measured and recorded (e.g., at clinical visit or by questionnaire) [28].
Resolution of Coexisting Conditions
In addition to significant and sustained weight loss, this study and most of the other LSG studies suggest that, at 7–15 years after the procedure, LSG provides meaningful resolution of obesity-related medical problems. Kraljevic et al. saw high rates of resolution in dyslipidemia (54.8%), hypertension (60.5%), and T2DM (61.0%) [20]. In the current study, good resolution of long-term coexisting conditions was found, including hypertension (51.7%), dyslipidemia (58.1%), and T2DM (72.2%). These rates were comparable to those reported by Capoccia et al. [14] and notably greater than those of Ben-Porat and Hauters, both including markedly smaller cohorts [13, 18].
In Felsenreich et al.’s study, although very few patients at baseline had any obesity-associated conditions, hypertension was resolved in 6/12 (50.0%) at ≥ 15 years of follow-up [15]. All patients in Ismail and co-worker’s cohort achieved complete resolution of hypertension, T2DM, and GERD between 6 months and 2 years but no long-term follow-up data was given [19]. Coexisting medical condition data provided in the included studies of the meta-analysis were too sparse for comparison, which is also true of other published work [16].
In Kraljevic et al., 52.0%, and in Hauters et al., 80.0% of the LSG cohort with baseline GERD experienced ongoing or worsened reflux at follow-up; of patients without preoperative reflux, 32.4% and 43.0%, respectively, developed de novo GERD in follow-up [18, 20]. While there are centers that have reported favorable mid-term LSG outcomes with respect to GERD [29, 30], the implications for complications due to both pre-existing and de novo GERD remain among the most important considerations for selection of this procedure. In the present study, about half of the patients reported that they suffered from heartburn or took any drugs to treat heartburn. However, this was reported subjectively with no available imaging test results to support it.
Patient Satisfaction and Reasons for Weight Regain
Several of the long-term LSG study comparators recorded quality of life (QoL) data using standardized assessment instruments [15, 16]. Others, as in the current study, assessed patient-reported characterization of the behaviors they believed influenced their weight regain through a questionnaire [13]. Study responses diverged in some of the parameters investigated, suggesting differing patient perspectives regarding whether maintaining a regimen of consulting with a dietitian or surgeon affected their LSG outcomes. The contrasting findings suggest the importance of further research to better clarify patient-reported experiences and behavioral choices in order to improve long-term BMS care.
Adherence to close follow-up of BMS patients is important but challenging beyond 2 years, where the rate is reported to be 40–62% [31–33]. We observed in the current study, as in most other BMS studies, that the rate of postoperative patient-reported dietitian interaction and routine blood testing was quite low. This finding confirmed the challenging nature of preserving interactive contact with patients over the long term. Designing studies that consider the input of patients regarding their weight regain over their years of long-term follow-up should provide insights that can be integrated into the longitudinal treatment regimen [34].
Limitations and Strengths
The major strengths of this study include the use of a large representative sample of LSG patients similar in baseline data composition to that of both a national and an international registry [35, 36]. However, there are some limitations to be acknowledged. First, data were collected mainly via a self-report questionnaire generated by the research team that has not been validated. Though it has become an acceptable method, initiated by the COVID-19 pandemic restrictions [37, 38], relying on patient-reported outcomes entails a potential for information, recall, and response bias. Nevertheless, Chao and colleagues observed in a study of 585 LSG and RYGB patients at 38 sites that patient-reported BMS outcomes had a high level of agreement with hospital records in early follow-up [39]. In this manner, to increase the validity of the data in the present study, some data were also collected by hospital charts and when needed data were verified by a telephone interview. Moreover, testing reliability of data collection is an important element of confidence in a study’s accuracy [11]. In our study, self-reported coexisting obesity-related conditions data at baseline presented fair to moderate agreement with the available data obtained from hospital records (Cohen’s kappa of 0.574, 0.559, 0.451, and 0.284 for reporting on T2DM, hypertension, obstructive sleep apnea, and dyslipidemia [p < 0.001], respectively). Nonetheless, reporting bias for follow-up data collection should be considered.
Second, the study lacked objective measurements such as a medication list, blood test results, and imaging test results, and it lacked a validated QoL instrument. Finally, study follow-up was performed during the COVID-19 pandemic which may have affected the study results. However, weight regain due to “Covid-19 consequences” was reported by only a few patients in the present study.
Conclusions
At long-term follow-up, primary LSG was associated with satisfactory weight and health outcomes. However, weight regain was notable. More efforts should be made to increase patients’ engagement in follow-up routines in the long-term following MBS.
Appendix
LSG Patient Experience and Satisfaction Questionnaire
Initials of your name and surname
Research code number
Age
Family status
- Married
- Single
- Divorced
- Widower
- Other5. Current weight (in kg)
6. Height (in cm)
7. Are you a smoker?
- Yes
- No
- I used to smoke in the past8. What is the minimal weight you achieved since the surgery? (in kg)
9. How long did it take to achieve the above weight? (in months)
10. What was the duration of the weight loss since the surgery? (in months)
11. Did you gain weight after the weight loss? (even if it’s few kilos)
- Yes
- No12. If yes, when did you started to gain weight? (in years)
13. If you gained weight since the surgery, what did you think was the main cause? (you can choose more than one of the options)
- I didn’t eat as recommended
- The sleeve enlarged
- I didn’t exercise
- I wasn’t followed by a dietitian
- Other14. Did you have to go to the ER in the post-operative period?
- Yes.
- No.15. If yes, explain what the reason was
16. Were you hospitalized in the post-operative period?
- Yes.
- No.17. If yes, explain what the reason was
18. Did you undergo another bariatric surgery before the sleeve gastrectomy?**
- Yes
- No19. If yes, which surgery did you undergo?**
- LSG – laparoscopic sleeve gastrectomy
- LAGB – Laparoscopic adjustable gastric banding
- SRVG—Silastic ring vertical gastroplasty20. If yes, how long before the sleeve gastrectomy? (in years)**
21. If yes, what was the reason to undergo the sleeve gastrectomy?**
- weight gain
- drug-resistant reflux and heartburn
- a complication such as a sleeve stenosis or a twisted sleeve
- vomiting and eating difficulties
- nutritional deficiencies
- other22. Did you undergo another surgery after the sleeve gastrectomy?**
23. If yes, how long after the sleeve gastrectomy? (in years)**
24. If yes, what surgery did you undergo?**
- sleeve reduction (an additional sleeve gastrectomy)
- bypass with a single anastomosis (mini-gastric bypass)
- bypass with two anastomoses (classic bypass)
- duodenal bypass with two anastomoses
- duodenal bypass with one anastomosis (SADI)
- addition of a gastric band
- other25. If yes, what was the reason to undergo another surgery?**
- weight gain.
- drug-resistant reflux and heartburn
- a complication such as a sleeve stenosis or a twisted sleeve
- vomiting and eating difficulties
- nutritional deficiencies
- other
Questions 26–29 are directed only to women
26. Did you get pregnant since the surgery?
- Yes
- No27. If yes, how many times have you gotten pregnant since the surgery?
28. Did you give birth after surgery?
- Yes
- No29. If yes, how many times did you give birth?
30. Did you suffer from gallstones before the surgery?
- Yes
- No
- I don’t know31. If yes, what is true for you?
- I suffered from gallstones before the surgery and I underwent a cholecystectomy when I underwent the sleeve gastrectomy
- I suffered from gallstones before the surgery and I underwent a cholecystectomy after the sleeve gastrectomy
- I suffered from gallstones before the surgery and I underwent a cholecystectomy before the sleeve gastrectomy32. Were you diagnosed with gallstones since the sleeve gastrectomy?
- Yes
- No
- I don’t know33. Did you undergo a cholecystectomy since the sleeve gastrectomy?
- Yes
- No34. Did you undergo any surgery since the sleeve gastrectomy?
- Yes
- No35. If yes, explain which surgery.
36. Do you suffer from heartburn or do you take any drugs to treat heartburn?
- Yes
- No37. Did you undergo a gastroscopy since the surgery?
- Yes
- No38. If yes, what were the findings?
39. Are you satisfied with the surgery? Grade from 1 to 10 (1 = not satisfied at all, 10 = very satisfied)
40. If you underwent another bariatric surgery, are you satisfied with the surgery? Grade from 1 to 10 (1 = not satisfied at all, 10 = very satisfied)**
41. If you could choose again the type of bariatric surgery, would you choose again a sleeve gastrectomy?
- Yes
- No42. If not, why? (you can choose more than one of the options)
- today there are more advanced procedures
- because I’m not satisfied from the surgery43. Did you suffer from diabetes type 2 before the surgery? (diagnosis by medical record/taking pills to treat diabetes)
- Yes
- No
- I don’t know44. If yes, what was the treatment? (you can choose more than one of the options)
- Diet
- Drugs
- Insulin
- Drugs and insulin
- Not treated
- I don’t know45. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement46. Did you suffer from hypertension before the surgery?
- Yes
- No
- I don’t know47. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement48. Did you suffer from sleep apnea or did you use a CPAP mask while asleep?
- Yes
- No
- I don’t know49. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement50. Did you suffer from hyperlipidemia before the surgery? (high cholesterol/triglycerides/LDL in your blood)
51. If yes, what is true for you?
- there was an improvement
- there was a total improvement (I was cured)
- I don’t know
- there was no improvement52. Were you followed by a dietitian after the surgery?
- No
- Only the first/second year after the surgery
- Intermittently during the years
- Regularly during the years, but not currently
- Regularly during the years to this day53. With which frequency did you undergo blood tests since the surgery?
-I didn’t undergo any blood test since the surgery
-I underwent blood tests only the first/second year after the surgery
-I underwent blood tests intermittently during the years
-I undergo blood tests regularly since the surgery
**These questions were asked to be marked only for patients who undergo another bariatric surgery in their past.
Acknowledgements
We thank TW McGlennon, Medwrite, Maiden Rock, WI, USA, for statistical analysis support.
Funding
Medwrite received a grant for assistance with manuscript development.
Declarations
Statement of Ethics
The study protocol and all procedures of the study were approved as compliant with the ethical standards of the Assuta Medical Center (IRB #43–20-ASMC).
Written Informed Consent
Written informed consent was obtained from all study participants.
Conflict of Interest
The authors declare no competing interests.
Human and Animal Rights
The study was performed in accord with the ethical standards of the 1964 Declaration of Helsinki and subsequent amendments.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
David Goitein and Asnat Raziel contributed equally to the study.
==== Refs
References
1. Angrisani L Santonicola A Iovino P Bariatric Surgery Survey 2018: similarities and disparities among the 5 IFSO chapters Obes Surg 2021 31 5 1937 1948 10.1007/s11695-020-05207-7 33432483
2. Angrisani L Santonicola A Iovino P IFSO Worldwide Survey 2016: primary, endoluminal, and revisional procedures Obes Surg 2018 10.1007/s11695-018-3450-2 30128644
3. English WJ DeMaria EJ Brethauer SA American Society for Metabolic and Bariatric Surgery estimation of metabolic and bariatric procedures performed in the United States in 2016 Surg Obes Relat Dis 2018 14 3 259 263 10.1016/j.soard.2017.12.013 29370995
4. Lo Menzo E Szomstein S Rosenthal RJ Changing trends in bariatric surgery Scand J Surg 2015 104 1 18 23 10.1177/1457496914552344 25269945
5. Clapp B Wynn M Martyn C Long term (7 or more years) outcomes of the sleeve gastrectomy: a meta-analysis Surg Obes Relat Dis 2018 14 6 741 747 10.1016/j.soard.2018.02.027 29625744
6. NIH conference Gastrointestinal surgery for severe obesity Consensus Development Conference Panel Ann Intern Med 1991 115 12 956 961 10.7326/0003-4819-115-12-956 1952493
7. Sakran N Raziel A Goitein O Laparoscopic sleeve gastrectomy for morbid obesity in 3003 patients: results at a high-volume bariatric center Obes Surg 2016 26 9 2045 2050 10.1007/s11695-016-2063-x 26757919
8. Telem DA Gould J Pesta C American Society for Metabolic and Bariatric Surgery: care pathway for laparoscopic sleeve gastrectomy Surg Obes Relat Dis 2017 13 5 742 749 10.1016/j.soard.2017.01.027 28416400
9. Sherf Dagan S Goldenshluger A Globus I Nutritional recommendations for adult bariatric surgery patients: clinical practice Adv Nutr 2017 8 2 382 394 10.3945/an.116.014258 28298280
10 Ranganathan P Pramesh CS Aggarwal R Common pitfalls in statistical analysis: measures of agreement Perspect Clin Res 2017 8 4 187 191 10.4103/picr.PICR_123_17 29109937
11 McHugh ML Interrater reliability: the kappa statistic Biochem Med (Zagreb) 2012 22 3 276 282 10.11613/BM.2012.031 23092060
12. Landis JR Koch GG The measurement of observer agreement for categorical data Biometrics 1977 33 1 159 174 10.2307/2529310 843571
13. Ben-Porat T Mashin L Kaluti D Weight loss outcomes and lifestyle patterns following sleeve gastrectomy: an 8-year retrospective study of 212 patients Obes Surg 2021 31 11 4836 4845 10.1007/s11695-021-05650-0 34403078
14 Capoccia D Coccia F Guarisco G Long-term metabolic effects of laparoscopic sleeve gastrectomy Obes Surg 2018 28 8 2289 2296 10.1007/s11695-018-3153-8 29497961
15. Felsenreich DM Artemiou E Steinlechner K Fifteen years after sleeve gastrectomy: weight loss, remission of associated medical problems, quality of life, and conversions to Roux-en-Y gastric bypass-long-term follow-up in a multicenter study Obes Surg 2021 31 8 3453 3461 10.1007/s11695-021-05475-x 34021882
16. Fiorani C Coles SR Kulendran M Long-term quality of life outcomes after laparoscopic sleeve gastrectomy and Roux-en-Y gastric bypass-a comparative study Obes Surg 2021 31 3 1376 1380 10.1007/s11695-020-05049-3 33064260
17. Gronroos S Helmio M Juuti A Effect of laparoscopic sleeve gastrectomy vs Roux-en-Y gastric bypass on weight loss and quality of life at 7 years in patients with morbid obesity: the SLEEVEPASS Randomized Clinical Trial JAMA Surg 2021 156 2 137 146 10.1001/jamasurg.2020.5666 33295955
18. Hauters P Dubart JW Desmet J Degolla R Roumain M Malvaux P Ten-year outcomes after primary vertical sleeve gastrectomy for morbid obesity: a monocentric cohort study Surg Endosc 2021 35 12 6466 6471 10.1007/s00464-020-08137-8 33140154
19. Ismail M Nagaraj D Rajagopal M Seven-year outcomes of laproscopic sleeve gastectomy in Indian patients with different classes of obesity Obes Surg 2019 29 1 191 196 10.1007/s11695-018-3506-3 30238216
20. Kraljevic M Cordasco V Schneider R Long-term effects of laparoscopic sleeve gastrectomy: what are the results beyond 10 years? Obes Surg 2021 31 8 3427 3433 10.1007/s11695-021-05437-3 33890225
21. Major P Stefura T Dziurowicz B Quality of life 10 years after bariatric surgery Obes Surg 2020 30 10 3675 3684 10.1007/s11695-020-04726-7 32535784
22. Athanasiadis DI Martin A Kapsampelis P Factors associated with weight regain post-bariatric surgery: a systematic review Surg Endosc 2021 35 8 4069 4084 10.1007/s00464-021-08329-w 33650001
23. El Ansari W Elhag W Weight regain and insufficient weight loss after bariatric surgery: definitions, prevalence, mechanisms, predictors, prevention and management strategies, and knowledge gaps-a scoping review Obes Surg 2021 31 4 1755 1766 10.1007/s11695-020-05160-5 33555451
24. Baig SJ Priya P Mahawar KK Shah S Indian Bariatric Surgery Outcome Reporting G Weight regain after bariatric surgery-a multicentre study of 9617 patients from Indian Bariatric Surgery Outcome Reporting Group Obes Surg 2019 29 5 1583 1592 10.1007/s11695-019-03734-6 30729366
25. Brethauer SA Kim J El Chaar M Standardized outcomes reporting in metabolic and bariatric surgery Obes Surg 2015 25 4 587 606 10.1007/s11695-015-1645-3 25802064
26. King WC Hinerman AS Belle SH Comparison of the performance of common measures of weight regain after bariatric surgery for association with clinical outcomes JAMA 2018 320 15 1560 1569 10.1001/jama.2018.14433 30326125
27. Voorwinde V Steenhuis IHM Janssen IMC Definitions of long-term weight regain and their associations with clinical outcomes Obes Surg 2020 30 2 527 536 10.1007/s11695-019-04210-x 31677016
28. Oved I Endevelt R Mardy-Tilbor L Health status, eating, and lifestyle habits in the long term following sleeve gastrectomy Obes Surg 2021 31 7 2979 2987 10.1007/s11695-021-05336-7 33829384
29. Garg H Aggarwal S Misra MC Mid to long term outcomes of Laparoscopic Sleeve Gastrectomy in Indian population: 3–7 year results - a retrospective cohort study Int J Surg 2017 48 201 209 10.1016/j.ijsu.2017.10.076 29122706
30. Melissas J Braghetto I Molina JC Gastroesophageal reflux disease and sleeve gastrectomy Obes Surg 2015 25 12 2430 2435 10.1007/s11695-015-1906-1 26428250
31. Owolabi EO Mac Quene T Louw J Telemedicine in surgical care in low- and middle-income countries: a scoping review World J Surg 2022 10.1007/s00268-022-06549-2 35428920
32. van Erkel FM Pet MJ Bossink EH Experiences of patients and health care professionals on the quality of telephone follow-up care during the COVID-19 pandemic: a large qualitative study in a multidisciplinary academic setting BMJ Open 2022 12 3 e058361 10.1136/bmjopen-2021-058361 35273062
33. Chao GF Bonham AJ Ross R Patient-reported comorbidity assessment after bariatric surgery: a potential tool to improve longitudinal follow-up Ann Surg 2021 10.1097/SLA.0000000000004841 35129487
34. Fried M Hainer V Basdevant A Interdisciplinary European guidelines for surgery for severe (morbid) obesity Obes Surg 2007 17 2 260 270 10.1007/s11695-007-9025-2 17476884
35. Israeli Society for Metabolic and Bariatric Surgery. National Center for Disease Control and Israeli Society for Metabolic and Bariatric Surgery Annual Summary 2020. https://www.health.gov.il/PublicationsFiles/Bariatric_2020.pdf
36. Brown W, Kow L, Shikora S, et al. Sixth IFSO Global Registry Report 2021. Henley-on-Thames: IFSO & Dendrite Clinical Systems Ltd; 2021. Available at: https://www.e-dendrite.com/IFSO6
37. Larjani S Spivak I Hao Guo M Preoperative predictors of adherence to multidisciplinary follow-up care postbariatric surgery Surg Obes Relat Dis 2016 12 2 350 356 10.1016/j.soard.2015.11.007 26806726
38. Wheeler E Prettyman A Lenhard MJ Adherence to outpatient program postoperative appointments after bariatric surgery Surg Obes Relat Dis 2008 4 4 515 520 10.1016/j.soard.2008.01.013 18586576
39. Aminian A Kashyap SR Wolski KE Patient-reported outcomes after metabolic surgery versus medical therapy for diabetes: insights from the STAMPEDE Randomized Trial Ann Surg 2021 274 3 524 532 10.1097/SLA.0000000000005003 34132694
| 0 | PMC9707170 | NO-CC CODE | 2022-12-01 23:20:21 | no | manag kinderopvang. 2022 Nov 28; 28(6):59-60 | latin-1 | null | null | null | oa_other |
==== Front
Aesthetic Plast Surg
Aesthetic Plast Surg
Aesthetic Plastic Surgery
0364-216X
1432-5241
Springer US New York
36443419
3182
10.1007/s00266-022-03182-1
Original Article
The Double Capsule Phenomenon in a Case Series and its Relationship with the Macro-Textured Breast Implant
http://orcid.org/0000-0001-9354-950X
Berry Miles G. [email protected]
Stanek Jan J.
Surgical Aesthetics, Flat 19 Milford House, 7 Queen Anne Street, London, W1G 9HN UK
28 11 2022
16
30 9 2022
5 11 2022
© Springer Science+Business Media, LLC, part of Springer Nature and International Society of Aesthetic Plastic Surgery 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Background
Silicone breast augmentation remains one of the most common aesthetic surgery procedures, and 2022 marks the 60th anniversary of the first case. Recent studies suggest a link between double capsule (DC) formation and macro-textured devices.
Methods
Between 2010 and 2015, 268 aesthetic patients underwent bilateral mammary prosthesis exchange for indications including PIP exchange, adverse capsular contracture and ultrasonographic evidence of rupture. All surgery, in the form of implant exchange and capsulectomy, was undertaken by the senior author using standard techniques. A retrospective review was undertaken, and data analysed with descriptive statistics and Fisher’s exact and Mann–Whitney U tests.
Results
Of 268 patients identified, 40 (14.9%) showed some degree of capsular duplication and bilateral involvement was marginally more common (52.5%). Two macroscopic patterns of duplication were observed: complete and subtotal. Complete DCs correlated with a clinical triad of extreme firmness, mobility and minimal-to-no pain. Whilst a wide range of manufacturers was represented, macro-textured devices were associated with the highest DC prevalence (58.3% vs. 5.6%) (Fisher’s exact test p < 0.00001). Patients with DC had been implanted for less than half the time, median 52 versus. 120 months (p = 0.0003) of those without.
Discussion
An elevated prevalence of duplicate capsules in macro-textured prostheses is reconfirmed in addition to a novel symptom constellation that may assist with clinical diagnosis. Our study reinforces the aetiopathogenic influence of the elastomer in DC formation and reports DC for the first time in non-macrotextured implants.Single-surgeon cohort of 268 consecutive patients with 532 implants
Statistically significant association of macro-textured devices with DC
Statistically significant reduced duration of implantation of macro-textured devices
First report of DC in non-macro-textured devices
Level of Evidence IV
This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266.
Keywords
Breast implants
Pseudocapsule
Dual capsule
Duplicate capsule
Macrotexture
==== Body
pmcIntroduction
Although stated to be ‘recently described and rare’ in a contemporary systematic review [1], double capsules (DC) were in fact first reported in the literature in 2002 [2]. Hall-Findlay published the first series a decade ago and noted a correlation with the Biocell macro-textured elastomer [3]. Van Slyke et al presented a larger study involving 539 implants over a 13-year period and reinforced the Biocell surface association. They also documented the shortest time to explantation and highest prevalence of failure in such devices [4].
The double—aka duplicate or pseudo-—capsule has been linked with late seroma (defined as symptomatic swelling at least one year after implantation) [3, 5] and possibly with BI-ALCL (breast implant associated anaplastic large-cell lymphoma) [6]. Its management mandates surgical intervention, thereby adding a degree of potential risk, morbidity and additional cost to patients.
Having observed the novel appearance of DCs after switching to Allergan implants for breast implant exchange, we took advantage of the coronavirus-induced lockdown to undertake an in-depth practice review of breast implant exchanges over a consecutive 5-year period.
Patients and Methods
In total, 268 patients underwent bilateral breast implant exchange by the senior author between January 2010 and December 2015, and their clinical records were independently assessed for the presence of duplicate capsule and any aetiopathological factors. Of the 268, 4 underwent unilateral surgery only; thus, 532 implants were available for analysis.
Statistical analysis involved a combination of descriptive tests, the two-tailed Fisher’s exact test for contingency analysis and the Mann–Whitney U test for independent group correlation. In all cases, a p value < 0.05 was considered significant.
Results
The median age of the cohort was 47 years (range 21–75) with comorbidity demographics summarised in Table 1. 187 patients had undergone a single previous operation, 53 two and 18 three or more breast implant-related procedures. The maximum was 7 previous operations. Active smoking was recorded in 45, and with Fisher’s exact test yielding a p value of 0.0712, this cohort showed no correlation between smoking and DC formation. The device was unidentifiable in 37 patients leaving 231 for further analysis.Table 1: Comorbidity within the cohort
Comorbidity Number
Nil 222
HRT/OCP 12
Depression 11
Hypertension 5
Hypothyroidism 4
Immunosuppression 2
Asthma 2
Psoriasis/eczema 2
Raynaud’s 1
261
HRT hormone replacement therapy
OCP oral contraceptive pill
Duplicate capsule formation was identified in 40 patients, being bilateral in 21. Patient prevalence is therefore 14.9% (40/268) with a device prevalence of 7.5% (40/532). Complete encasement of the prosthesis (Fig. 1) was observed in 19 and partial in 21 (Fig. 2). Figure 3 shows examples of both types of capsule in situ after implant extraction.Fig. 1 Typical example of a complete duplicated capsule in a 40-year-old woman 57 months after primary breast augmentation.
Fig. 2 The subtotal capsule of a 54-year-old, 15 months after Natrelle implants had been used to replace Siltex devices, without DC, inserted 16 years earlier. Note also the lack of capsule formation over the filling disc
Fig. 3 Intraoperative example of standard () and duplicated () capsule. Absence of capsule formation over the filling disc ()
A wide range of implant manufacturers was represented (Table 2), the highest rates being observed in macro-textured devices. Statistical analysis strongly correlated macro-textured surfaces with duplicate encapsulation (Fisher’s exact test p < 0.00001).Table 2 Number of implants and DC by manufacturer
Implant Total Unilateral DC Bilateral DC %
PIP 149 4 1 3.4
Natrelle 21 4 9 61.9
CUI 19 3 – 15.8
McGhan 13 3 4 53.8
Siltex/Mentor 10 1 1 20
Hydrogel 8 – – –
Nagor 3 – 1 33
Saline (McGhan) 2 1 – 50
Eurosilicone 2 – – –
Smooth 2 – – –
Trilucent 2 – – –
231 16 16 13.9
Unknown 37 3 5 21.6
Devices with duplicate capsules were also found to have been implanted for a significantly shorter time—median 52 versus. 120 months—than non-DC implants (Mann–Whitney U test Z score of 3.62, p = 0.0003).
Discussion
Whilst axiomatic that any foreign body too large for the immune system to either remove or destroy is walled off with a protective fibrous tissue barrier, adverse capsular contraction (ACC) continues to impact mammary prosthesis implantation. The host-to-device interface is important, and evidence exists for a beneficial reduction with textured silicone elastomers, although most obviously in the subglandular plane [7]. The rough topography of polyurethane devices precipitated the development of aggressively textured silicone elastomers, but we are now aware that devices at the more aggressive end of the spectrum may both reduce ACC less than anticipated [3, 8] and contribute to additional problems.
Studies suggest an overall duplicate capsule prevalence of 2.2–8.3% [3, 4]; however, both studies saw DC only in Biocell macro-textured devices where the prevalence was 13.3% in Hall-Findlay’s series. Van Slyke et al. did not detail the breakdown of their 45 DCs in 123 Biocell patients so their rate sits somewhere between 18.3% (45/246) and 36.6% (45/123), extrapolating unilateral and bilateral, respectively. Although not yet peer review published, Sforza et al. presented a 6.7% (53/794) prevalence of DC in a 3-year series of patients with Biocell devices in primary breast augmentation [9]. Our rates of patient and device prevalence of 14.9% and 7.5% are broadly similar and provide important supporting evidence for the high rates in macro-textured devices. Thus far, there have been no reports of DC with non-macro-textured mammary implants; however, as Table 2 shows, other topographies are not immune.
Duplicate capsule formation is an incompletely understood entity first documented with two cases in association with the now defunct Trilucent device [2]. Being known to stimulate a significant inflammatory reaction, causation was duly implicated possibly in combination with trans-elastomer filler bleed [10]. A third case prompted the mechanical aetiology hypothesis: that is, shear force disruption of an original capsule particularly strongly adherent to an aggressively textured elastomer [8]. Haemorrhage and subsequent organisation lead to the formation of the second capsule and low friction between the two capsular surfaces further facilitates malrotation of the original capsule-implant construct within the neo-capsule. A small study of Biocell-associated DCs has recently shown the inner lamina to range between a thin layer of nascent capsule and an identifiable organised one. Histological analysis revealed hypocellular fibrosis with synovial metaplasia and provided a clinicopathological correlation between DCs and failed tissue adherence [11].
With macro-textured devices acknowledged by their proponents to harbour more bacteria [12], an infective element may also be part of the multifactorial pathogenesis of DC. Implants inserted into a porcine model were co-inoculated with S epidermidis, and after 20 weeks, two of the six were found to have duplicate capsules and biofilm [13]. Moreover, scanning electron microscopy has shown significantly greater bacterial loads and biofilm deposition at the implant-capsule interface, as compared to the inter-capsular surface, suggesting that a degree of impaired capsule formation might facilitate delamination [14, 15]. Lately, strong biofilm formation has been demonstrated with a range of Gram-positive micro-organisms with thicker growths on textured (micro- and standard) rather than smooth implant surfaces. [16]
DC has been reported after both primary [3] and secondary breast augmentation [5], although the latter far less frequently. One patient with 3 prior sets of implants presented with seroma 4 years after the exchange and capsulectomy of a ruptured aggressively textured (Biocell Style 110) implant [5]. Our cohort comprised predominantly primary augmentations, but 28.4% had undergone at least one previous implant exchange (range 1–7 procedures). A Fisher’s exact test of 0.1861 (p < 0.05) suggests implant texture itself to be far more important a determinant of DC formation than ‘surgical load’.
Whilst no precise definition of implant texture has hitherto been available due to proprietary considerations, a recent classification (Table 3) has been proposed based on characteristics including implant hydrophobicity, macrophage-based biocompatibility assay and surface roughness characteristic (Sa) [17]. The coarsest surface, with pore sizes in excess of 300 µm diameter, is known as macro-textured. The underlying premise is one of causality of ACC, or not, through texture-mediated inflammation at the prosthesis interface. Allergan’s devices employ a ‘salt-elution’ technique, and their intention was to maximise host tissue ingrowth in order to resist rotation. The main alternative texturisation process is the ‘imprint’ technique as favoured by manufacturers such as Mentor [6]. Interestingly, although the macro-texture was invented to mimic the contracture-minimising topography of polyurethane, reduced contracture has perhaps counter-intuitively been associated with less aggressively textured surfaces [8].Table 3 Characteristics of mammary prosthesis surface textures. After Barr et al. [17] and Maxwell GP et al. [19]
Surface Sa Pore diameter Devices
Macro- Sa > 75 µm 600 – 800 µm Biocell (Allergan); Sebbin
Micro- 10 µm < Sa < 75 µm 70 – 150 µm Siltex (Mentor); Eurosilicone; PIP; Sientra
Meso- Sa < 15 µm Silk surface™
Nano- Sa < 5 µm SM smooth (Mentor)
Sa = roughness
We observed two broad macroscopic patterns of DC: subtotal and complete encasement. The latter is self-explanatory and represented well by the case in Fig. 1. ‘Subtotal’ is incomplete so tends to deform implants to a lesser extent (Fig. 2). Of course, such division may well be arbitrary and simply reflect a temporal continuum. We have been unable to correlate time with degree of encapsulation not least because the date of surgery rarely coincides with actual DC appearance, merely recording its previous occurrence.
We also investigated the duration of implantation. Those without DC had been in situ for a median of 120 months (range 7–408) as compared to 52 months (6–252). Mann–Whitney U test Z score of 3.62 yielded a highly significant p value of 0.0003. Many of the macro-textured prostheses had been used in exchange of PIP devices so the resulting DCs had occurred within 5 years; thus, speed of onset may equally have prompted exchange. This latter point certainly generated a considerable degree of patient complaint.
We wonder whether complete DCs may be an entity. The key symptom triad includes a very firm and mobile construct. It almost appears separate and, unlike other high-grade capsular contractures, is pain-free. This hard ‘ball in a pocket’ effect is presumably explained by the contractile forces of the neo-capsule being applied to the insensate prosthesis rather than the patient’s parenchyma as with standard capsular contracture of the new, external capsule. With many of the DCs occurring within 5 years, the speed of onset would be considered rapid as alluded to previously [13]. Whilst some authors have noted an association between DC formation and delayed seroma collection [3, 14], this was not observed in our cohort.
There were 6 patients with at least one ruptured device in association with DC. Whilst such small numbers preclude worthwhile statistical analysis, it is worth noting that whilst 4 were bilateral and the degree of encapsulation varied, 5 (83.3%) were macro-textured. Moreover, they had been in situ for less than 5 years, much shorter than the single ruptured PIP device. Time will tell whether the severe contracture of the peri-prosthetic capsule is so much more forceful than a standard contracture that it actually precipitates elastomer rupture. An alternative explanation might be that patchy or subtotal DCs produce a localised concentration of force at the point of adherence. This then exaggerates fold-flaw defects predisposing to consequent elastomeric fragility.
Another interesting phenomenon has been the frequent observation of a ‘mesentery’ on the posterior surface. Additionally, and as observed by other authors [3, 18], the filling disc is devoid of any encapsulation (Fig. 3). With this element being smooth, one might conclude that some form of texturisation is fundamental to DC formation.
There are inherent limitations with any non-prospective study, for example, incomplete information about prior device history; however, we feel this to be considerably ameliorated by the bias minimisation associated with a single, highly experienced surgeon and good subject retention. It is, however, frequently the situation in which surgeons find themselves when undertaking breast prosthesis exchange so has clinical relevance.
It is worth noting that a defence of the aggressively, or macro-textured, implant surface was posited by a group of KOLs who reiterated that large pore diameters and depths promoted implant immobilisation through enhanced adhesion [19]. Whilst acknowledging a ‘slightly higher risk of double capsule and later seroma’, their defence regarding adverse capsular contracture (ACC) included only those studies that compared the Biocell (aggressive) with smooth devices. Direct comparisons of the 4 main grades of textured surface prostheses have yet to be made and will undoubtedly make interesting reading when they eventually appear.
There are no official figures for the prevalence of DC, other than Allergan’s own rate of 0.02%, although this figure has yet to be subjected to peer review [19]. The consensus group also observed that DC is ‘not necessarily associated with complications’ and, given that a proportion are discovered incidentally, this may be true in some cases. However, when they do present, our experience is of rather dissatisfied patients who not only have to undergo further intervention, but significant cost, not long after their last surgical episode. Surgery is also not entirely risk-free, and no doubt contributes to the risk of infection, with device loss, future capsular contracture, and so on.
Finally, there was no mention in the consensus paper [19] as to whether any differences exist between primary and secondary cases vis-a-vis recurrence and our series materially contributes. Previous implantation has clearly changed the local milieu both irrevocably and in an incompletely understood fashion. The relative contributions to the genesis of a second capsule are at present unknown and unknowable; however, similar series of PIP implants have reported far less ACC over a much longer period [20].
Conclusion
Our series reports an entity more prevalent than most are aware of. Our figures support the previous studies showing both a higher prevalence of DC in macro-textured devices along with their earlier failure rate. Our higher rate may well reflect a greater proportion of secondary cases, and we report for the first time the phenomenon of breast implant capsule duplication in non-macrotextured devices.
We also characterise 3 key clinical features of the complete duplicate capsule—a particularly firm and mobile, yet pain-free, construct that appears to move independently of the breast parenchyma on palpation.
Acknowledgements
Neither author has any additional financial or corporate affiliations to disclose.
Declaration
Human or Animal Rights
This article does not contain any studies with human participants or animals performed by any of the authors.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Zogheib S Hanna C Daou B Breast implant-associated double capsules: what do we know so far? A systematic review of the literature Aesthet Plast Surg 2022 46 1 35 42 10.1007/s00266-021-02443-9
2. Colville RJI Dickson MG True double capsules in oil-based (Trilucent) breast implants Br J Plast Surg 2002 55 3 270 271 10.1054/bjps.2002.3813 12041998
3. Hall-Findlay EJ Breast implant complication review: double capsules and late seromas Plast Reconstr Surg 2011 127 1 56 66 10.1097/PRS.0b013e3181fad34d 21200201
4. Van Slyke AC Carr M Carr NJ Not all breast implants are equal: a 13-year review of implant longevity and reasons for explantation Plast Recon Surg 2018 142 3 281 289 10.1097/PRS.0000000000004678
5. Park BY Lee D-H Sy L Is late seroma a phenomenon related to textured implants? A report of rare complications and a literature review Aesth Plast Surg 2014 38 1 139 145 10.1007/s00266-013-0232-z
6. Brody GS Deapen D Taylor CR Anaplastic large cell lymphoma occurring in women with breast implants: analysis of 173 cases Plast Reconstr Surg 2015 135 3 695 705 10.1097/PRS.0000000000001033 25490535
7. Collis N Coleman D Foo IT Sharpe DT Ten-year review of a prospective randomized controlled trial of textured versus smooth subglandular silicone gel breast implants Plast Reconstr Surg 2000 106 6 786 791 10.1097/00006534-200009020-00005 11007389
8. Pandya AN McLean NR Cross PA Capsule within a capsule: an unusual entity Br J Plast Surg 2002 55 5 455 456 10.1054/bjps.2002.3864 12372386
9. Sforza M, Andjelkov K, Zaccheddu R, et al (2014). Complaints, high revision rates and liability: what to expect after using the aggressively textured silicone breast implants? Presentation - The Aesthetic Meeting, San Francisco April 28th 2014
10. Cross PA Double capsule or capsule within a capsule: is there a difference? Br J Plast Surg 2003 56 1 76 10.1016/S0007-1226(02)00471-X 12706168
11. Van Slyke AC Roller JM Alaghehbandan R Defining double capsules: a clinical and histological study Aesthet Surg J 2021 41 11 NP1437 NP1444 10.1093/asj/sjab045 33655295
12. Adams WP Jr Culbertson EJ Deva AK Macrotextured breast implants with defined steps to minimize bacterial contamination around the device: experience in 42,000 implants Plast Reconstr Surg 2017 140 3 427 431 10.1097/PRS.0000000000003575 28841597
13. Allan JM Jacombs AS Hu H Detection of bacterial biofilm in double capsule surrounding mammary implants: findings in human and porcine breast augmentation Plast Reconstr Surg 2012 129 3 578e 580e 10.1097/PRS.0b013e3182419c82 22374027
14. Giot J-P Paek LS Nizard N The double capsules in macro-textured breast implants Biomaterials 2015 67 65 72 10.1016/j.biomaterials.2015.06.010 26210173
15. Danino MA Nizard N Paek LS Do bacteria and biofilm play a role in double capsule formation with macrotextured implants? Plast Reconstr Surg 2017 140 5 878 883 10.1097/PRS.0000000000003767 29068919
16. Rezende-Pereira G Albuquerque JP Souza MC Biofilm formation on breast implant surfaces by major gram-positive bacterial pathogens Aesthet Surg J 2021 41 10 1144 1151 10.1093/asj/sjaa416 33378420
17. Barr S Hill EW Bayat A Functional biocompatibility testing of silicone breast implants and a novel classification system based on surface roughness J Mech Behav Biomed Mater 2017 75 75 81 10.1016/j.jmbbm.2017.06.030 28697402
18. Robinson H Breast implant complication review: double capsules and late seromas: letter Plast Reconstr Surg 2011 128 3 818 10.1097/PRS.0b013e3182221513 21866021
19. Maxwell GP Scheflan M Spear S Benefits and limitations of macrotextured breast implants and consensus recommendations for optimizing their effectiveness Aesthet Surg J 2014 34 6 876 881 10.1177/1090820X14538635 25024450
20. Berry MG Stanek JJ The PIP mammary prosthesis: a product recall study J Plast Reconstr Aesthet Surg 2012 65 6 697 704 10.1016/j.bjps.2012.02.019 22405818
| 36443419 | PMC9707171 | NO-CC CODE | 2022-12-01 23:20:21 | no | Aesthetic Plast Surg. 2022 Nov 28;:1-6 | utf-8 | Aesthetic Plast Surg | 2,022 | 10.1007/s00266-022-03182-1 | oa_other |
==== Front
Nat Rev Clin Oncol
Nat Rev Clin Oncol
Nature Reviews. Clinical Oncology
1759-4774
1759-4782
Nature Publishing Group UK London
36443594
707
10.1038/s41571-022-00707-0
Review Article
Criteria for the translation of radiomics into clinically useful tests
http://orcid.org/0000-0002-8458-655X
Huang Erich P. [email protected]
1
O’Connor James P. B. 2
http://orcid.org/0000-0001-8195-3206
McShane Lisa M. 1
Giger Maryellen L. 3
http://orcid.org/0000-0001-7961-0191
Lambin Philippe 4
Kinahan Paul E. 5
Siegel Eliot L. 6
Shankar Lalitha K. 1
1 grid.94365.3d 0000 0001 2297 5165 Division of Cancer Treatment and Diagnosis, National Cancer Institute, National Institutes of Health, Rockville, MD USA
2 grid.18886.3f Division of Radiotherapy and Imaging, Institute of Cancer Research, London, UK
3 grid.170205.1 0000 0004 1936 7822 Department of Radiology, University of Chicago, Chicago, IL USA
4 grid.5012.6 0000 0001 0481 6099 Department of Precision Medicine, Maastricht University, Maastricht, Netherlands
5 grid.34477.33 0000000122986657 Department of Radiology, University of Washington, Seattle, WA USA
6 grid.411024.2 0000 0001 2175 4264 Department of Diagnostic Radiology, University of Maryland, Baltimore, MD USA
28 11 2022
114
2 11 2022
© This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Computer-extracted tumour characteristics have been incorporated into medical imaging computer-aided diagnosis (CAD) algorithms for decades. With the advent of radiomics, an extension of CAD involving high-throughput computer-extracted quantitative characterization of healthy or pathological structures and processes as captured by medical imaging, interest in such computer-extracted measurements has increased substantially. However, despite the thousands of radiomic studies, the number of settings in which radiomics has been successfully translated into a clinically useful tool or has obtained FDA clearance is comparatively small. This relative dearth might be attributable to factors such as the varying imaging and radiomic feature extraction protocols used from study to study, the numerous potential pitfalls in the analysis of radiomic data, and the lack of studies showing that acting upon a radiomic-based tool leads to a favourable benefit–risk balance for the patient. Several guidelines on specific aspects of radiomic data acquisition and analysis are already available, although a similar roadmap for the overall process of translating radiomics into tools that can be used in clinical care is needed. Herein, we provide 16 criteria for the effective execution of this process in the hopes that they will guide the development of more clinically useful radiomic tests in the future.
Despite a considerable increase in research output over the past decades, the translation of radiomic research into clinically useful tests has been limited. In this Review, the authors provide 16 key criteria to guide the clinical translation of radiomics with the hope of accelerating the use of this technology to improve patient outcomes.
Key points
Despite tens of thousands of radiomic studies, the number of settings in which radiomics is used to guide clinical decision-making is limited, in part owing to a lack of standardization of the radiomic measurement extraction processes and the lack of evidence demonstrating adequate clinical validity and utility.
Processes to acquire and process source images and extract radiomic measurements should be established and harmonized.
A radiomic model should be tested on external data not used for its development or, if no such dataset is available, tested using proper internal validation techniques.
Model outputs should be shown to guide disease management decisions in a way that leads to a favourable risk–benefit balance for patients.
Clinical performance should be assessed periodically in its intended clinical setting (task and population) after model lockdown.
A list of 16 criteria for the optimal development of a radiomic test has been compiled herein and should hopefully guide the implementation of future radiomic analyses.
Subject terms
Translational research
Medical imaging
Cancer screening
Diagnostic markers
Machine learning
==== Body
pmcIntroduction
For decades, computer-aided diagnosis (CAD) algorithms have made use of computer-extracted tumour characteristics for improved disease detection and diagnosis, treatment planning, and follow-up1, with some particularly notable developments in breast and lung cancer screening2,3. More recently, radiomics, involving high-throughput computer-extracted quantitative characterization of healthy or pathological structures and processes as captured by in vivo medical imaging, has emerged as an extension of CAD4. Similar to other ‘omics’ technologies, the extraction of such large quantities of information from images obtained during standard clinical workflows enables extensive tumour characterization and facilitates assessments of both within-tumour and between-tumour heterogeneity and longitudinal changes1. Interest in both CAD and radiomics (two terms that are occasionally used interchangeably) has increased substantially within the past two decades; a PubMed search for “(computer-aided diagnosis) OR CAD OR radiomic OR radiomics AND (cancer OR tumours OR tumours)” yields over 44,000 publications since 1967, over 85% of which are from 2005 onwards (Fig. 1).Fig. 1 Number of publications per year since 1967.
A PubMed search of “(computer-aided diagnosis) OR CAD OR radiomic OR radiomics AND (cancer OR tumours OR tumours)” was performed. The number of items published each year is presented as of 20 September 2022.
Similar to CAD, radiomics can assist with clinical decision-making. Radiomic features, namely measurements extracted from medical images (currently usually CT, MRI or digital radiography), are combined with data on clinical characteristics and from other omics analyses to detect disease, predict the likelihood of death, disease progression and/or recurrence by a specific time point, evaluate response to therapy or identify an appropriate course of treatment. The ultimate goal of radiomic analyses should be the development of a test, defined by the FDA–NIH Biomarker Working Group as a system comprising materials for measurement, procedures for measurement, and methods or criteria for interpretation5, that can be used to guide medical decision-making as in disease diagnosis and management.
Despite a dramatic increase in research output over the past two decades (Fig. 1), the vast majority of radiomic studies have not yet led to clinically useful tests. Across all medical indications, 343 artificial intelligence and machine learning-based tests currently have FDA clearance, only a small proportion of which are based on radiomics6. This lack of clinical translation might be attributable to several factors. The vast majority of radiomic studies assess correlations between certain radiomic features and a biological or clinical end point of interest; therefore, the added value of the radiomic test (such as improved clinical performance or reduced invasiveness) is often neglected as is clinical utility, namely that acting upon the information provided leads to a favourable benefit–risk balance for the patient. Additionally, as established in the statistical and machine learning literature, analyses of high-throughput data, such as those obtained using radiomics, are fraught with potential issues, including insufficient data for development and validation and improper application of statistical methodology for the specific purpose of the test. Furthermore, different studies have used widely varying protocols for image acquisition and feature extraction. Several studies have shown the effects of differences in data acquisition, image reconstruction and image post-processing on downstream analyses; different software platforms or even different versions of the same software can produce widely varying results regarding the strength and direction of the associations between features and outcomes7.
Existing guidelines on the acquisition and analysis of radiomic data include a radiomic quality score to evaluate the completeness and appropriateness of such an analysis8, computational procedures for commonly used types of features9, and protocols for image acquisition, feature extraction and statistical analysis10,11. However, radiomics would also benefit from a roadmap for the entire process of translating radiomic data into clinically useful tools for guiding clinical care, encompassing not only recommendations for image acquisition and processing, feature extraction, and statistical analysis but also aspects such as test lockdown and demonstrating clinical utility. Such a roadmap has yet to be published for radiomics, although similar criteria and guidelines have been compiled for other omics technologies12.
Herein, we present a 16-point list of criteria for the translation of radiomics into clinically useful tests. These criteria (Box 1) were developed by radiologists, physicists and statisticians with extensive experience with radiomics and other omics technologies, and are based on analogous recommendations developed for other omics technologies12. These criteria are also adapted to accommodate issues that are unique to radiomics, such as vendor-driven changes in imaging technology and software and the dynamic nature of certain models, and are intended to help researchers to navigate the translation process and catalyse an increase in the number of clinically useful radiomic tests.
Box 1 The 16 criteria for the translation of radiomics into a clinically useful tool Clinical use of the test Criterion 1: the intended role of the radiomic test in clinical decision-making and the target population is clearly specified.
Criterion 2: use of the test in clinical care is expected to lead to patient benefit beyond that resulting from a standard-of-care testing procedure.
Imaging and feature extraction Criterion 3: protocols for image acquisition, processing and normalization are standardized to optimize image quality and minimize variability owing to factors such as device or operator.
Criterion 4: list of quantities to extract from the images and computational procedures to do so are specified.
Criterion 5: the technical validity of feature measurements is adequately robust.
Criterion 6: procedures to correct for technical artefacts in the feature measurements are in place.
Model development and validation Criterion 7: images, outcomes and other relevant data are obtained for patients in the target population through prospective clinical trials or retrospectively via completed studies, databases or image repositories.
Criterion 8: the radiomic model is developed with guards against overfitting.
Criterion 9: the ability of the model to predict an end point of interest is shown to be sufficiently robust using proper model validation techniques.
Criterion 10: imaging, feature extraction, data preprocessing and computational procedures are fully specified and locked down to the greatest extent possible.
Criterion 11: each possible value of the test output is associated with an unambiguous interpretation with regard to clinical management decisions.
Criterion 12: the reproducibility of the test output with regard to variables, such as imaging centre or operator, is sufficiently high.
Criterion 13: measures are in place to monitor drift, permit lockdown relaxation as appropriate and re-evaluate test performance as needed.
Justifying the use of the test in clinical care Criterion 14: the test outputs are shown to be clinically valid.
Criterion 15: prospective or prospective–retrospective studies are conducted to directly evaluate the performance of the radiomic test in the context of its intended use.
Criterion 16: the patient benefit resulting from acting upon the result of the radiomic test is sufficiently substantial to warrant its use in clinical care.
Clinical application
Prior to any formal development and validation, the intended clinical use of the radiomic test and the target population should be established (criterion 1). The use of the test in clinical care should be expected to guide disease assessment and management decisions in a way that leads to a favourable benefit–risk tradeoff and offers advantages over other tests designed to serve the target population in the same role (criterion 2). The intended clinical use will have important implications for the subsequent stages of development and validation, including which features to extract from the imaging data, the optimal imaging time points and the design of the clinical trial to directly assess the performance of the test in its intended role.
Criterion 1: intended role and target population
Radiomics is often used for either screening or cancer diagnosis. For example, MRI radiomics is useful for the diagnosis of breast abnormalities13 and CT radiomics for the detection of lesions in various organs, including lungs, brain and prostate14. The use of radiomics in prognostication, namely predicting the clinical outcomes of patients undergoing standard therapy, is an area of increasing research interest15; for example, CT-based radiomics might be a useful method of predicting the outcomes of patients with head and neck squamous cell carcinomas or non-small-cell lung cancer receiving standard-of-care therapies16. Radiomic tests can also be used for treatment selection, namely as assays designed to indicate benefit, or lack thereof, from a specific class of therapies; for example, a model of oestrogen receptor expression based on tumour size, shape and entropy features on dynamic contrast-enhanced MRI (DCE-MRI) has been developed to inform treatment selection for patients with breast cancer17. Radiomic tests might also be used to assess response to treatment and monitor disease status18–20.
Roles in which radiomic tests could serve have been summarized comprehensively elsewhere21. In certain scenarios, the same radiomic test can have more than one role; for example, the aforementioned model of oestrogen receptor expression might also be useful for prognostication17. However, ‘off label’ use of radiomic tests, namely application in a role other than the one for which the test has been shown to be clinically useful, is discouraged. The criteria for clinical performance depend strongly on the intended role (see criterion 14) as is typical in the regulatory clearance and approval processes applied to both new drugs and medical devices. Diagnostic radiomic tests should have an adequate level of accuracy in detecting disease. Prognostic radiomic tests should have an adequate ability to predict death, disease recurrence or progression depending on the intended role of the test. Tests designed for therapy selection should also be sufficient to predict outcomes, such as death or disease progression, in patients receiving the therapy of interest. If the goal is to guide the choice between one treatment and a designated alternative approach, the outcomes of patients receiving each therapy need to be studied. However, if the predictive goal is merely to identify those patients who are most likely to respond to a particular therapy, then the test should have adequate ability to predict either a response or a level of expression of an established predictive biomarker sufficient to indicate a response to the treatment of interest. The translation process outlined in this Review should therefore be applied for each role in which a specific radiomic test is likely to be useful.
Aspects of the target population to specify include those pertaining to disease characteristics (such as primary tumour types and grades, disease stage, molecular subtypes, risk groups and receptor expression status) and treatment history. A radiomic test might also be useful in multiple target populations; the test based on the model described by Aerts et al.16, for example, might be useful for predicting the outcomes of patients with head and neck cancer or non-small-cell lung cancer receiving standard-of-care therapies. However, researchers are encouraged not to assume, without appropriate evidence, that the utility of a radiomic test extends across target populations because the technical performance of the imaging device and feature extraction software and the clinical performance of the test might not be consistent across different populations.
Criterion 2: patient benefit from use of the test in clinical care
The benefit of using a radiomic test should be clearly specified in the context of available treatments for the target population and access to other tests serving similar roles. A radiomic test might be used to stratify patients to optimize the choice of therapy for each individual, thus sparing patients from receiving ineffective or unnecessary treatments. A predictive test designed to guide treatment selection might differentiate between patients who are likely to derive clinical benefit (such as a longer median progression-free survival (PFS) or overall survival duration) from a specific therapy or class of therapies and those that will not. A prognostic test could identify patients with particularly poor outcomes on standard-of-care therapy who might consider a more intensive regimen; however, such a test will probably only be useful if a suitable, alternative treatment is available22. Moreover, a radiomic test might help to direct clinical management in a way that treatment-related toxicities, including financial toxicities, are reduced; prognostic tests might also identify patients whose outcomes on standard well-tolerated regimens are so good that they need not consider additional highly aggressive or toxic treatments or may consider treatment de-escalation.
The decision to use a radiomic test over other tests addressing the same clinical problem should be supported by a compelling reason. The radiomic test could have superior clinical performance to a standard test serving in the same role. The radiomic test might be able to identify underlying characteristics that cannot be detected as easily using other means; for example, assessing intratumour and intertumour heterogeneity of oestrogen receptor expression might be much less difficult when using radiomic tests compared with immunohistochemistry assays. Alternatively, the radiomic test might have a similar level of clinical performance but reduced invasiveness (such as biopsy avoidance), a reduced financial burden, greater convenience, or a reduction of one or more associated risks (potential harms, discomforts or exposures inherent to the testing procedure).
Imaging and feature extraction
Standard operating procedures for imaging, including protocols for the administration of contrast or imaging agents, specifications for image acquisition, procedures for image processing, and the timing of the scans should be in place (criterion 3) as should those for feature extraction, including a list of quantities to compute from the imaging data, segmentation algorithms, and computational algorithms and software to compute these quantities (criterion 4). The resulting feature measurements should also have been shown to have adequate technical validity (criterion 5). In most cases, this would entail each feature exhibiting strong repeatability and reproducibility or, if feasible, robust agreement with a standardized reference measurement of the underlying characteristic. A procedure to correct feature measurements for technical artefacts (the effects of factors such as imaging centre, device, operator or device-calibration settings on the distribution of the feature measurements) should also have been developed (criterion 6).
Criterion 3: standard operating procedures for image acquisition and processing
Image acquisition parameters should be specified in order to optimize image quality (for example, by keeping imaging noise to an acceptably low level or ensuring that the spatial, contrast and/or temporal resolution is adequate) and should be standardized to maximize reproducibility across imaging centres, devices and operators. Numerous studies have demonstrated the strong dependency of the resulting feature measurements on the imaging protocol23,24. Standard operating procedures for image acquisition could be based on established imaging guidelines such as those provided by the American College of Radiology25, the Society of Nuclear Medicine and Molecular Imaging26, the European Association of Nuclear Medicine27 or the Quantitative Imaging Biomarker Alliance28.
Image acquisition protocols will depend on the intended use as well as on the imaging modality and features that will be extracted. If the radiomic test is intended for diagnosis, spatial resolution will be an important consideration29. In theory, tests involving the analysis of morphological features will depend more on spatial resolution30, whereas kinetic features, such as those derived from fast DCE-MRI, will depend more on temporal resolution31. In practice, perfect standardization is infeasible as is optimization of the protocol with respect to all the features to be extracted. The optimal resolution in DCE-MRI for breast cancer diagnosis often involves a compromise between spatial and temporal parameters to obtain measurements of morphological and kinetic features with adequate technical validity32. Furthermore, image acquisition protocols, particularly those applied to standard-of-care imaging approaches, are often determined in an ad hoc manner.
The time points at which patients undergo imaging should also be specific to the intended use. Radiomic tests intended for treatment selection will involve scans obtained prior to intervention. Those intended for response assessment will involve scans obtained not only prior to the intervention but also at specified time points during and after therapy (often termed ‘delta-radiomics’33). The timing of response assessment can vary substantially; for example, radiomic tests designed to measure early metabolic response could involve imaging at baseline and then at a certain number of days to weeks following the initiation of treatment34 whereas assessment of the effects of certain classes of therapies, such as anti-vascular agents, might occur in a timescale of hours to days31.
Standard operating procedures should include processes designed to normalize the intensity values of images obtained from different patients and from the same patient. Normalization techniques include image resampling with filtering35, normalizing voxel intensity values relative to a histogram or global and local intensities on a reference image36,37, or harmonizing across different scans obtained from different populations or acquisition sites38. For certain features (such as second-order textural features), discretization through methods such as grey-level resampling and histogram binning is also needed11,39. Although the grey-level and standardized uptake value discretization methods used vary from centre to centre, these values can be normalized relative to a reference set of measurements40,41. Alternatively, standardized image preprocessing methods can be applied42. A comprehensive summary of imaging harmonization methods is provided elsewhere43.
Criterion 4: standard operating procedures for feature extraction
Prior to formal test development, a list of quantities that will be extracted from the imaging data should be established. Traditionally, radiomic features are human-engineered and are extracted through delineation of the tumour from surrounding tissues using manual, semi-automated or fully automated segmentation44–46 followed by application of pre-specified computational procedures to the voxel data within the region of interest10. Human-engineered features include those quantifying size (tumour dimension), shape (3D geometry), morphology (margin characteristics), enhancement texture (the extent of heterogeneity within the texture of the tumour and/or contrast uptake), quantifications of kinetic curves (shape of the curve and quantifications of the physiological process of uptake and washout of the contrast agent) and enhancement-variance kinetics (such as the time course of spatial variance of enhancement within the tumour)47–50.
Extraction of such features will typically involve conversion and harmonization of the imaging data (criterion 6), post-processing (such as interpolation to cubic voxels, denoising, and correction of intensity and partial volume effects), image segmentation, region-of-interest extraction, and feature computation9. Existing guidelines and recommendations can serve as a starting point for the development of a standard operating procedure for feature extraction but will often require adaptation to suit both the target population and the imaging modality51.
Alternatively, features of interest can be computer learned, namely extracted by direct application of computer algorithms to voxel data without the need for human intervention such as those computed using deep learning networks52,53. In this approach, a deep learning network can be applied to the voxel-level data and the last layer of the underlying convolutional neural network is taken as a set of features, similar to those used by Li et al. to predict IDH1 mutation status in patients with low-grade gliomas54. An illustration of the differences between such features and human-engineered ones is provided in Fig. 2. Computer-learned features have been considered in conjunction with operator-dependent features55 or even as a replacement. Such features are often less transparent in their computation and less interpretable; nonetheless, they might capture information that human-engineered features cannot, often resulting in more reproducible feature extraction and models with improved performance54. Fully automated extraction of such features enables the processing and computation of larger volumes of data with reductions in the variability of test output values owing to the elimination of human error during processes such as manual delineation and segmentation52.Fig. 2 Types of radiomic analysis.
a, Analyses using human-engineered features. Different types of features (such as histogram, shape or texture) are extracted from the images according to a pre-specified computational procedure. Variable selection techniques are used to identify which of these features are important in diagnosing a medical condition. The values of these selected variables are combined into a model to produce a diagnosis. b, Analyses using machine learning and artificial intelligence algorithms. The voxel-level data are fed into a convolutional neural network consisting of multiple hidden layers whose output is used to produce a diagnosis.
Criterion 5: technical validity of the feature measurements
Adequate technical validity typically entails assessing the repeatability and reproducibility of the feature measurements. Repeatability describes the precision when a specific imaging and feature extraction standard operating procedure are applied multiple times to the same patient at the same centre by the same operators within a short period of time. Reproducibility describes the precision of repeat measurements when factors such as imaging centre and operator are allowed to vary56–58. Study designs and the statistical methodology for studies assessing repeatability and reproducibility have been summarized in detail elsewhere59. Strong technical validity is important for model development and the establishment of the clinical utility of a radiomic test given that poor feature reproducibility, as mentioned previously, can produce widely varying results regarding the strength and direction of the association between features and outcomes7 and result in models with insufficient levels of performance60.
Ideally, repeatability and reproducibility would be assessed using clinical data. In such clinical studies, patients undergo repeat scans with the feature extraction standard operating procedure then applied to each image. Such studies have been conducted61,62, although they are often difficult in practice as patients can be reluctant to participate owing to a lack of direct benefit, the inconvenience of undergoing multiple scans and, with certain techniques, additional exposure to contrast agents or ionizing radiation. An alternative approach involves different operators extracting features from the same set of images, possibly at different centres; however, this approach, although also feasible as a retrospective method, only enables the assessment of variability attributed to the feature extraction process51,56,58.
As an alternative approach, some components of technical validation can be conducted using in vitro or in silico phantoms, simulated digital reference images or synthetic data such as those produced by generative adversarial network systems40,63. However, conclusions based on data obtained using phantoms and digital reference images will be overly optimistic regarding their technical validity given that they cannot fully capture the complexity of actual patients. Several authors have provided recommendations on the minimum technical validity requirements of phantoms and digital reference images59,64.
Technical validity can also be assessed using the level of agreement between feature measurements and certain comparator quantities (for example, with a measurement of the underlying biological characteristic according to an independent in vitro assay), bias (the mean difference between the measurement and the true value of the characteristic being measured), and the linearity of the relationship between the feature measurement and the true value59. However, assessing agreement is often not possible for computer-learned features owing to a lack of an appropriate biological correlate. Assessing bias or the relationship between the measurement and the true value of the feature being measured is generally only possible with phantoms and digital reference images.
Repeatability and reproducibility can be used as screening criteria to immediately eliminate features with poor technical validity from further consideration for inclusion in the model. Filtering out features in this manner has been shown to improve the level of power in settings with large numbers of features, of which only a small proportion are associated with the outcome of interest65. Such filtering must be done solely on the basis of technical validity and must not use outcome data that will also be used to assess performance of the model under development (criterion 9).
Technical validity criteria are much less well developed for computer-learned features such as those described by Li et al.54; their methodology produced 16,384 dimensional descriptors arranged in 128 × 64 × 2 arrays, for which applying the technical validity assessment methods described above is clearly not feasible. Regardless of the type of feature used, researchers are encouraged to assess the technical validity of the output of any radiomic models based on these features (criterion 12).
Criterion 6: feature measurement correction for technical artefacts
Technical artefacts, namely the effects of factors related to variables such as imaging centre, operator and/or device configurations on the distributions of the feature measurements, can potentially confound the results of subsequent radiomic analyses. For example, a feature with no association with survival might seem to predict outcome if patients who undergo imaging in one location have substantially better outcomes than those undergoing imaging in another centre and if the median feature measurement differs between the two sites owing to variations in image acquisition and processing. Thus, procedures designed to correct the variations in feature measurements created by such factors should be established prior to the development and validation of a radiomic model.
In addition to the image normalization methods described previously (criterion 3), the feature measurements themselves can also be standardized following extraction. These measurements can be normalized relative to a reference set of measurements66 or according to a harmonization model67,68, similar to the approaches used in other omics settings. Features strongly associated with variation from these technical artefacts might then be removed from consideration before model construction69,70.
Model development and validation
Patient-level data, including images, outcomes, standard clinical variables, measurements of in vitro biomarkers and other relevant data, should be obtained from the target population; these data can be obtained prospectively or retrospectively from already completed studies, imaging repositories or health-care databases (criterion 7). A radiomic model should be developed using appropriate statistical or machine learning techniques incorporating safeguards designed to avoid overfitting (criterion 8). The performance of a model in predicting an end point of interest must be shown to be adequately robust using proper model validation techniques (criterion 9). By the end of its development, all aspects of the radiomic test, including the feature preprocessing steps, mechanisms of imputing missing data, the underlying computational procedures, any cut points in the feature measurements themselves and/or the model outputs, must be fully specified (criterion 10). Each possible output value of the test is then linked to an unambiguous interpretation with regard to clinical care (criterion 11) and the reproducibility of these outputs should be shown to be sufficiently strong (criterion 12). Processes designed to address drift in the performance of the radiomic test, which refers to changes arising from factors such as the evolution of image acquisition and processing protocols and feature extraction procedures over time, software upgrades and obsolescence, and replacement of devices with newer models, should be established, including monitoring processes and procedures to perform further technical validation and model adjustment as necessary (criterion 13).
Criterion 7: imaging, outcome and other relevant data from the target population
Data on the performance of radiomic analyses can be acquired prospectively, most often as part of a clearly stated secondary objective in a phase II or phase III trial involving the target population, with standard operating procedures for image acquisition and processing at the desired time points and a feature extraction protocol, guided by the points described previously, written into the protocol. Alternatively, data can be acquired retrospectively from imaging data repositories, health-care databases, or datasets from completed clinical trials, subject to inclusion and/or exclusion criteria involving image acquisition and processing protocols, image quality, and the availability of images acquired at the relevant time points. For example, The Cancer Genome Atlas Breast Imaging Research Group identified patients from The Cancer Imaging Archive repository71,72 for whom gene expression analysis and pretreatment standard-of-care breast MRIs obtained with 1.5 Tesla GE Medical Systems devices were available17,18,73. Any clinical data to be obtained should be matched with the images via unique patient ID numbers.
Sample sizes should be determined according to factors such as the number of events (patients with disease versus without, or observed number of deaths), the type of model to be fitted to the data, the expected strength of the relationship between the features and the outcome, the desired standard error of the performance metric, the variance of the model outputs and their concordance with observed event probabilities74–76. Logistic and Cox regression models constructed using data from too few patients often have lower performance relative to models constructed using larger sample sizes60. Deep learning classifiers can require data from thousands of patients per class owing to their complexity (in preprint77); however, dataset sizes can be reduced with the use of transfer learning through feature extraction or fine-tuning methods78. Smaller numbers of patients can be used for model fitting if the relationship between the features and the outcome is particularly strong. Notwithstanding, sample sizes are often constrained by the amount of data available from the completed studies, image repositories or databases from which they were acquired or, if the radiomic study is a secondary objective of a clinical trial assessing a therapeutic intervention, by the number of patients required to meet the primary objective, which will often be much smaller than what is needed for the radiomic analysis.
Ideally, prospective studies should involve multiple centres and retrospectively acquired data should be obtained from multiple studies or repositories and then combined. Using multiple imaging centres, as opposed to a single one, not only facilitates more rapid accrual of data and accumulation of a sufficient number of patients for reliable statistical modelling and validation but can also result in the acquisition of data from a broader population. However, this approach comes with the risk of introducing technical artefacts into the data that will need to be corrected prior to model development and validation (criterion 6).
Criterion 8: development of the radiomic model with guards against overfitting
The range of model-fitting techniques proposed in the statistical and machine learning literature has been described in detail elsewhere79. The literature suggests that no single model-fitting technique is uniformly superior to any other80 although, regardless of the approach used, care should always be taken to avoid overfitting, that is, fitting an overly complex model to noise in the data and thus producing a model that is only poorly predictive when applied to completely new data. Overfitting risk is high when using more complex models, such as those based on neural networks81 or non-parametric regression, as opposed to simpler ones such as those based on logistic or Cox regression. These simpler models have also been shown to often perform as well, if not better, than their more complex counterparts, especially when the number of variables is large and the underlying relationship between the radiomic features and the end point is neither strong nor complex45,60,82,83.
Inclusion of too many features in the model, which can be viewed as another form of model complexity, is another common cause of overfitting. Models based on high-dimensional data, such as those typically encountered in radiomic settings, are particularly prone to this issue. Eliminating any features with subpar levels of technical validity (such as poor reproducibility) or those associated with batch processing before any formal model development takes place (criteria 5 and 6) might reduce the likelihood of overfitting as will the use of variable selection techniques. Several authors have described common variable selection techniques in greater detail elsewhere79,84. Of note, many of these techniques require the selection of a tuning parameter controlling the stringency of the inclusion criteria for variables in the model (such as a P value below which univariate associations between individual features and the outcome must lie to be included, the number of variables to be included, or regularization parameters in LASSO regression techniques85). The optimal tuning parameter value is typically identified using the data (see criterion 9).
Criterion 9: model validation
Once a model has been developed, with mitigation against possible overfitting, the model should then be shown to be capable of predicting an end point of interest, be it a clinical event or state or a biological characteristic, with a sufficient level of accuracy. Robust model performance does not necessarily imply usefulness in guiding medical decision-making; for example, as mentioned previously, a radiomic test with a high level of diagnostic accuracy or a robust ability to predict treatment response or an end point of interest will not be clinically useful if the improvement in clinical performance is not substantial enough to justify its use over standard-of-care diagnostic workups. The broad principles described in this subsection, as well as those regarding lockdown, clinical validity and clinical utility in subsequent sections, apply to both more traditional human-engineered features and computer-learned features.
The area under the receiver operating characteristic curve86 (AUC) of the model outputs or their sensitivity and specificity can be used to quantify the ability of the model to discriminate between patients with a specific health condition from those without. A related metric to the AUC is the c-index87, which quantifies the ability of the model to predict survival (the probability that among two randomly chosen patients, the one with the higher model output has the shorter survival time). Additionally, assessments of model performance should include calibration, namely the concordance between the predicted and expected probabilities of an event of interest88–90. Calibration curves, namely plots of the observed frequencies of the event versus predicted probabilities, are also used to examine whether the model predictions are consistently either too high or too low91. As emphasized during the discussion of criterion 1, the most appropriate performance metric will depend on the intended use of the radiomic test.
Ideally, model validation should be accomplished by applying the newly developed model, without any alterations to any aspect, to a completely external dataset that was not used in any part of the model development process. External data should be acquired from patients in the target population from whom imaging data were obtained under similar imaging, processing and feature extraction protocols to the data used in model development. Variations in imaging centre, operating personnel, scan acquisition date, and certain methods of imaging and feature extraction (such as device and software version) between the training and validation datasets might be permitted to enable evaluation of the robustness of the model to variability in these factors.
However, adequate external validation is not always performed, primarily owing to the logistical challenges associated with accessing data from an independent cohort. In our experience, the performance of the model is often assessed through internal validation, namely the use of a single dataset for both model development and evaluation. Internal validation involves carefully splitting or subsampling the data to avoid overlap with the data used to develop the model (the training set) and those used to evaluate the performance of the model (the validation set). Internal validation can provide reasonable estimates of the predictive accuracy of the radiomic model, although results obtained in this way might not necessarily be generalizable to completely new data. If model development and internal validation were performed on data that were obtained using obsolete image acquisition and processing protocols or that involved a cohort that was not completely representative of the entire target population (such as patients from a location at which a disproportionate percentage had a poor prognosis), then the results will reflect performance in this setting; performance might be diminished in other settings such as those with updated image acquisition and processing protocols12.
Internal validation methods include split-sample validation92, cross-validation93 or bootstrap validation94; these various techniques have been summarized in detail elsewhere79,95. Cross-validation is usually preferable to split-sample validation when only small sample sizes are available; the latter produces estimates of model performance that are often pessimistically biased (that is, estimates of model performance that are substantially lower than those obtained from external validation) when sample sizes are of about 200 or fewer individuals60,96.
Appropriate internal validation requires the maintenance of strict separation of data used to specify any aspect of the model from those used to evaluate its performance. Any violation of this strict separation results in overly optimistic estimates of the performance97,98. In this regard, full resubstitution, in which the entire dataset is used for both development and validation of the same model, provides the most egregious example. Partial cross-validation, in which the entire dataset is used to select features based on their significant univariate association with outcome followed by cross-validation of the model using only this restricted feature set, is another variant of this inappropriate approach to validation. In a comprehensive review of internal validation approaches, data from simulation studies are presented indicating that, even in a scenario in which the variables have no relationship with an outcome, inappropriate internal validation techniques can still produce an AUC estimate of 0.7–0.8 (ref.97).
The selection of tuning parameters during model development (criterion 8) is yet another stage at which problems in model validation can arise. Often, for each candidate from a list of tuning parameter values, the model is fitted using the training set and then applied to the validation set to obtain a performance metric estimate. The tuning parameter value associated with the optimal performance metric estimate is then identified and this metric estimate is then reported. However, in this approach, some aspects of the model development (the identification of the tuning parameter) took place on data used to estimate the performance metric. Such approaches can lead to biased estimates of the performance metric98. Appropriate validation techniques for use when tuning parameter selection is also involved include a three-way split of the data into training, validation and test sets (the training and validation sets are used to identify the tuning parameter and fully specify the model, which is then applied to the test set to obtain a performance metric estimate)92 or nested cross-validation98.
Criterion 10: radiomic test lockdown
Once the model has been developed and shown to have reasonable predictive accuracy, all components of the test, as described in the Introduction of this Review, should be locked down. In radiomics, procedures for measurement will include both standard operating procedures for image acquisition and processing (criterion 3) as well as those for feature extraction (criterion 4) and calculation of model output. Outputs are then associated with specific clinical interpretations (criterion 11).
All computational aspects of the model (for example, the mathematical expression, including regression coefficients, weightings, cutoffs and any other parameters) should be locked down to the greatest extent possible. In situations in which concise model descriptions are not feasible, such as for those based on deep learning, the underlying computational algorithm and software platform should be closed to further changes and any crucial components, such as the random number generator seeds used to generate the model or the output, should be fixed. Interpretations of the inputs of the model (for example, the variables included in a logistic or Cox regression model involving human-engineered features) are often of interest to researchers as they can provide insights into the degree of importance of each feature in predicting an outcome. For computationally derived model inputs, such as features obtained using deep learning algorithms, methods to aid interpretability include visualizing the latent space discovered through the learning process, post hoc highlighting of the regions of the input images that the model labelled as important and visualization of features from different filters in the convolutional neural network99.
The locked-down model could still be affected by any remaining biases inherent to the data on which it was fitted and validated (such as technical artefacts and distributions of radiomic feature values and outcomes that differ substantially from those of the target population). Allowing the model to evolve over time as new data become available will alleviate some of these effects (criterion 13).
Criterion 11: interpretation of test outputs
Models based on techniques such as support vector machines will produce outputs consisting of discrete categories78, each of which can be linked to a specific clinical interpretation and decision. However, models constructed via most other techniques will produce a quantitative output such as the predicted probability of a specific event of interest. Binning these continuous outputs into a limited number of discrete categories might be desired for the purposes of interpretation and clinical decision-making. For example, a test output value that falls below a prescribed cutoff value might indicate a good prognosis and that additional treatment will not be needed and/or that the likelihood of a response to a treatment is high. Alternatively, a test output value above a prescribed cutoff could indicate a high risk of mortality and that the patient might survive longer on an alternative regimen.
Sometimes, these cutoffs are set arbitrarily to specific quantiles, such as the median, in order to define high-risk versus low-risk groups; however, this approach ignores associations with clinical outcomes. Cutoff optimization and comparisons of the outcomes of patients in each category defined by the cutoffs should be done on separate datasets so as not to violate the principle of separation of data used for model development from those used for validation. When cutoff optimization and outcome comparisons are done using the same data (for example, by applying various cutoffs to a dataset, computing the log-rank test P values of the resulting groups and choosing the cutoffs associated with the lowest P values), the risk of a type I error is increased100. To ensure the test can be applied to one patient at a time, cutoff values should be specified as absolute values rather than as percentiles that would need to be recalculated on the availability of new data.
Analytical approaches that consider the consequences of specific treatment decisions based on the test output have also been proposed as a method for cutoff selection. These methods aim to balance the risks (adverse consequences) of incorrect test results against the benefits (positive consequences) of correct test results. The risk–benefit balance can then be compared to that of the standard-of-care approach for the specific clinical indication or any other competing tests or to the use of no test at all. Such approaches include the decision curve analysis method101. This methodology has been applied to a radiomic study involving features obtained from preoperative CT images in conjunction with images from intraoperative frozen sections and clinical data to differentiate invasive lung adenocarcinomas from preinvasive lesions or minimally invasive adenocarcinomas102.
Criterion 12: test output reproducibility
The reproducibility of the test outputs should be shown to be sufficiently robust to ensure that the radiomic test will produce similar results regardless of where it is performed or by whom. One approach involves having patients undergo repeat scans using an established standard operating procedure without interventions in between. Multiple operators, also possibly at different imaging centres, would then apply feature extraction according to standard operating procedures and the radiomic model to the repeat scans independently of one another. Finally, the model or algorithm underlying the test is applied to the images and feature data. Reproducibility metrics for individual features can also be considered at this stage.
This assessment of reproducibility encompasses variability potentially owing to all aspects of image acquisition and processing, feature extraction, and application of the model; however, this approach is rarely feasible in practice for reasons that include a lack of availability of repeat imaging in many scenarios and the unwillingness of many patients to undergo multiple scans within a short space of time. Alternatively, both the feature extraction process and the model can be applied repeatedly to the same set of images, possibly by different operators at different locations. This approach can be applied to retrospectively acquired data but can only produce an assessment of reproducibility that encompasses variability owing to feature extraction and application of the model (and not factors that influence raw data acquisition). If estimates of the repeatability and reproducibility of individual features are known, error propagation models and simulation approaches can be used to estimate the reproducibility of the test output60.
Criterion 13: processes to address data and radiomic test drift
The computational procedures underlying most radiomic tests are likely to evolve over time. Imaging hardware and computational software are likely to be upgraded. Furthermore, the model itself could change after fitting to new data102. Monitoring for such changes in a way that enables their effects to be assessed should be in place. Certain changes might also require a return to previous steps in model development and validation. Changes not related to drift, such as application of the test in a different patient population or indication or the addition of new features, should necessitate a return to model development and validation and might also require the re-establishment of standard operating procedures for feature extraction with re-assessments of the technical validity of individual features.
Assessments of technical and clinical validity and clinical utility (criteria 14 and 15) should be performed periodically for tests for which the underlying computational procedure is expected to evolve over time. Changes to the standard operating procedures for image acquisition and processing or upgrades to the feature extraction software should be followed by assessments of the level of agreement between feature measurements obtained under the previous and the new versions. Researchers should proceed with the new versions of the standard operating procedures and software platforms based on the degree of agreement; however, empirical guidelines on what constitutes a sufficiently strong level of agreement are not available and are probably dependent on both the feature itself and the context. If this agreement in feature measurements is inadequate, the level of concordance between the test outputs computed using the two versions can be assessed (for example, by demonstrating that the mean squared difference of the outputs from the two versions is lower than some meaningful threshold). High concordance between the test outputs indicates that the two versions produce similar results and that the new one could therefore safely replace the previous one, although poor concordance might also reflect the superior clinical performance of the new version. In some scenarios, the model might need to be refitted; such changes can alter the significance and sometimes even the direction of the association of the features with an outcome of interest7.
Justifying use in clinical care
Robust performance of the underlying model in predicting an end point of interest does not automatically mean that the test will be clinically useful (meaning that acting upon the results of the radiomic test leads to patient benefit via improved outcomes or quality of life, reduction in toxicity, invasiveness, risk of complications, financial burden, or the avoidance of ineffective or unnecessary treatments). After the radiomic model has been validated and the test has been locked down, its clinical validity, namely the ability of the outputs to provide information regarding the presence or absence of a condition or the risk of an event of interest103 (for example, sensitivity, specificity, or positive and negative predictive values in detecting disease or the proportion of patients classified as low-risk who remain progression free at 5 years), should be assessed in the context of its intended use and clinical setting (criterion 14). The clinical utility of the test should then be assessed using a prospective study or an appropriately designed prospective–retrospective study, in which the performance of the test in its intended clinical setting is directly assessed (criterion 15) and the risk–benefit balance for the patient when acting upon the results of the radiomic test is shown to be sufficiently favourable to justify use in clinical care (criterion 16). Note that such scenarios often do not reflect the standalone performance of the radiomic model but rather how the test influences the end user (for example, the clinician) when they make clinical decisions with and without the test results as was often used in assessing CAD systems104,105.
Criterion 14: clinical validity of the test
Clinical validation goes beyond model validation (criterion 9) in that the former involves the evaluation of model performance with greater specificity to the clinical setting and intended use. For example, model validation of a prognostic radiomic test might involve showing that the level of concordance between overall survival and model outputs is above some pre-specified and meaningful threshold. Clinical validation, meanwhile, might involve demonstrating that patients who have been classified in a low-risk category have a very high (>90%) 5-year PFS on a well-tolerated standard therapy regimen whereas those in other risk categories have substantially worse outcomes. This may suggest that patients in the low-risk category may potentially consider foregoing additional highly invasive or toxic treatments. Alternatively, showing clinical validity of a prognostic radiomic test might entail demonstrating that the association between test output and clinical outcome remains statistically significant even after adjusting for standard clinical or pathological variables with known prognostic value. The robustness of such a finding to the effects of potential confounders, such as variations in the operator of the feature extraction or the imaging centre in which the extraction and test were performed, should also be established. Different approaches have been summarized in detail elsewhere21.
The radiomic test should be fully locked down and the data used to determine clinical validity should be independent from any data used in model development and validation. Such data could come from prospective clinical trials. For example, to estimate the 5-year PFS of patients with low-risk disease according to the radiomic test, such a cohort could receive standard-of-care therapy and comparisons of the outcomes of the different risk groups could be made after 5 years of follow-up monitoring. Alternatively, data might also be acquired retrospectively from completed clinical trials or imaging data repositories such as The Cancer Imaging Archive71 or the sequestered commons from Medical Imaging and Data Resource Center106, from which testing data can be drawn based on the clinical question and population of interest. Again, this approach assumes that imaging data for a sufficient number of patients from the target population were acquired using protocols similar to the previously established standard operating procedures (criterion 3).
Criterion 15: direct evaluation of performance of the test in its clinical use
The optimal design, end points and statistical analyses to assess the benefits of using a radiomic test to guide clinical disease management differ widely depending on the intended use of the test21. For example, for a radiomic test expected to outperform an in vitro prognostic assay currently in widespread use, patients whose treatment decisions were based on the radiomic test should be shown to have substantially improved outcomes compared to those of patients for whom clinical care was dictated by the in vitro assay.
Prospective studies have numerous desirable qualities, including enabling researchers to have full control over the features to measure, image acquisition and processing, the study design, and sample size. However, such studies are likely to be time consuming and costly, particularly for disease settings with already favourable outcomes that require a large sample size and/or lengthy follow-up duration to observe sufficient events (such as death, disease recurrence or progression) for adequate statistical power. Prospective–retrospective studies can reduce or even eliminate many of the delays and costs associated with image acquisition and follow-up assessments107. For prospective–retrospective studies, data from standard-of-care images, clinical outcomes and other data, such as standard clinical variables, are acquired from patients in completed clinical trials that satisfy the appropriate inclusion and/or exclusion criteria regarding the patient population, treatment approach, image acquisition and processing specifications, and availability of the necessary images. Both the feature extraction and the test are applied prospectively. Similar to a prospective study, the radiomic test, the statistical analysis plan, sample size, level of power, and the inclusion and exclusion criteria should be fully specified in a protocol before the initiation of a prospective–retrospective study. Criteria for establishing clinical utility through prospective–retrospective studies for other omics approaches have already been published12,107. These criteria include the stipulation that two such studies must produce similar results, an approach that can also be adapted for radiomic tests. In silico clinical trials using patient-specific models to develop a simulated cohort might provide an alternative approach108, although these simulated patients might not entirely reflect the complexities of real-life patients.
Criterion 16: benefit versus risk balance from use of the radiomic test
The benefit–risk balance associated with use of a radiomic test will encompass not only the risks and benefits associated with performing the test but also those associated with the clinical decisions directed by the test results. If the intended use of a test is to choose a therapy that provides superior clinical outcomes compared with other available options, then the improvement in clinical outcome should not only be statistically significant but also large enough in magnitude to justify use of the radiomic test. Alternatively, a favourable benefit–risk balance might emerge when use of the radiomic test leads to non-inferior outcomes while being associated with reduced risks, including those inherent in the standard testing procedure, or if the toxicities of unnecessary or ineffective treatment can be avoided. For example, even if the radiomic test leads to treatment decisions that are similar to those based on standard diagnostic workups, the former might nevertheless have clinical utility if the information it provides enables patients to undergo fewer subsequent scans or biopsies while still leading to similar outcomes.
Finally, a radiomic test does not have clinical utility if it separates patients into groups for which the outcomes are statistically different but the recommended clinical management would be the same. Even if one patient group has an inferior outcome on standard therapy, another treatment might be available that is more effective for that group.
Conclusions
The 16 recommended criteria provided herein aim to guide the translation of radiomic tests into clinically useful tools and are expected to be relevant across a range of imaging modalities and scenarios. Many of these recommendations share common themes with other published guidelines for radiomics; adherence to these recommendations addresses many components of the radiomic quality score4, for example.
The statistical considerations regarding model development and validation and the design of studies for the assessment of clinical utility have numerous parallels to those for in vitro test considerations12,109. Several components of our recommendations are based on these sources. However, some important and consequential differences specific to radiomics also merit consideration. Radiomic approaches increasingly utilize multiple machine learning and deep learning methods, which introduces new issues regarding standard operating procedures for feature extraction, test lockdown, machine learning interpretability, correlations with biology, regulatory considerations and assessments of analytical validity. These criteria are likely to further evolve in the future as researchers become aware of additional issues and as more radiomic models become locked down, validated and evaluated for clinical utility. We emphasize that these recommendations pertain to the conduct and analysis of radiomic studies and are not intended as reporting guidelines for radiomic and CAD studies in the vein of REMARK for tumour prognostic studies110 or other reporting guidelines catalogued by the EQUATOR project111. However, some of these recommendations are expected to serve as the basis of such radiomic-specific reporting guidelines.
Radiomics is increasingly likely to involve full machine learning-based image analysis such as deep learning-based features or the application of artificial intelligence and machine learning algorithms directly to voxel-level data. Such a transformation, as mentioned before, is expected to eliminate much of the variability created by human error and improve model performance in many scenarios, although it will also benefit from integration with clinical information to better personalize the test result to each patient. For example, this type of test might be used not only to detect cancer but also to do so in the presence of additional comorbidities (for example, examining a renal finding in the presence of diabetes mellitus, chronic inflammatory processes and/or hypertension). The increased availability of different types of data should facilitate these types of improvements.
Acknowledgements
P.L. acknowledges support for the publication of this work from the European Union’s Horizon 2020 Research and Innovation Programme under grant agreement CHAIMELEON No. 952172, EuCanImage No. 952103, IMI-OPTIMA No. 101034347 and ERC advanced grant (ERC-ADG-2015 No. 694812 – Hypoximmuno). P.K. acknowledges support for the publication of this work from NCI grant P50 CA228944.
Peer review
Peer review information
Nature Reviews Clinical Oncology thanks K. Bera. J.-E. Bibault, J. Tian and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Competing interests
M.G. has acted as a scientific adviser of Quantitative Insights (now Qlarity Imaging), is the contact Principal Investigator for MIDRC (funded by NIBIB COVID-19 Contract 75N92020D00021), receives royalties from Hologic, GE Medical Systems, MEDIAN Technologies, Riverain Medical, Mitsubishi and Toshiba, holds stocks in R2/Hologic, is a shareholder in Qview, and is a co-founder of and equity holder in Quantitative Insights (now Qlarity Imaging). P.L. is a co-founder, minority shareholder and member of the advisory board of Oncoradiomics, and is listed as a co-inventor on several licensed patents in radiomics. E.P.H., J.P.B.O.-C., L.M.M., P.E.K., E.L.S. and L.K.S. declare no competing interests.
Glossary
Biomarker A characteristic indicating non-pathological or pathological biological processes and/or an increased likelihood of a response to an exposure or intervention5.
Clinical utility The degree to which acting upon the results of the radiomic test leads to a favourable benefit–risk balance for the patient.
Clinical validity The adequacy of the clinical performance of the radiomic test for its intended purpose.
Deep learning A class of machine learning based on neural networks.
Model A computational algorithm applied to extracted image features or voxel-level image data themselves.
Model outputs The result of a computational algorithm applied to the extracted image features or voxel-level data themselves; a quantity to be used in guiding clinical management.
Model validation Establishment of the ability of a model to predict an outcome of interest when applied to new data.
Neural network A type of computational algorithm based on the operation of biological neural systems in animals that feeds the input (in this context, feature measurements or voxel-level data) through a series of nodes that perform mathematical operations on the outputs of preceding nodes to produce an output. In a convolutional neural network, these mathematical operations involve applying convolutional kernels to the outputs of preceding nodes.
Normalization A process for adjusting the voxel intensity values of an image for differences resulting from variability in image acquisition and processing parameters.
Omics The study of related sets of biological molecules in a comprehensive fashion with examples including genomics, transcriptomics, proteomics, metabolomics and epigenomics109. Radiomics naturally extends this definition to include quantification of radiological imaging features for the purposes of characterization and measurement of structure, function and interaction between biological molecules in a comprehensive and high-throughput manner.
Overfitting The process of fitting an overly complex model to noise in the data, thus producing a model that is only poorly predictive when applied to completely new data.
Performance metric A quantity indicating the ability of a model to predict an outcome of interest.
Phantoms An object that is imaged to measure the technical performance of an imaging device.
Radiomic features Quantities computed from voxel-level image data.
Radiomic test A system comprising materials, methods and procedures for image acquisition, processing and feature extraction, and methods or criteria for interpretation of the image data for use in guiding clinical management.
Technical artefacts The effects of factors, such as imaging centre, device, operator or device-calibration settings, on the distribution of the feature measurements.
Technical validity The quality of the feature measurements in terms of their accuracy in assaying an underlying characteristic of interest or their variability when the feature extraction process is applied repeatedly to the same patient.
Test lockdown Full specification of all image acquisition, processing and feature extraction procedures, all aspects of the underlying model, and interpretations of the output.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Gillies RJ Kinahan PE Hricak H Radiomics: images are more than pictures, they are data Radiology 2016 278 563 577 10.1148/radiol.2015151169 26579733
2. Giger ML Update on the potential of computer-aided diagnosis for breast cancer Fut. Oncol. 2010 6 1 4 10.2217/fon.09.154
3. Doi K Computer-aided diagnosis in medical imaging: historical review, current status, and future potential Comput. Med. Imaging Graph. 2007 31 198 211 10.1016/j.compmedimag.2007.02.002 17349778
4. Lambin P Radiomics: extracting more information from medical images using advanced feature analysis Eur. J. Cancer 2012 48 441 446 10.1016/j.ejca.2011.11.036 22257792
5. FDA-NIH Biomarker Working Group. BEST (Biomarkers, EndpointS, and other Tools) Resource (Food and Drug Administration and National Institutes of Health, 2016).
6. FDA. Artificial Intelligence and Machine Learning (AI/ML)-Enabled Devices https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-aiml-enabled-medical-devices. (2022).
7. Fornacon-Wood IM Reliability and prognostic value of radiomic features are highly dependent on choice of feature extraction platform Eur. Radiol. 2020 30 6241 6250 10.1007/s00330-020-06957-9 32483644
8. Radiomics. Radiomics Quality Score – RQS 2.0 https://www.radiomics.world/rqs2 (2022).
9. Zwanenburg A The image biomarker standardization initiative: standardized quantitative radiomics for high throughput image-based phenotyping Radiology 2020 295 328 338 10.1148/radiol.2020191145 32154773
10. Kumar V Radiomics: the process and the challenges Magn. Reson. Imaging 2012 30 1234 1248 10.1016/j.mri.2012.06.010 22898692
11. Fournier L Incorporating radiomics into clinical trials: expert consensus endorsed by the European society of radiology on considerations for data-driven compared to biologically driven quantitative biomarkers Eur. Radiol. 2021 31 6001 6012 10.1007/s00330-020-07598-8 33492473
12. McShane LM Criteria for the use of omics-based predictors in clinical trials: explanation and elaboration BMC Med. 2013 11 220 10.1186/1741-7015-11-220 24228635
13. Jiang Y Edwards AV Newstead GM Artificial intelligence applied to breast MRI for improved diagnosis Radiology 2021 298 39 46 10.1148/radiol.2020200292
14. Data Science Institute, American College of Radiology. FDA Cleared AI Algorithms https://www.acrdsi.org/DSI-Services/FDA-Cleared-AI-Algorithms, (2022).
15. Clark GM Prognostic factors versus predictive factors: examples from a clinical trial of erlotinib Mol. Oncol. 2008 1 406 412 10.1016/j.molonc.2007.12.001 19383314
16. Aerts HJWL Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach Nat. Commun. 2014 5 4006 10.1038/ncomms5006 24892406
17. Li H Quantitative MRI radiomics in the prediction of molecular classifications of breast cancer subtypes in the TCGA/TCIA data set NPJ Breast Cancer 2016 2 16012 10.1038/npjbcancer.2016.12 27853751
18. Li H MRI radiomics signatures for predicting the risk of breast cancer recurrence as given by research versions of gene assays of MammaPrint, Oncotype DX, and PAM50 Radiology 2016 281 382 391 10.1148/radiol.2016152110 27144536
19. Cha KH Bladder cancer treatment response assessment in CT using radiomics with deep learning Nat. Sci. Rep. 2017 7 8738
20. Drukker K Most-enhancing tumor volume by mri radiomics predicts recurrence-free survival “Early On” in neoadjuvant treatment of breast cancer Cancer Imaging 2018 18 12 10.1186/s40644-018-0145-9 29653585
21. Huang EP Lin FI Shankar LK Beyond correlations, sensitivities, and specificities: a roadmap for demonstrating utility of advanced imaging in oncology treatment and clinical trial design Acad. Radiol. 2017 24 1036 1049 10.1016/j.acra.2017.03.002 28456570
22. Subramanian J Simon R What should physicians look for in evaluating prognostic gene-expression signatures? Nat. Rev. Clin. Oncol. 2010 7 327 334 10.1038/nrclinonc.2010.60 20421890
23. Shafiq-Ul-Hassan M Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels Med. Phys. 2017 44 1050 1062 10.1002/mp.12123 28112418
24. Berenguer R Radiomics of CT features may be nonreproducible and redundant: influence of CT acquisition parameters Radiology 2018 288 407 415 10.1148/radiol.2018172361 29688159
25. American College of Radiology. ACR Appropriateness Criteria https://www.acr.org/Clinical-Resources/ACR-Appropriateness-Criteria (2022).
26. Society of Nuclear Medicine and Medical Imaging. Procedure Standards https://www.snmmi.org/ClinicalPractice/content.aspx?ItemNumber=6414. (2022).
27. European Association of Nuclear Medicine. Guidelines https://www.eanm.org/publications/guidelines/ (2022).
28. QIBQ Wiki. Profiles http://qibawiki.rsna.org/index.php/Profiles (2022).
29. Fass L Imaging and cancer: a review Mol. Oncol. 2008 2 115 152 10.1016/j.molonc.2008.04.001 19383333
30. Zhao B Exploring intra- and inter-reader variability in unidimensional, bidimensional, and volumetric measurements of solid tumors on CT scans reconstructed at different slice intervals Eur. J. Radiol. 2013 82 959 968 10.1016/j.ejrad.2013.02.018 23489982
31. O’Connor JPB Jackson A Parker GJM Roberts C Jayson GC Dynamic contrast-enhanced MRI in clinical trials of anti-vascular therapies Nat. Rev. Clin. Oncol. 2012 9 167 177 10.1038/nrclinonc.2012.2 22330689
32. Tudorica LA QIN: a feasible high spatiotemporal resolution breast DCE-MRI protocol for clinical settings Magn. Reson. Imaging 2012 30 1257 1267 10.1016/j.mri.2012.04.009 22770687
33. Nardone V Delta radiomics: a systematic review Radiol. Med. 2021 126 1571 1583 10.1007/s11547-021-01436-7 34865190
34. Pinker K Riedl C Weber WA Evaluating tumor response with FDG-PET: updates on PERCIST, comparison with EORTC criteria and clues to future development Eur. J. Nucl. Med. Mol. Imaging 2017 44 55 66 10.1007/s00259-017-3687-3 28361188
35. Mackin D Harmonizing the pixel size in retrospective computed tomography radiomics studies PLoS ONE 2017 12 e0178524 10.1371/journal.pone.0178524 28934225
36. Madabhushi A Udupa JK Souza A Generalized scale: theory, algorithms, and application to image inhomogeneity correction Comput. Image Vis. Underst. 2006 101 100 121 10.1016/j.cviu.2005.07.010
37. Madabhushi A Udupa JK New methods of MR image intensity standardization via generalized scale Med. Phys. 2006 33 3426 3434 10.1118/1.2335487 17022239
38. Whitney HM Harmonization of radiomic features of breast lesions across international DCE-MRI datasets J. Med. Imaging 2020 7 012707 10.1117/1.JMI.7.1.012707
39. Duron L Gray-level discretization impacts reproducible MRI radiomics texture features PLoS ONE 2019 14 e0213459 10.1371/journal.pone.0213459 30845221
40. Larue RTHM Influence of gray level discretization on radiomic feature stability for different CT scanners, tube currents, and slice thicknesses: a comprehensive phantom study Acta Oncol. 2017 56 1544 1553 10.1080/0284186X.2017.1351624 28885084
41. Leijenaar RT The effect of SUV discretization in quantitative FDG-PET radiomics: the need for standardized methodology in tumor texture analysis Nat. Sci. Rep. 2015 5 11075
42. Willemink MJ Preparing medical imaging data for machine learning Radiology 2020 295 4 15 10.1148/radiol.2020192224 32068507
43. Mali SA Making radiomics more reproducible across scanner and imaging protocol variations: a review of harmonization methods J. Per. Med. 2021 11 842 10.3390/jpm11090842
44. Lin Y Deep learning for fully automated tumor segmentation and extraction of magnetic resonance radiomics features in cervical cancer Eur. Radiol. 2020 30 1297 1305 10.1007/s00330-019-06467-3 31712961
45. Parmar C Grossman P Bussink J Lambin P Aerts HJWL Machine learning methods for quantitative radiomic biomarkers Nat. Sci. Rep. 2015 5 13087
46. Primakov SP Automated detection and segmentation of non-small cell lung cancer computed tomography images Nat. Commun. 2022 13 3423 10.1038/s41467-022-30841-3 35701415
47. Gilhuijs KGA Giger ML Bick U Automated analysis of breast lesions in three dimensions using dynamic magnetic resonance imaging Med. Phys. 1998 25 1647 1654 10.1118/1.598345 9775369
48. Chen W Giger ML Lan L Bick U Computerized interpretation of breast MRI: investigation of enhancement-variance dynamics Med. Phys. 2004 31 1076 1082 10.1118/1.1695652 15191295
49. Chen W Giger ML Bick U Newstead G Automatic identification and classification of characteristic kinetic curves of breast lesions on DCE-MRI Med. Phys. 2006 33 2878 2887 10.1118/1.2210568 16964864
50. Chen W Giger ML Li H Bick U Newstead G Volumetric texture analysis of breast lesions on contrast-enhanced magnetic resonance images Magn. Reson. Med. 2007 58 562 571 10.1002/mrm.21347 17763361
51. van Timmeren JE Test-retest data for radiomics feature stability analysis: generalizable or study-specific? Tomography 2016 2 361 365 10.18383/j.tom.2016.00208 30042967
52. Afshar P Mohammadi A Plataniotis KN Oikonomou A Benali H From hand-crafted to deep learning-based cancer radiomics: challenges and opportunities IEEE Signal. Process. Mag. 2019 36 132 160 10.1109/MSP.2019.2900993
53. Sahiner B Deep learning in medical imaging and radiation therapy Med. Phys. 2019 46 e1 e36 10.1002/mp.13264 30367497
54. Li Z Wang Y Yu J Guo Y Cao W Deep learning based radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma Nat. Sci. Rep. 2017 7 1 11
55. Antropova N Huynh BQ Giger ML A deep feature fusion methodology for breast cancer diagnosis demonstrated on three imaging modality datasets Med. Phys. 2017 44 5162 5171 10.1002/mp.12453 28681390
56. International Organization for Standardization. Guidance for the Use of Repeatability, Reproducibility, and Trueness Estimates in Measurement Uncertainty Evaluation https://www.iso.org/obp/ui/#iso:std:iso:21748:ed-2:v1:en (2020).
57. Drukker K Pesce L Giger ML Repeatability in computer-aided diagnosis: application to breast cancer diagnosis on sonography Med. Phys. 2010 37 2659 2669 10.1118/1.3427409 20632577
58. Kessler LG The emerging science of quantitative imaging biomarkers terminology and definitions for scientific studies and regulatory submissions Stat. Methods Med. Res. 2015 24 9 26 10.1177/0962280214537333 24919826
59. Raunig DL Quantitative imaging biomarkers: a review of statistical methods for technical performance assessment Stat. Methods Med. Res. 2015 24 27 67 10.1177/0962280214537344 24919831
60. Huang EP Multiparametric quantitative imaging in risk prediction: recommendations for data acquisition, technical performance assessment, and model development and validation Acad. Radiol. 2022 10.1016/j.acra.2022.09.018 36411153
61. McHugh DJ Image contrast, image preprocessing, and T1-mapping affect MRI radiomic feature repeatability in patients with colorectal cancer liver metastases Cancers 2021 13 240 10.3390/cancers13020240 33440685
62. Jha AK Repeatability and reproducibility study of radiomic features on a phantom and human cohort Sci. Rep. 2021 11 2055 10.1038/s41598-021-81526-8 33479392
63. Bissoto, A., Perez, F., Valle, E. & Avila, S. Skin lesion synthesis with generative adversarial networks. OR 2.0 Context-Aware Operating Theaters, Computer Assisted Robotic Endoscopy, Clinical Image-Based Procedures, and Skin Image Analysis. OR 2.0 First International Workshop, CARE Fifth International Workshop, CLIP Seventh International Workshop, ISIC Third International Workshop. Springer Lecture Notes in Computer Science (Springer, 2019).
64. Sullivan DC Metrology standards for quantitative imaging biomarkers Radiology 2015 277 813 825 10.1148/radiol.2015142202 26267831
65. Hackstadt AJ Hess AM Filtering for increased power for microarray data analysis. BMC Bioinformatics 2009 10 11 10.1186/1471-2105-10-11 19133141
66. Luo J A comparison of batch effect removal methods for enhancement of prediction performance using MACQ-II microarray gene expression data Pharmacogenomics J. 2010 10 278 291 10.1038/tpj.2010.57 20676067
67. Johnson WE Li C Rabinovic A Adjusting batch effects in microarray expression data using empirical bayes methods Biostatistics 2007 8 118 127 10.1093/biostatistics/kxj037 16632515
68. Orlhac F A post-reconstruction harmonization method for multicenter radiomic studies in PET J. Nucl. Med. 2018 59 1321 1328 10.2967/jnumed.117.199935 29301932
69. Parker HS Leek JT The practical effect of batch on genomic prediction Stat. Appl. Genet. Mol. Biol. 2012 11 10 10.1515/1544-6115.1766 22611599
70. Robinson K Li H Lan L Schacht D Giger M Radiomics robustness assessment and classification evaluation: a two-stage method demonstrated on multivendor FFDM Med. Phys. 2019 46 2145 2156 10.1002/mp.13455 30802972
71. The Cancer Imaging Archive http://cancerimagingarchive.net (2020).
72. Clark K The Cancer Imaging Archive (TCIA): maintaining and operating a public information repository J. Digital Imaging 2013 26 1045 1057 10.1007/s10278-013-9622-7
73. Zhu Y Deciphering genomic underpinnings of quantitative MRI-based radiomic phenotypes of invasive breast carcinoma Nat. Sci. Rep. 2015 5 17787
74. Riley RD Minimum sample size for developing a multivariable prediction model: part II — binary and time-to-event outcomes Stat. Med. 2018 38 1276 1296 10.1002/sim.7992 30357870
75. Riley RD Minimum sample size for external validation of a clinical prediction model with a binary outcome Stat. Med. 2021 40 4230 4251 10.1002/sim.9025 34031906
76. Riley RD Minimum sample size calculations for external validation of a clinical prediction model with a time-to-event outcome Stat. Med. 2022 41 1280 1295 10.1002/sim.9275 34915593
77. Cho, J., Lee, K., Shin, E., Choy, G. & Do, S. How much data is needed to train a medical image deep learning system to achieve necessary high accuracy? Preprint at 10.48550/arXiv.1511.06348 (2015).
78. Whitney H Li H Ji Y Liu P Giger ML Comparison of breast MRI tumor classification using human-engineered radiomics, transfer learning from deep convolutional neural networks, and fusion methods Proc. IEEE 2020 108 163 177 10.1109/JPROC.2019.2950187
79. Hastie, T., Tibshirani, R. & Friedman, J. The Elements of Statistical Learning: Data Mining, Inference and Prediction 2nd edn (Springer, 2009).
80. Deist TM Machine learning algorithms for outcome prediction in (chemo)radiotherapy: an empirical comparison of classifiers Med. Phys. 2018 45 3449 3459 10.1002/mp.12967 29763967
81. Haykin S. Neural Networks: A Comprehensive Foundation (Prentice Hall, 1994).
82. Ben-Dor A Tissue classification with gene expression profiles J. Comput. Biol. 2000 7 559 583 10.1089/106652700750050943 11108479
83. Dudoit S Fridlyand J Speed TP Comparison of discrimination methods for the classification of tumors using gene expression data J. Am. Stat. Assoc. 2002 97 77 87 10.1198/016214502753479248
84. Heinze G Wallisch C Dunkler D Variable selection — a review and recommendations for the practicing statistician Biom. J. 2018 60 431 449 10.1002/bimj.201700067 29292533
85. Tibshirani R Regression shrinkage and selection via the LASSO J. R. Stat. Soc. Ser. B 1996 58 267 288
86. Hanley JA McNeil BJ The meaning and use of the area under a receiver operating characteristic (ROC) curve Radiology 1982 143 29 36 10.1148/radiology.143.1.7063747 7063747
87. Harrell FE Jr. Califf RM Pryor DB Lee KL Rosati RA Evaluating the yield of medical tests J. Am. Med. Assoc. 1982 247 2543 2546 10.1001/jama.1982.03320430047030
88. Hosmer DW Lemeshow S Goodness of fit tests for the multiple logistic regression model Commun. Stat. Theory Methods 1980 9 1043 1069 10.1080/03610928008827941
89. Lemeshow S Hosmer D A review of goodness of fit statistics for use in the development of logistic regression model Am. J. Epidemiol. 1982 115 92 106 10.1093/oxfordjournals.aje.a113284 7055134
90. van Calster, B. & Steyerberg, E. W. Wiley StatsRef: Statistics Reference Online (John Wiley and Sons, Ltd., 2018).
91. Bröcker J Smith LA Increasing the reliability of reliability diagrams Weather Forecast. 2007 22 651 661 10.1175/WAF993.1
92. McLachlan, G. J. Discriminant Analysis and Statistical Pattern Recognition (John Wiley and Sons, 2002).
93. Stone M Cross-validatory choice and assessment of statistical predictions J. R. Stat. Soc. Ser. B 1974 36 111 147
94. Breiman L Bagging predictors Mach. Learn. 1996 24 123 140 10.1007/BF00058655
95. Molinaro AM Simon R Pfeffer RM Prediction error estimation: a comparison of resampling methods Bioinformatics 2005 21 3301 3307 10.1093/bioinformatics/bti499 15905277
96. Dobbin KK Simon RM Optimally splitting cases for training and testing high-dimensional classifiers BMC Med. Genomics 2011 4 31 10.1186/1755-8794-4-31 21477282
97. Sachs MC McShane LM Issues in developing multivariable molecular signatures for guiding clinical care decisions J. Biopharm. Stat. 2016 26 1098 1110 10.1080/10543406.2016.1226329 27561083
98. Varma S Simon R Bias in error estimation when using cross-validation for model selection BMC Bioinformatics 2006 7 91 10.1186/1471-2105-7-91 16504092
99. Salahuddin Z Woodruff HC Chatterjee A Lambin P Transparency of deep neural networks for medical image analysis: a review of interpretability methods Comput. Biol. Med. 2022 140 105111 10.1016/j.compbiomed.2021.105111
100. Hilsenbeck SG Clark GM McGuire WL Why do so many prognostic factors fail to pan out? Breast Cancer Res. Treat. 1992 22 197 206 10.1007/BF01840833 1391986
101. Vickers AJ Elkin EB Decision curve analysis: a novel method for evaluating prediction models Med. Decis. Mak. 2006 26 565 574 10.1177/0272989X06295361
102. Wu G Preoperative CT-based radiomics combined with intraoperative frozen section is predictive of invasive adenocarcinoma in pulmonary nodules: a multicenter study Eur. Radiol. 2020 30 2680 2691 10.1007/s00330-019-06597-8 32006165
103. Hayes DF Defining clinical utility of tumor biomarker tests: a clinician’s viewpoint J. Clin. Oncol. 2021 39 238 249 10.1200/JCO.20.01572 33326253
104. Saha A Hosseinzadeh M Huisman H End-to-end prostate cancer detection in bpmri via 3d cnns: effects of attention mechanisms, clinical priori and decoupled false positive reduction Med. Image Anal. 2021 73 102155 10.1016/j.media.2021.102155 34245943
105. Hosseinzadeh M Deep learning-assisted prostate cancer detection on bi-parametric MRI: minimum training data size requirements and effect of prior knowledge Eur. Radiol. 2022 32 2224 2234 10.1007/s00330-021-08320-y 34786615
106. Baughan, N. et al. Sequestration of Imaging Studies in MIDRC: A Multi-institutional Data Commons. Medical Imaging 2002; Image Perception, Observer Performance, and Technology Assessment, vol. 12035 (SPIE, 2022).
107. Simon RM Paik S Hayes DF Use of archived specimens in evaluation of prognostic and predictive biomarkers J. Natl Cancer Inst. 2009 101 1446 1452 10.1093/jnci/djp335 19815849
108. Pappalardo F Gusso G Tshinanu FM Viceconti M In silico clinical trials: concepts and early adoptions Brief. Bioinforma. 2019 20 1699 1708 10.1093/bib/bby043
109. Committee on the Review of Omics-Based Tests for Predicting Patient Outcomes in Clinical Trials, Board on Health Care Services, Board on Health Sciences Policy, Institute of Medicine. Evolution of Translational Omics: Lessons Learned and the Path Forward (The National Academies Press, 2012).
110. Altman DG McShane LM Sauerbrei W Taube SE Reporting recommendations for tumor marker prognostic studies (REMARK): explanation and elaboration PLoS Med. 2012 9 e1001216 10.1371/journal.pmed.1001216 22675273
111. Equator Network. Enhancing the Quality and Transparency of Health Research (EQUATOR) https://www.equator-network.org/ (2022).
| 36443594 | PMC9707172 | NO-CC CODE | 2022-12-01 23:20:21 | no | Nat Rev Clin Oncol. 2022 Nov 28;:1-14 | utf-8 | Nat Rev Clin Oncol | 2,022 | 10.1038/s41571-022-00707-0 | oa_other |
==== Front
Adv Fiber Mater
Adv Fiber Mater
Advanced Fiber Materials
2524-7921
2524-793X
Springer Nature Singapore Singapore
234
10.1007/s42765-022-00234-8
Research Article
Rapid Ferroelectric-Photoexcited Bacteria-Killing of Bi4Ti3O12/Ti3C2Tx Nanofiber Membranes
Wang Zhiying 1Zhiying Wang
obtained her B.E. from Hebei University of Technology in 2020, majoring in Metal Materials Engineering. She is studying for a master’s degree at Tianjin University, majoring in materials science under the guidance of Prof. Shuilin Wu. Her current research interests focus on antibacterial biomaterials.
Li Jianfang 1
Qiao Yuqian 2
Liu Xiangmei 3
Zheng Yufeng 25
Li Zhaoyang 1
Shen Jie 4
Zhang Yu 5
Zhu Shengli 1
Jiang Hui 1
Liang Yanqin 1
Cui Zhenduo 1
Chu Paul K. 6
Wu Shuilin [email protected]
[email protected]
12
1 grid.33763.32 0000 0004 1761 2484 The Key Laboratory of Advanced Ceramics and Machining Technology by the Ministry of Education of China, School of Materials Science and Engineering, Tianjin University, Tianjin, 300072 China
2 grid.11135.37 0000 0001 2256 9319 School of Materials Science and Engineering, Peking University, Beijing, 100871 China
3 grid.412030.4 0000 0000 9226 1013 School of Health Science and Biomedical Engineering, Hebei University of Technology, Tianjin, 300401 China
4 grid.440601.7 0000 0004 1798 0578 Shenzhen Key Laboratory of Spine Surgery, Department of Spine Surgery, Peking University Shenzhen Hospital, Shenzhen, 518036 China
5 grid.410643.4 Department of Orthopedics, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Guangzhou, 510080 China
6 grid.35030.35 0000 0004 1792 6846 Department of Physics, Department of Materials Science and Engineering, and Department of Biomedical Engineering, City University of Hong Kong, Hong Kong, 999077 China
28 11 2022
113
1 9 2022
7 11 2022
© Donghua University, Shanghai, China 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
In this study, an antibacterial nanofiber membrane [polyvinylidene fluoride/Bi4Ti3O12/Ti3C2Tx (PVDF/BTO/Ti3C2Tx)] is fabricated using an electrostatic spinning process, in which the self-assembled BTO/Ti3C2Tx heterojunction is incorporated into the PVDF matrix. Benefiting from the internal electric field induced by the spontaneously ferroelectric polarization of BTO, the photoexcited electrons and holes are driven to move in the opposite direction inside BTO, and the electrons are transferred to Ti3C2Tx across the Schottky interface. Thus, directed charge separation and transfer are realized through the cooperation of the two components. The recombination of electron–hole pairs is maximumly inhibited, which notably improves the yield of reactive oxygen species by enhancing photocatalytic activity. Furthermore, the nanofiber membrane with an optimal doping ratio exhibits outstanding visible light absorption and photothermal conversion performance. Ultimately, photothermal effect and ferroelectric polarization enhanced photocatalysis endow the nanofiber membrane with the ability to kill 99.61% ± 0.28% Staphylococcus aureus and 99.71% ± 0.16% Escherichia coli under 20 min of light irradiation. This study brings new insights into the design of intelligent antibacterial textiles through a ferroelectric polarization strategy.
Graphical Abstract
Supplementary Information
The online version contains supplementary material available at 10.1007/s42765-022-00234-8.
Keywords
Ferroelectric polarization
Antibacterial nanofiber
Ti3C2Tx
Photocatalysis
Schottky heterojunction
http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 51871162 52173251 82002303 http://dx.doi.org/10.13039/501100005153 China National Funds for Distinguished Young Scientists 51925104 Central Guidance on Local Science and Technology Development Fund of Hebei Province226Z1303G Scientific Research Foundation of Peking University Shenzhen HospitalKYQD2021064 Guangdong Basic and Applied Basic Research Foundation2021A1515220093 2022A1515011536 NSFC-Guangdong Province Joint ProgramU21A2084
==== Body
pmcIntroduction
Bacterial infections are the major cause of morbidity and mortality in intensive care units (ICU) worldwide [1]. They have become a threat to global public health, with multiple secondary bacterial infections [2, 3]. Since the outbreak of COVID-19 in December 2019, the course of COVID-19 patients has been complicated by bacterial infections [4, 5]. The cross-infection of bacteria also aggravates the fatality rates of COVID-19. Medical textiles, including coveralls, aprons, isolation gowns, surgical masks, and gloves [6], play an important role in protecting medical staff from external bacteria and effectively preventing cross-infection between medical workers and patients. Therefore, the demand for antibacterial textiles has dramatically increased.
The antimicrobial agents used in textiles work by either inhibiting cell growth or killing pathogens [7]. However, the long-term and systemic safety of common commercial antimicrobial agents, such as metal nanoparticles, remains a major challenge [8–10]. Therefore, various innovative antibacterial strategies are developed to rapidly eradicate bacteria, including materioherbology, photocatalytic disinfection, and photothermal therapy [11–15]. Antibacterial materials that respond to external stimuli, especially light, have been widely studied for their rapid and reusable antibacterial effect, broad antibacterial spectrum, and non-antibiotic resistance. Photo-responsive nanomaterials are excited by light with corresponding wavelengths to provide photothermal and/or photocatalytic performance with a locally increased temperature and/or reactive oxygen species (ROS) to kill bacteria [15]. Thus, the loading of photoexcited antibacterial materials on textiles can provide a new idea for designing novel antibacterial textiles.
Bismuth-based photocatalysts, such as BiOX (X = Cl, Br, I) [16] and Bi2MO6 (M = W, Mo) [17] have been widely utilized for catalytic applications owing to their unique layer structure and high stability. Unlike them, Bi4Ti3O12 (BTO) is a ferroelectric semiconductor with an Aurivillius phase [18]. It has a strong ferroelectric property due to its layered perovskite structure and high Curie temperature of 675 °C [19]. BTO is composed of fluorite-like [Bi2O2]2+ slices and perovskite-type TiO6 octahedral layers stacked along the c-axis, which form an inner electric field and promote the separation of photoexcited electrons and holes [20, 21]. The polarization effect in ferroelectric materials has been proven to be responsible for enhanced photocatalytic activity [22]. However, this activity is limited by the large bandgap and poor light absorption [18–20, 23]. Constructing heterojunctions and interface engineering can be considered an effective strategy to improve photocatalytic performance.
Ti3C2Tx MXene, a typical representative material among the emerging two-dimensional layered transition metal carbides and/or nitrides family, has multiple advantages, including excellent metallic conductivity, remarkable light absorption ability, high photothermal conversion efficiency and superior biocompatibility [24, 25]. These advantages have made Ti3C2Tx a promising candidate for realizing functional composite materials for wearing. In addition, Ti3C2Tx has abundant surface functional groups containing F, O, and OH terminations, which can interact closely with textile substrates [26]. More importantly, it can be combined with the semiconductor of BTO to create a Schottky heterojunction. Currently, there is no literature review on BTO/Ti3C2Tx, and the potential mechanism of ferro-photocatalytic sterilization is also unclear.
As a basic method for fabricating nanofibers, electrospinning is a facile and versatile technique. Electrospinning membranes have a high specific surface area, adequate mechanical strength, a complex porous structure, simple scalable fabrication, and low cost [27, 28]. Currently, more than 100 different kinds of organic polymers, such as silk fibroin, chitosan, poly(Ɛ-caprolactone), poly(lactic acid), polystyrene, poly(vinyl chloride), polyaniline, polypyrrole, and polyvinylidene fluoride (PVDF), have been successfully electrospun into nanofibers for various applications [29]. Among them, PVDF is a promising polymer matrix due to its high flexibility, ferroelectric properties, and high thermal stability [30]. Note that PVDF has a broad potential in wound dressings due to its applicability [31], anti-adhesion [32], coagulation [33], and non-neutralization with ROS [34].
Therefore, we propose the hypothesis of whether ferroelectric-photoresponsive nanofibers can be electrospun into textiles with ferroelectric polarization-strengthened photoexcited bactericidal ability to prevent pathogenic infections. To verify this, the PVDF/BTO/Ti3C2Tx membrane was prepared through self-assembly and electrostatic spinning. The built-in electric field induced by the spontaneous polarization of BTO promoted the separation of charge carriers. BTO and Ti3C2Tx formed a Schottky junction with a tight contact interface, in which BTO could be excited to generate electrons and holes under illumination, and electrons rapidly transferred to the side of Ti3C2Tx. During light exposure, Ti3C2Tx enhanced the light absorption range and photothermal conversion ability of BTO and accelerated the oriented electron transfer. The trapped electrons and holes could be used to produce more ROS (Scheme 1). Consequently, the PVDF/BTO/Ti3C2Tx membrane could effectively kill pathogenic bacteria in vitro and in vivo through the synergistic effect of ROS and heat under 20 min of light irradiation.Scheme 1 Schematic diagram of ferroelectric polarization strategy for treating wound infection by light responsive nanofiber membrane of PVDF/BTO/Ti3C2Tx
Experimental Section
Preparation of BTO/Ti3C2Tx
BTO nanostructures were self-assembled on Ti3C2Tx nanosheets according to the following method. 80 mg of BTO was dispersed into 25 mL THF. The Ti3C2Tx colloid solution with different mass ratios (20, 40, and 60 wt% of BTO) were immediately transferred to the above solution. The mixed solution was sonicated at 100 W for 4 h in an ice bath. After sonication, the solid mixtures were washed with THF for three times. The obtained materials were vacuum-dried, ground, and stored before use. These materials were named by BT20, BT40, and BT60, respectively.
Fabrication of PVDF and PVDF/BT40 Nanofiber Membranes
All nanofiber membranes were fabricated by the electrospinning method. PVDF powder was dissolved in DMF and stirred for 1.5 h to obtain PVDF solution (18 wt%). Then different amount of BT40 (10, 30, and 50 wt% of PVDF) was added to the PVDF solution under stirring for 2.5 h. Each homogeneous mixture was poured into a syringe capped with a 21-gauge stainless steel needle, which connected with a 17 kV voltage difference. The distance between the needle tip and the collector was 15 cm. The feed rate was maintained at 1 mL h−1 and the spinning time was set to 3 h. All collected nanofiber membranes were dried in the vacuum oven at 25 °C overnight. These nanofiber membranes were named PVDF, PVDF/BT40-10, PVDF/BT40-30, and PVDF/BT40-50, respectively.
Characterization
The morphologies and microstructures of synthesized materials were observed by field emission scanning electron microscopy (FESEM, S4800, Japan) and transmission electron microscopy (TEM, JEOL JEM-2100F, Japan). The element distributions were executed by energy dispersive spectrometer (EDS, X-MAX20, UK). The phase structure was investigated by X-ray diffraction (XRD, D8 Advance, Germany). The UV–Vis–NIR absorption spectra were measured by a UV–Vis spectrophotometer (UV-2700, Shimadzu). The elemental analysis of the samples was examined by X-ray photoelectron spectroscopy (XPS, ThermoFisher Scientific 250Xi, USA). Ultraviolet photoemission spectroscopy (UPS) was conducted on an Escalab 250Xi spectrometer using He I resonance lines (21.22 eV). Photoluminescence (PL) spectra were measured by a fluorescence spectrophotometer (Fluorolog-3, USA). Raman spectroscopy was detected by a Raman microscope (DXR2, Thermo Scientific, USA). The surface functional groups were tested by Fourier transform infrared (FTIR, Thermo Scientific Nicolet iS10, USA). The simulated sunlight came from a Xenon lamp (PLS-SXE300).
Results and Discussion
Synthesis and Characterization of Ti3C2Tx, BTO, and BTO/Ti3C2Tx
The synthesis process of the self-assembled BTO/Ti3C2Tx heterostructure was shown in Fig. S1. The prepared Ti3C2Tx and BTO were dispersed into tetrahydrofuran. Compared to Ti3C2Tx, BTO had a small surface energy due to the presence of an oleic acid layer, resulting in good dispersibility. During the preparation, Ti3C2Tx nanosheets were wrapped around the surface of the BTO nanostructure via van der Waals interactions, further minimizing the surface energy needed to form a heterostructure [35].
Ti3C2Tx nanosheets were obtained from the etching and exfoliating of bulk Ti3AlC2 (Fig. S2a, b). The XRD pattern of Ti3C2Tx (Fig. S2c) showed a prominent diffraction peak at 2θ = 5.76° corresponding to the (002) plane [36]. The wrinkled surface was characterized using TEM (Fig. S2d) to reveal the ultrathin structure of Ti3C2Tx nanosheets. The high-resolution TEM (HRTEM) image (Fig. S2e) showed a corresponding lattice spacing of 0.260 nm, which was assigned to the (0 1¯1 0) lattice plane of Ti3C2Tx [37] with a hexagonal structure in the selected area electron diffraction. Elemental mapping (Fig. S2f) showed that the Ti, C, O, and F elements were evenly distributed on the surface of Ti3C2Tx.
The SEM image of BTO showed a microspherical structure composed of myriad overlapping nanosheets, with an average particle size of about 1 μm (Fig. S3a). The microstructure of the nanosheet was determined using TEM (Fig. S3b). Interplanar distances of 0.270 and 0.272 nm were indexed to (020) and (200) crystal faces, respectively (Fig. S3c) [38].
BTO/Ti3C2Tx composites were labeled BT20, BT40, and BT60, and the mass of Ti3C2Tx was 20, 40, and 60 wt% of BTO, respectively. SEM examination showed the morphologies of the three composites (Fig. S4a–c). The TEM and elemental mapping (Fig. 1a) showed the direct contact between Ti3C2Tx and BTO. The HRTEM image (Fig. 1b) exhibited a compact interface between Ti3C2Tx and BTO, as shown by the orange dashed line. The lattice fringes of 0.260, 0.270, and 0.272 nm could be attributed to the (0 1¯1 0), (020), and (200) faces of Ti3C2Tx and BTO, respectively. The diffraction peaks of the (006), (008), (111), (115), (117), (020), (208), (220), (0214), and (137) planes of orthorhombic BTO are shown in Fig. 1c [20]. The XRD patterns of the BTO/Ti3C2Tx composites showed the (002) peak of Ti3C2Tx and all the peaks of BTO. The Raman spectra (Fig. 1d) showed that the peaks at 169 cm−1 (A1g), 412 cm−1 (Eg), and 639 cm−1 (Eg) matched well with the Raman features of Ti3C2Tx [39, 40], and the peaks at 250–300 cm−1 represented the vibration band of the TiO6 octahedron [41]. These data confirmed the combination of Ti3C2Tx and BTO. The structures of BTO, Ti3C2Tx, and BTO/Ti3C2Tx are shown in Fig. 1e. The Aurivillius-type BTO is an asymmetric material, which can be described as (Bi2O2)2+(Bi2Ti3O10)2−. Ti3C2Tx provided surface termination groups, resulting in abundant active sites on the surface and promoting the combination with BTO.Fig. 1 Characterization of BTO/Ti3C2Tx. a TEM and element mapping of BT40. b HRTEM image of BT40. c XRD patterns of materials. d Raman spectra of materials. e Crystal structures of Ti3C2Tx, BTO, and BTO/Ti3C2Tx. f The survey spectra of BTO, Ti3C2Tx, and BTO/Ti3C2Tx. g The high-resolution of Ti 2p. h The high-resolution of Bi 4f
The chemical states of the elements of Ti3C2Tx, BTO, and BTO/Ti3C2Tx were obtained using XPS (Fig. 1f–h and Fig. S5). All spectra were calibrated by the C 1 s peak at 284.8 eV. The elements of Bi, Ti, C, and O co-existed in BTO/Ti3C2Tx (Fig. 1f), suggesting a successful combination of BTO and Ti3C2Tx. As shown in Fig. 1g, the Ti 2p spectra in Ti3C2Tx could be deconvoluted into eight peaks of Ti–C, Ti2+, Ti3+, and Ti–O [42]. BTO/Ti3C2Tx only showed peaks of Ti–O and Ti–C, and the binding energies shifted to lower positions compared to those of Ti3C2Tx, consistent with the binding energy change in the Ti–C peak in the C 1 s spectra (Fig. S5a). The Ti 2p profiles in both BTO and BTO/Ti3C2Tx partially overlapped with those of Bi 4d3/2 at 466.4 eV. For the Bi 4f spectra (Fig. 1h), two main peaks at 158.9 and 164.3 eV in pure BTO were specified as Bi 4f7/2 and Bi 4f5/2 of Bi3+ [38], while they became 159.2 and 164.6 eV in BTO/Ti3C2Tx, respectively. This proved that the electron density of Bi in BTO/Ti3C2Tx decreased due to the strong interaction between BTO and Ti3C2Tx. The O 1s (Fig. S5b) and F 1s spectra (Fig. S5c) showed the peaks of the surface oxygen substances in BTO/Ti3C2Tx and the surface terminal group of F in Ti3C2Tx. These results confirmed that the interfacial electrons were transferred from BTO to Ti3C2Tx due to the strong interface interaction between BTO and Ti3C2Tx, which led to the formation of the Schottky heterojunction of BTO/Ti3C2Tx.
Ferroelectric Property and Photocatalytic Performance of BTO/Ti3C2Tx
Ferroelectricity was characterized using a piezo-response force microscope (PFM) and Kelvin probe force microscopy (KPFM). Figure 2a, b showed the PFM amplitude image and phase image of BTO/Ti3C2Tx. The typical amplitude voltage butterfly loops and well-defined 180° phase reversal hysteresis loops were obtained under the ± 10 V DC bias field, confirming the ferroelectric feature of BTO/Ti3C2Tx (Fig. 2c). The KPFM images under dark and light conditions were shown in Fig. 2d, e. The surface potential with illumination was lower than that without illumination. This could be interpreted as more photogenerated carriers migrating in the opposite direction under illumination driven by the potential difference. The difference in surface potential between dark and light was 46.73 mV (Fig. 2f), indicating that the polarization electric field was conducive to promoting the separation and migration of photogenerated carriers.Fig. 2 Ferroelectric property of BTO/Ti3C2Tx. a PFM amplitude image. b PFM phase image. c Amplitude-voltage curve and phase-voltage curve of BTO/Ti3C2Tx. d KPFM image in the dark. e KPFM image under illumination. f Surface potential distribution of BTO/Ti3C2Tx on the selected line under dark (black) and light (cyan)
To investigate the effect of ferroelectric polarization on photogenerated charge migration, the photoelectrochemical properties were measured. As shown in Fig. 3a, the photocurrent density of BTO/Ti3C2Tx hybrids was much larger than that of individual BTO and Ti3C2Tx, indicating that more free electrons were transferred under illumination. BT40 had the highest photocurrent density. Furthermore, BT40 had the smallest semicircle radius (Fig. 3b), demonstrating the lowest electrical impedance. Thus, BT40 had the best photoelectric performance, which was ascribed to the optimal doping amount of Ti3C2Tx. The effective interfacial transfer of photogenerated charges was realized using the most compact interface between BTO and Ti3C2Tx [43]. Conversely, BTO a presented great resistance to electron transfer in darkness, proving that spontaneous polarization alone could not rapidly migrate electrons. These results showed the efficient separation of photogenerated charges under the synergy of spontaneous polarization and Schottky heterojunction. Furthermore, the peak intensity of BT40 in the PL spectra (Fig. 3c) was obviously lower than that of BTO, implying that the recombination of photoexcited carriers was greatly restrained. The separated electrons and holes were captured by the surrounding oxygen species to produce ROS, which were detected using electron spin resonance (ESR). No obvious signs of hydroxyl radical (⋅OH) or superoxide radical (⋅O2−) were detected without light (Fig. 3d). Under illumination, the characteristic peaks of ⋅OH and ⋅O2− occurred, indicating the production of ⋅OH and ⋅O2−. This result confirmed the efficient separation and transfer of photogenerated carriers in BT40 under illumination, which was consistent with the photocurrent results (Fig. 3a).Fig. 3 Photocatalytic performance of BTO/Ti3C2Tx and the mechanism of ferroelectric polarization enhanced photocatalysis. a Photocurrent response under illumination. b EIS spectra with or without light. c Steady-state PL spectra of BT40 and BTO. d ESR spectra of ⋅OH and ⋅O2−. e The bandgap of BTO, BT20, BT40, and BT60. f Mott–Schottky curves of BTO at different frequencies. g Secondary electron cutoff of Ti3C2Tx. h Energy band structure diagram before illumination and charge directed migration and ROS production mechanism under illumination
In addition, the photothermal ability of the materials was tested under Xenon lamp irradiation (0.15 W cm−2) for 15 min (Fig. S6). Ti3C2Tx exhibited an excellent photothermal effect, with the highest temperature of 59.4 °C, which could be attributed to its localized surface plasmon resonance (LSPR) effect [44]. However, pure BTO only increased from 29.1 to 49.7 °C, which was almost the same as the Control group. The temperatures of BT20, BT40, and BT60 were 56.9, 58.1, and 57.4 °C, respectively, which were significantly higher than that of BTO. Compared to BT20 and BT60, BT40 had the highest temperature. This proved that the moderate addition of Ti3C2Tx could effectively improve the photothermal effect of BTO.
Compared to BTO, the photo-responsive range and photothermal conversion efficiency of BTO/Ti3C2Tx was enlarged (Fig. S7, 8), indicating the good light absorption of the composites under visible light. The energy gap (Eg) was estimated in the Tauc plot (Fig. 3e). The bandgap of BTO was 3.21 eV, whereas BT40 exhibited a narrower bandgap of 2.83 eV. The Mott–Schottky plot (Fig. 3f) showed that BTO was an n-type semiconductor with a flat band potential (Efb) of − 0.88 eV. Thus, the conduction band (CB) and valence band (VB) of BTO was − 0.68 and 2.53 eV, respectively. As shown in Fig. 3g, the secondary electron cutoff edge of Ti3C2Tx was 15.36 eV, and the corresponding work function (WF) was 5.86 eV, which was determined by subtracting from the excitation energy of He I (21.20 eV). The Fermi level (EF) of BTO was reported to be − 0.088 eV [23]. The band structure diagram of BTO/Ti3C2Tx was shown in Fig. 3h (left). As the WF of Ti3C2Tx was higher than that of BTO, a Schottky heterojunction was formed when BTO came into contact with Ti3C2Tx. The electrons were transferred from BTO to Ti3C2Tx until the EF reached equilibrium, accompanied by an upward bending of the energy band of BTO and the formation of the Schottky barrier (φSB).
The mechanism of the enhanced photocatalytic activity is shown in Fig. 3h (right). Upon visible light irradiation, electrons transitioned from VB to CB of BTO. Spontaneous polarization arose from the distortion of the [TiO6] octahedra of ferroelectric BTO [38]. Photogenerated carriers were diffused in the opposite direction under the internal electric field induced by the polarization charges. Thus, more photoelectrons crossed the φSB into EF of Ti3C2Tx. The φSB inhibited the backflow of electrons, which efficiently prohibited the recombination of electrons and holes. The photoinduced electrons directionally migrated to Ti3C2Tx, and holes were trapped in BTO. Additionally, Ti3C2Tx endowed BTO with a wider light response range. The increased light absorption excited more photogenerated carriers. The charge carriers rapidly reacted with O2 and H2O to produce abundant ROS in Ti3C2Tx and BTO, respectively. In sum, ferroelectric polarization promoted the directional separation and transfer of charge carriers, reduced the recombination of carriers, and further improved photocatalytic performance.
BT40-Based Light-Responsive Antibacterial Nanofiber Membrane
Based on the optimal light response, antibacterial property, and biocompatibility (Fig. 3a, b and Fig. S9, 10), BT40 was used in electrostatic spinning to design smart antibacterial textiles for killing pathogenic bacteria. The preparation of the electrospinning nanofiber membrane is illustrated in Fig. 4a. BT40 was integrated into PVDF through electrospinning with different percentages (i.e., 0, 10, 30, and 50 wt%), forming PVDF/BT40 composite membranes. The SEM images and photographs of PVDF, PVDF/BT40-10, PVDF/BT40-30, and PVDF/BT40-50 were shown in Fig. 4b–e. A pure PVDF membrane was composed of uniformly distributed nanofibers with a diameter of about 300 nm. The spheres in PVDF/BT40-50 contained Bi, Ti, and O elements, identified as BTO (Fig. S11). However, the Ti3C2Tx nanosheets were dissolved in the solvent during the preparation of the mixed electrospinning solution, which could be the reason for the improved tensile property of composite membranes. BT40 was both aggregated and partially dispersed in the PVDF matrix, and both states could respond to light. The FTIR spectra (Fig. S12) showed absorption bands at 840 and 1180 cm−1, which corresponded to the β-crystal phase of PVDF. PVDF/BT40 membranes had similar peaks to pristine PVDF membranes, indicating that the BT40 filler had no significant effect on the crystal phase of the PVDF substrate.Fig. 4 Synthesis, characterization, and properties of nanofiber membranes. a The schematic diagram of synthesis of electrospinning nanofiber membrane. b–e SEM images of PVDF (b), PVDF/BT40-10 (c), PVDF/BT40-30 (d), PVDF/BT40-50 (e), and insets show the photographs of the nanofiber membranes. f Digital images of contact angles of water droplets on the surface of nanofiber membranes. g Mechanical tests of the tensile strength versus elongation and inset shows the photograph of bend and torsion of PVDF/BT40-50. h Digital photographs of gas permeability of PVDF/BT40-50. i Water vapor transmittance rate. j UV–Vis–NIR absorption spectra of membranes from 250 to 800 nm. k Photothermal curves of different membranes under light irradiation (0.1 W cm−2, 20 min). l ESR spectra of ⋅OH. m ESR spectra of ⋅O2−
The hydrophobicity, flexibility, and breathability of the nanofiber membranes were also investigated. The contact angles of PVDF, PVDF/BT40-10, PVDF/BT40-30, and PVDF/BT40-50 were 139°, 143°, 141.7°, and 140.4°, respectively (Fig. 4f). They were far larger than 90°, demonstrating their remarkable hydrophobicity. The tensile strength of PVDF/BT40 membranes was greater than that of the pure PVDF membranes (Fig. 4g). The improved strength was attributed to the reinforcement of Ti3C2Tx in BT40 phase. However, the tensile strength of PVDF/BT40-60 decreased due to the excessive doping amount of BT40. The photographs of bend and torsion imply the flexibility of PVDF/BT40-50. Furthermore, the stability of BT40 in the membranes before and after stretching was characterized using XRD (Fig. S13). These results suggested that the PVDF/BT40-50 had enough strength and stability to be woven into textiles. The white smoke produced by volatile HCl and NH3⋅H2O indicated the good air permeability of PVDF/BT40-50 (Fig. 4h). Moreover, the water vapor transmittance rate (WVTR) was tested under the ASTM E96 inverse cup standard [45]. The WVTR of PVDF, PVDF/BT40-10, PVDF/BT40-30, and PVDF/BT40-50 was 66, 77, 82, and 92 g m−2 h−1 (Fig. 4i). This showed that the air permeability of PVDF/BT40 membranes was improved, which arose from the increase of inter-fiber pore size after doping BT40 [46]. The best air permeability of PVDF/BT40-50 implied good comfort when attached to human skin.
The UV–Vis–NIR absorption spectra showed that the light absorption intensity was enhanced with the doping of BT40 (Fig. 4j). The temperature changes in the nanofiber membranes under illumination (0.1 W cm−2, 20 min) were recorded to assess the photothermal effect (Fig. 4k and Fig. S14a). The temperature of PVDF/BT40 membranes was higher than that of PVDF, indicating a better photothermal conversion ability. PVDF/BT40-50 presented repeatable temperature changes after three cycles, proving its excellent photothermal stability (Fig. S14b). Moreover, the temperature remained almost constant after five washing cycles, indicating the photothermal durability of PVDF/BT40-50 (Fig. S14c). The photothermal conversion efficiencies of PVDF/BTO, PVDF/Ti3C2Tx, and PVDF/BT40-50 were calculated in Fig. S15. The ROS generation ability of PVDF/BT40-50 was confirmed by the ESR spectra (Fig. 4l, m). The results showed that PVDF/BT40-50 could produce ⋅OH and ⋅O2− under visible light irradiation. For the photocatalytic properties of all membranes, rhodamine B photodegradation was performed (Fig. S16). The degradation of PVDF/BT40-60 was not greater than that of PVDF/BT40-50, which proved that 50 wt% of BT40 was the best doping amount.
Antibacterial Activity Test and Biosafety Evaluation
Representative Gram-positive Staphylococcus aureus and Gram-negative Escherichia coli were chosen as the experimental strains to evaluate the antibacterial performance of nanofiber membranes. Antibacterial activity was determined using the spread plate method under simulated sunlight irradiation with an air mass 1.5 (AM 1.5) filter. After treatment with simulated sunlight (0.1 W cm−2, 20 min), the antibacterial effect of the PVDF/BT40 membrane was particularly significant, with dramatically reduced bacterial counts, while the colonies of S. aureus or E. coli of all groups almost had no variation without illumination (Fig. 5a, b). The antibacterial rates of PVDF/BT40-50 against S. aureus and E. coli were 99.61% ± 0.28% and 99.71% ± 0.16%, respectively (Fig. 5c, d). This implied that PVDF/BT40-50 had highly effective and broad-spectrum antibacterial activity. The fast bacteria-killing behavior of the materials and composite membrane was superior to other works [47–49]. However, due to excessive doping of BT40, the antibacterial effect of PVDF/BT40-60 was not further improved (Fig. S17). Therefore, PVDF/BT40-50 had the best antibacterial activity. The excellent antibacterial activity could be derived from the synergistic effect of combining ROS and hyperthermia induced by light irradiation. To explain the antibacterial mechanism, the bacterial morphologies in the different membranes were observed using SEM (Fig. 5e, f). Under illumination, the morphologies of both S. aureus and E. coli in the PVDF/BT40 group were differently destroyed with plicated and broken membranes (red arrows), whereas the PVDF and Control groups showed intact membranes and smooth surfaces. More severe damage was found in PVDF/BT40-50. This result further verified that PVDF/BT40-50 had high-efficiency antibacterial activity under illumination, consistent with the spread plate results.Fig. 5 Antibacterial performance and biocompatibility of nanofiber membranes. a, b Spread plates of S. aureus (a) and E. coli (b) treated with Control, PVDF, PVDF/BT40-10, PVDF/BT40-30, and PVDF/BT40-50 without and with illumination (0.1 W cm−2) for 20 min. c, d The corresponding statistical chart of strain counts of S. aureus (c) and E. coli (d). e, f The morphologies of S. aureus (e) and E. coli (f) in the dark and light. g Schematic diagram of antibacterial mechanism of nanofiber membrane. h Cell viability of nanofiber membranes after coculturing for 1 and 3 days. i Hemolysis ratio of different membranes. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001
Figure 5g shows the antibacterial mechanism of the nanofiber membrane. When contacting the membrane, the bacteria were killed by ROS and heat under light. That is, the BT40 composites on the surface of the nanofiber membrane responded to visible light and photogenerated carriers were excited in the semiconductor BTO. The inner electric field induced by the polarization charge of the BTO facilitated the separation and transfer of electrons and holes. Due to the formation of the Schottky heterojunction of BTO/Ti3C2Tx, the photoexcited electrons were transferred from BTO to Ti3C2Tx to generate ROS. Simultaneously, the photothermal effect was activated in BTO/Ti3C2Tx to produce heat. Based on ROS and heat, the cell membranes of the bacteria were damaged. The bacteria were sensitive to heat under the attack of ROS. Heat increased the permeability of the cell membrane, allowing ROS to enter the cell membrane more easily. Thus, the ROS and heat were synergetic in destroying the structure of bacteria and achieving the purpose of sterilization.
The cell viability assay of nanofiber membranes was performed using a Cell Counting Kit 8 (CCK-8). NIH-3T3 cells were incubated with different membranes for one and three days, and the results showed that the cell activity increased with the doping content of BT40 (Fig. 5h). After one day, PVDF/BT40-50 could promote cell proliferation compared to the Control group. After three days, all groups showed no obvious cytotoxicity. In addition, the hemolysis rates of PVDF, PVDF/BT40-10, PVDF/BT40-30, and PVDF/BT40-50 were lower than 5% (Fig. 5i), revealing that the fabricated nanofiber membranes had good blood compatibility and biological safety in vivo.
In Vivo Wound Infection Treatment with PVDF/BT40-50
A model rat skin wound, infected with S. aureus was developed to evaluate the therapeutic effects of PVDF/BT40-50 in vivo, as shown in Fig. 6a. The thermal images of PVDF/BT40-50 + Light group under illumination for 20 min were recorded in Fig. S18. The wound of the PVDF/BT40-50 + Light group was almost completely healed after 12 days, whereas the wounds of the Control, 3 M, and PVDF/BT40-50 + Dark groups still did not coalesce (Fig. 6b). The statistics of wound areas showed that PVDF/BT40-50 + Light group had the best therapeutic efficacy (Fig. 6c). After two and five days of treatment, the number of bacteria in the infected tissue was visibly reduced after PVDF/BT40-50 + Light treatment (Fig. 6d). Hematoxylin and eosin (H&E) staining showed that the number of neutrophils in PVDF/BT40-50 + Light group was significantly lower than that in the other groups, reflecting the minimum level of inflammation (Fig. 6e). The above data suggested that PVDF/BT40-50 had an excellent antibacterial effect irradiated by light and showed good application prospects in the treatment of tissue infection. Moreover, the area of collagen fibers in the PVDF/BT40-50 + Light group was highest (Fig. 6f), showing strong tissue regeneration ability and promoting wound healing ability. The heart, liver, spleen, lung, and kidney of the rats were analyzed using H&E staining (Fig. 6g). No abnormality was observed in the main organs, indicating that PVDF/BT40-50 was biosafe in vivo.Fig. 6 Antibacterial assay in vivo. a The flow chart of animal experiments. b, c The actual images (b) and area statistics (c) of the infected wounds on 0, 2, 5, and 12 days for Control, 3 M, PVDF/BT40-50 + Dark, and PVDF/BT40-50 + Light. d Giemsa staining for infected wound tissues collected on 2 and 5 days. Scale bars: 50 μm. e H&E staining of infected tissues after 2, 5, and 12 days of treatment. Scale bars: 50 μm. f Masson staining on 12 days. Scale bars: 50 μm. g H&E staining of the heart, liver, spleen, lung, and kidney tissues. Scale bars: 50 μm. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (Bacteria and neutrophils were labeled by red arrows in d and e)
Conclusions
In summary, the self-assembled BTO/Ti3C2Tx heterostructure was successfully constructed and embedded into the nanofibers through electrospinning. The PVDF/BT40-50 nanofiber membrane had prominent hydrophobicity, flexibility, and permeability, meeting the requirements of wearability. The spontaneous polarization of ferroelectric BTO first separated the inner photogenerated electrons and holes, and then the electrons migrated to Ti3C2Tx under the Schottky heterojunction of BTO/Ti3C2Tx. The charge carriers were reversely separated and utilized to the maximum extent, thus boosting the improvement in photocatalytic ability. Combining the generated ROS and heat, the PVDF/BT40-50 membrane exhibited high-efficiency antibacterial efficacies of 99.61% ± 0.28% and 99.71% ± 0.16% against S. aureus and E. coli under simulated solar light irradiation for 20 min, respectively. Moreover, the membrane showed good biocompatibility and strong wound healing ability. This study provides new insights into developing smart functional textiles incorporated with light-responsive antibacterial nanomaterials.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 9760 KB)
Acknowledgements
This work is jointly supported by the National Natural Science Foundation of China (Nos. 51871162, 52173251, 82002303), the China National Funds for Distinguished Young Scholars (No. 51925104), the Central Guidance on Local Science and Technology Development Fund of Hebei Province (226Z1303G), Scientific Research Foundation of Peking University Shenzhen Hospital (KYQD2021064), Guangdong Basic and Applied Basic Research Foundation (2021A1515220093, 2022A1515011536) and NSFC-Guangdong Province Joint Program (Key program no. U21A2084).
Declarations
Conflict of interest
The authors declare no competing financial interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Vincent JL Rello J Marshall J Silva E Anzueto A Martin CD Moreno R Lipman J Gomersall C Sakr Y Reinhart K International study of the prevalence and outcomes of infection in intensive care units J Am Med Assoc 2009 302 2323 10.1001/jama.2009.1754
2. Nelson RE Hatfield KM Wolford H Samore MH Scott RD Reddy SC Olubajo B Paul P Jernigan JA Baggs J National estimates of healthcare costs associated with multidrug-resistant bacterial infections among hospitalized patients in the United States Clin Infect Dis 2021 72 S17 10.1093/cid/ciaa1581 33512523
3. Jernigan JA Hatfield KM Wolford H Nelson RE Olubajo B Reddy SC McCarthy N Paul P McDonald LC Kallen A Fiore A Craig M Baggs J Multidrug-resistant bacterial infections in US hospitalized patients, 2012–2017 N Engl J Med 2020 382 1309 10.1056/NEJMoa1914433 32242356
4. Moore JB June CH Cytokine release syndrome in severe COVID-19 Science 2020 368 473 10.1126/science.abb8925 32303591
5. Moreno-Torres V de Mendoza C de la Fuente S Sanchez E Martinez-Urbistondo M Herraiz J Gutierrez A Gutierrez A Hernandez C Callejas A Mainez C Royuela A Cuervas-Mons V Bacterial infections in patients hospitalized with COVID-19 Intern Emerg Med 2021 17 431 10.1007/s11739-021-02824-7 34406633
6. CDC. Personal protective equipment (PPE). 2020. https://www.cdc.gov/coronavirus/2019-ncov/hcp/using-ppe.html. Accessed 19 Aug 2020.
7. Karim N Afroj S Lloyd K Oaten LC Andreeva DV Carr C Farmery AD Kim ID Novoselov KS Sustainable personal protective clothing for healthcare applications: a review ACS Nano 2020 14 12313 10.1021/acsnano.0c05537 32866368
8. Makabenta JMV Nabawy A Li CH Schmidt-Malan S Patel R Rotello VM Nanomaterial-based therapeutics for antibiotic-resistant bacterial infections Nat Rev Microbiol 2021 19 23 10.1038/s41579-020-0420-1 32814862
9. Matatkova O Michailidu J Miskovska A Kolouchova I Masak J Cejkova A Antimicrobial properties and applications of metal nanoparticles biosynthesized by green methods Biotechnol Adv 2022 58 107905 10.1016/j.biotechadv.2022.107905 35031394
10. Guo ZC Chen Y Wang YH Jiang H Wang XM Advances and challenges in metallic nanomaterial synthesis and antibacterial applications J Mater Chem B 2020 8 4764 10.1039/D0TB00099J 32207511
11. Zhou ZA Li J Tan L Liu XM Zheng YF Cui ZD Li CY Yeung KWK Li ZY Liang YQ Zhu SL Wu SL Using tea nanoclusters as β-lactamase inhibitors to cure multidrug-resistantbacterial pneumonia: a promising therapeutic strategy by Chinese materioherbology Fundam Res 2022 2 496 10.1016/j.fmre.2021.11.019
12. Li J Zhou ZA Liu XM Zheng YF Li CY Cui ZD Yeung KWK Zhou HB Zou JH Li ZY Zhu SL Liang YQ Wang XB Wu SL Material-herbology: an effective and safe strategy to eradicate lethal viral-bacterial pneumonia Matter 2021 4 3030 10.1016/j.matt.2021.07.001
13. Qiao YQ Xu YD Liu XM Zheng YF Li B Han Y Li ZY Yeung KWK Liang YQ Zhu SL Cui ZD Wu SL Microwave assisted antibacterial action of garcinia nanoparticles on gram-negative bacteria Nat Commun 2022 13 2461 10.1038/s41467-022-30125-w 35513402
14. Jin LG Liu XM Zheng YF Li ZY Zhang Y Zhu SL Jiang H Cui ZD Chu PK Wu SL Interface polarization strengthened microwave catalysis of MoS2/FeS/Rhein for the therapy of bacteria-infected osteomyelitis Adv Funct Mater 2022 32 2204437 10.1002/adfm.202204437
15. Liu HP Li JF Liu XM Li ZY Zhang Y Liang YQ Zheng YF Zhu SL Cui ZD Wu SL Photo-sono interfacial engineering exciting the intrinsic property of herbal nanomedicine for rapid broad-spectrum bacteria killing ACS Nano 2021 15 18505 10.1021/acsnano.1c08409 34739223
16. Guo JY Li X Liang J Yuan XZ Jiang LB Yu HB Sun HB Zhu ZQ Ye SJ Tang N Zhang J Fabrication and regulation of vacancy-mediated bismuth oxyhalide towards photocatalytic application: development status and tendency Coordin Chem Rev 2021 443 214033 10.1016/j.ccr.2021.214033
17. Li HP Hou WG Tao XT Du N Conjugated polyene-modified Bi2MO6 (M=Mo or W) for enhancing visible light photocatalytic activity Appl Catal B Environ 2015 172 27 10.1016/j.apcatb.2015.02.005
18. Das K Majhi D Bhoi YP Mishra BG Combustion synthesis, characterization and photocatalytic application of CuS/Bi4Ti3O12 p-n heterojunction materials towards efficient degradation of 2-methyl-4-chlorophenoxyacetic acid herbicide under visible light Chem Eng J 2019 362 588 10.1016/j.cej.2019.01.060
19. Zhang Q Shi YY Shi XJ Huang TT Lee SC Huang Y Cao JJ Constructing Pd/ferroelectric Bi4Ti3O12 nanoflake interfaces for O2 activation and boosting NO photo-oxidation Appl Catal B: Environ 2022 302 120876 10.1016/j.apcatb.2021.120876
20. Liu YB Zhu GQ Gao JZ Hojamberdiev M Zhu RL Wei XM Guo QM Liu P Enhanced photocatalytic activity of Bi4Ti3O12 nanosheets by Fe3+-doping and the addition of Au nanoparticles: photodegradation of phenol and bisphenol A Appl Catal B Environ 2017 200 72 10.1016/j.apcatb.2016.06.069
21. Chen ZW Jiang H Jin WL Shi CK Enhanced photocatalytic performance over Bi4Ti3O12 nanosheets with controllable size and exposed 0 0 1 facets for rhodamine B degradation Appl Catal B Environ 2016 180 698 10.1016/j.apcatb.2015.07.022
22. Yu HJ Chen F Li XW Huang HW Zhang QY Su SQ Wang KY Mao EY Mei B Mul G Ma TY Zhang YH Synergy of ferroelectric polarization and oxygen vacancy to promote CO2 photoreduction Nat Commun 2021 12 4594 10.1038/s41467-021-24882-3 34321482
23. Wang XY Wang YS Gao MC Shen JN Pu XP Zhang ZZ Lin HX Wang XX BiVO4/Bi4Ti3O12 heterojunction enabling efficient photocatalytic reduction of CO2 with H2O to CH3OH and CO Appl Catal B Environ 2020 270 118876 10.1016/j.apcatb.2020.118876
24. Mohammadi AV Rosen J Gogotsi Y The world of two-dimensional carbides and nitrides (MXenes) Science 2021 372 1165
25. Li YZ Zhang XT Electrically conductive, optically responsive, and highly orientated Ti3C2Tx MXene aerogel fibers Adv Funct Mater 2022 32 2107767 10.1002/adfm.202107767
26. Seyedin S Uzun S Levitt A Anasori B Dion G Gogotsi Y Razal JM MXene composite and coaxial fibers with high stretchability and conductivity for wearable strain sensing textiles Adv Funct Mater 2020 30 1910504 10.1002/adfm.201910504
27. Ding JX Zhang J Li JN Li D Xiao CS Xiao HH Yang HH Zhuang XL Chen XS Electrospun polymer biomaterials Prog Polym Sci 2019 90 1 10.1016/j.progpolymsci.2019.01.002
28. Zhang X Lv SS Lu XC Yu H Huang T Zhang QH Zhu MF Synergistic enhancement of coaxial nanofiber-based triboelectric nanogenerator through dielectric and dispersity modulation Nano Energy 2020 75 104894 10.1016/j.nanoen.2020.104894
29. Xue JJ Wu T Dai YQ Xia YN Electrospinning and electrospun nanofibers: methods, materials, and applications Chem Rev 2019 119 5298 10.1021/acs.chemrev.8b00593 30916938
30. Chen HY Zhou LL Fang Z Wang SZ Yang T Zhu LP Hou XM Wang HL Wang ZL Piezoelectric nanogenerator based on in situ growth all-inorganic CsPbBr3 perovskite nanocrystals in PVDF fibers with long-term stability Adv Funct Mater 2021 31 2011073 10.1002/adfm.202011073
31. Lu LJ Ding WQ Liu JQ Yang B Flexible PVDF based piezoelectric nanogenerators Nano Energy 2020 78 105251 10.1016/j.nanoen.2020.105251
32. Sharma A Panwar V Mondal B Prasher D Bera MK Thomas J Kumar A Kamboj N Mandal D Ghosh D Electrical stimulation induced by a piezo-driven triboelectric nanogenerator and electroactive hydrogel composite, accelerate wound repair Nano Energy 2022 99 107419 10.1016/j.nanoen.2022.107419
33. Zhang XW Liang YY Ni CJ Li YJ Anti-biofouling microfiltration membranes based on 1-vinyl-3-butylimidazolium chloride grafted PVDF with improved bactericidal properties and vitro biocompatibility Mater Sci Eng C 2021 118 111411 10.1016/j.msec.2020.111411
34. Sun J Fan Y Ye W Tian LM Niu SC Ming WH Zhao J Ren LQ Near-infrared light triggered photodynamic and nitric oxide synergistic antibacterial nanocomposite membrane Chem Eng J 2021 417 128049 10.1016/j.cej.2020.128049
35. Liu YT Zhang P Sun N Anasori B Zhu QZ Liu H Gogotsi Y Xu B Self-assembly of transition metal oxide nanostructures on MXene nanosheets for fast and stable lithium storage Adv Mater 2018 30 1707334 10.1002/adma.201707334
36. Yang LY Cui J Zhang L Xu XR Chen X Sun DP A moisture-driven actuator based on polydopamine-modified MXene/bacterial cellulose nanofiber composite film Adv Funct Mater 2021 31 2101378 10.1002/adfm.202101378
37. Hou TT Luo QQ Li Q Zu HL Cui PX Chen SW Lin Y Chen JJ Zheng XS Zhu WK Liang SQ Yang JL Wang LB Modulating oxygen coverage of Ti3C2Tx MXenes to boost catalytic activity for HCOOH dehydrogenation Nat Commun 2020 11 4251 10.1038/s41467-020-18091-7 32843636
38. Liu LZ Huang HW Chen ZS Yu HJ Wang KY Huang JD Yu H Zhang YH Synergistic polarization engineering on bulk and surface for boosting CO2 photoreduction Angew Chem Int Ed Engl 2021 60 18303 10.1002/anie.202106310 34058045
39. Li R Ma XY Li JM Cao J Gao HZ Li TS Zhang XY Wang LC Zhang QH Wang G Hou CY Li YG Palacios T Lin YX Wang HZ Ling X Flexible and high-performance electrochromic devices enabled by self-assembled 2D TiO2/MXene heterostructures Nat Commun 2021 12 1587 10.1038/s41467-021-21852-7 33707439
40. Sarycheva A Gogotsi Y Raman spectroscopy analysis of the structure and surface chemistry of Ti3C2Tx MXene Chem Mater 2020 32 3480 10.1021/acs.chemmater.0c00359
41. Liu LZ Huang HW Chen F Yu HJ Tian N Zhang YH Zhang TR Cooperation of oxygen vacancies and 2D ultrathin structure promoting CO2 photoreduction performance of Bi4Ti3O12 Sci Bull 2020 65 934 10.1016/j.scib.2020.02.019
42. Li JF Li ZY Liu XM Li CY Zheng YF Yeung KWK Cui ZD Liang YQ Zhu SL Hu WB Qi YJ Zhang TJ Wang XB Wu SL Interfacial engineering of Bi2S3/Ti3C2Tx MXene based on work function for rapid photo-excited bacteria-killing Nat Commun 2021 12 1224 10.1038/s41467-021-21435-6 33619276
43. Ran JR Gao GP Li FT Ma TY Du AJ Qiao SZ Ti3C2 MXene co-catalyst on metal sulfide photo-absorbers for enhanced visible-light photocatalytic hydrogen production Nat Commun 2017 8 13907 10.1038/ncomms13907 28045015
44. Wu XH Wang JH Wang ZY Sun F Liu YZ Wu KF Meng XY Qiu JS Boosting the electrocatalysis of MXenes by plasmon-induced thermalization and hot-electron injection Angew Chem Int Ed Engl 2021 60 9416 10.1002/anie.202016181 33480124
45. Hsu PC Song AY Catrysse PB Liu C Peng YC Xie J Fan SH Cui Y Radiative human body cooling by nanoporous polyethylene textile Science 2016 353 6303 10.1126/science.aaf5471
46. Li ZJ Cheng BW Ju JG Kang WM Liu Y Development of a novel multi-scale structured superhydrophobic nanofiber membrane with enhanced thermal efficiency and high flux for membrane distillation Desalination 2021 501 114834 10.1016/j.desal.2020.114834
47. Deka S Devi MB Khan MR Keerthana Venimadhav A Choudhury B Piezo-photocatalytic and photocatalytic bismuth vanadate nanorods with antibacterial property ACS Appl Nano Mater 2022 5 10724 10.1021/acsanm.2c02072
48. Jia YN Zhan SH Ma SL Zhou QX Fabrication of TiO2-Bi2WO6 binanosheet for enhanced solar photocatalytic disinfection of E. coli: insights on the mechanism ACS Appl Mater Interfaces 2016 8 6841 10.1021/acsami.6b00004 26910210
49. Wang XL Li ZH Zhang Y Li QL Du H Liu F Zhang XL Mu HB Duan JY Enhanced photocatalytic antibacterial and degradation performance by p–n–p type CoFe2O4/CoFe2S4/MgBi2O6 photocatalyst under visible light irradiation Chem Eng J 2022 429 132270 10.1016/j.cej.2021.132270
| 36466134 | PMC9707173 | NO-CC CODE | 2022-12-01 23:20:21 | no | Adv Fiber Mater. 2022 Nov 28;:1-13 | utf-8 | Adv Fiber Mater | 2,022 | 10.1007/s42765-022-00234-8 | oa_other |
==== Front
J Autism Dev Disord
J Autism Dev Disord
Journal of Autism and Developmental Disorders
0162-3257
1573-3432
Springer US New York
36441431
5817
10.1007/s10803-022-05817-3
S:I: .Developmental Approach and Targeted Treatment of Sensory Alterations
Longitudinal Relations Between Early Sensory Responsiveness and Later Communication in Infants with Autistic and Non-autistic Siblings
http://orcid.org/0000-0002-5723-5834
Feldman Jacob I. [email protected]
12
Garla Varsha 3
Dunham Kacie 45
Markfeld Jennifer E. 5
Bowman Sarah M. 16
Golden Alexandra J. 7
Daly Claire 5
Kaiser Sophia 8
Mailapur Nisha 9
Raj Sweeya 7
Santapuram Pooja 710
Suzman Evan 1112
Augustine Ashley E. 1314
Muhumuza Aine 3
Cascio Carissa J. 241516
Williams Kathryn L. 17
Kirby Anne V. 18
Keceli-Kaysili Bahar 1
Woynaroski Tiffany G. 12416
1 grid.412807.8 0000 0004 1936 9916 Department of Hearing and Speech Sciences, Vanderbilt University Medical Center, 8310 South Tower, 1215 21St Avenue South, Nashville, TN 37232 USA
2 grid.152326.1 0000 0001 2264 7217 Frist Center for Autism & Innovation, Vanderbilt University, Nashville, TN USA
3 grid.152326.1 0000 0001 2264 7217 Neuroscience Undergraduate Program, Vanderbilt University, Nashville, TN USA
4 grid.152326.1 0000 0001 2264 7217 Vanderbilt Brain Institute, Vanderbilt University, Nashville, TN USA
5 grid.152326.1 0000 0001 2264 7217 Department of Hearing and Speech Sciences, Vanderbilt University, Nashville, TN USA
6 grid.239573.9 0000 0000 9025 8099 Present Address: Department of Pediatrics, Cincinnati Children’s Hospital, Cincinnati, OH USA
7 grid.152326.1 0000 0001 2264 7217 Vanderbilt School of Medicine, Vanderbilt University, Nashville, TN USA
8 grid.152326.1 0000 0001 2264 7217 Cognitive Studies Undergraduate Program, Vanderbilt University, Nashville, TN USA
9 grid.152326.1 0000 0001 2264 7217 Economics Undergraduate Program, Vanderbilt University, Nashville, TN USA
10 grid.21729.3f 0000000419368729 Present Address: Department of Anesthesiology, Columbia University Irving Medical Center, New York, NY USA
11 grid.152326.1 0000 0001 2264 7217 Master’s Program in Biomedical Science, Vanderbilt University, Nashville, TN USA
12 grid.267313.2 0000 0000 9482 7121 Present Address: University of Texas Southwestern Medical School, University of Texas, Dallas, TX USA
13 grid.152326.1 0000 0001 2264 7217 Biological Sciences Undergraduate Program, Vanderbilt University, Nashville, TN USA
14 grid.241104.2 0000 0004 0452 4020 Present Address: Department of Pediatrics, University Hospitals Cleveland, Cleveland, OH USA
15 grid.412807.8 0000 0004 1936 9916 Department of Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center, Nashville, TN USA
16 grid.412807.8 0000 0004 1936 9916 Vanderbilt Kennedy Center, Vanderbilt University Medical Center, Nashville, TN USA
17 grid.265122.0 0000 0001 0719 7561 Department of Occupational Therapy and Occupational Science, Towson University, Towson, MD USA
18 grid.223827.e 0000 0001 2193 0096 Department of Occupational and Recreational Therapies, University of Utah, Salt Lake City, UT USA
28 11 2022
113
4 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Early differences in sensory responsiveness may contribute to difficulties with communication among autistic children; however, this theory has not been longitudinally assessed in infants at increased familial versus general population-level likelihood for autism (Sibs-autism vs. Sibs-NA) using a comprehensive battery of sensory responsiveness and communication. In a sample of 40 infants (20 Sibs-autism, of whom six were later diagnosed with autism; 20 Sibs-NA), we tested (a) associations between sensory responsiveness at 12–18 months and communication 9 months later and (b) evaluated whether such associations were moderated by sibling group, autism diagnosis, or age. We found negative zero-order correlations between sensory responsiveness (i.e., caregiver reported hyperresponsiveness and hyporesponsiveness; an observational measure of hyperresponsiveness) and later communication. Additionally, caregiver reported sensory seeking was negatively associated with later expressive communication only in Sibs-NA. Limitations include our relatively small sample size of infants diagnosed with autism. Implications for future research are discussed.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10803-022-05817-3.
Keywords
Sensory responsiveness
Sensory reactivity
Communication
Infant siblings
Baby sibs
Predictors
http://dx.doi.org/10.13039/100000002 National Institutes of Health P50HD103537 http://dx.doi.org/10.13039/100000055 National Institute on Deafness and Other Communication Disorders R21 DC016144 DC020129-01A1 Dunham Kacie Woynaroski Tiffany G. http://dx.doi.org/10.13039/100006108 National Center for Advancing Translational Sciences KL2TR000446 Woynaroski Tiffany G. http://dx.doi.org/10.13039/100007206 Vanderbilt Institute for Clinical and Translational Research VR14759 VR53990 Garla Varsha Woynaroski Tiffany G. http://dx.doi.org/10.13039/501100008982 National Science Foundation NRT grant DGE 19-22697 http://dx.doi.org/10.13039/100007429 Nancy Lurie Marks Family Foundation National Institutes of HealthTL1TR002244 R01DC020186 Woynaroski Tiffany G. http://dx.doi.org/10.13039/100018481 Vanderbilt Kennedy Center, Vanderbilt University Medical Center
==== Body
pmcAutism is a neurodevelopmental condition that affects approximately two million people in the United States and tens of millions of people worldwide (Baio et al., 2018; Maenner et al., 2021). With differences in sensory responsiveness now recognized as a core feature of autism (American Psychiatric Association, 2013), there has been an increased interest in sensory function in autistic individuals1 (e.g., Cascio et al., 2016; Robertson & Baron-Cohen, 2017; Schauder & Bennetto, 2016). It has been proposed that differences in sensory responsiveness, particularly early in life, may impact a child’s ability to engage with and learn from their environment, thereby producing “cascading effects” on development across domains and ultimately causing or contributing to core and related features of autism (Bradshaw et al., 2022; Cascio et al., 2016; Wallace et al., 2020), in particular differences in communication (Bahrick & Todd, 2012; Damiano-Goodwin et al., 2018; Santapuram et al., 2022). The acquisition of communication skills in early childhood has been repeatedly linked with quality of life and long-term outcomes of autistic children (e.g., social, academic, and vocational success; Billstedt et al., 2007; Eisenberg, 1956; Gillberg & Steffenburg, 1987; Kobayashi et al., 1992); thus, identifying early predictors of communication in this population is imperative.
There is a large literature supporting differences in sensory responsiveness, specifically hyperresponsiveness (i.e., exaggerated responses to sensory stimuli), hyporesponsiveness (i.e., reduced or absent responses to sensory stimuli), and sensory seeking (i.e., craving of or fascination with certain sensory experiences), in autistic children and adults compared to non-autistic peers (see Ben-Sasson et al., 2009, 2019; Kirby et al., 2022). Increased report of these patterns of sensory responsiveness is associated with lower social communication scores in autistic children and adults (Foss-Feig et al., 2012; Horder et al., 2014; Lane et al., 2010; Nowell et al., 2020; Williams et al., 2018) and lower language scores in autistic preschoolers and children (Baranek et al., 2013; Feldman et al., 2020; Nowell et al., 2020; Watson et al., 2011), lending some support to the cascading effects hypothesis.
However, testing theories that rely on observations of early development in autistic children, such as the cascading effects theory, is challenging given that autism cannot always be reliably diagnosed until at least the second year of life (e.g., Luyster et al., 2009; Woolfenden et al., 2012). Historically, researchers have utilized retrospective analyses to study sensory development in infancy (e.g., Baranek, 1999a; Freuler et al., 2012), but more recent efforts have examined the longitudinal impacts of early sensory development by focusing on infants at increased likelihood for developing autism (e.g., Baranek et al., 2018; Damiano-Goodwin et al., 2018; Nowell et al., 2020; Wolff et al., 2019). By prospectively following these infants, we can study the emergence and the sequelae of altered sensory functioning in the earliest stages of development in children who will go on to be diagnosed with autism. In one branch of this research focused on community-referred infants with autistic features early in life, differences in sensory responsiveness, specifically elevated hyporesponsiveness and sensory seeking, during the second year of life has been associated with lower communication scores later in life (Baranek et al., 2018; Grzadzinski et al., 2021).
Another line of research has focused on infants who are known to be at elevated likelihood of autism because they have an autistic older sibling (Sibs-autism; Ozonoff et al., 2011); approximately one-third of these infants will be diagnosed with autism or a language impairment (Charman et al., 2017; Ozonoff et al., 2011). For example, in a large-scale analysis of Sibs-autism, both hyporesponsiveness and sensory seeking were associated with concurrent communication at 24 months (Wolff et al., 2019). Although Wolff et al. (2019) included all three patterns of sensory responsiveness in their analyses, they only utilized caregiver report measures of sensory responsiveness (i.e., the Sensory Experiences Questionnaire [SEQ]; Baranek et al., 2006) and communication (i.e., the Vineland Adaptive Behavior Scales [VABS-2]; Sparrow et al., 2005) and did not carry out any longitudinal analyses. In another large-scale study of both Sibs-autism and infants at lower familial likelihood for autism (i.e., infants with only non-autistic older siblings; Sibs-NA), Narvekar et al. (2022) found that perceptual sensitivity at 14 and 24 months were related to social communication differences at 36 months; however, they did not comprehensively evaluate sensory processing differences associated with autism.
More recent work in our laboratory (Feldman et al., 2021) has focused on differences between Sibs-autism and Sibs-NA. We found that some alterations in sensory responsiveness were detectable in the developmental window between 12 and 18 months of age. Additionally, we found that select indices of sensory responsiveness were associated with concurrent communication outcomes, either unconditionally (e.g., link between caregiver reported hyporesponsiveness and both receptive and expressive communication) or only in older infants [e.g., associations between expressive communication and both caregiver reported sensory seeking and hyperresponsiveness as measured by the Test of Sensory Functions in Infants (TSFI)]. Though the aforementioned work, collectively, provided preliminary empirical support for theorized links between early sensory responsiveness and communication skills across the 12–18 month period in Sibs-autism, conclusions about directionality or causality of observed associations between atypical sensory responsiveness and communication skill could not be drawn, given the cross-sectional and concurrent correlational nature of the designs.
The present study, thus, seeks to expand upon our previous work by exploring the potential longitudinal links between early sensory responsiveness and later communication skills. To expand upon our previous study, we prospectively followed the same sample of 20 Sibs-autism and 20 Sibs-NA. For this study, we added to our comprehensive evaluation of infants’ sensory responsiveness at 12–18 months of age by measuring their communication skills 9 months later (Time 2; i.e., at 21–27 months). We sought to evaluate (a) whether sensory responsiveness in infancy, as indexed using several previously developed and validated caregiver report and observational measures, is associated with later receptive and expressive communication, and whether the aforementioned associations vary according to (b) sibling group (Sibs-autism and Sibs-NA) and/or (c) chronological age when sensory measures were collected, given our previous findings that concurrent correlations were moderated by age (see Feldman et al., 2021). We additionally explored whether relations of interest varied according to preliminary diagnostic outcome (i.e., Sibs-autism who were diagnosed with autism during the study period [Sibs-autism-dx] versus Sibs-autism who were not diagnosed with autism during the study period [Sibs-autism-nodx] and Sibs-NA).
Method
Study Design
The Vanderbilt University Institutional Review Board approved all recruitment and study procedures. Parents provided written informed consent, and families were compensated for their participation.
Participants were recruited when they were between the ages of 12–18 months (± 30 days; resulting in one infant who was 11 months old at the time of the initial evaluation) from advertisements, at outreach events at the Adventure Science Center, and through flyers handed out at Vanderbilt outpatient clinics and preschools for children with autism. We followed these infants longitudinally. In this study, we will be reporting results relevant to their first outcome visit, 9 months after study entry (Time 2; i.e., 21–27 months). At each time point, infants completed all study measures in one to three visits to Vanderbilt University Medical Center scheduled over the course of a 2-week period.
Participants
Analyses were conducted on 40 infants and toddlers (20 Sibs-autism, 20 Sibs-NA). This sample was previously included in Feldman et al. (2021), which focused on concurrent correlations between sensory responsiveness and communication at the entry time-point in our study. Groups were matched on chronological age and biological sex (see Table 1). Inclusion criteria for infants in both groups were (a) full term birth, (b) no concomitant genetic disorders, (c) no known adverse neurological history, (d) primarily English-speaking household, and (e) at least one older sibling. For the Sibs-autism group, infants were required to have at least one older sibling with autism as diagnosed by a licensed clinician according to DSM-5 criteria (American Psychiatric Association, 2013). The records of older siblings with autism were reviewed by a member of our research team to confirm diagnostic status (n = 15) at the time of the infant sibling’s entry to the study; when records were not available, a licensed clinician administered the ADOS-2 (Lord et al., 2012) and independently confirmed the diagnosis (n = 5). For the Sibs-NA group, infants were required to have (a) only non-autistic older siblings, as confirmed by screening below the threshold for autism concern (i.e., < 15) on the Social Communication Questionnaire (Rutter et al., 2003) and a screening interview for developmental delay, (b) no first-degree relatives diagnosed with autism, (c) no prior history or present indicators of developmental delays or disorders per caregiver report, and (d) an Early Learning Composite on the Mullen Scales of Early Learning (MSEL; Mullen, 1995) greater than 70.Table 1 Means, standard deviations, and group differences for selected variables by group
Sibs-autism Sibs-NA
Total
(n = 20) Sibs-autism-dx
(n = 6) Sibs-autism-nodx
(n = 14) Total
(n = 20)
M (SD) M (SD) M (SD) M (SD)
Time 1 variables
Chronological Age (Months) 13.75 (1.9) 12.67 (0.8) 14.21 (2.0) 13.75 (2.0)
Mental Age (Months) 13.30 (1.4) 12.54 (1.5) 13.63 (1.3) 14.49 (2.2)
Mullen ELC 89.93 (12.6) 86.83 (16.3) 91.26 (11.1) 99.85 (9.4)
Time 2 variables
Mullen Expressive Lang. AEQ 19.05 (4.7) 15.00 (2.7) 20.79 (4.3) 25.95 (7.8)
Mullen Receptive Lang. AEQ 17.90 (6.7) 11.17 (2.2) 20.79 (5.8) 27.03 (6.9)
VABS Expressive Com. AEQ 20.95 (6.0) 15.17 (5.1) 23.43 (4.4) 24.14 (5.2)
VABS Receptive Com. AEQ 21.30 (8.3) 14.00 (5.5) 24.43 (7.3) 26.61 (4.8)
MCDI Vocabulary† 110.67 (106.7) 45.83 (36.1) 138.45 (115.6) 230.39 (158.2)
Demographic variables n N n n
Biological sex 11 Male
9 Female
4 Male
2 Female
7 Male
7 Female
11 Male
9 Female
Race 20 White 6 White 14 White 18 White
2 Multiple
Ethnicity 1 Hispanic or Latino
19 Non-Hispanic or Latino
6 Non-Hispanic or Latino 1 Hispanic or Latino
13 Non-Hispanic or Latino
20 Non-Hispanic or Latino
Sibs-autism = infant siblings of autistic children, Sibs-autism-dx = Sibs-autism who were diagnosed with autism during the study period, Sibs-autism-nodx = Sibs-autism who were not diagnosed with autism during the study period, Sibs-NA = infant siblings of non-autistic, otherwise typically developing children. Sibs-autism information is summarized both across and within (i.e., for autism and non-autism) diagnostic outcome subgroups. Time 1 = 12–18 months, Time 2 = 9 months later (i.e., 21–27 months). Mental Age = average of Visual Reception, Fine Motor, Expressive Language, and Receptive Language age equivalency scores from the Mullen Scales of Early Learning (Mullen, 1995), Mullen ELC = Early Learning Composite standard score from the Mullen Scales of Early Learning (Mullen, 1995), Lang. = Language, AEQ = age equivalency scores, VABS = Vineland Adaptive Behavior Scales, second edition (Sparrow et al., 2005), Com. = Communication, MCDI = MacArthur-Bates Communicative Development Inventories (Fenson et al., 2007). Sibs-autism and Sibs-NA groups were non-significantly different on biological sex and chronological age; groups differed on Mullen ELC and mental age at Time 1 and all variables at Time 2 (p values < .05). †Values represent back-transformed values, as this variable was transformed with a square-root transformation prior to analyses
Measures of Sensory Responsiveness
Several observational and caregiver report measures of sensory responsiveness were utilized in the present study in an attempt to increase the stability, and thereby the potential construct validity, of indices of hyporesponsiveness, hyperresponsiveness, and sensory seeking (Cronbach & Meehl, 1955; Rushton et al., 1983). See Table 2 for a list of variables derived from these measures.Table 2 Summary of component variables derived for use in analyses
Assessment Type Component variables
Measures of sensory responsiveness
Sensory Experiences Questionnaire version 2.1 (Baranek et al., 2006) Caregiver Report Sensory Seekinga, Hyperresponsivenessb, and Hyporesponsivenessc mean scores
Sensory Profile
(Dunn, 1999)
Caregiver Report Low Registration (Hypo)c, Sensation Seeking (Seeking)a, Sensory Sensitivity (Hyper)b, and Sensation Avoiding (Hyper)b scores
Sensory Processing Assessment (Baranek, 1999b) Observational Sensory Seeking Intensityd and Inventoryd, Avoidance (Hyper), and Orientation (Hypo) mean scores
Test of Sensory Functions in Infants (DeGangi & Greenspan, 1989) Observational Hyporesponsiveness and Hyperresponsiveness mean scores
Measures of communication
Mullen Scales of Early Learning (Mullen, 1995) Observational Receptivee and Expressivef Language age equivalency scores
MacArthur-Bates Communicative Development Inventories (Fenson et al., 2007) Caregiver Report Raw number of words child understands and says (expressive vocabulary)f
Vineland Adaptive Behavior Scales, second edition (Sparrow et al., 2005) Caregiver Report Receptive Communicatione and Expressive Communicationf age equivalency scores
The construct purportedly tapped by each variable is indicated in parentheses when not transparent. Seeking = Sensory Seeking, Hyper = Hyperresponsiveness, Hypo = Hyporesponsiveness. Variables were aggregated to measure: acaregiver reported sensory seeking, bcaregiver reported hyperresponsiveness, ccaregiver reported hyporesponsiveness, dsensory seeking as measured by the Sensory Processing Assessment, ereceptive communication, and fexpressive communication
Sensory Experiences Questionnaire (SEQ)
The SEQ version 2.1 (Baranek et al., 2006) is a caregiver report measure that characterizes behaviors across a range of sensory modalities, response patterns, and social and non-social contexts. Mean scores for hyporesponsiveness, hyperresponsiveness, and sensory seeking were derived from this measure for use in analyses.
Infant/Toddler Sensory Profile Caregiver Questionnaire (SP)
The SP (Dunn, 1999) is an 81-item caregiver report measure that characterizes early sensory processing. From this measure, we derived the Low Registration (hyporesponsiveness), Sensation Seeking (sensory seeking), Sensory Sensitivity and Sensation Avoiding (hyperresponsiveness) indices for use in analyses. Scores on this measure were reflected (i.e., raw scores were subtracted from the maximum observed score + 1) to ensure consistency of interpretation with other measures (i.e., wherein higher scores are indicative of increased presence of the behaviors of interest).
Sensory Processing Assessment (SPA)
The SPA (Baranek, 1999b) is a 15-min observational assessment that evaluates a child’s responses (seeking and/or avoiding behaviors, orienting and habituation responses) to novel toys and environmental sensory stimuli that are either social or non-social in nature. From this measure, we derived the sensory seeking intensity mean rating score and the sensory seeking inventory to index sensory seeking behavior (see Damiano-Goodwin et al., 2018), the avoidance mean score to index hyperresponsiveness, and the orienting mean score to index hyporesponsiveness. Higher scores indicated greater presence of these patterns of sensory responsiveness.
Coders were naïve to sibling status. Eleven samples (26.8%) were randomly selected for coding by a second rater; intraclass correlation coefficients (ICC) for SPA variables were 0.907 for orienting (hyporesponsiveness), 0.820 and 0.810 for sensory seeking intensity and sensory seeking inventory, respectively, and 0.894 for avoidance (hyperresponsiveness), indicating good to excellent interrater reliability for these observational variables (see Feldman et al., 2021 for more detailed information).
Test of Sensory Functions in Infants (TSFI)
The TSFI (DeGangi & Greenspan, 1989) is a brief observational measure of sensory processing and reactivity for infants. The Tactile Deep Pressure, Visual-Tactile Integration, and Vestibular Stimulation subscales measure infants’ responses to being rubbed on the arm, stomach, foot, and mouth by the examiner, having objects with different sensory properties (e.g., a furry mitt, a sticky piece of tape, a rubber ball) placed on different parts of their bodies, and being lifted and turned through different axes. For the purposes of this study, these three subtests were scored on a hyperresponsiveness scale and a hyporesponsiveness scale. Behaviors scored on the hyperresponsiveness scale included various adverse responses to stimuli (e.g., withdrawing from, pushing away from, or kicking away stimuli); behaviors on the hyporesponsiveness scale were based on lack of reaction to stimuli (e.g., not orienting or looking to stimuli, displaying neutral affect throughout stimulation). Higher scores indicated greater presence of these patterns of sensory responsiveness. The coding manual for this measure is available upon request from the corresponding author.
As with the SPA, eleven samples were randomly selected for coding by a second naïve coder. Interrater reliability for variables derived from this measure was high (intraclass correlation coefficients [ICC] = 0.986 and 0.800 for mean hyperresponsiveness and hyporesponsiveness, respectively; see Feldman et al., 2021 for more detailed information).
Creation of Sensory Responsiveness Aggregates
Indices from the SEQ and SP purported to tap sensory seeking, hyporesponsiveness, and hyperresponsiveness were highly intercorrelated (r values indexing covariation between component variables ≥ 0.4; see Table S1), allowing us to create caregiver report aggregates following z-score transformation (Rushton et al., 1983; see Feldman et al., 2021 for more complete information on intercorrelations). Higher scores indicated greater presence of these patterns of sensory responsiveness. However, indices from observational measures were not sufficiently correlated with one another or with indices derived from the aforementioned caregiver report measures to be aggregated and, thus, were analyzed separately.
Measures of Receptive and Expressive Communication
The following communication measures were collected at study entry and at Time 2 (21–27 months; the outcome measurement period that is the focus of the present study).
Mullen Scales of Early Learning (MSEL)
The MSEL (Mullen, 1995) is a standardized test that assesses development in several domains, including expressive and receptive language, for children birth-68 months. From this measure, we characterized participants by calculating the early learning composite standard score and mental age (i.e., average of age equivalency scores across Fine Motor, Visual Reception, Receptive Language, and Expressive Language scales). The age equivalency scores from the receptive and expressive language scales were derived for use in analyses.
MacArthur-Bates Communicative Development Inventories (MCDI)
The MCDI (Fenson et al., 2007) is a caregiver report measure that assesses early vocabulary and broader spoken language ability. We utilized the MCDI: Words and Sentences version to calculate raw scores for the number of words each child understands and says (i.e., expressive vocabulary) across the age range of interest. Note that the Words and Sentences version of this measure does not yield a receptive vocabulary score.
Vineland Adaptive Behavior Scales, Second Edition (VABS-2)
The VABS-2 (Sparrow et al., 2005) is a caregiver report measure that assesses adaptive function in several domains, including receptive and expressive communication. The age equivalency scores from the receptive and expressive communication scales were derived for use in analyses.
Creation of Communication Aggregates
All measures purported to tap expressive and receptive communication, respectively, were correlated at r ≥ 0.4 at Time 2, supporting the creation of communication aggregate scores using the corresponding component variables (following z-score transformation) from the MCDI, VABS-2, and MSEL (see Table S2).
Autism Diagnostic Assessment
All infants in both groups received a comprehensive autism diagnostic assessment by a licensed clinician at Time 2. This assessment included a research reliable administration of the ADOS-2 toddler module (Luyster et al., 2009) and a caregiver interview. Four infants in the Sibs-autism group were diagnosed at Time 2.
As a part of the larger ongoing study (see Study Design), most infants have also now been re-assessed after their third birthday (i.e., at 36–47 months) via a diagnostic battery that included an appropriate module of the ADOS-2 (Lord et al., 2012) and a clinical interview. At this later timepoint, three of the infants diagnosed with autism at Time 2 continued to meet criteria for a diagnosis, while the fourth did not complete their in-person evaluation at Time 3 due to the COVID-19 pandemic (see Missing Data); however, this infant’s caregivers reported continued features consistent with an autism diagnosis. An additional two infants in the Sibs-autism group had been diagnosed with autism at the Time 3 measurement period, for a total of six infants with a preliminary diagnosis of autism in the longitudinal dataset (i.e., in the Sibs-autism-dx group; see Table 1). No infants in the Sibs-NA group were diagnosed with autism at either timepoint.
Missing Data
Missing data were imputed using the missForest package (Stekhoven & Bühlmann, 2012) in R (R Core Team, 2022). To aid data imputation, all available communication data from longitudinal study visits (which are ongoing as part of the larger project) were included as auxiliary variables in the datasheet. Prior to imputing missing data, all variables of interest were evaluated for normality, specifically for skewness >|1.0| and kurtosis >|3.0| (see Tabachnick & Fidell, 2001). Three variables included in analyses (i.e., TSFI hyporesponsiveness, SPA sensory seeking inventory, MCDI expressive vocabulary) were corrected for positive skew with a square root transformation (Osborne, 2002). One additional variable included in analyses (i.e., SP sensory sensitivity) was corrected for negative skew with a square transformation.
Analytic Plan
To answer our first research question, zero-order correlations were evaluated for each index of sensory responsiveness with expressive communication and receptive communication aggregates. Then, to assess whether sibling group moderated associations between sensory responsiveness and communication, regression models were run with expressive and receptive communication aggregates as the dependent variables and the index of sensory responsiveness of interest, group, and sensory index*group interaction term as the independent variables. This method was also used to assess whether the associations of interest were moderated by age and preliminary diagnostic outcome (i.e., Sibs-autism-dx versus Sibs-autism-nodx and Sibs-NA).
Interaction effects were probed at p ≤ .1 in PROCESS (Hayes, 2017). This slightly lower threshold for Type I errors was used in testing interaction effects to decrease our risk of making Type II errors, as interaction effects are often difficult to detect with small sample sizes (Aiken & West, 1991; Fairchild & MacKinnon, 2009). For this reason, an alpha level < 0.1 threshold is employed for flagging significant interaction effects by default in statistics programs (e.g., R Core Team, 2022) and is commonly utilized in testing moderated effects in the autism literature (e.g., Feldman et al., 2021; Sandbank et al., 2020). For our only continuous putative moderator (chronological age at Time 1), we also utilized Johnson-Neyman tests to interpret significant moderation models.
Results
Unconditional Links with Communication
See Table 3 for zero-order correlations between early sensory responsiveness and later communication. Although most of the correlations trended in the anticipated direction, only three were significant. Time 2 receptive communication was significantly predicted by caregiver reported hyporesponsiveness (r = − .47, p = .002; see Fig. 1A) and caregiver reported hyperresponsiveness (r = − .33, p = .036; see Fig. 1B). Additionally, TSFI hyperresponsiveness significantly predicted Time 2 expressive communication (r = − .46, p = .003; see Fig. 1C).Table 3 Zero-order associations between indices of sensory responsiveness and later communication
Time 1
Sensory Index Time 2
Expressive communication Time 2
Receptive communication
Caregiver reported seeking − 0.16 − 0.16
Caregiver reported hypo − 0.15 − 0.47**
Caregiver reported hyper 0.13 − 0.33*
SPA seeking − 0.02 0.01
SPA orienting (hypo) − 0.26 − 0.26
SPA avoidance (hyper) − 0.08 − 0.04
TSFI hypo 0.00 0.07
TSFI hyper − 0.46** − 0.30
Time 1 = Initial visit (i.e., 12–18 months), Time 2 = 9 months later (i.e., 21–27 months), Hypo = Hyporesponsiveness, Hyper = Hyperresponsiveness, SPA = Sensory Processing Assessment (Baranek, 1999b), TSFI = Test of Sensory Functions in Infants (DeGangi & Greenspan, 1989)
Bolded values are statistically significant, *p < .05, **p < .01
Fig. 1 Selected unconditional relations between sensory responsiveness and later communication. Note Figures depict unconditional associations between patterns of early sensory responsiveness at Time 1 (i.e., at 12–18 months) and communication at Time 2 (9 months later; i.e., at 21–27 months). Time 2 receptive communication was significantly predicted by A caregiver reported hyporesponsiveness (r = − .47, p = .002) and B caregiver reported hyperesponsiveness (r = −.33, p = .036). C Time 2 expressive communication was significantly predicted by hyperresponsiveness scores from the Test of Sensory Functions in Infants (TSFI; DeGangi & Greenspan, 1989; r = − .46, p = .003). Sibs-autism-dx (orange triangles) = infants with at least one autistic older sibling who did receive an autism diagnosis, Sibs-autism-nodx (light grey circles) = infants with at least one autistic older sibling who did not receive an autism diagnosis, Sibs-NA (dark grey circles) = infants with only non-autistic older siblings
Links with Communication as Moderated by Sibling Group
Multiple regression analyses indicated that one relation between indices of sensory responsiveness and later expressive communication was moderated by sibling group (see Table 4 for a summary of models testing associations according to sibling group and Tables S3-S4 for full multiple regression output). Specifically, the relation between caregiver reported sensory seeking and expressive communication at Time 2 varied according to sibling group (βinteraction = − 1.05, p = .020, Cohens’ f2 = 0.12; see Fig. 2). Caregiver reported sensory seeking significantly predicted expressive communication at Time 2 in the Sibs-NA (β for Sibs-NA = –0.53, p = .016) but not in the Sibs-autism (β for Sibs-autism = 0.15, p = .52).Table 4 Relations between sensory responsiveness and later communication as moderated by sibling group, preliminary diagnostic outcome, and age
Time 1 Sensory Index Moderation by sibling group Moderation by outcome Moderation by age
Expressive
βint (f2) Receptive
βint (f2) Expressive
βint (f2) Receptive
βint (f2) Expressive
βint (f2) Receptive
βint (f2)
Caregiver reported seeking − 1.05* (0.12) − 0.47 (0.02) − 0.12 (0.00) 0.07 (0.00) 0.14 (0.00) − 0.48 (0.00)
Caregiver reported hypo − 0.31 (0.01) − 0.37 (0.01) − 0.30 (0.01) − 0.50 (0.02) 0.15 (0.00) 0.40 (0.00)
Caregiver reported hyper 0.61 (0.04) 0.61 (0.04) − 0.54 (0.02) − 0.78 (0.03) 1.43 (0.04) 1.48 (0.05)
SPA seeking 0.25 (0.02) 0.01 (0.00) − 0.61 (0.02) − 0.31 (0.00) 0.42 (0.00) 1.61 (0.06)
SPA orienting (Hypo) 0.38 (0.00) 0.07 (0.00) − 0.13 (0.00) − 0.28 (0.00) 1.06 (0.02) 0.48 (0.00)
SPA avoidance (Hyper) − 0.56 (0.03) − 0.04 (0.00) 0.22 (0.00) 0.10 (0.00) − 1.63† (0.06) − 0.17 (0.00)
TSFI hypo 0.09 (0.00) − 0.35 (0.01) − 0.13 (0.00) − 0.09 (0.00) − 1.63 (0.04) 0.30 (0.00)
TSFI hyper − 0.42 (0.01) 0.00 (0.00) − 0.24 (0.00) 0.01 (0.00) − 1.46 (0.04) − 0.69 (0.01)
Time 1 = Initial visit (i.e., 12–18 months), Expressive = expressive communication composite scores at Time 2 (9 months later; i.e., 21–27 months), Receptive = receptive communication composite scores at Time 2, βint = Standardized regression coefficient for the sensory*putative moderator interaction term in the multiple regression model predicting the communication index of interest (i.e., sensory*sibling group for Sibling Group, sensory*autism diagnosis or not for Outcome, sensory*age for Age), f2 = Cohen’s f2, Hypo = Hyporesponsiveness, Hyper = Hyperresponsiveness, SPA = Sensory Processing Assessment (Baranek, 1999b), TSFI = Test of Sensory Functions in Infants (DeGangi & Greenspan, 1989)
Bolded values are statistically significant, †p < .1, *p < .05
Fig. 2 The relation between caregiver reported sensory seeking and later expressive communication is moderated by sibling group. Note The relation between caregiver reported sensory seeking at Time 1 (i.e., at 12–18 months) and expressive communication at Time 2 (9 months later; i.e., at 21–27 months) was moderated by sibling group (β for the sibling group × sensory seeking term in the multiple regression model = − 1.05, p = .020, Cohen’s f2 = 0.12). The relation was significant and in the anticipated direction for the infant siblings of non-autistic children (Sibs-NA; dark grey dots and solid line; β = − 0.53, p = .016) but not in the infant siblings of autistic children (Sibs-autism; dotted line, β = 0.15, p = .52). In this figure, Sibs-autism who did and did not go on to receive an autism diagnosis are denoted by orange triangles (Sibs-autism-dx) and light grey circles (Sibs-autism-nodx), respectively
Links with Communication as Moderated by Preliminary Diagnostic Outcome
Multiple regression analyses indicated that none of the relations between indices of sensory responsiveness and communication at Time 2 were moderated by preliminary diagnostic outcome status (i.e., Sibs-autism-dx versus Sibs-autism-nodx and Sibs-NA; see Table 4 for a summary and Tables S5–S6 for full output).
Links with Communication as Moderated by Age
Only one of the associations (i.e., the relation between SPA avoidance and expressive communication at Time 2) was moderated by chronological age at Time 1, βinteraction = − 1.63, p = .091, Cohens’ f2 = 0.06 (see Tables 4, S7–S8). However, there was no significant result for the Johnson-Neyman test for the moderated relation, so we cannot interpret this significant result.
Post-hoc Analyses
A series of post-hoc analyses was run to evaluate whether our significant findings were robust to controlling for (a) chronological age and (b) communication at study entry. All significant zero-order associations (i.e., TSFI hyperresponsiveness and expressive communication, caregiver reported hyporesponsiveness and receptive communication, caregiver reported hyperresponsiveness and receptive communication) remained statistically significant when controlling for age. These relations were also robust to controlling for communication at study entry.
None of the significant moderated relations were robust to controlling for communication at study entry. However, when controlling for age, there was a persistent trend towards a significant moderated relation between caregiver reported sensory seeking and expressive communication according to sibling group (p for the interaction term = .108). Notably, we have limited power to detect interaction effects when including covariates in regression models in the present sample.
Discussion
This study sought to expand our previous work on concurrent associations between early patterns of sensory responsiveness and later receptive and expressive communication by evaluating longitudinal links between these domains in the same sample of infants at increased and general population-level familial likelihood for a later diagnosis of autism (Sibs-autism and Sibs-NA). Findings suggest that indices of sensory responsiveness derived during the 12–18 month developmental period predict later communication in different ways.
Three sensory indices were significant predictors of subsequent communication outcomes, unconditionally (i.e., not covarying for or conditional upon any other variables, such as putative moderators), across both groups of infants. Specifically, increased caregiver reported hyporesponsiveness and hyperresponsiveness significantly predicted lower receptive communication scores at Time 2 (i.e., 9 months later), while increased hyperresponsiveness as measured via the TSFI significantly predicted lower expressive communication scores at Time 2. These relations held when controlling for several additional factors, including chronological age and entry-level communication, suggesting that selected indices of sensory responsiveness are promising as value-added predictors of communication across Sibs-autism and Sibs-NA (Yoder et al., 2015).
The aforementioned findings, on the whole, are consistent with the cascading effects theory (Cascio et al., 2016). The present results provide increased empirical support for the notion that hyporesponsiveness is associated with poor communication outcomes across autistic children and infants at elevated likelihood for the condition (e.g., Baranek et al., 2013; Feldman et al., 2020; Grzadzinski et al., 2021), likely because this particular pattern of sensory responsiveness manifests via reduced orienting towards, and likely subsequently engaging with and learning from, one’s environment. In contrast, this is the first study to our knowledge to find unconditional, longitudinal links between early hyperresponsiveness and later communication. It is possible that early hyperresponsiveness to sensory stimuli may cause infants to become hyperselective or hyperrestrictive to certain sensory aspects of social experiences, contributing to social withdrawal and disrupting communication development. Alternatively, hyperresponsiveness may cause infants to become distracted by irrelevant information in social settings, leading to interference with opportunities for engagement and thereby contributing to communication delay and disorder.
These results also accord with some of the cross-sectional work previously conducted on Sibs-autism (Feldman et al., 2021; Wolff et al., 2019), though most longitudinal studies to date on infants at an increased likelihood for later autism diagnosis, notably, have not assessed all patterns of sensory responsiveness (i.e., Baranek et al., 2018; Damiano-Goodwin et al., 2018; Grzadzinski et al., 2021; Narvekar et al., 2022). Additionally, most studies on sensory responsiveness in infants have only utilized one type of measure (i.e., caregiver report or observational) or have not attempted to aggregate caregiver report and observational measures.
Though there is limited evidence for associations between observational and caregiver-report measures of the same patterns of sensory responsiveness in prior work (i.e., Boyd et al., 2010), it is notable that the caregiver-report and observational measures purported to tap the same patterns of sensory responsiveness were not sufficiently intercorrelated in this sample to warrant aggregation. We suspect that the lack of strong covariation may be attributable to the fact that the various measures differ in the extent to which they tap patterns of sensory responsiveness within specific sensory modalities. The caregiver reports index infants’ behavioral responses to sensory stimuli from a broad range of modalities (i.e., auditory, visual, tactile, gustatory, olfactory, vestibular, proprioceptive, multisensory) encountered in everyday settings. In contrast, variables derived from the SPA index infants’ observed responses to sensory stimuli in primarily auditory, visual, and tactile modalities, and variables derived from the TSFI index infants’ observed responses to predominantly visual, tactile, and vestibular stimuli. Emerging work suggests that considerable variance is accounted for by modality-specific response patterns versus supra-modal response patterns (Ausderau et al., 2014; Tillmann et al., 2020). Given our differential findings for theorized links with later communication, we recommend that future studies employ multiple measures with at least some prior psychometric support for tapping all patterns of sensory responsiveness across the broadest possible range of sensory modalities in infants and/or autistic children (e.g., Eeles et al., 2013; Schaaf & Lane, 2015; Yeung & Thomacos, 2020). Doing so will further our understanding of which sensory responsiveness scores have the highest validity or clinical utility for predicting communication outcomes.
Only one of the associations tested in this study varied according to sibling group: the relation between caregiver reported sensory seeking and later expressive communication. In this case, the association was stronger in Sibs-NA relative to Sibs-autism and only in the anticipated direction for the Sibs-NA. It is unclear why these associations were not present or in the hypothesized direction in the Sibs-autism group. Notably, this interaction was not robust to controlling for other factors, such as entry-level communication and chronological age; however, the present study was somewhat underpowered to include additional covariates in regression models. Additionally, this study is already at an increased risk of making a type I error due to the number of analyses run. Thus, our findings may be best considered preliminary and beg for replication in future studies testing more complex moderation models with larger samples.
It is also possible that relations of interest may differ between Sibs-autism-dx, Sibs-autism-nodx, and Sibs-NA. Although our prospective design allowed us to follow infants over the course of their early development and to assess whether the relations between sensory differences and communication skill were moderated by preliminary diagnostic outcome, we did not find evidence for differences between infants diagnosed with autism (Sibs-autism-dx) compared to the other infants (Sibs-autism-nodx and Sibs-NA). Our relatively small number of six participants diagnosed with autism in the present sample, though consistent with published literature on diagnostic rates in Sibs-autism (Ozonoff et al., 2011), is arguably not sufficient to comprehensively assess moderation according to diagnostic outcome, thus limiting our ability to draw firm conclusions regarding the extent to which longitudinal relations of interest vary according to this factor. Additionally, because some of the Sibs-autism who were not diagnosed with autism at this time may still receive a diagnosis later in life (e.g., around their fifth or eighth birthday; Ozonoff et al., 2018), we may need to follow a larger sample of infants for a more extended period of time to more fully evaluate the degree to which these relations differ according to outcome status.
Although we did find evidence for moderated longitudinal associations between early sensory responsiveness and later communication in the present study, the predictive relations of interest were not significantly moderated by chronological age. This result is somewhat surprising, given that our prior work on Sibs-autism suggested that concurrent relations between sensory responsiveness and communication did vary according to age (Feldman et al., 2021), such that concurrent correlations trended in the anticipated (negative) direction and were stronger in older infants. In the previous report, we noted that at least some of the previously observed moderated associations may have been driven by reduced variance (i.e., a truncated range) in communication when measured concurrently, in particular at younger ages (i.e., before 13 months). Alternatively, it is still possible that, at younger ages, certain patterns of sensory responsiveness may be adaptive (or at least not maladaptive) in the short-term.
A notable limitation of this study was that our sample was largely white and non-Hispanic or Latino, which may limit our ability to generalize our results to the entire autistic population. Thus, future research recruiting samples of infants from more diverse racial and ethnic backgrounds is needed.
It is also important to note that our study does not permit conclusions regarding how sensory responsiveness as measured early in life may influence later communication. Theory and past research, however, point towards factors that may mediate the relations that we have observed. Specifically, theory suggests that early disruptions in sensory function may influence downstream developmental skills by interfering with an infant’s engagement with their environment (e.g., Baranek et al., 2018). Santapuram et al. (2022) provided some empirical support for this theorized mechanism, by demonstrating that an early multisensory milestone (looking to audiovisual speech) influenced communication via higher-order supported joint engagement in infants at both high and low likelihood for autism (see Bottema-Beutel et al., 2014, 2019). In a related study of infants identified as being at high likelihood for autism via community screening, Grzadzinski et al. (2021) found that a transactional construct (i.e., caregiver responsivity) mediated selected associations between early indices of sensory responsiveness and communication outcomes. These findings, collectively, suggest that hyporesponsiveness, hyperresponsiveness, and/or sensory seeking may interfere with typical dyadic interactions, thus decreasing the opportunities that caregivers have to engage with and respond to their child during play interactions and thereby indirectly influencing later communication. Further research, employing larger samples, longitudinal designs, and advanced analytic approaches such as moderation and mediation will provide additional insights into precisely when, for whom, and by what mechanisms early sensory responsiveness is useful for predicting communication development in infants at high and low likelihood for a future diagnosis of autism.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 28 kb)
Author Contributions
JIF, CJC, and TGW posed the research questions. JIF, VG, KD, JEM, SMB, AJG, CD, SK, NM, SR, PS, ES, AEA, AM, BKK, and TGW collected and scored the data. KLW and AVK scored the data. JIF and TGW analyzed the data. JIF, VG, and TGW drafted the manuscript and interpreted the results. All authors read and approved the final manuscript.
Funding
The work described was supported by NIH grants P50HD103537 (PI: Neul), R21 DC016144 (PI: Woynaroski), R01DC020186 (PI: Woynaroski), KL2TR000446 (PI: Woynaroski), F31DC020129 (PI: Dunham), and TL1TR002244 (PI: Hartmann); Vanderbilt Institute for Clinical and Translation Research awards VR14759 (PIs: Cascio & Woynaroski) and VR53990 (PI: Garla); NSF NRT grant DGE 19-22697 (PI: Wallace); COVID Recovery grants from Vanderbilt Kennedy Center (Nicholas Hobbs Award; PI: Woynaroski) and Vanderbilt Edge for Scholars (PI: Woynaroski); and the Nancy Lurie Marks Family Foundation (PI: Schaaf). We would like to thank the Adventure Science Center in Nashville, TN for their assistance in participant recruitment.
Declarations
Conflict of interest
The authors declare no competing interests.
Ethical Approval
All procedures were performed in accordance with the ethical standards of the Vanderbilt University Institutional Review Board and with the 1964 Declaration of Helsinki and its later amendments.
Informed Consent
All participants or their parents provided written informed consent. All minor participants provided written assent prior to their participation.
1 We have opted to use identity-first language (e.g., autistic individuals) in this report to align with the preferences of the community and current recommendations for researchers (Bottema-Beutel et al., 2021; Bury et al., 2020).
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Aiken LS West SG Multiple regression: Testing and interpreting interactions 1991 Sage Publications
American Psychiatric Association Diagnostic and statistical manual of mental disorders-5 2013 American Psychiatric Association
Ausderau K Sideris J Furlong M Little LM Bulluck J Baranek GT National survey of sensory features in children with ASD: Factor structure of the Sensory Experience Questionnaire (3.0) Journal of Autism and Developmental Disorders 2014 44 915 925 10.1007/s10803-013-1945-1 24097141
Bahrick LE Todd JT Stein BE Multisensory processing in autism spectrum disorders: Intersensory processing disturbance as atypical development The new handbook of multisensory processes 2012 MIT Press 657 674
Baio J Wiggins L Christensen DL Maenner MJ Daniels J Warren Z Kurzius-Spencer M Zahorodny W Rosenberg CR White T Prevalence of autism spectrum disorder among children aged 8 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2014 MMWR Surveillance Summaries 2018 67 6 1 23 10.15585/mmwr.ss6706a1
Baranek GT Autism during infancy: A retrospective video analysis of sensory-motor and social behaviors at 9–12 months of age Journal of Autism and Developmental Disorders 1999 29 213 224 10.1023/A:1023080005650 10425584
Baranek, G. T. (1999b). Sensory processing assessment for young children (SPA) [Unpublished manuscript]. University of North Carolina at Chapel Hill.
Baranek GT David FJ Poe MD Stone WL Watson LR Sensory Experiences Questionnaire: Discriminating sensory features in young children with autism, developmental delays, and typical development Journal of Child Psychology and Psychiatry 2006 47 591 601 10.1111/j.1469-7610.2005.01546.x 16712636
Baranek GT Watson LR Boyd BA Poe MD David FJ McGuire L Hyporesponsiveness to social and nonsocial sensory stimuli in children with autism, children with developmental delays, and typically developing children Development and Psychopathology 2013 25 307 320 10.1017/S0954579412001071 23627946
Baranek GT Woynaroski TG Nowell S Turner-Brown L DuBay M Crais ER Watson LR Cascading effects of attention disengagement and sensory seeking on social symptoms in a community sample of infants at-risk for a future diagnosis of autism spectrum disorder Developmental Cognitive Neuroscience 2018 29 30 40 10.1016/j.dcn.2017.08.006 28869201
Ben-Sasson A Gal E Fluss R Katz-Zetler N Cermak SA Update of a meta-analysis of sensory symptoms in ASD: A new decade of research Journal of Autism and Developmental Disorders 2019 49 12 4974 4996 10.1007/s10803-019-04180-0 31501953
Ben-Sasson A Hen L Fluss R Cermak SA Engel-Yeger B Gal E A meta-analysis of sensory modulation symptoms in individuals with autism spectrum disorders Journal of Autism and Developmental Disorders 2009 39 1 11 10.1007/s10803-008-0593-3 18512135
Billstedt E Gillberg IC Gillberg C Autism in adults: Symptom patterns and early childhood predictors. Use of the DISCO in a community sample followed from childhood Journal of Child Psychology and Psychiatry 2007 48 1102 1110 10.1111/j.1469-7610.2007.01774.x 17995486
Bottema-Beutel K Kapp SK Lester JN Sasson NJ Hand BN Avoiding ableist language: Suggestions for autism researchers Autism in Adulthood 2021 3 1 18 29 10.1089/aut.2020.0014
Bottema-Beutel K Kim SY Crowley S Augustine A Kecili-Kaysili B Feldman JI Woynaroski TG The stability of joint engagement states in infant siblings of children with and without ASD: Implications for measurement practices Autism Research 2019 12 495 504 10.1002/aur.2068 30618181
Bottema-Beutel K Yoder PJ Hochman JM Watson LR The role of supported joint engagement and parent utterances in language and social communication development in children with autism spectrum disorder Journal of Autism and Developmental Disorders 2014 44 9 2162 2174 10.1007/s10803-014-2092-z 24658867
Boyd BA Baranek GT Sideris J Poe MD Watson LR Patten E Miller H Sensory features and repetitive behaviors in children with autism and developmental delays Autism Research 2010 3 78 87 10.1002/aur.124 20437603
Bradshaw J Schwichtenberg AJ Iverson JM Capturing the complexity of autism: Applying a developmental cascades framework Child Development Perspectives 2022 16 1 18 26 10.1111/cdep.12439 36407945
Bury SM Jellett R Spoor JR Hedley D “It defines who I am” or “It’s something I have”: What language do [autistic] Australian adults [on the autism spectrum] prefer? Journal of Autism and Developmental Disorders, Advance Online Publication 2020 10.1007/s10803-020-04425-3
Cascio CJ Woynaroski T Baranek GT Wallace MT Toward an interdisciplinary approach to understanding sensory function in autism spectrum disorder Autism Research 2016 9 920 925 10.1002/aur.1612 27090878
Charman T Young GS Brian J Carter A Carver LJ Chawarska K Curtin S Dobkins K Elsabbagh M Georgiades S Hertz-Picciotto I Hutman T Iverson JM Jones EJ Landa R Macari S Messinger DS Nelson CA Ozonoff S Saulnier C Stone WL Tager-Flusberg H Webb SJ Yirmiya N Zwaigenbaum L Non-ASD outcomes at 36 months in siblings at familial risk for autism spectrum disorder (ASD): A baby siblings research consortium (BSRC) study Autism Research 2017 10 1 169 178 10.1002/aur.1669 27417857
Cronbach LJ Meehl PE Construct validity in psychological tests Psychological Bulletin 1955 52 4 281 302 10.1037/h0040957 13245896
Damiano-Goodwin CR Woynaroski TG Simon DM Ibañez LV Murias M Kirby A Newsom CR Wallace MT Stone WL Cascio CJ Developmental sequelae and neurophysiologic substrates of sensory seeking in infant siblings of children with autism spectrum disorder Developmental Cognitive Neuroscience 2018 29 41 53 10.1016/j.dcn.2017.08.005 28889988
DeGangi GA Greenspan SI Test of Sensory Functions in Infants (TSFI) 1989 Western Psychological Services
Dunn W The sensory profile: User’s manual 1999 Psychological Corporation
Eeles AL Spittle AJ Anderson PJ Brown N Lee KJ Boyd RN Doyle LW Assessments of sensory processing in infants: A systematic review Developmental Medicine & Child Neurology 2013 55 4 314 326 10.1111/j.1469-8749.2012.04434.x 23157488
Eisenberg L The autistic child in adolescence American Journal of Psychiatry 1956 112 8 607 612 10.1176/ajp.112.8.607 13292547
Fairchild AJ MacKinnon DP A general model for testing mediation and moderation effects Prevention Science 2009 10 2 87 99 10.1007/s11121-008-0109-6 19003535
Feldman JI Cassidy M Liu Y Kirby AV Wallace MT Woynaroski TG Relations between sensory responsiveness and features of autism in children Brain Sciences 2020 10 11 775 10.3390/brainsci10110775 33114357
Feldman JI Raj S Santapuram P Golden AJ Daly C Dunham K Suzman E Augustine AE Garla V Muhumuza A Cascio CJ Williams KL Kirby AV Keceli-Kaysili B Woynaroski TG Sensory responsiveness is linked with communication in infants siblings of children with and without autism Journal of Speech, Language, and Hearing Research 2021 64 6 1964 1976 10.1044/2021_JSLHR-20-00196 34003699
Fenson L Marchman VA Thal DJ Dale PS Reznick JS MacArthur-Bates communicative development inventories: User’s guide and technical manual 2007 Brookes
Foss-Feig JH Heacock JL Cascio CJ Tactile responsiveness patterns and their association with core features in autism spectrum disorders Research in Autism Spectrum Disorders 2012 6 337 344 10.1016/j.rasd.2011.06.007 22059092
Freuler A Baranek GT Watson LR Boyd BA Bulluck JC Precursors and trajectories of sensory features: Qualitative analysis of infant home videos The American Journal of Occupational Therapy 2012 66 5 e81 e84 10.5014/ajot.2012.004465 22917133
Gillberg C Steffenburg S Outcome and prognostic factors in infantile autism and similar conditions: A population-based study of 46 cases followed through puberty Journal of Autism and Developmental Disorders 1987 17 2 273 287 10.1007/BF01495061 3610999
Grzadzinski R Nowell SW Crais ER Baranek GT Turner-Brown L Watson LR Parent responsiveness mediates the association between hyporeactivity at age 1 year and communication at age 2 years in children at elevated likelihood of ASD Autism Research 2021 14 9 2027 2037 10.1002/aur.2557 34128355
Hayes AF Introduction to mediation, moderation, and conditional process analysis: A regression-based approach 2017 2 Guilford Press
Horder J Wilson CE Mendez MA Murphy DG Autistic traits and abnormal sensory experiences in adults Journal of Autism and Developmental Disorders 2014 44 1461 1469 10.1007/s10803-013-2012-7 24305777
Kirby AV Bilder DA Wiggins LD Hughes MM Davis J Hall-Lande JA Lee L-C McMahon WM Bakian AV Sensory features in autism: Findings from a large population-based surveillance system Autism Research 2022 10.1002/aur.2670 35040592
Kobayashi R Murata T Yoshinaga K A follow-up study of 201 children with autism in Kyushu and Yamaguchi areas, Japan Journal of Autism and Developmental Disorders 1992 22 395 411 10.1007/BF01048242 1383189
Lane AE Young RL Baker AEZ Angley MT Sensory processing subtypes in autism: Association with adaptive behavior Journal of Autism and Developmental Disorders 2010 40 112 122 10.1007/s10803-009-0840-2 19644746
Lord C Rutter M DiLavore P Risi S Gotham K Bishop SL Autism Diagnostic Observation Schedule, second edition (ADOS-2) manual (Part I): Modules 1–4 2012 Western Psychological Services
Luyster R Gotham K Guthrie W Coffing M Petrak R Pierce K Bishop S Esler A Hus V Oti R Richler J Risi S Lord C The Autism Diagnostic Observation Schedule—Toddler Module: A new module of a standardized diagnostic measure for autism spectrum disorders Journal of Autism and Developmental Disorders 2009 39 9 1305 1320 10.1007/s10803-009-0746-z 19415479
Maenner MJ Shaw KA Bakian AV Bilder DA Durkin MS Esler A Furnier SM Hallas L Hall-Lande J Hudson A Hughes MM Patrick M Pierce K Poynter JN Salinas A Shenouda J Vehorn A Warren Z Constantino JN DiRienzo M Fitzgerald RT Grzybowski A Spivey MH Pettygrove S Zahorodny W Ali A Andrews JG Baroud T Gutierrez J Hewitt A Lee LC Lopez M Mancilla KC McArthur D Schwenk YD Washington A Williams S Cogswell ME Prevalence and characteristics of autism spectrum disorder among children aged 8 years—Autism and developmental disabilities monitoring network, 11 sites, United States, 2018 MMWR Surveillance Summaries 2021 70 11 1 16 10.15585/mmwr.ss7011a1
Mullen EM Mullen scales of early learning 1995 American Guidance Service
Narvekar N Carter Leno V Pasco G Johnson MH Jones EJH Charman T A prospective study of associations between early fearfulness and perceptual sensitivity and later restricted and repetitive behaviours in infants with typical and elevated likelihood of autism Autism, Advance Online Publication 2022 10.1177/13623613211068932
Nowell SW Watson LR Crais ER Baranek GT Faldowski RA Turner-Brown L Joint attention and sensory-regulatory features at 13 and 22 months as predictors of preschool language and social-communication outcomes Journal of Speech, Language, and Hearing Research 2020 63 9 3100 3116 10.1044/2020_JSLHR-20-00036 32810416
Osborne J Notes on the use of data transformations Practical Assessment, Research, and Evaluation 2002 8 6 10.7275/4vng-5608
Ozonoff S Young GS Brian J Charman T Shephard E Solish A Zwaigenbaum L Diagnosis of autism spectrum disorder after age 5 in children evaluated longitudinally since infancy Journal of the American Academy of Child & Adolescent Psychiatry 2018 57 11 849 857 10.1016/j.jaac.2018.06.022 30392626
Ozonoff S Young GS Carter A Messinger D Yirmiya N Zwaigenbaum L Bryson S Carver L Constantino JN Dobkins K Hutman T Iverson JM Landa R Rogers S Sigman M Stone W Recurrence risk for autism spectrum disorders: A baby siblings research consortium study Pediatrics 2011 128 3 e488 e495 10.1542/peds.2010-2825 21844053
R Core Team. (2022). R: A language and environment for statistical computing. In (Version 4.2.1) Vienna, Austria. https://www.R-project.org/
Robertson CE Baron-Cohen S Sensory perception in autism Nature Reviews Neuroscience 2017 18 671 684 10.1038/nrn.2017.112 28951611
Rushton JP Brainerd CJ Pressley M Behavioral development and construct validity: The principle of aggregation Psychological Bulletin 1983 94 18 38 10.1037/0033-2909.94.1.18
Rutter M Bailey A Lord C The Social Communication Questionnaire 2003 Western Psychological Services
Sandbank M Bottema-Beutel K Crowley S Cassidy M Feldman JI Canihuante M Woynaroski TG Intervention effects on language in children with autism: A Project AIM meta-analysis Journal of Speech, Language, and Hearing Research 2020 63 5 1537 1560 10.1044/2020_JSLHR-19-00167 32384865
Santapuram P Feldman JI Bowman SM Raj S Suzman E Crowley S Kim SY Keceli-Kaysili B Bottema-Beutel K Lewkowicz DJ Wallace MT Woynaroski TG Mechanisms by which early eye gaze to the mouth during multisensory speech influences expressive communication development in infant siblings of children with and without autism Mind, Brain, and Education 2022 16 1 62 74 10.1111/mbe.12310
Schaaf RC Lane AE Toward a best-practice protocol for assessment of sensory features in ASD Journal of Autism and Developmental Disorders 2015 45 1380 1395 10.1007/s10803-014-2299-z 25374136
Schauder KB Bennetto L Toward an interdisciplinary understanding of sensory dysfunction in autism spectrum disorder: An integration of the neural and symptom literatures Frontiers in Neuroscience 2016 10 268 10.3389/fnins.2016.00268 27378838
Sparrow SS Cicchetti DV Bella DA Vineland Adaptive Behavior Scales 2005 2 Pearson
Stekhoven DJ Bühlmann P MissForest—Non-parametric missing value imputation for mixed-type data Bioinformatics 2012 28 112 118 10.1093/bioinformatics/btr597 22039212
Tabachnick B Fidell L Using multivariate statistics 2001 4 Allyn and Bacon
Tillmann J Uljarevic M Crawley D Dumas G Loth E Murphy D Buitelaar J Charman T the AIMS-2-TRIALS LEAP group Dissecting the phenotypic heterogeneity in sensory features in autism spectrum disorder: A factor mixture modelling approach Molecular Autism 2020 11 1 67 10.1186/s13229-020-00367-w 32867850
Wallace MT Woynaroski TG Stevenson RA Multisensory integration as a window into orderly and disrupted cognition and communication Annual Review of Psychology 2020 71 193 219 10.1146/annurev-psych-010419-051112 31518523
Watson LR Patten E Baranek GT Poe M Boyd BA Freuler A Lorenzi J Differential associations between sensory response patterns and language, social, and communication measures in children with autism or other developmental disabilities Journal of Speech, Language, and Hearing Research 2011 54 1562 1576 10.1044/1092-4388(2011/10-0029) 21862675
Williams KL Kirby AV Watson LR Sideris J Bulluck J Baranek GT Sensory features as predictors of adaptive behaviors: A comparative longitudinal study of children with autism spectrum disorder and other developmental disabilities Research in Developmental Disabilities 2018 81 103 112 10.1016/j.ridd.2018.07.002 30060977
Wolff JJ Dimian AF Botteron KN Dager SR Elison JT Estes AM Hazlett HC Schultz RT Zwaigenbaum L Piven J Network TIBIS A longitudinal study of parent-reported sensory responsiveness in toddlers at-risk for autism Journal of Child Psychology and Psychiatry 2019 60 3 314 324 10.1111/jcpp.12978 30350375
Woolfenden S Sarkozy V Ridley G Williams K A systematic review of the diagnostic stability of autism spectrum disorder Research in Autism Spectrum Disorders 2012 6 1 345 354 10.1016/j.rasd.2011.06.008
Yeung LHJ Thomacos N Assessments of sensory processing in infants and children with autism spectrum disorder between 0 and 12 years old: A scoping review Research in Autism Spectrum Disorders 2020 72 101517 10.1016/j.rasd.2020.101517
Yoder P Watson LR Lambert W Value-added predictors of expressive and receptive language growth in initially nonverbal preschoolers with autism spectrum disorders Journal of Autism and Developmental Disorders 2015 45 5 1254 1270 10.1007/s10803-014-2286-4 25344152
| 36441431 | PMC9707174 | NO-CC CODE | 2022-12-01 23:20:21 | no | J Autism Dev Disord. 2022 Nov 28;:1-13 | utf-8 | J Autism Dev Disord | 2,022 | 10.1007/s10803-022-05817-3 | oa_other |
==== Front
AIDS Behav
AIDS Behav
AIDS and Behavior
1090-7165
1573-3254
Springer US New York
36441409
3933
10.1007/s10461-022-03933-w
Original Paper
Global Online Interest in HIV/AIDS care Services in the time of COVID-19: A Google Trends Analysis
http://orcid.org/0000-0001-6300-0292
Ornos Eric David B. [email protected]
1
http://orcid.org/0000-0002-4535-8837
Tantengco Ourlad Alzeus G. [email protected]
1
http://orcid.org/0000-0002-5183-8239
Abad Cybele Lara R. [email protected]
[email protected]
2
1 grid.11159.3d 0000 0000 9650 2179 College of Medicine, University of the Philippines Manila, Manila, Philippines
2 grid.11159.3d 0000 0000 9650 2179 Section of Infectious Diseases, Department of Medicine, Philippine General Hospital, University of the Philippines, Manila, Philippines
28 11 2022
17
12 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The Covid-19 pandemic has compounded the challenge of HIV/AIDS elimination, creating difficulties in accessing HIV care services such as early testing and treatment. This paper characterized the global online interest in HIV care services-related search terms before and during the pandemic. Global online search interest for HIV was measured using the Google Trends™ database. Spearman’s rank-order correlation correlated country-specific characteristics and HIV prevalence data with the search volume index (SVI). We found a significant decrease in the global online search interest for HIV/AIDS care services-related search terms during the Covid-19 pandemic. The top countries with the highest online interest for “HIV/AIDS” search terms were Zambia, Eswatini, Malawi, Lesotho, and Zimbabwe. In addition, search volume indices for HIV correlated positively with HIV prevalence and negatively with GDP, GDP per capita, and the number of physicians. This result highlights that resource-poor countries with a high prevalence of HIV have a high online interest in HIV/AIDS. Therefore, there is a need to improve internet access, the quality of HIV-related health information, and online health literacy to improve health-seeking behavior, especially in areas with a high disease burden. Overall, our study shows that the infodemiologic approach through Google Trends™ can be used to assess the online interest of the public toward HIV infection and related healthcare services.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10461-022-03933-w.
Keywords
HIV
Google trend
Infodemiology
Covid-19
Internet
==== Body
pmcIntroduction
Immunodeficiency Virus (HIV) and Acquired Immune Deficiency Syndrome (AIDS) is a significant public health problem globally, with 36.3 million deaths to date and over 37 million current HIV infections [1–3]. In the post-2015 Sustainable Development Goals, the international community declared its commitment to end the HIV/AIDS epidemic by 2030 [4, 5]. However, there are several barriers to realizing this goal –first, there are disparities and fragmentation in health systems, leading to inadequate coverage of HIV health care services[6]; 2) there is insufficient access to antiretroviral therapy (ART), a key element in decreasing HIV/AIDS mortality and 3) HIV testing is limited by social factors such as stigma, discrimination, and violence [6, 7].
The Covid-19 pandemic has compounded the challenge of eliminating HIV/AIDS. It has caused almost 400 million infections and 5.75 million deaths worldwide [8]. The emergence of variants of concern (VOC), such as delta and omicron, has caused recurrent Covid-19 surges in several countries [9, 10]. The ongoing pandemic has strained healthcare systems worldwide, causing disruptions in routine health services, including those for HIV [11]. Several countries have recorded a decrease in HIV clinic visits, HIV testing, and ART initiation [12, 13]. Disruptions in these services have raised concerns regarding the possible consequences on HIV incidence and mortality [13].
People living with HIV (PLHIV) use the internet as a source of disease and health services information [14, 15]. Internet usage was shown to influence the initiation of care and treatment decisions of PLHIV [16]. Moreover, the search for health-related information increased during the pandemic [17, 18]. Google Trends analysis has been utilized to assess public awareness of different diseases and public health interventions and analyze the health-seeking behavior of patients [19–22]. This research characterized the online global search interest for HIV/AIDS care services before and during the pandemic. We hypothesized that the global pandemic would decrease online interest in HIV/AIDS care services. Additionally, we identified the possible factors associated with online interest for HIV/AIDS.
Methods
We used the Google Trends™ database to measure global online search interest for HIV (http://trends.google.com). Google Trends reports search volume index (SVI) unit to measure search trends. This unit represents the search interest relative to the highest point on the chart for the given region and time. For instance, an SVI of 100 correlates with the peak popularity of the search term. A value of 50 signifies that the term is half as popular. A value of zero means that the search term has insufficient data. Previous studies have shown that SVI can give insight into collective health trends and population health-seeking behavior (SVI) [20–23].
We accessed Google Trends™ by visiting http://trends.google.com. We chose search terms related to the disease, diagnosis, prevention, and treatment of HIV as reported in the literature [24–26]. The authors also added terms as deemed appropriate in this study. The search terms included, “HIV”, “AIDS”, “HIV/AIDS”; HIV prevention and diagnosis terms such as “HIV clinic”, “HIV doctor”, “HIV test” “HIV PrEP”, “HIV PEP”, “HIV ART”, “HIV prevention”, “condom”; and common medications for HIV such as “triumeq”, “tivicay”, “atripla”, “complera”, “genvoya”, “stribild”, “isentress”, “norvir”, “prezista”, “viread”. We limited the results to 4 years from the Search Query (January 1, 2018 - December 31, 2021). We obtained SVI and related queries from Google Trends™.
We investigated the difference in the HIV-related search terms between the pre-pandemic and pandemic period. December 31, 2019, was chosen as a delineation between two time periods since WHO was first alerted of the case on this date. We used the Mann-Whitney U test (p-value < 0.05 considered significant) to compare SVIs before and after the time point. We also determined the correlation of HIV-related search term SVIs with COVID-19 disease SVIs throughout the specified period using Spearman’s correlation (p < 0.05 considered significant).
We also determined country-specific variables that may correlate with SVIs. These variables include HIV prevalence and socio-economic factors shown in multiple studies to be correlated with HIV risks ([27–30]). The data on HIV prevalence, GDP, GDP per capita, Physician per 1000 population, and poverty indices were taken from World Bank (The World Bank, 2021). The correlation between country-specific characteristics and SVI was determined using Spearman’s rank-order correlation. A Spearman’s correlation coefficient (ρ) with a p-value of less than 0.05 was considered significant. This statistical analysis was done using GraphPad Prism version 7 (GraphPad Software, San Diego, CA).
Results
Search volume indices on HIV search terms for 2018–2021 are shown in Fig. 1. A global online interest drop for HIV search terms was observable in early 2020. The nadir in the global online interest for HIV search terms coincided with the World Health Organization (WHO) declaration of Covid-19 as a pandemic.
Fig. 1 Search volume indices for HIV related terms for the year 2018–2021. SVIs were calculated separately for each search term such that SVIs are normalized within each term
To determine if Covid-19 affected global online interest in HIV, we compared the median SVI between 2018 and 2019 (pre-pandemic) and 2020–2021 (pandemic period). The SVIs for all HIV-related search terms significantly decreased during the pandemic compared to the pre-pandemic season (Table 1). In addition, we observe the highest drop in online search interest for search terms Atripla (43.18% drop, Mann-Whitney U = 2066, p < 0.0001) and Stirbild (42.50% drop, Mann-Whitney U = 2884, p < 0.0001), which are common HIV medications. The biggest drop was shown in medications, which might indicate a decrease in online interest in HIV treatment. Moreover, we found that SVIs for several HIV and care search terms are correlated negatively with COVID-19 SVI. This negative correlation indicates high COVID-19 online interest amidst low online interest in HIV (Supplementary Table I; Supp. Figure 1).
Table 1 Comparison of search volume indices for HIV search terms in 2018–2019 (before the pandemic) vs. 2020–2021 (during the pandemic). (Mann Whitney U test)
General Search Term Classification Search Terms Median SVI Percent Change in SVI p-value
Pre-pandemic
(2018–2019) Pandemic
(2020–2021) Mann-Whitney U
General HIV/AIDS
terms
HIV 67 51 -23.88% 871 < 0.0001
AIDS 50 43.5 -13.00% 1965 < 0.0001
HIV/AIDS 57 44 -22.81% 1111 < 0.0001
HIV prevention and diagnosis HIV clinic 57.5 38.5 -33.04% 1958 < 0.0001
HIV doctor 54 43 -20.37% 3493 < 0.0001
HIV test 72 54 -25.00% 618.5 < 0.0001
HIV PrEP 70 54 -22.86% 2983 < 0.0001
HIV PEP 54 41 -24.07% 3492 < 0.0001
HIV ART 49 43.5 -11.22% 4632 0.0737
HIV prevention 52.5 39 -25.71% 2992 < 0.0001
condom 80 68 -15.00% 342 < 0.0001
HIV/AIDS treatment triumeq 26 17 -34.62% 2778 < 0.0001
tivicay 46 35 -23.91% 3491 < 0.0001
Atripla 44 25 -43.18% 2066 < 0.0001
Complera 41 31.5 -23.17% 4048 0.0017
Genvoya 54.5 31.5 -42.20% 1447 < 0.0001
Stribild 40 23 -42.50% 2884 < 0.0001
Isentress 28.5 21 -26.32% 3867 0.0004
Norvir 29 24.5 -15.52% 3717 < 0.0001
Prezista 24 15.5 -35.42% 3647 < 0.0001
viread 39 26 -33.33% 1904 < 0.0001
We also determined the Youtube SVI for search terms “HIV”, “AIDS” and “HIV/AIDS” (Supp. Figure 2). We observed a drop in global online search interest for HIV-related terms. Furthermore, Youtube SVIS for these HIV-related search terms significantly decreased during the pandemic compared to the pre-pandemic period (Supplementary Table III).
Fig. 2 Search frequencies for HIV related terms by country from 2018–2021. The color intensity represents the percentage of searches for the leading search term in a particular region. Search term popularity is relative to the total number of Google searches performed at a specific time, in a specific location
We also identified the countries where HIV search terms were most popular for 2018–2021. The countries with the highest online search interest for “HIV” search terms were Malawi, Lesotho, Zambia, Eswatini, and Zimbabwe. The top countries that used the “AIDS” search term were Eritrea, Zimbabwe, Malawi, Liberia, and Papua New Guinea. Lastly, the top countries with the highest online interest for the “HIV/AIDS” search topic were Zambia, Eswatini, Malawi, Lesotho, and Zimbabwe. The Philippines ranked 45th for “HIV” search term, 43rd in “AIDS”, and 66th “HIV/AIDS”.
We determined country-specific factors that may correlate with online interest in HIV. Our correlation analysis showed that GDP correlated negatively with search terms “HIV” (r=-0.394, p < 0.0001), “AIDS” (r=-0.396, p < 0.0001), and “HIV/AIDS” (r=-0.403, p < 0.0001). Similarly, GDP per capita also showed a negative correlation with search terms “HIV” (r=-0.14, p = 0.04), “AIDS” (r=-0.07, p = 0.28), and “HIV/AIDS” (r=-0.522, p < 0.0001). The number of physicians in the country is also negatively correlated with “HIV” (r=-0.399, p < 0.0001), “AIDS” (r=-0.392, p < 0.0001), and “HIV/AIDS” (r=-0.68, p < 0.0001). On the other hand, the poverty headcount ratio showed a positive correlation with “HIV” (r = 0.187, p = 0.030), “AIDS” (r = 0.147, p = 0.088), and HIV/AIDS(r = 0.401 p < 0.0001). Furthermore, HIV prevalence is correlated positively with the search terms “HIV” (r = 0.427, p < 0.0001), “AIDS” (r = 0.356, p < 0.0001), and “HIV/AIDS” (r = 0.601, p < 0.0001).
Discussion
This study showed a significant decrease in the global online interest in HIV/AIDS-related healthcare services during the Covid-19 pandemic. This reduction was mirrored in Youtube, another popular platform. This drop in online search interest might be due to shifting concerns from HIV to Covid-19 during the pandemic. This reduction parallels the decrease in online interest in other non-covid chronic diseases such as obesity, cancer, and rheumatic and ophthalmologic diseases [31–35]. Cancer screening online searches have significantly reduced during the pandemic [32]. Online interest in invasive surgery has also decreased [22].
The decrease in online interest is paralleled by an evident reduction in physical visits to HIV clinics during the Covid-19 pandemic. Several countries have shown reductions in HIV testing/screening performed, leading to decreased HIV diagnoses and timely initiation of treatment[11, 12]. The decline in testing has been attributed to nationwide lockdown, transportation difficulty, travel requirements, and patients’ fear of contracting covid-19 in HIV clinics [36, 37]. For example, in the Philippines, monthly numbers from the Department of Health HIV and AIDS Registry showed a decline from a median of 1092 reported cases/month in 2019 to 620 cases/month in 2020 (Supplementary Table VI) [38–40]. A decrease in HIV diagnosis might be due to an actual reduction in incidence because of a decline in risky behaviors. Risky sexual behaviors, including multiple sexual partners and condomless sex, decreased during the period with heightened pandemic restrictions [41, 42]. However, there is concern that many HIV diagnoses are missed or delayed because of Covid-19 associated lockdowns [43]. In addition, there is evidence of reduced access to and utilization of STD and HIV services despite continued risky sexual behavior [44]. Moreover, a Japanese survey showed that HIV cases with AIDS diagnosis increased during the pandemic when seasonality effects were removed [45].
Our results also showed decreased online search interest for commonly used HIV medications. Internet health information-seeking behavior was previously shown to be associated with adherence to ART [46]. Thus, the observed reduction in online search interest for HIV medications may reflect the decrease in adherence to ART in different countries. This was observed in several countries, which showed declines in ART initiation during the pandemic [11]. In addition, a decrease in adherence to ART in patients living with HIV has also been noted [47]. Early initiation of and adherence to ART is known to decrease HIV mortality; thus, decreases in ART treatment are alarming because this jeopardizes the health status of patients living with HIV [6].
There is an over-representation of low to middle-income (LMIC) nations in the top countries for HIV/AIDS-related search terms. Furthermore, HIV SVIs correlate positively with HIV prevalence. SVIs and HIV prevalence correlate negatively with GDP, GDP per capita, and the number of physicians per country (Table 2; Supplementary Table IV). This negative correlation was also reflected in the geographical distribution of online interest in HIV search terms. These terms were most popular among LMIC countries. These results can be explained by the relatively high burden of HIV in these countries [2, 3]. The correlation with physicians per country also highlights people’s desire for HIV information in countries with limited resources, especially regarding health care services.
Table 2 Correlations between country-specific socioeconomic indicators and search volume indices for HIV search terms. (Spearman’s correlation)
Country specific characteristics Search Terms Spearman r p-value
HIV prevalence “HIV” 0.4273251 < 0.0001
“AIDS” 0.355592746 < 0.0001
“HIV/AIDS” 0.600515185 < 0.0001
GDP “HIV” -0.393784924 < 0.0001
“AIDS” -0.395379888 < 0.0001
“HIV/AIDS” -0.403432255 < 0.0001
GDP per capita “HIV” -0.147535477 0.041
“AIDS” -0.078655427 0.279
“HIV/AIDS” -0.522355369 < 0.0001
Physicians per 1000 population “HIV” -0.398875982 < 0.0001
“AIDS” -0.392127563 < 0.0001
“HIV/AIDS” -0.681560063 < 0.0001
Poverty headcount ratio at $1.90 a day (2011 PPP) (% of the population) “HIV” 0.186838184 0.030
“AIDS” 0.147188438 0.088
“HIV/AIDS” 0.400725982 < 0.0001
Multidimensional poverty headcount ratio (% of the total population) “HIV” -0.045179434 0.734
“AIDS” -0.063932815 0.637
“HIV/AIDS” 0.255086366 0.051
Unemployment, total (% of the total labor force) (modeled ILO estimate) “HIV” 0.016358631 0.833
“AIDS” 0.013221514 0.865
“HIV/AIDS” -0.004636209 0.951
Government leaders can utilize the findings of our study in framing policies amidst global pandemics. Drops in online interest in HIV care search terms during the COVID-19 pandemic, paralleled with a reduction in HIV clinic visits, testing, and ART initiation signifies that healthcare personnel should be proactive in engaging PLHIV in their healthcare. In addition, systems should be in place to provide necessary services such as diagnostics, medications, and treatment in the face of pandemic restrictions.
This study has several limitations. Although the most popular search tool, Google is not the sole web search engine. This online search engine is also not available in all countries. Countries without Google trends data were not included in the analysis. The data generated from Google Trends is subject to the inherent nonrepresentative sampling bias in Google search trends. This study only included data collected from people with internet access and may have inadvertently excluded lower-income groups with limited access or areas with issues with freedom of speech. Despite these limitations, we have successfully shown a decline in global online interest in HIV-related care services during the Covid-19 pandemic, which is concerning.
Conclusion
This Google Trends analysis demonstrated a significant drop in online interest in HIV care and services during the Covid-19 pandemic. This decrease in online interest mirrored the observed reduction in HIV clinic visits, testing, and ART initiation. Thus, healthcare personnel should be proactive in engaging PLHIV in their healthcare. Systems should be in place to provide necessary services such as medications and treatment in the face of pandemic restrictions. In addition, we showed that SVI for HIV-related search terms was positively correlated with HIV prevalence and negatively correlated with GDP, GDP per capita, and physicians per 1000 population. These findings highlight the high burden of disease of HIV in LMIC and the desire for HIV-related information among their citizens. Therefore, there is a need to improve access to the internet and the quality of HIV-related health information online due to its importance in HIV patient care. Furthermore, online health literacy and health-seeking behavior, especially in areas with a high disease burden, are needed to realize the goal of HIV/AIDS elimination. Overall, we showed the utility of Google Trends analysis in assessing the dynamics of online interest in HIV care services before and during the Covid-19 pandemic.
Electronic Supplementary Material
Below is the link to the electronic supplementary material.
Supplementary Material 1
Author Contributions
EDBO and OAGT contributed to conception and design. Data collection and analysis were performed by EDBO and OAGT. First draft of the manuscript was written by EDBO and all authors commented on previous version of the manuscript. All authors read and approved the final manuscript.
Funding
The study did not receive any funding.
Data Availability
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material.
Code Availability
Not Applicable.
Declarations
Competing Interests
The research was conducted in the absence of any commercial or financial relationships that could be construed as a potential competing interest.
Ethics Approval
No Ethics Approval are required since all data are publicly available.
Consent to Participate
Not Applicable.
Consent for publication
Not Applicable.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Govender RD Hashim MJ Khan MA Mustafa H Khan G Global epidemiology of HIV/AIDS: a resurgence in North America and Europe J Epidemiol Glob Health NLM (Medline) 2021 11 3 296 301 10.2991/jegh.k.210621.001
2. World Health Organization. Global Health Observatory HIV AIDS. 2021.
3. United Nations. Global HIV & AIDS statistics - Fact Sheet. 2022.
4. 70/1. Transforming our world: the 2030 Agenda for Sustainable Development Transforming our world: the 2030 Agenda for Sustainable Development Preamble. 2015.
5. THE 17 GOALS. | Sustainable Development [Internet]. [cited 2022 September 8]. Available from: https://sdgs.un.org/goals.
6. WHO. 2016 Global Health Sector Strategy on HIV.
7. Gyamerah AO, Taylor KD, Atuahene K, et al. Stigma, discrimination, violence, and HIV testing among men who have sex with men in four major cities in Ghana. AIDS Care - Psychological and Socio-Medical Aspects of AIDS/HIV. Routledge; 2020; 32(8):pp. 1036–44.
8. World Heatlh Organization. WHO Coronavirus (COVID-19) Dashboard. 2022.
9. Dyer O Covid-19: South Africa’s surge in cases deepens alarm over omicron variant BMJ NLM (Medline) 2021 375 n3013 10.1136/bmj.n3013
10. Mohapatra RK, Tiwari R, Sarangi AK, et al. Twin combination of Omicron and Delta variant triggering a Tsunami wave of ever high surges in COVID-19 cases: a challenging global threat with a special focus on Indian sub‐continent. J Med Virol. Wiley; 2022.
11. Beusenberg M. Disruption in HIV, Hepatitis and STI services due to COVID-19.
12. Dorward J, Khubone T, Gate K, et al. The impact of the COVID-19 lockdown on HIV care in 65 south african primary care clinics: an interrupted time series analysis. Lancet HIV. Elsevier Ltd; 2021; 8(3):pp. e158–65.
13. Simões D, Stengaard AR, Combs L, Raben D. Impact of the COVID-19 pandemic on testing services for HIV, viral hepatitis and sexually transmitted infections in the WHO european region, March to August 2020. Eurosurveillance. European Centre for Disease Prevention and Control (ECDC); 2020.
14. Threats M, Bond K. Hiv information acquisition and use among young black men who have sex with men who use the internet: Mixed methods study. J Med Internet Res. JMIR Publications Inc.; 2021; 23(5).
15. Passaro RC, Haley CA, Sanchez H, Vermund SH, Kipp AM. High HIV prevalence and the internet as a source of HIV-related service information at a community-based organization in Peru: a cross-sectional study of men who have sex with men. BMC Public Health. BioMed Central; 2016; 16(1).
16. Perazzo J, Haas S, Webel A, Voss J. Role of the internet in Care initiation by people living with HIV. Res Nurs Health. John Wiley and Sons Inc.; 2017; 40(1):pp. 43–50.
17. Du H, Yang J, King RB, Yang L, Chi P. COVID-19 increases online searches for emotional and health-related terms. Appl Psychol Health well being. Wiley-Blackwell; 2020; 12(4):pp. 1039–53.
18. Starcevic V Schimmenti A Billieux J Berle D Cyberchondria in the time of the COVID-19 pandemic Hum Behav Emerg Technol John Wiley and Sons Inc 2021 3 1 53 62 10.1002/hbe2.233
19. Savelkoel J, Claushuis TAM, Engelen TSR van, Scheres LJJ, Wiersinga WJ. Global impact of World Sepsis Day on digital awareness of sepsis: an evaluation using Google Trends. Crit Care. BioMed Central Ltd.; 2018.
20. Schootman M Toor A Cavazos-Rehg P The utility of Google Trends data to examine interest in cancer screening BMJ Open 2015 5 6 e006678 10.1136/bmjopen-2014-006678 26056120
21. Berlinberg EJ, Deiner MS, Porco TC, Acharya NR. Monitoring interest in herpes zoster vaccination: analysis of google search data. JMIR Public Health Surveill. 2018; 4(5).
22. Krishnan N, Anand S, Sandlas G. Evaluating the impact of COVID-19 pandemic on Public Interest in minimally invasive surgery: an Infodemiology Study using Google Trends. Cureus. 2021; 13(10).
23. Tantengco OAG. Increased global online interest in diabetes during the COVID-19 pandemic: An infodemiology study. Obes Med. Elsevier Ltd; 2021; 28.
24. Mahroum N, Bragazzi NL, Brigo F, et al. Capturing public interest toward new tools for controlling human immunodeficiency virus (HIV) infection exploiting data from Google Trends. Health Informatics J [Internet]. SAGE Publications Ltd; 2019 [cited 2022 September 1]; 25(4):1383–1397. Available from: https://journals.sagepub.com/doi/full/10.1177/1460458218766573.
25. Chiu APY, Lin Q, He D. News trends and web search query of HIV/AIDS in Hong Kong. PLoS One [Internet]. Public Library of Science; 2017 [cited 2022 September 1]; 12(9):e0185004. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0185004.
26. Jena AB, Karaca-Mandic P, Weaver L, Seabury SA. Predicting New Diagnoses of HIV Infection Using Internet Search Engine Data. Clinical Infectious Diseases [Internet]. Oxford Academic; 2013 [cited 2022 September 1]; 56(9):1352–1353. Available from: https://academic.oup.com/cid/article/56/9/1352/293757.
27. Mondal MNI, Shitan M. Factors affecting the HIV/AIDS epidemic: an ecological analysis of global data. Afr Health Sci [Internet]. 2013 [cited 2022 Sep 1]; 13(2):301–310. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/pmid/24235928/?tool=EBI.
28. Essig A, Kang S, Sellers R. The Relationship between HIV Infection Rates and GDP Per Capita in African Countries. Georgia Institute of Technology; 2015 [cited 2022 September 1];. Available from: https://smartech.gatech.edu/handle/1853/53303.
29. Bärnighausen T, Hosegood V, Timaeus IM, Newell ML. The socio-economic determinants of HIV incidence: evidence from a longitudinal, population-based study in rural South Africa. AIDS [Internet]. Europe PMC Funders; 2007 [cited 2022 September 1]; 21(Suppl 7):S29. Available from: /pmc/articles/PMC2847257/.
30. Dorrucci M, Regine V, Pezzotti P, Mammone A, Girardi E, Suligoi B. Demographic and socio-economic determinants of poor HIV-risk perception at first HIV diagnosis: analysis of the HIV Surveillance data, Italy 2010–2016. Ann Ist Super Sanita [Internet]. Ann Ist Super Sanita; 2020 [cited 2022 September 1]; 56(3):267–276. Available from: https://pubmed.ncbi.nlm.nih.gov/32959792/.
31. Adelhoefer S Berning P Solomon SB Decreased public pursuit of cancer-related information during the COVID-19 pandemic in the United States Cancer Causes and Control Springer Science and Business Media Deutschland GmbH 2021 32 6 577 85 10.1007/s10552-021-01409-1
32. Snyder A, Jang S, Nazari IS, et al. Google search volume trends for cancer screening terms during the COVID-19 pandemic. J Med Screen. SAGE Publications Ltd; 2021.
33. Kardeş S Kuzu AS Raiker R Pakhchanian H Karagülle M Public interest in rheumatic diseases and rheumatologist in the United States during the COVID-19 pandemic: evidence from Google Trends Rheumatol Int Springer Science and Business Media Deutschland GmbH 2021 41 2 329 34
34. Gupta R Pakhchanian H Raiker R Asahi M Raparla N Belyea D Public interest in refractive Diseases and Treatments during the COVID-19 pandemic: a Google Trends Analysis Cureus 2021 13 8 e17207 34540434
35. Tantengco OAG. Decreased global online interest in obesity from 2004 to 2021: An infodemiology study. Obes Med [Internet]. 2022; 30:100389. Available from: https://www.sciencedirect.com/science/article/pii/S245184762200001X.
36. Odinga MM, Kuria S, Muindi O, et al. HIV testing amid COVID-19: Community efforts to reach men who have sex with men in three Kenyan counties. Gates Open Res. F1000 Research Ltd; 2020.
37. Ponticiello M, Mwanga-Amumpaire J, Tushemereirwe P, Nuwagaba G, King R, Sundararajan R. “Everything is a Mess”: how COVID-19 is impacting Engagement with HIV Testing Services in Rural Southwestern Uganda. AIDS Behav. Springer; 2020. pp. 3006–9.
38. of Health Philippines D. HIV/AIDS and ART Registry of the Philippines: June 2021. 2021.
39. Antonio Albornoz RM, Roque VG, Palaypayon NS, et al. SPECIFIC POPULATION Overseas Filipino Workers (OFW) map-mark Chief, SMED HIV/AIDS & ART Registry of the Philippines (HARP) Editorial Team: M&E and Strategic Information Specialist Electronic Database Manager [Internet]. Available from: https://doh.gov.ph/.
40. Fig. 4: Number of HIV cases and advanced infections reported in the Philippines Sex and Age. 1984.
41. Hyndman I, Nugent D, Whitlock GG, McOwan A, Girometti N. COVID-19 restrictions and changing sexual behaviours in HIV-negative MSM at high risk of HIV infection in London, UK. Sex Transm Infect [Internet]. BMJ Publishing Group Ltd; 2021 [cited 2022 September 9]; 97(7):521–524. Available from: https://sti.bmj.com/content/97/7/521
42. Hammoud MA, Maher L, Holt M, et al. Physical Distancing Due to COVID-19 Disrupts Sexual Behaviors Among Gay and Bisexual Men in Australia: Implications for Trends in HIV and Other Sexually Transmissible Infections. J Acquir Immune Defic Syndr [Internet]. NLM (Medline); 2020 [cited 2022 September 9]; 85(3):309–315. Available from: https://journals.lww.com/jaids/Fulltext/2020/11010/Physical_Distancing_Due_to_COVID_19_Disrupts.9.aspx
43. Chow EPF, Hocking JS, Ong JJ, Phillips TR, Fairley CK. Sexually Transmitted Infection Diagnoses and Access to a Sexual Health Service before and after the National Lockdown for COVID-19 in Melbourne, Australia. Open Forum Infect Dis. Oxford University Press; 2021; 8(1).
44. Pampati S, Emrick K, Siegler AJ, Jones J. Changes in Sexual Behavior, PrEP adherence, and Access to Sexual Health Services Because of the COVID-19 Pandemic Among a Cohort of PrEP-Using MSM in the South. J Acquir Immune Defic Syndr [Internet]. NIH Public Access; 2021 [cited 2022 September 9]; 87(1):639. Available from: /pmc/articles/PMC8533029/
45. Ejima K, Koizumi Y, Yamamoto N, et al. HIV Testing by Public Health Centers and Municipalities and New HIV Cases During the COVID-19 Pandemic in Japan [Internet]. 2021. Available from: www.jaids.com
46. Samal L, Saha S, Chander G, et al. Internet Health information seeking behavior and antiretroviral adherence in persons living with HIV/AIDS. AIDS Patient Care STDS. 2011; 25(7):445–449.
47. Hochstatter KR, Akhtar WZ, Dietz S, et al. Potential Influences of the COVID-19 Pandemic on Drug Use and HIV Care Among People Living with HIV and Substance Use Disorders: Experience from a Pilot mHealth Intervention. AIDS Behav. Springer; 2021; 25(2):354–359.
| 36441409 | PMC9707175 | NO-CC CODE | 2022-12-01 23:20:21 | no | AIDS Behav. 2022 Nov 28;:1-7 | utf-8 | AIDS Behav | 2,022 | 10.1007/s10461-022-03933-w | oa_other |
==== Front
Stud Russ Econ Dev
Stud Russ Econ Dev
Studies on Russian Economic Development
1075-7007
1531-8664
Pleiades Publishing Moscow
7339
10.1134/S1075700722060120
Economic Policy
Is Economic Planning Our Future?
Jacques Sapir [email protected]
Higher School of Social Sciences and Centre for Industrial Models, Paris, France
28 11 2022
2022
33 6 583597
11 1 2022
14 6 2022
16 6 2022
© Pleiades Publishing, Ltd. 2022, ISSN 1075-7007, Studies on Russian Economic Development, 2022, Vol. 33, No. 6, pp. 583–597. © Pleiades Publishing, Ltd., 2022.Russian Text © The Author(s), 2022.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The article is devoted to the topic of economic planning at the state level. The historical excursus shows the origins of the planned economy elements on the example of three countries: Japan, India and France, as well as the evolution of its planned policy in the 20th century. The article provides a rationale for the relevance of planning, including as a tool to overcome market failures. The ESG agenda, in particular the issue of reducing the negative anthropological impact on the environment, is seen as one of the drivers for strengthening the role of planning.
Keywords:
public policy
economic development
planned economy
market failures
energy transfer
decarbonization
economic planning
indicative planning
issue-copyright-statement© Pleiades Publishing, Ltd. 2022
==== Body
pmcPlan and planning are back in fashion. That may seem surprising more than thirty years after the dissolution of Gosplan USSR (1991) or the interruption of French plans (1992). Yet the Plan was far from having completely disappeared. China continues to have a Plan, even though it has long differed from the Soviet Plan. India, which formally dissolved the planning body, the Planning Commission, in 2014, has maintained a very similar body, the NITI Aayog, whose relationships are eerily similar to the body it replaces [1].
The return of the idea of planning is now an indisputable reality. We can see this in the economic debates on the energy transition and on the consequences of the COVID-19 pandemic [2] and particularly in France where President Emmanuel Macron recreated a “High Commission for Planning” in September 2020.1
The planning we are talking about is understood in the sense of so-called “indicative” or “decentralized” planning. This indicative planning has a history; the French contribution, if not the only one, was far from being negligible. It is therefore necessary to understand which were the situations, which imposed the planning. We therefore propose, in the following lines, to study three cases of indicative planning, India, Japan and France, by observing the origin of planning institutions and their results. This leads to a discussion of the relevance of planning procedures and the reasons for their apparent lack of support in these three countries. The conditions for the return to grace of the plan will finally be discussed based on the case of France.
Why planning? The so-called “indicative” or “decentralized” planning is a mechanism that allows existing resources to be concentrated on certain objectives and activities, which do not seem profitable enough in the short term to be of interest to the market. In these situations, state intervention helps to overcome market failures, and this intervention is all the more effective if it is part of general planning. Decentralized planning mobilized some of the greatest scientists of the twentieth century. It has attracted the greatest economists and statisticians, from Wassily Leontieff to Edmond Malinvaud [3] and François Perroux, from Roy Radner to Prasanta Chandra Mahalanobis [4]. The COVID-19 pandemic will also have been the occasion of an extraordinary strengthening of the role of the State, a strengthening which could refer to exceptional powers [5], and which raises the question of a coherent organization of this role.
Several arguments are added to justify the use of forms of planning in the economies for the years to come, whether it is to ensure a hypothetical pharmaceutical sovereignty, to reindustrialize the economy, to make it less dependent on chains production facilities that have been hit hard by the COVID-19 crisis, or to ensure the energy transition to a carbon-free economy.
Planning as a response to market failures. Planning is a process, which enables to concentrate the existing means towards certain objectives and activities when these do not seem profitable enough in the short term to interest the market, but are nevertheless socially necessary. This constitutes a first type of market failure [6]. This situation is fairly typical of developing economies [7] but can be extended to developed economies in the case of a new technology presenting strong uncertainties in terms of the rate of profit, or of a strong specificity of the assets, or finally when the profit does not may be fully appropriated in one activity and diffuse widely to other activities . An interesting exаmple is given in [8, 9].
Planning is a procedure for regulating competition in a situation of absolute scarcity of resources by allowing them to be allocated to priority users, which is a second type of market failure.
Planning is also a response to radical uncertainty, whether it is dominant only in certain activities or expressed in the entire economic context. Here we are facing a third type of market failure.
This radical uncertainty arises from surprise, in the sense defined by G.L.S. Shackle, which comes along with the perfectly unexpected event [10]. This surprise therefore invalidates economic calculation results and brings out the act of political decision as central in line with bounded rationality [11] while the markets are unable to manage this type of event. In this situation, discretionary political decision-making turns out to be more effective than economic calculation. This is also the case with heuristic decision-making [12], when it is opposed to restricted economic calculation, because it becomes impossible to calculate in advance the consequences of the unforeseeable event. In this situation, Otto Neurath had shown, as early as 1909, that it led to a return to the “war” economy, what he also called an “economy in kind” [13], and in 1919–1920 that planning could facing it prove to be superior to the market.
Planning as an instrument of the return to rationality. Beyond the question of market failures, the question of uncertainty raises the problem of the shift from substantial rationality, which can only work in a world governed entirely by probability, to procedural rationality [14]. Procedural rationality, which is the form of rationality that prevails when strong informational constraints dominate [15], plays a decisive role in strategic decision-making [16].
Planning therefore makes it possible to overcome uncertainty when it is likely to paralyze the market by reconstructing, through the publication of its objectives and the means implemented to achieve them, a predictable universe for decentralized agents as stated in Le Plan ou l’Anti-Hasard written by Pierre Massé [17]. For the latter, humanity is swept along in the game of what François Perroux calls “the overlapping waves of innovation” [18]. These induce a radical uncertainty that overturns the certainties of the present. The implementation of these waves of innovations actually supposes considerable and prolonged flows of resources. This the tensions between scarce goods persist in modern developed societies; they are even perhaps and paradoxically more alive than ever as we see today on the issue of rare earths. The role of the economist, of the planner is then to foresee and reduce the most notable tensions. As we can see, planning is everything except the mechanical pursuit of the past. It can be a place of debate and meeting par excellence of private and public decisions [19]; it is in itself a decision-making process that engages the future of the nation. Here we are: planning is linked with sovereignty. If this can be conceived without planning, at the risk of getting lost, planning cannot be conceived without sovereignty.
This view of the role of planning has attracted strong criticism. The key question was how the allocation choices should be made. For von Mises, apart from the calculation of costs there were no possibility of rational choice, a point hotly disputed by Neurath who considered that any economic choice in reality involves social, moral and ethical values [20]. This gave rise to one of the most important debates of the 1920s and 1930s about the possibility of the planned economy, a debate in which Ludwig von Mises [21], von Hayek [22] or Oskar Lange [23–26] made a name for themselves.
Neurath was undoubtedly right in the debate against von Mises [21], especially when he argued that the unpredictability of the future radically limits the cognitive role of price. So in the real world von Mises was referring to, the market price of primary goods could not match their relative degree of scarcity. Under these conditions, von Mises’s argument collapses, as O’Neill has shown [27].
More generally, the various procedures that tend to associate the actors are fundamental in the logic of planning through consultation, which was the French planning tradition and was implemented from the start by the French post-war General Planning Commission. They do not simply identify local priorities but help to form a consensus between the various actors concerned. They allow a harmonization of points of view and anticipations, whether this harmonization is conscious or unconscious. This harmonization or convergence is precisely the situation that the market must, in theory, create. This point is important because, in the 1930s, F.A. Hayek had quite rightly developed this argument of the convergence or harmonization of expectations as the intrinsic force of the market [28, 29].
The plan in capitalist economy: were origins stand? The idea of planning is largely linked to the two world wars of the 20th century. War economies have made economists, but also politicians and union leaders aware, that capitalism can be in a sense “organized” and served for ends that are not those dictated by mere profit. War economies made planning a reality [30]. This stopped to be vague speculations stemming from the writings of Marx and Engels, an ideal carried by the socialist militants of the time, or even an intellectual exercise—certainly stimulating—as in the work by Enrico Barone [31]. They exerted a profound influence in the decades that followed.
War as origin of planning. The 1914–1918 war imposed a brutal restructuring of economic activities on all the belligerents [32]. The phenomenon of the «war economy» can’t be reduced to the economy in wartime. It also covers the brutal restructuring of institutions and activities which, coupled with a single buyer, the state [33], tends to limit market mechanisms in favour of organized or even centralized allocation methods. Germany was the first of the belligerent countries to take measures to control the purchase and distribution of raw materials. The German war economy can be seen as the «prototype» of planning. The reason for this was the blockade put in place by the Franco-British [34–36], which no one could ignore would hit hard an economy highly dependent on the import of certain raw materials. The reorganization of the German economy thus began practically in the early days of the conflict [37].
France was confronted too with a strong increase in industrial demands but one that was combined to an equally strong reduction in production capacities. The German Army occupied 3.7% of the national territory with 8.2% of the population [38]. This amounted to 14% of the labour force, representing 58% of steel production and 40% of coal [33]. Men like Albert Thomas [39], Victor Dalbiez, Louis Loucheur [40] or Étienne Clémentel [41] organized, thanks to new laws (Dalbiez law of June 4, 19152) or decrees, a real mobilization of the war economy, with requisition of companies and workforce. This policy, slowly implemented in 1915, developed in 1916–1917 to lead to an original model of planning through consultation.
The question of economic mobilization was raised in 1917 in the United States. The late entry into the war, however, largely limited the scope of measures taken despite initial ambitions [42, 43]. These measures consisted mainly in attempts to control the economy through centralized control of prices [44, 45] and production [46]. Existed then during World War I a wide range of government agencies charged with influencing or controlling economic activities, which operated under the authority of the Council on National Defence established by President Wilson in 1916 to prepare for entry into war of the United States [47]. However, the impact of World War I was relatively limited.
This nevertheless ignored the creation in 1924 of the Army Industrial College, which would play a decisive role in preparing men for the mobilization of World War II [48]. From the mid-1920s to the 1940s, the Army Industrial College developed a series of “plans” and concepts for the mobilization of the American economy in the event of major conflict; these plans were largely inspired by the French experience of planning through consultation [49]. This impact of the French experience, concentrated in the 1920s, is not surprising given the prestige then acquired by the French Army in 1918, but also the experience of many American officers who had seen how the military-industrial—with which some had been associated—system worked.
In all countries, states have to varying degrees governed the economy. They did this first by securing vital supplies, but also by seeking to avoid waste and finally by organizing trade and commerce. The rationalization of production was largely implemented. States did not hesitate to constitute themselves as producers. But, the logic of the industrial division of labor imposed to extend this system to many basic industries creating, at the end of the process, a sector hegemonized by the State on which it depended, if not for property, for financial reasons and priorities in the allocation of raw materials. This extension of state control sometimes extended even to agriculture, especially in Germany.
There was therefore a whole planning experiment in “capitalist” economies that aroses out of the First World War. This experience was reinforced and consolidated by the Second World War. It profoundly affected decision-makers and post-war economic policy, as historian Alan S. Milward notes: “The successes achieved left those concerned with the management of Western economies with profoundly different attitudes towards economy (…). Henceforth, they would no longer feel helpless victims of economic forces beyond their control. The economies were directed, positively and successfully, on different trajectories during the war period” [50].
The origins of post-war planning in India, Japan and France. The origins of the planning process in the post-WWII era are manifold. The comparison between India, Japan and France allows us to better appreciate this, but also to discern points of convergence.
India. The case of India is typical of a country gaining independence [51]. Planning appears as a means of rapidly obtaining strong economic development [52, 53], a guarantee of economic and political sovereignty [54].
But the debate actually took shape there in the early 1930s. If great intellectuals influenced by communist ideas intervened in this debate, their weight was less important than that of intellectuals from the nationalist movement [55]. We can note Meghnad Saha [56], an astrophysicist well known in his discipline and especially Mokshagundam Visvervaraya. Born in 1860, he is considered the pioneer of planning in India. His book “Planned Economy for India” published in 1934 and again in 1936 [57] proposed a ten-year plan and his ideas were firmly anchored in the concept of “democratic capitalism”.
At the same time, the idea of planning was emerging within the main party demanding independence, the Congress Party. In 1938, the Indian National Congress appointed a National Planning Committee (NPC) to prepare an economic development plan. In October 1938, the National Planning Committee (NPC) was therefore created under the leadership of Nehru and its members included a mix of scientists and top businessmen of the time including Meghnad Saha and M. Visvervaraya (with the double cap of promoter of the first plan project in India and representative of Tata Steel). However, with the outbreak of World War II in September 1939 and the imprisonment of the leaders of the Congress Party, all this was called into question. The NPC was suspended in 1940.
Indian industrialists then took up the planning torch. They were working as part of the British war effort. From the end of 1943, some of these industrialists came together to discuss the development of the Indian economy after the war. This gave birth to one of the most discussed plans in India during the 1940s: the “Bombay Plan” [58]. In this plan, planning and industrialization were largely synonymous. A key tenet of the Bombay plan was that the economy could not grow without government intervention and regulation, a point on which Indian industrialists in fact joined the left of the Congress Party.
Jawaharlal Nehru, the first Prime Minister of India from 1948, adopted some of the ideas of the Bombay plan, in particular those of a substantially interventionist state and an economy with a large public sector [59]. The Bombay plan has then become a witness to the developmental and nationalist aspiration of the national capitalist class [60]. Planning then emerges as one of the indispensable instruments for Indian sovereignty.
Japan. The adoption of planning in the immediate postwar Japan provides a second example. The expansion of the Japanese economy after 1945 was carried out within the framework of a planning [61], admittedly qualified as «indicative», but in which the means of controls by the state were far from being negligible. Like many Japanese economic institutions of the 1950s and 1960s, the plan has its origins in the period of the war economy [62] and economic thinking of the late 1930’s.
Among thinkers and intellectuals, Narumi Hijikata defended the idea that Japan should have an “Economic Headquarters” in order to organize the interests of the Kereitsu and large Japanese companies in a convergent manner with those of the state [63]. Another important intellectual influence was the Showa Research Association (Shōwa Kenyukai), formed under the inspiration of Prince Konoye. It set itself the goal of developing exchanges between university economists, entrepreneurs, military journalists and senior officials [63]. This association brought together academics from the left, even from the far left, but also ultra-nationalists and was one of the centers of the so-called “reformist” movement which coincided with Konoye’s rise to power. As early as the 1930s, technocrats had sought to combine state planning and private initiative. Learning from the lessons of industrialization in Manchuria and Manchoukuo since the early 1930s [64], these technocrats—under the leadership of Nobusuke Kishi3 who had played an important role in the administration of Manchoukuo [65]—they had recast their policies in more business friendly terms.
With the outbreak of conflict with China in 1937, the economy shifted completely to a war economy [66]. Most Japanese companies remained private, but they were heavily regulated to help with the war effort [67]. The key steps to establishing the war economy converged in the constitution of the Cabinet Planning Board (CPB) [68]. The most brilliant bureaucrats from various ministries were assembled for this purpose. A one-party system was advocated and adopted in 1940. However, we cannot speak of “Japanese fascism” as shown by Peter Duus and Daniel Okimoto [69]. The CPB planned the National General Mobilization Plans (Kokkasodoin-keikaku), which were at the heart of the Material Mobilization Plans (Butsudo-keikaku), and it spearheaded the application of the National General Mobilization Law [70, 71].
The planning dreams of Japanese technocrats were quickly dashed when the tide of the war turned against Japan. But their project survived the defeat. From the late 1940s, banking on the reversal of the occupation policy by the United States following the start of the Korean War in order to make Japan the bulwark against communism in Asia, the technocrats have emerged as the main architects of the fast-growing Japanese system [72]. The policies of the late 1930s to early 1940s [73] were largely retained after World War II and worked into the 1950s and 1960s as Japan developed rapidly.
France. In January 1946, France created the Commissariat Général au Plan, which proved to be extremly effective for many years. The context seemed to dictate this planning. The country was emerging from World War II ravaged both by war and by the multiple looting of the German occupation. However, the origins of planning thought are multiple and complex.
The Vichy regime was to put in place the first planning instruments [74], taking up the ideas of the group called X-Crise [75, 76]. The Delegation Generale à l’Equipement National (General Delegation for National Equipment or DGEN) presented by May 1942, a document which described a plan of 10 years aimed at the reconstruction of the country with a strong industrial and technological catch-up and controlled urbanization [77]. This plan was implemented at the beginning of 1944, in the context one can guess. In total, only the two-year “start-up phase” of the plan was actually launched. It will be taken over by the de Gaulle government in the fall of 1944 [77]. For its part, the DGEN will serve as the basis for the future Commissariat General au Plan (CGP) [77].
Planning naturally occupied a significant place in the economic projects of Free France too. Two names here emerge, that of Mendes France and Georges Boris, co-authors of a planning project dating from 1938. Boris arrived in London on June 19, 1940. He was to participate in the work of the Economic, Financial and Social Commission created by General de Gaulle in December 1941, well before the Conseil National de la Résistance (CNR). It was in this commission that post-war reform plans began to be discussed. The CNR program was published at the beginning of 1944. It allusively evoked the prospect of planning [78]. With the gradual liberation of the territory, these ideas turned into economic decisions. The action of Mendes France was notable here [79]. When he returned to the Commissariat aux Finances in November 1943, it was by defending an economic policy that included planning [79]. As soon as Mendes-France took office as Minister of the Economy in the Provisional Government (June 1944), Georges Boris was appointed to deal with the Plan. But the implementation of a Plan came up against the conflicts aroused by Mendes-France monetary reform project, which ultimately was rejected by General de Gaulle. Mendes-France resigned and this opened the door to the third source of influence, American influence.
On December 13, 1945, in a memorandum addressed to General de Gaulle, Jean Monnet therefore in turn proposed the development of a national plan. Jean Monnet was linked in London with a former deputy of the Socialist (SFIO) party, André Philip [80], who had voted against full powers in Pétain in 1940. The two men found themselves in a common opposition to de Gaulle, which they nevertheless served during and after the war, and by their proximity to American positions [81]. They were the main channel of American influence with Free France, in London as well as in Algiers [82]. In 1944 and 1945, while he was still in Washington, Monnet undertook to define, with the help of his American friends, the main lines of a French plan. Robert Nathan in particular, who had headed the Plan Committee of the War Production Board, was very directly involved [83].
Monnet and his American collaborators were trying to apply the American experience of wartime planning for postwar France [83]. Monnet then described the Plan in these terms, in a note that probably dates from the end of spring or summer 1946: “Nostalgic people, with their eyes turned to the past, deplore the intrusion of the State into economic life. (…) The ambitious, considering certain foreign achievements or the city of their dreams, underline, on the contrary, the insufficiency of the ends that the State proposes to itself and of the means which it gives itself. The Monnet Plan will disappoint everyone.
It must, on the contrary, seduce the Frenchmen of good sense for whom freedom is not anarchy, nor order, the advent of bureaucracy. (…) From liberalism, the 1st Plan rejects profit as the only guide to the economy, but resorts to it as a stimulus. From bureaucracy, the Plan rejects intrusive and finicky interventions. But, from interventionism he retains the need to discipline reconstruction in the name of the sole criterion of the general interest”.4 These words aptly sum up the balanced position he intended to maintain.
Planning results. The results of these planning experiments were mixed. They were very positive in India until 1962, in France and Japan until the early 1970s.
India, half a failure, half a success? Indian planning spanned a long period, from 1950 to 2017, and continues currently, in a very residual form, to this day. This makes it the longest planning experience, longer than that of the USSR and Eastern European countries actually, on par with China. The crucial point here is that this planning never used all the instruments of “centralized planning” of the Soviet model [84, 85]. The differences with the Soviet “model” have always been substantial [86], and India retained its mixed-economy nature as defined by Nehru. The place of the State in the economy was certainly important, especially from 1965 to 1990. But, was it not in the logic of an economy starting with a strong problem of underdevelopment and seeking to catch up [87]?
Has planning helped India emerge from underdevelopment faster than if India had not had planning? The question is delicate, because there is no “control group” in Asia, other countries (Japan, South Korea, Taiwan) having also adopted forms of planning. Compared with the developed countries of the OECD, India’s growth was initially slower (Table 1).
Table 1. India’s years of planning, %
Years Prime-minister in power Growth rate target Average actual rate of growth
1st plan 1951–1956 J. Nehru 2.1 3.6
2nd plan 1956–1961 J. Nehru 4.5 4.3
3rd plan 1961–1966 J. Nehru/L.B. Shatri 5.6 2.8
4th plan 1969–1974 I. Gandhi 5.7 3.3
5th plan 1974–1979 I. Gandhi/M. Desai/C. Singh 4.4 4.8
6th plan 1980–1985 I. Gandhi 5.2 5.7
7th plan 1985 –1990 R. Gandhi 5.0 6.0
8th plan 1992 –1997 P. V. Narasimha Rao 5.6 6.8
9th plan 1997–2002 S. A. Bihari Vajpayee 6.5 5.4
10th plan 2002–2007 S. A. Bihari Vajpayee 8.0 7.6
11th plan 2007 –2012 Dr. M. Singh 9.0 8.0
Source: Indian Planning Commission.
But the weight of the military conflicts India faced from 1962 to 1972 arguably explains this. Globally, India was experiencing several problems: insufficient savings, backwardness in agrarian structures and equipment in rural areas, dependence on Great Britain and significant technical backwardness.
The first problem involves an inflation rate (Table 2) that corresponds to the “inflationary gap” described by Keynes [88]. It can also be argued that with planning India set a target for aggregate demand, which also had inflationary implications [89, 90].
Table 2. India’s rate of inflation
Years Average inflation rate Standard deviation Mean deviation
1958–1967 6.70 5.24 3.93
1968–1977 6.98 10.28 7.37
1978–1987 7.54 3.83 2.97
1988–1997 9.35 2.62 1.92
1998–2007 5.27 3.91 2.37
2008–2017 7.93 3.70 2.95
Total 1958–2017 7.29 5.44 3.96
Source: India’s Central Bank, author’s computation.
In Table 2, we can see that the standard deviation and the mean deviation increase considerably over the decade 1968–1977. This is, quite logically, the effect of the military conflicts of the period, but also of the food crises. The reforms of 1985–1991 also resulted in significant inflationary pressures.
India has gradually broken with the situation of dependence vis-à-vis Great Britain and it has caught up with its technical backwardness in many sectors ranging from the nuclear industry [91] to the pharmaceutical industry to electronics. The role of the state planning has been decisive here. The same could be said for infrastructures development. There remains the question of agriculture. Despite the various priorities of the plans, it has only developed slowly. The impact of planning was therefore small on this sector, as it was on reducing income inequalities, although the latter narrowed from 1950 to 1985.
In fact, like many other countries around the world, India has been hit by the wave of neoliberalism. The 1991 external debt crisis brought India closer to default. This was the reason explicitly cited for the implementation of the reforms. This new policy adopted by India was a break with the previous policy and involved increased reliance on market forces. In addition, the role of the state in the sphere of economic development has been considerably reduced. Finally, the degree of openness of the economy has been significantly increased [92].
Inequalities exploded in India then (Fig. 1), and in particular between cities and the countryside [93]. The harmful effects of these neoliberal policies have paradoxically led to a worsening of the debt crisis, but also to the degradation of the environment and the rise in unemployment, accompanied by the rise in food and fuel prices and the fall of salaries.
Fig. 1. Evolution of inequalities in India: – – – share of the poorest 50%; — share of the wealthiest 10%
Source: World Inequality Database.
Social harms have been exacerbated by the imposition of «austerity measures» of cutting subsidies to the poor and cutting public spending on health, social protection and education. The period of planning decomposition, which begins with Rajiv Gandhi’s reforms, while accompanied by strong growth, has also been accompanied by a dramatic increase in inequality.
Japan: A spectacular success? Two periods here are important. The Plans which drawn up from the 1950s to the late 1970s were tremendous successes, marked by very high growth rates (Table 3). On the other hand, the Tanaka plan of 1975 was a failure. Designed in reaction to the oil crisis of 1973, it proved unable to provide growth of more than 3.1% when a rate of 9.4% was expected.
Table 3. List of the 6 first Japanese plans, %
Plan’s name Five years plan
for economic autonomy New
economic
plan Doubling national
income plan Middke-term economic plan Economic
and social
development plan New plan for economic and social development
Date of publishing Dec.-55 Dec.-57 Dec.-60 Jan.-65 March-67 Apr.-70
Prime-minister in charge when the plan was adopted Hatoyama Kishi Ikeda Sato Sato Sato
Theoretical plan span 1956–1960 1958–1962 1961–1970 1964–1968 1967–1971 1970–1975
Average growth before the plan 8.6 7.3 9.1 11.3 10.0 12.7
Growth target 5.0 6.5 7.2 8.1 8.2 10.6
Forecasted growth spread –3.6 –0.8 –1.9 –3.2 –1.8 –2.1
Plan actual growth 9.1 10.1 10.9 10.8 10.9 6.1
Gap between forecasted and actual growth 4.1 3.6 3.7 2.7 2.7 –4.5
Source: [94].
The Maki plan targeted growth of 6% per year. The latter did not exceed 4.5%. From the 1980s, the planning process lost its reality and importance, and average annual growth did not exceed 2.7% over the period 1980–1999. With the impact of the “lost decade”, there is a tendency to view the Japanese economy as stuck in weak growth. This led some observers to speak of the end of the Japanese “miracle” [95].
But, weren’t there more of the building and then the destruction of the institutions that allowed this country to experience very strong growth? Moreover, are the difficulties encountered by the Japanese economy due to planning institutions or, on the contrary, to their gradual dismantling from the early 1970s to the 1980s?
An objective assessment of planning in Japan must therefore be based on several criteria. The first is, of course, that of growth. For this, it is necessary to look precisely at the characteristics of this growth.
Whether for the level of average growth or for that of the reduction in the magnitude of fluctuations, the so-called “high planning” period (1956–1971) comes out on top (Table 4).
Table 4. Growth rates and fluctuations of Japanese economy: 1885–1999*, %
Years Average growth Mean deviation Average/Mean deviation
1885–1903 Meiji’s time 2.95 3.4 87.3
1906–1914 From 1904-05 war to 1914 war 2.03 2.9 70.1
1915–1922 Expansion linked to the first world war 4.86 2.6 187.6
1923–1932 Japan “roaring twenties” 1.69 2.7 62.0
1933–1940 Build-up toward a war economy 5.31 6.0 89.0
1946–1955 Post-war reconstruction 3.93 9.9 39.7
1956–1971 High planning 9.34 2.1 440.2
1972–1984 Decay of planning 4.08 1.8 228.8
1985–1999 Financiarisation and lost decade 2.72 2.1 131.4
* The average deviation must be compared to the average growth to take into account the magnitude of the latter. The Average Growth / Average Deviation ratio gives a more accurate view of the fluctuations, which are all the more important the smaller the percentage.
Sources: Management and Coordination Agency, Statistics Bureau, Historical Statistics of Japan on CD-ROM (Tokyo: Japan Statistical Association, 1999), Tableaux 13-03, 13-04 et 13-10.
If we now observe more qualitative criteria, the period of “high planning” is also characterized by a much stronger investment, but also much more balanced, than the previous periods [96, 97]. If the investment was particularly important in the following period, which can be described as “planning decadence”, this was the product of unregulated competition between large companies and resulted in a financial crisis which reflected the significant accumulated imbalances [98]. Likewise, the so-called “high planning” period enabled Japan to manage its dilemma between investment, domestic consumption and the balance of payments in a way that was far less unbalancing for the world economy than in later years.
This planning also had an obvious and significant impact on other Asian countries, such as South Korea (which adopted a five-year plan regime) [99–101], Taiwan [102], Malaysia [103] (from 1966) and Singapore [104]. It was an element of Japanese “soft power” in the Far East. From this point of view, too, it has undoubtedly proved to be one of the most successful models of planning in a “capitalist” country.
Success and decay of planning in France. France’s economic trajectory in the post-war years was spectacular, leading to the years 1945–1975 being called the “Trente Glorieuses”. How much of this is due to planning [105]? The latter, far from having been the sole prerogative of France, has been imitated in several countries [106]. Charles Kindleberger drew a glowing review in 1962 [107]. Indeed, assessment ihat can be drawn from the planning in the 1950s and up to the 1970s [108, 109]. can be regarded as positive.
The object of an assessment is not, however, to know if France has rebuilt and developed itself in the twenty-five years which followed the war [110] but, rather, to know if the French model of planning conferred a comparative advantage. The next question is how to measure this advantage and what were the constraints that faced, as well as the advantages enjoyed by different countries. We will therefore take 5 countries, France and Italy (which have had forms of planning), Germany in the FRG sense (country which has not seen fit to adopt a form of planning) and the United States and the United Kingdom, countries that were then considered as targets for countries under reconstruction (Table 5). France was constrained by colonial wars but was able to benefit from the Marshall Plan (or ERP) ERP, like Italy, the FRG and the United Kingdom. The latter was constrained by its international role, where it wanted to maintain a posture equal to the United States. The FRG, for its part, benefited from the spending of the US military on its territory (like Japan) and the flow of skilled labor fleeing the GDR until the late 1950s.
Table 5. Rate of growth, comparison, %
Plan Years France Italy Germany (FRG) United Kingdom United States
1st Plan (partial) 1949–1953 4.91
2nd Plan 1954–1957 5.22 5.27 8.39 2.21 2.59
3rd Plan 1958–1961 4.21 6.49 6.40 2.54 2.34
4th Plan 1962–1965 5.40 4.30 4.81 3.02 5.16
5th Plan 1966–1970 5.22 6.13 4.26 2.77 2.92
1952–1961 Average 4.12 5.85 7.78 2.24 2.75
1962–1970 Average 5.30 5.32 4.50 2.88 3.92
1954–1970 Mean deviation 1952–1970 0.81 1.02 2.05 1.18 2.01
1954–1970 Variance × 100 0.99 1.76 7.15 2.07 5.83
Source: INSEE and different countries statistical offices.
One can see that France and Italy have done better than the United Kingdom and the United States and even Germany from 1960 onwards. We also observe that French growth, if it lags a little behind in the 1950s, no doubt due to the overvaluation of the exchange rate mentioned by Kindleberger, accelerated in the 1960s (once the Franc depreciated and the colonial question) and is more stable than in the other four countries. Qualitatively, the productivity gains were spectacular (Table 6).
Table 6. Average annual productivity gains by sector (1st to 5th Plan), %
Plan Years Global productivity Agricultural productivity Industrial productivity Construction productivity
1st Plan (partial) 1949–1953 3.0 6.3 3.4 0.6
2nd Plan 1954–1957 6.0 5.0 5.0 3.8
3rd Plan 1958–1961 5.7 10.9 6.6 5.0
4th Plan 1962–1965 5.9 3.4 7.5 3.0
5th Plan 1966–1970 7.8 9.3 9.3 3.7
Source: INSEE.
We can also see that the efforts to modernize the various French plans have borne fruit if we look at the productivity gains. The gains were significant in agricultural activities at large and in industry where they are even tending to accelerate. This is the result of the various modernization actions undertaken from the 2nd to the 5th Plan. As for investments, the results of France appear very good. They should be compared with the eight percentage point increase in investment GDP recorded in France (Fig. 2). Moreover, from 1952 to 1967, this increase is almost linear. This reflects the effectiveness of investment support policies under the conditions of the time.
Fig. 2.Investment in GDP percent.
Source: INSEE.
Planning: from decline to resurrection? However, French planning will be weakened and then deconstructed by a series of factors, both internal and external.
The latter are generally well known: major rise in the price of oil from the end of 1973 to the beginning of the 1980s, pressure from increasing international competition, increasingly heavy impact of the European framework, in particular with the signature of the Single European Act which entered into force in 1987 and the progressive financialization of the economy which took place from 1984–1986.
But, these external factors, except in the case of the oil crisis, could only have had these destructive effects because of internal decisions that gave them decisive weight, even anticipated and aggravated them. The impact of international competition can only be explained by successive free trade factor decisions and the decision to maintain an overvalued exchange rate. The significance of the European Single Act was made much important in France, due to the pro-European turn taken in 1983 by François Mitterrand. This ushered in an era of financial rigor. Finally, financialization was a deliberate choice of the French administrative elites [111]. The three successive plans after 1980 will each represent an additional stage in the planning breakdown process.
The causes of the turnaround that took place in the 1970s are, it has been said, multiple. Some are linked to the collapse of the alliances and political blocs that carried the idea of planning in the early 1940s. This is evidently the case in India. Others are linked to struggles within the administrations from which the various planning bodies originated. This was the case in France and Japan. In these struggles for administrative preeminence and power enter deeply ideological elements, but also the personal interests of the actors. Finally, the key elements of the international context have changed. But, is this change purely exogenous or was it not prepared from within certain countries? The border between exogenous change and endogenous change appears here to fluctuate widely.
Economic or political crisis? For Japan as for France, the immediately visible break was the oil crisis. It is probable, and even certain in the French case, that it was accompanied by other ruptures. A lot of economics book have been published on this matter, along them [112–116]. Planning will not recover from the shock, less because of its inability to adapt to the new situation and more because of a breakdown in the consensus surrounding the broad directions of the economy that this shock exacerbated.
This rupture led to the gradual dismantling of the instruments and institutions on which planning was based. To what extent did the administrative elites in France but also in Japan rely on the new context to justify this dismantling and to what extent they created the new context, or at least an institutional framework, making planning less and less practicable in its classic form, remains to be studied in more detail. The case of trade liberalization and financialization practices is a good example.
However, the main measures were national, and they made planning less and less practicable. We see this for France in the case of financial liberalization, which was largely the work of senior officials from the Finance Inspectorate [117–119]. For France in particular, it is clear that there was a will on the part of the administration to put an end to planning from the beginning of the 1970s and even before.
This section of the administration will use the consequences of the oil crisis and then the dynamic of opening up the economy (and the growing power of the European framework) to dismantle the instruments of planning and reduce it to impotence. It found its advantages when the most eminent members were appointed to head public enterprises in the process of privatization. This explains why the arrival in 1981 of a Left Union government will bring no change. The Ministry of Planning, which was given to Michel Rocard, was indeed a sinecure without means.
The context of a resurrection. Yet this planning, which one might have thought buried, relegated to limbo, is now making a dramatic comeback. This is done in a context marked by the significance of the concept of sovereignty. In his speech of July 12, 2021, Emmanuel Macron repeatedly referred to the independence of France.5 When he talks about the decisions to be taken, it is to put this in mind: “The will that France must carry is indeed this: regain control of our destiny…”.6 When he concludes, it is with these words: “So, at the start of the school year, we will have an appointment with our future. To build an independent France (…)”.7 It will take up again this theme during the presentation of the “France-2030 Plan” on October 12, 2021.8
If we go back a little, we find this declaration of the President, when he inaugurated on July 2, 2021 the exhibition “Made in France”: “The made in France, is sovereignty, independence, that is to say the ability also, as several of you have done, to relocate know-how or production shares on French soil”.9 Or again, in the speech delivered on June 29, on the occasion of the presentation of the health innovation strategy 2030, where he used the word “sovereignty” four times10, a word that was also spoken in the address to the French of November 24, 2020.
There is therefore no doubt that the notions of sovereignty, of national independence, are regularly mentioned in the vocabulary of the Head of State, even if one can wonder about the real meaning he gives them. Let us add to this a questioning of the omnipotence of the market, in his speech of November 9, 2021: “To meet all these challenges, to control our destiny, the market alone is not enough. We must assume a strong public intervention with, in a few key areas, significant investments”.11
And, Emmanuel Macron is far to be alone to regularly invoke these notions in his statements. Arnaud Montebourg, former Minister of the Industry and putative father of the “made in France” movement, did the same.12 We are therefore not out of the “sovereignist moment” that could be diagnosed as early as 2016 [120] and which continues to permeate French politics. However, we have seen, whether with regard to India, Japan and of course France in 1945, that planning was central in the struggle to regain or assert sovereignty. It is then understandable, that the question of planning is again approached.
The current need for a return to planning: the case of France. We must therefore try to understand the underlying reasons, and the arguments used for this purpose, for the return to favor of the idea of planning. We must also ask ourselves whether the instruments of planning can again be mobilized.
Is there an actual “downgrading” of France? In the case of France, it is clear that the shock induced by the health crisis played a big role. This shock revealed a pre-existing situation of deindustrialization of the country, everything did not date from the COVID-19 far from it, but it revealed it with a force and a cruelty that marked the spirits deeply. From this episode ensued a return of the idea of Plan to the public debate.
The image that the French will retain of this epidemic is that of the great downgrading of France, reduced in many ways to the situation of a developing country. Of course, there is injustice in this picture. France was not alone in his case and some of his reactions have lived up to it. After the initial mistakes, the vaccination campaign went rather well. But, the part of truth which exists in this formula of great downgrading should not be ignored either.
France, therefore, found itself dependent on foreigners to the point that the masks delivered from China were, in April 2020, guarded by impressive police forces. But this dependence and downgrading has manifested itself in other areas. They range from medical drug shortages to the shutdown of auto factories in the summer of 2021 due to semiconductor shortages.
We must recognize Emmanuel Macron and his government for having taken the measure of this downgrading. One of its first reactions was to recreate, in September 2020, a High Commission for Planning.13 He also presented a year later, in a speech on October 12, 2021, the outline of a plan for 2030.14 One might think that these initiatives are linked. It doesn’t. Certainly, the word «plan» is frequently cited in this speech. But it is in this general form or plan simply becomes synonymous with project. Emmanuel Macron also quotes the “European plan”, but there again there is an abuse of language. The “plan” he speaks of is just a funding program, moreover not very rigorous in its methods and far too limited in its resources. It should be noted that the High Commissioner for Planning is mentioned only once, in the preliminary greetings to the speech.
The HCP finds itself much more in coordinating various forecasting bodies, such as France-Stratégie, than in the function which should be its own, defining priorities and implementing, in permanent consultation with administrations and companies, the means to reach them. However, one should not minimize the foresight work that the HCP provided in its first year of existence. Note nо. 2 on vital products and strategic sectors is a prospective work of good standard.15 It reads: “We discovered that the supplies of essential drugs for treatment and for the health of our compatriots were called into question (…) We checked that the stocks did not exist or were not sufficient. But this risk can also be proven for other products essential to the agricultural world, to the industrial world or to our life in society. These risks of scarcity reveal our country’s dependence on distant production chains, dependent on decisions over which we have no control”.16
In the introduction signed by François Bayrou, this note calls for the development of mobilization plans so that, in the event of a new crisis, the disposal of products and vital equipment could be ensured. It calls too for a reflection on the definition of strategic sectors on which depends the national defense but also the critical supplies necessary for the sovereignty of France and its security. The note analyzes France’s dependence on drugs on the basis of a former Senate report [121]. It clearly shows the problem posed by deindustrialisation in France.
The energy transition. The debate on planning is then resurfacing, and it is resurfacing in particular around the ecological question of energy transition. Several journals have devoted issues to it in recent months [122]. Coming thirty years after the collapse of the USSR and the Soviet bloc, and what could be considered the collapse of one of the traditions of planning, even if the latter had not been relevant for the France, this return of planning to the center of the debate is particularly significant. Indeed, the use of planning will most likely be required if we are to face the climate emergency and truly lead an energy transition leading to carbon neutrality. The question of transition processes, in particular when they involve deep social, technical and economic transformations [123], requires articulated policies that only planning allows [124]. But, the risk of seeing this process being conducted in an undemocratic manner is real [125]. From this point of view, the French model of planning through consultation provides guarantees for democratic procedures. The ability to conduct the debate that will lead to an order of priority in a transparent manner is a guarantee that the energy transition is not captured by vested interests [126].
Laure Després, a former specialist in Soviet economics, wonders about the need for planning to achieve the ecological transition [127]. This was the theme of one of his recent previous articles [128]. Laure Després starts from the observation of the inability of “green growth” strategies, at least in their current form, to respond to ecological and social crises. The inconsistency between the various documents that have been published on this subject in recent years in France shows, according to her, the need to establish a comprehensive procedure. Effective planning should guide investments, and therefore question their financing procedures and therefore break with the neoliberal functioning of the administration. The idea of planning applied to ecological transition [129] is also today defended by Gaël Giraud [130], “who was the former chief economist of the French Agency for Development”.
Another author, Dominique Plihon, starts from the observation that our societies are poorly prepared for ecological and social transition because they suffer from both a problem of temporal inconsistency, between the long-term horizon of the transition and the he short horizon for most economic and political actors, and a problem of democratic deficit which hinders the mobilization of society around precisely this objective of ecological transition [131]. Analyzing the question of uncertainty as an endogenous product, as Keynes argued [132], of the decisions of decentralized actors, but also that of the irreversible and cumulative nature of climate change [131], he ended up with the impossibility for market mechanisms to resolve the problem of temporal inconsistency. He then writes: “If they do not have clear information on these future policies, these actors may be led to postpone or abandon their decisions. As economic analysis shows, uncertainty is one of the main obstacles to investment” [131]. He then shows that the French planning model, because it sheds light on the future in order to overcome, or loosen, the constraints of immediacy, would allow the challenges of the medium and long term to be defined.
This model appears particularly well suited to the treatment of ecological transition [131]. Plihon then wrote, about Pierre Massé, who was the author of a fundamental work on planning: [133] “To justify the usefulness of planning, Pierre Massé starts from a critique of the generalized market theory in particular by Gérard Debreu, by showing that all future eventualities cannot be known in advance. This is especially the case for long-term investments, such as the construction of a dam, for which there is no market or price reflecting future services” [131]. In fact, Dominique Plihon uses the terms of the debate between Neurath and von Mises on the subject of planning.
Michel Aglietta and Étienne Espagne have published an article along the same lines [134]. Of particular importance is the article by Aglietta and Espagne. As one of the authors of the Fifi model, Aglietta is uniquely placed to talk about a return of the plan. This article immediately raises the question of planning in the context of the energy transition. In these two important articles, the question of time horizons, how to articulate short and long-term responses, appears to be the driving force behind this return to the plan.
The question remains, however, whether planning could be done at European Union level or whether national planning is possible within the framework of European institutions. There is no doubt that this will be the subject of many future debates.
1 https://www.gouvernement.fr/haut-commissariat-au-plan/presentation.
2 http://genealogie.dalbiez.eu/Loi%20Dalbiez.htm.
3 Nobusuke Kishi, 1896–1987, was Shinzo Abe, the Japanese Prime-minister from 2012 to 2020, great-father.
4 AMF 005/002/040, Fonds Jean Monnet, Fondation Jean Monnet, Lausanne.
5 https://www.elysee.fr/emmanuel-macron/2021/07/12/adresse-aux-francais-12-juillet-2021.
6 See note 5.
7 See note 5.
8 https://www.elysee.fr/emmanuel-macron/2021/10/12/-presentation-du-plan-france-2030.
9 https://www.elysee.fr/emmanuel-macron/2021/07/02/-inauguration-de-la-grande-exposition-du-fabrique-en-france.
10 https://www.elysee.fr/emmanuel-macron/2021/06/29/-faire-de-la-france-la-1ere-nation-europeenne-innovante-et-souveraine-en-sante.
11 https://www.elysee.fr/emmanuel-macron/2021/11/09/-adresse-aux-francais-9-novembre-2021.
12 https://www.lepoint.fr/politique/montebourg-appelle-a-une-reconquete-de-notre-souverainete-et-une-reconstruction-ecologique-08-04-2020-2370666_20.php.
13 https://www.gouvernement.fr/haut-commissariat-au-plan/les-textes.
14 https://www.elysee.fr/index.php/emmanuel-macron/-2021/10/12/-presentation-du-plan-france-2030.
15 https://www.gouvernement.fr/les-grands-enjeux-de-demain.
16 Note No. 2 du HCP du 18 décembre 2020. Р. 2.
==== Refs
REFERENCES
1 Voir, Government of India – NITI Aayog, ANNUAL REPORT 2014-15 (Government of India Press, New Delhi, 2015). http://www.niti.gov.in.
2 Sapir J. “Has the COVID-19 crisis put planning back on the agenda?,” Nauch. Tr. Vol’n. Ekon O-va Ross. 2021 230 259 274
3 W. Andreff, “Edmond malinvaud et la planification décentralisée,” Revue d’histoire de la pensée économique, No. 9, 43–90 (2020).
4 Rudra A. Prasanta Chandra Mahalanobis: A Biography 1996 London and New York Oxford Univ. Press
5 R. Savy, “Les pouvoirs économiques exceptionnel,” Pouvoirs, No. 10 (1979).
6 Ott J. S. Dicke L. A. The Nature of the Non-Profit Sector 2021 New York Routledge
7 Stiglitz J. E. Markets, market failures, and development Amer. Econ. Rev. 1989 79 197 203
8 Borenstein S. The trouble with electricity markets: understanding California’s restructuring disaster J. Econ. Perspective 2002 16 191 211 10.1257/0895330027175
9 Percebois J. Wright. Electricity consumers under the state and the private sector: comparing price performance of the French and UK electricity industries, 1990–2000 Utilities Policy, No. 2001 10 167 179 10.1016/S0957-1787(02)00039-5
10 G. L. S. Shackle, Expectation in Economics (Cambridge Univ. Press Cambridge, 1949).
11 Simon H. A. The New Palgrave: A Dictionary of Economics 1987 London Macmillan
12 Simon H. A. Rationality as process and as product of thought Amer. Econ. Rev. 1978 68 1 16
13 O. Neurath, in War Economy. Economic Writings – Selections 1904–1945, Ed. by T. Uebel and R. Cohen (Kluwer, New York, 2005), pp. 12, 193
14 Simon H. A. “From Substantive to Procedural Rationality,” in 25 Years of Economic Theory 1976 Boston, MA Springer
15 Laville F. Foundations of procedural rationality: cognitive limits and decision processes Econ. Philos. 2000 16 117 138 10.1017/S026626710000016X
16 J. W. Dean, Jr. and M. P. Sharfman, “Procedural rationality in the strategic decision-making process,” P. J. Manag. Stud., No. 30, 587–610 (1993).
17 P. Massé, Le Plan ou l’Anti-hasard (Gallimard, NRF Idées, Paris, 1965), in F. Perroux, L’Économie du XX e Siècle (Presses Universitaires de Grenoble, Grenoble, 1961).
18 F. Perroux, Ed. L'Economie du XXe Siècle (Presses Universitaires de Grenoble, Grenoble, 1961).
19 Fondation Nationale des Sciences Politiques (Paris) and Institut d’Etudes Politiques de l’Université de Grenoble. La Planification comme Processus de Decision. Cahiers de la Fondation Nationale des Sciences Politiques Paris (Armand Colin, Paris, 1965).
20 Sapir J. Les Trous Noirs de la Pensée Économique 2000 Paris Albin Michel
21 Von Mises L. Collectivist Economic Planning 1935 London Routledge
22 Hayek F. A. Collectivist Economic Planning 1935 London Routledge
23 O. Lange, “On the Economic Theory of Socialism,” Rev. Econ. Stud. 4, (1–2) (1936).
24 Lange O. Papers in Economics and Sociology 1970 Warsaw and London PWN and Pergamon Press
25 A. P. Lerner, “A note on socialist economies,” Rev. Econ. Stud. 4 (1) (1936).
26 A. P. Lerner, “Statics and dynamics in socialist economies,” Econ. J. 47 (1937).
27 O’Neill J. Who Won the Socialist Calculation Debate History of Political Thought 1996 17 431 442
28 Hayek F. A. Profit, Interest and Investment 1939 London Routledge
29 Monetary Theory and the Trade Cycle (Jonathan Cape, London, 1933).
30 Sapir J. Planning in capitalist economies: On the example of military economies Questions Polit. Econ. 2021 27 120 136
31 Barone E. “Ministro della Produzione nello Stato Collettivista,” Giornale degli Economisti, September-October, 267–293, 392–414 (1908), republished in Le Opere Economiche Zanichelli, Bologna 1936 1 229 297
32 Broadberry S. Harrison M. The Economics of World War I 2005 Cambridge Cambridge Univ. Press
33 G. Hardach, “The First World War, 1914–1918,” in History of the World Economy in the Twentieth Century (Univ. of California Press, Berkeley and Los Angeles, 1977), No. 2.
34 Vincent C. P. The Politics of Hunger. The Allied Blockade of Germany (1915–1919) 1985 Athens, OH Ohio Univ. Press
35 Siney M. C. The Allied Blockade of Germany (1914–1916) 1957 Ann Arbor, MI Univ. of Michigan Press
36 Farrar M. M. Conflict and Compromise. The Strategy, Politics and Diplomacy of the French Blockade (1914–1918) 1974 The Hague Mouton
37 Dallas G. 1918: War and Peace 2000 London John Murray
38 Nivet P. La France Occupee. 1914–1918 2011 Paris Armand Colin
39 Hennebicque A. “Albert Thomas et le régime des usines de guerre (1915–1917),” in 1914–1918. L’autre Front 1977 Paris Editions Ouvrières
40 Barjot D. Deux Guerre Totales, 1914–1918, 1939–1945. La Mobilisation de la Nation 2012 Paris Economica
41 Étienne Clémentel 1864–1936: Politique et Action Publique sous la Troisième République, Ed. by M.-C. Kessler and G. Rousseau (P.I.E-Peter Lang S.A. Soll. “France contemporaine,” Bruxelles, 2018).
42 Clarkson G. B. Industrial America in the World War 1923 New York Houghton Mifflin
43 Baruch B. M. American Industry in the War 1941 New York Prentice Hall
44 Rockoff H. Drastic Measures: A History of Wage and Price Controls in the United States 1984 New York Cambridge Univ. Press
45 Taussig F. W. Price-fixing as seen by a price-fixer Quarterly J. Econ. 1919 33 205 241 10.2307/1884733
46 Kerr K. A. Decision for federal control: Wilson, McAdoo, and the railroads, 1917 J. Amer. History 1967 54 550 560 10.2307/2937406
47 Breen W. J. Uncle Sam at Home: Civilian Mobilization, Wartime Federalism, and the Council of National Defense (1917–1919) 1984 Westport CT Greenwood Press
48 R. E. Smith, The Army and Economic Mobilization (Center of Military History, United States Army, US-GPO Washington D.C., 1959).
49 H. W. Thatcher, Planning for Industrial Mobilization (1920–1940) (General Administrative Services Division, Office of the Quartermaster General, US-GPO, Washington D.C., 1948).
50 Milward A. S. War, Economy and Society. 1939–1945 1979 Berkeley and Los Angeles Univ. of California Press
51 Gregory P. R. Stuart R. C. Comparing Economic Systems in the Twenty-First Century 2003
52 J. N. Bhagwati and P. Desai, India: Planning for Industrialization (Oxford Univ. Press London, 1970).
53 Badhuri S. Yangla N. Seventy five years of national planning committee (1938–2013): A recollection Science and Culture 2015 81 66 70
54 Bettelheim C. L’Inde Indépendante 1962 Paris Colin
55 Haithcox J. P. Communism and Nationalism in India 2015 Princeton Princeton Univ. Press
56 Chatterjee S. J. Gupta A. Meghnad Saha in Parliament 1993 Calcutta The Asiatic Society
57 Visvesvaraya M. Planned Economy for India 1936 Bangalore The Bangalore Press
58 P. Thakurdas, A Brief Memorandum Outlining a Plan of Economic Development for India (2 Vols.) (Penguin, London, 1945).
59 Ananth V. K. Religion, Power and Violence: Expression of Politics in Contemporary Times 2005 New Delhi SAGE
60 Chibber V. Locked in Place: State-building and Late Industrialization in India 2003 Princeton Princeton Univ. Press
61 Komiya R. Economic Planning – East and West 1975 Cambridge, MA Harvard University Press
62 Okimoto D. I. Competitve Edge. The Semiconductor Industry in the US and Japan 1984 Stanford, CA Stanford University Press
63 Morris-Suzuki T. A History of Japanese Economic Thought 1989 London Routledge
64 Kapur N. Japan at the Crossroads: Conflict and Compromise after Anpo 2018 Cambridge, MA Harvard University Press
65 J. Maiolo, Cry Havoc: How the Arms Race Drove the World to War, 1931–1941 (Basic Books, New York, 2010), p. 29.
66 Allen G. C. Japanese Industry: Its Recent Development and Present Condition 1940 New York Inst. of Pacific Relations
67 Jansen M. Hall J. W. Kanai M. Twitchett D. The Cambridge History of Japan 1989 Cambridge Cambridge University Press
68 Pauer E. Japan’s War Economy 1999 London Routledge
69 Duus P. Okimoto D. Comment: Fascism and the history of pre-war Japan: The failure of a concept J. Asian Stud. 1979 39 65 76 10.2307/2053504
70 T. A. Bisson, Japan’s War Economy (International Secretariat, Institute of Pacific Relations, The Macmillan Co., New York, 1945), pp. XV, 267.
71 Cohen J. B. Japan’s Economy in War 1949 Minneapolis Univ. of Minnesota Press
72 J. Mimura, “Japan’s new order and greater East Asia co-prosperity sphere: Planning for empire,” Asia-Pacific J. Japan Focus 9 (49), No. 3 (2011).
73 Dower J. W. Japan in War & Peace 1993
74 R. F. Kuisel, Le Capitalisme et l’État en France, Chap. I: “Les Brouillons de la Période de Vichy” (Gallimard, Paris, 1984).
75 O. Dard, “Voyage à l’intérieur d’X-crise,” Vingtième Siècle, revue d’histoire, No. 47, 132–146 (1995).
76 M. E. Fischman, Lendjel, De X-Crise (1931–1939) à X-Sursaut (2005–?). L’apport des Polytechniciens à la réflexion sur le rôle de l’Etat dans la vie économique. https://halshs.archives-ouvertes.fr/halshs-00269386/document.
77 Kuisel R. F. Vichy et les origines de la planification économique (1940–1946) Le Mouvement Social, No. 1977 98 77 101 10.2307/3778070
78 Andrieu C. Le Programme Commun de la Resistance 1984 Paris Éditions de l’Érudit
79 P. Mioche, “La planification comme ‘Réforme de structure’: l’action de Pierre Mendès-France, de 1943 à 1945,” Histoire, économie et société. 1e année, No. 3, 471–488 (1982).
80 Ehrmann H. La Politique du Patronat Français 1959 Paris Armand Colin
81 Maximin E. R. The United States and the French left, 1945–1949: The view from the State Department J. Modern History 1984 19 729 74
82 M. Margairaz, “Autour des accords Blum-Byrnes. Jean Monnet entre le consensus national et le consensus atlantique,” Histoire, économie et société. 1e année, No. 3, 439-470 (1982).
83 M-L. Djelic, “Genèse et fondements du plan Monnet: l’inspiration américain,” Revue Française d’Etudes Américaines, No. 68, 77–86, 80 (1996), Sexualités: expression et répression.
84 Zaleski E. La Planification Stalinienne. Croissance et Fluctuations Économiques en URSS (1933–1952) 1984 Paris Economica
85 Nove A. L’Économie Soviétique 1981 Paris Economica
86 Hutchings R. The Structural Origins of Soviet Industrial Expansion 1984 London Macmillan
87 Gerschenkron A. Economic Backwardness in Historical Perspective 1962 Cambridge, MA The Belknap Press of Harvard Univ. Press
88 Keynes J. M. How to Pay for the War: A Radical Plan for the Chancellor of the Exchequer 1940 London Macmillan
89 P. R. Tcherneva, Keynes’s Approach to Full Employment: Aggregate or Targeted Demand?, Levy Economics Institute Working Paper, No. 542, Bard College, The Levy Economics Institute, 2008.
90 Kregel J. A. The effective demand approach to employment and inflation J. Post Keynesian Econ. 1987 10 133 145 10.1080/01603477.1987.11489665
91 Gupta A. India’s nuclear energy programme: Prospects and challenges Strategic Analysis 2011 35 373 380 10.1080/09700161.2011.559963
92 K. Siddiqui, “Globalisation and Neoliberal Economic Reforms in India: A Critical Review,” in Globalization in India: New Frontiers and Emerging Challenges (Prentice Hall India, New Delhi, 2010), pp. 219–243.
93 V. Negi, Where have all the farmers gone! The story of agriculture and the small Indian farmer, Focus on the global South, India, 2014. http://focusweb.org/sites/www.focusweb.org/files/Where%-20have%20all%20the%20small%20farmers%20-gone.pdf.
94 M. Baba, “Le Caractère de la Planification Économique au Japon et le Rôle Tenu par le Gouvernement,” in La Planification en France et au Japon, Les Collections de l’INSEE. Série C. No. 61, Ed. by Ch. Sautter and M. Baba (INSEE, Paris, 1978), No. 261, Tab. 1, p. 37.
95 Katz R. Japan, The System That Soured 1998 Armonk, New York and London MF Sharpe
96 A. J. Alexander, “Japan’s economy in the 20th century,” JEI report, No. 3 (2000).
97 Ohkawa K. Rosovsky H. Japanese Economic Growth 1973 Stanford, CA Stanford Univ. Press
98 A. Maddison, “Standardized Estimates of Fixed Capital Stock: A Six-Country Comparison,” in Explaining the Economic Performance of Nations (Edward Elgar Publ. Co, Brookfield, VA, 1995).
99 C. Wolf, Jr., Economic Planning in Korea (The Rand Corporation, Santa-Monica, CA, 1962), Paper P-2655.
100 Chung Kae H. Industrial progress in South Korea Asian Survey 1974 14 439 455 10.2307/2642849
101 Amsden A. Asia’s Next Giant: South Korea and Late Industrialization 1989 New York Oxford Univ. Press
102 Wade R. Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization 1990 Princeton, NJ Princeton Univ. Press
103 J. W. Henderson, N. Vreeland, G. B. Dana, G. B. Hurwitz, P. Just, P. W. Moeller, and R. S. Shinn, Area Handbook for Malaysia (American Univ., Foreign Area Stud., Washington D.C., 1977).
104 Huff G. What is the Singapore model of economic development? Cambridge J. Econ. 1995 19 735 59
105 S. Wickham, “French Planning: Retrospect and Prospect,” Rev. Econ. Stat. 45 (1963).
106 Wellisz S. Economic planning in the Netherlands, France, and Italy J. Polit. Econ. 1960 68 252 283 10.1086/258322
107 C. P. Kindleberger, “French Planning,” in National Economic Planning, Ed. by M. F. Milikan (Columbia University Press and NBER, Cambridge, New-York and London, 1967).
108 J. Hacket and A-Marie Hackett, Economic Planning in France (Harvard Univ. Press, Cambridge, MA, 1963).
109 Sheahan J. Promotion and Control of Industry in Postwar France 1963 Cambridge, MA Harvard Univ. Press
110 Estrin S. Holmes P. J. The performance of French planning 1952–1978 Econ. Planning 1980 16 1 19 10.1007/BF00351462
111 R. Abdelal, “Le consensus de Paris: la France et les règles de la finance Mondiale,” Critique international 28 (3), 87–115 (2005).
112 Aglietta M. Régulation et Crises du Capitalisme 1976 Paris Calmann-Lévy
113 Aglietta M. Régulation et Crises du Capitalisme 1982 Paris Calmann-Lévy
114 Boyer R. Mistral J. Accumulation, Inflation, Crises 1978 Paris PUF
115 Boyer R. Saillard Y. Théorie de la Régulation. État des Savoirs 1995 Paris La Découverte
116 Lipietz A. Crise et Inflation, Pourquoi? 1979 Paris Maspéro
117 N. Véron, “Le ‘French Paradox’ de la libéralisation financière,” La Vie des Idées, September (2007).
118 Cusset F. La Décennie. Le Grand Cauchemar des Années 1980 2006 Paris La Découverte
119 Abdelal R. Capital Rules: The Construction of Global Finance 2007 Cambridge Harvard Univ. Press
120 Sapir J. Souveraineté, Démocratie, Laïcité 2016 Paris Michalon
121 Sénat. Rapport de la mission d’information sur la pénurie de médicaments et de vaccins (Yves Daudigny, président, Jean-Pierre Decool, rapporteur), September 27, 2018. https://www.senat.fr/commission/missions/ penurie_de_medicaments_et_de_vaccins/index.html.
122 “La planification aujourd’hui,” Actuel Marx 2019/1 (65) (2020), Issue cordinated by Cédric Durand and Razmig Keucheyan, Les Possibles, No. 23 (2020).
123 Rotmans J. Loorbach D. Complexity and transition management J. Industr. Ecol. 2009 13 184 196 10.1111/j.1530-9290.2009.00116.x
124 Grin J. Rotmans J. Schot J. Transitions to Sustainable Development; New Directions in the Study of Long Term Transformative Change 2010 New York Routledge
125 Hendriks C. M. Policy design without democracy? Making democratic sense of transition management Policy Sci. 2009 42 341 368 10.1007/s11077-009-9095-1
126 J. Gaede and J. Meadowcroft, “A question of authenticity: Status quo bias and the International Energy Agency’s World Energy outlook,” J. Envir. Policy & Planning, 1–20 (2015).
127 L. Després, “Une planification écologique et sociale: un impératif!,” Actuel Marx 2019/1 (65), 103–118 (2019).
128 L. Després, “La transition écologique. Vers un capitalisme régulé ou vers un changement de système économique,” Van Lang A. Penser et Mettre en Oeuvre les Transitions Écologiques: Colloque, 2015 (Mare & Martin, 2018) pp. 55–73.
129 A. Delmas, “Planification écologique: La formule magique de Mélenchon,” Liberation, 8 février 2017.
130 M. Midena Gaël Giraud, pour une planification écologique, Forbes France, 15 janvier 2021. https://www.forbes.fr/politique/adr-gael-giraud-po-ur-une-planification-ecologique/.
131 Plihon D. La planification écologique: Une approche institutionnaliste Les Possibles, No. 2020 23 88 96
132 Keynes J. M. Collected Writings 1973 London Macmillan
133 P. Massé, Le Plan ou L’anti-hasard (Gallimard, NRF Idees, Paris, 1965).
134 M. Aglietta and E. Espagne, “L’‘ardente obligation’ de l’écologie politique,” L’Economie Politique, No. 89 (2021).
| 36466733 | PMC9707176 | NO-CC CODE | 2022-12-01 23:20:21 | no | Stud Russ Econ Dev. 2022 Nov 28; 33(6):583-597 | utf-8 | Stud Russ Econ Dev | 2,022 | 10.1134/S1075700722060120 | oa_other |
==== Front
Curr Psychiatry Rep
Curr Psychiatry Rep
Current Psychiatry Reports
1523-3812
1535-1645
Springer US New York
36445636
1398
10.1007/s11920-022-01398-1
Complex Medical-Psychiatric Issues (M Riba, Section Editor)
Violence Against Physicians in the Workplace: Trends, Causes, Consequences, and Strategies for Intervention
Caruso Rosangela 12
Toffanin Tommaso 1
Folesani Federica [email protected]
1
Biancosino Bruno 2
Romagnolo Francesca 1
Riba Michelle B. 3
McFarland Daniel 4
Palagini Laura 12
Belvederi Murri Martino 12
Zerbinati Luigi 12
Grassi Luigi 12
1 grid.8484.0 0000 0004 1757 2064 Institute of Psychiatry, Department of Neuroscience and Rehabilitation, University of Ferrara, Via Fossato di Mortara 64a, 44121 Ferrara, Italy
2 grid.416315.4 Integrated Department of Mental Health and Pathological Addictions, S. Anna University Hospital and Local Health Trust, Ferrara, Italy
3 grid.214458.e 0000000086837370 Department of Psychiatry, University of Michigan, Ann Arbor, MI USA
4 grid.415895.4 0000 0001 2215 7314 Department of Medicine, Northwell Health Cancer Institute, Lenox Hill Hospital, New York, NY USA
29 11 2022
114
7 10 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose
Violence against healthcare professionals has become an emergency in many countries. Literature in this area has mainly focused on nurses while there are less studies on physicians, whose alterations in mental health and burnout have been linked to higher rates of medical errors and poorer quality of care. We summarized peer-reviewed literature and examined the epidemiology, main causes, consequences, and areas of intervention associated with workplace violence perpetrated against physicians.
Recent Findings
We performed a review utilizing several databases, by including the most relevant studies in full journal articles investigating the problem. Workplace violence against doctors is a widespread phenomenon, present all over the world and related to a number of variables, including individual, socio-cultural, and contextual variables. During the COVID-19 pandemic, incidence of violence has increased. Data also show the possible consequences in physicians’ deterioration of quality of life, burnout, and traumatic stress which are linked to physical and mental health problems, which, in a domino effect, fall on patients’ quality of care.
Summary
Violence against doctors is an urgent global problem with consequences on an individual and societal level. This review highlights the need to undertake initiatives aimed at enhancing understanding, prevention, and management of workplace violence in healthcare settings.
Keywords
Violence
Physicians
Workplace
Burnout
Psychiatry
Emergency departments
==== Body
pmcIntroduction
The World Health Organization (WHO) defines physical and psychological workplace violence as “incidents where staff are abused, threatened or assaulted in circumstances related to their work, including commuting to and from work, involving an explicit or implicit challenge to their safety, well-being or health” [1].
Workplace violence is a worldwide problem with a higher incidence in sectors requiring intense human interactions, of which the healthcare sector is a prominent example. For the World Medical Association, violence against healthcare professionals is “an international emergency that undermines the very foundations of health systems and impacts critically on patient’s health” [2]. This is not a new phenomenon, as written in 1892 by the anonymous author of a landmark article on violence against doctors, which at that time seemed to be an infrequent phenomenon “[…] no physician, however conscientious or careful, can tell what day or hour he may not be the object of some undeserved attack, malicious accusation, black mail or suit for damages; ‘ for sufferance is the badge of all our race” [3].
During the last years, the number of violent acts against healthcare workers has incredibly increased [4••], although the real dimension is not known because of significant under-reporting [5•]. Recent research points out that as much as 48% of non-fatal workplace violence incidents take place in healthcare settings, and up to 50% of healthcare workers experience some form of violence in their career [6], especially but not only from patients with psychiatric disorders [7, 8•, 9].
In general, healthcare professionals (mainly nurses) have a 16 times increased risk of violence as compared to workers in other sectors and are four times more likely to require time away from work as a result of violence [10], with some more recent data concerning specifically violence against physicians.
While the majority of incidents are verbal, a considerable amount is constituted by physical assault, battery, stalking, or sexual harassment (Table 1). Also, a distinction should be made between affective violence (or defensive violence) which, according to several authors, mobilizes the emotional “fight or flight” response system, towards evading a stimulus that is perceived as a threat; while predatory violence (or intentional, premeditated violence) involves planning a deliberate attack on a target, often one who is the subject of some type of grievance [11].Table 1 Type of violence against physicians in the different settings
Physical violence
Physical assaults (e.g., open violent behavior, use of weapons)
Aggression (e.g., kicking, spitting, scratching, hitting, grabbing, biting, throwing objects)
Sexual violence
Harassment
Assault
Intimidation
Verbal remarks
Psychological violence
Verbal abuse
Intimidation behaviour
Verbal threats
Ethnical and racial harassment
Reputation smearing
Mobbing
Bullying
Aggressions in the workplace have been associated with somatic injuries, but also with psychological consequences, such as burnout, post-traumatic stress disorder, depression, and anxiety. Literature in this area has mainly focused on experiences of violence against nurses (e.g., [12]), while there are less studies on physicians, in whom the secondary emotional distress and burnout are important [13, 14•] and have been linked to higher rates of medical errors and poorer quality of care [15•]. The figures of patient-initiated violence against doctors are globally impressively high, to the point that some authors begin to refer to it as a “viral epidemic” [16], if not a “pandemic” [17].
On this background, the main aim of this narrative review is to focus attention on workplace violence against physicians perpetrated by patients or families, which falls into the wider category of type II workplace violence, referred to as client-on-worker-violence [18]. We examined both the prevalence in different countries (and across specialties) to account for possible culture- or healthcare system-driven differences and prevention and management strategies. As a second aim, we dedicated a section to the current COVID-19 pandemic, to explore whether this unexpected emergency and the associated widespread disinformation and politicization has increased hostile reactions and influenced the patterns of patient-initiated violence against physicians.
Methods
We searched electronic databases of PubMed and Google Scholar, with various combinations of keywords, [(hospital * OR healthcare OR health* OR doctor * OR physician * OR surgeon *) AND (violence OR aggression * OR harassment)]. Sources of information used for literature search were PubMed, Cochrane Library, Excerpta Medica dataBASE (EMBASE), and PsychINFO. In addition, a manual search was conducted based on literature references. We focused on recent articles in English language from all over the world. Full-text review of the included studies was carried out, and data were extracted on study’s characteristics and outcomes. All the possible disagreements were resolved with consensus in the presence of senior authors (RC, LG).
Results
The analysis of the literature allowed us to extrapolate data that can be separated into the following: (1) epidemiological data regarding the prevalence of violence against physicians, (2) causes and risk factors for violence in these settings, (3) consequences of healthcare related violence, and (4) strategies to prevent or limit violence against physicians.
Epidemiology of Aggression and Violence Against Physicians
Analysis of rates of violence against doctors is limited by the variability of findings and inconsistences in the definition of violence categories (e.g., “verbal aggression,” “threats,” “physical assault,” “battery”), as well as heterogeneity of measures, which may compromise reliability among studies. Therefore, reports of workplace violence against physicians vary greatly, although there is a consensus of high prevalence in critical care services, where rates can reach over 85% [19]. This is especially true for emergency departments (EDs), intensive care units (ICUs), psychiatry, and mental health units (MHUs), which represent worldwide the most frequent theaters of patient-initiated violence [20–22] (Table 2).Table 2 Primary studies of prevalence of workplace violence in different areas of the world
Authors and year Country Healthcare setting Number of participants Main findings
American College of Emergence Physicians 2018 USA EDs > 3,500 physicians 47% physically assaulted; 60% of them in the previous year
Ferri et al. 2016 Italy General hospital 745 HCPs 45% exposed to violence; 12% of them physicians
Vorderwülbecke et al. 2015 Germany Primary care 831 physicians 91% exposed to violence in their career; 73% in the previous year
De Jager et al. 2019 Belgium Hospital and community 3726 physicians 37% exposed to violence in the previous year; 33% verbal, 30% psychological, 14% physical, 10% sexual
ONAM Workgroup 2018 Spain Private and public hospital and community 2419 physicians Of 2419 physicians reporting some form of violence, 51% were men. Primary care most interested area (54%). Public sector most affected (89%)
Jatic et al. 2019 Bosnia Herzegovina Primary care 558 HCPs
181 physicians
Overall violence prevalence: 90.3%
Physicians more exposed than nurses to indirect physical violence
Sharma et al. 2019 India Tertiary care hospital 295 HCPs 53% of junior residents and 61% senior residents exposed to verbal violence
Physical violence mostly against residents with less than 5-year experience
Jain et al. 2021 India General hospital 307 residents 86% exposed to violence. 94% of episodes: verbal
Most frequent reason: patient’s death
Kaur et al. 2019 India Hospital and community 617 physicians 77% exposed to violence in their career
Singh et al. 2019 Uttar Pradesh General hospital 305 residents 69.5% exposed to violence; 70% verbal; 47.2% physical; 20% threats
Ahmed et al. 2017 Pakistan Primary care 524 physicians 85% exposed to mild, 62% to moderate, 38% to severe violence in the previous year
Fang et al. 2020 Northern China General hospital 884 physicians and 537 residents 73% of physicians and 24.8% of residents exposed to psychological violence; 10.9% of physicians and 1.5% of residents exposed to physical violence in previous year
Tian et al. 2020 China General hospital 934 physicians 16% exposed to physical violence; 50% to emotional violence; 30% to threats; 19.1 to verbal sexual harassment; 7.8% to sexual assault
Cheung et al. 2020 Macau Hospital and community 107 physicians 38.3 exposed to verbal violence; 3.7 to physical violence; 12.1% to bullying; 3.7 to sexual harassment; 3.7 to racial harassment
Kaya et al. 2016 Turkey EDs 112 physicians 76.6% exposed to violence
Oğuz et al. 2020 Turkey Paediatric clinic 75 physicians 56% exposed to violence in the previous year. 8.8% incidents physical. Physicians more exposed than other HCPs
Çevik et al. 2020 Turkey Hospital and community 948 physicians 83% exposed to violence in their career. 72% verbal
Hamdam et al. 2015 Palestine EDs 142 physicians 88% exposed to verbal violence and 29% to physical violence in the previous year
Nevo et al. 2017 Israel Hospital and community 145 physicians 59% exposed to verbal violence and 9% to physical violence in the previous year
Shafran-Tivka et al. 2017 Israel General hospital 230 physicians 27% exposed to verbal threats; 74% to other forms of verbal violence in the previous 6 months
Mohamad et al. 2021 Syria General hospital 1226 residents 85% exposed to violence in the previous year
Alhamad et al. 2021 Jordan General hospital 969 physicians 63% exposed to violence in the previous year. 62% verbal;6% physical; 32% non specified
Al Amazi et al. 2020 Saudi Arabia General hospital 351 HCPs 59% of physicians exposed to violence. Physicians the most exposed HCPs
Kasai et al. 2018 Myanmar General hospital 196 physicians 8.7% exposed to verbal violence and 1% to physical violence in the previous year
Elamin et al. 2021 Sudan General hospital 387 physicians 50.4% exposed to in the previous year. 92% verbal
Seun-Fadipe et al. 2019 Nigeria General hospital 99 physicians 44.4% exposed to violence in a 7-month time span
Akami et al. 2019 Nigeria Psychiatric hospital 30 physicians 29% exposed to physical violence in the previous year. 73% in their career
Olashore et al. 2018 Botswana Psychiatric hospital 10 physicians 40% exposed to violence in their career
EDs emergency departments, HCPs healthcare professionals, ONAM National Observatory of Aggression to Physicians
North America
In U.S. general hospitals, violence from patients constitutes a serious occupational hazard as confirmed by the Federal Bureau of Labor Statistics reports, outlining that between 2011 and 2013 nearly 75% of 24,000 workplace assaults per year occurred in the healthcare sector (6). These data, however, might reflect only the tip of the iceberg, as it has been found that the actual prevalence of workplace violent incidents was as much as three times that reported by the federal survey, since verbal incidents are not recorded [23]. As mentioned above, violent incidents increase further when focusing on high-exposure settings, including EDs, as confirmed by the American College of Emergency Physicians (5), and MHUs, where rates of workplace physical violence against physicians appear even higher than those in EDs, with 40% of psychiatrists reporting physical assaults over the course of their careers [24].
Europe
In Europe, violence rates against physicians are equally worrisome.
In Germany, data show that about 90% of 831 doctors had faced some form of aggression in their career and about 70% in the previous year [25]. A similar overall prevalence of violence was found among 558 primary healthcare professionals in Bosnia and Herzegovina. Compared to nurses, physicians were more often subjected to humiliation, false allegations, rumors, etc., and among physicians, women experienced verbal violence more often than their male counterparts [26].
In Italy, hospital violence is common [27], especially among psychiatrists, who have been reported at high risk for verbal aggression, injuries, threats with dangerous objects, stalking, and physical aggression [28]. In contrast, Italian ED physicians were reported to experience violence less frequently than other professionals [29]. The risk of death of doctors should be also considered. In a 30-year review of homicides of Italian doctors in the workplace, 21 were registered: in about half of the cases, the perpetrator was not affected by any mental disorders and the killer was one of the doctor’s patients; in about one-third, he/she was a patient’s relative, and in one-fifth, a patient at first consultation [30•].
In Spain, 2,419 episodes of different forms of aggression against physicians were registered during a 5-year period. The scant denunciation by physicians in Spain contributes, as for the other countries, to the underestimation of the real dimensions of the phenomenon [31].
Asia
In the developing countries, alarming incidents of all forms of violence against doctors have been reported in the last decades, to the point that in some states, safety of doctors has become a critical issue [32, 33]. Globally, India is the “leading country” regarding violence against physicians [34] with data indicating that up to 75% of doctors face some form of violence in their working life [35] and a higher prevalence among physicians with less than 5 years of experience [36–38]. Similar data were reported in Pakistan [39].
High prevalence rates of both physical and psychological violence against physicians have been documented in Chinese hospital settings, where aggressive episodes appear rising over time [40••], including murder of physicians [41], although accurate statistics are lacking [42•, 43–47]. China is also the theater of a culture-based phenomenon known as Yi Nao, literally defined as “health care disturbance,” which is perpetrated by gangs with a designated leader who threatens and assaults hospital professionals, damage equipment, and prevent normal medical activities. The most frequent aim of Yi Nao is to force hospitals to reduce medical care costs. Yi Nao has increased tenfold over the years with 118,000 Yi Nao incidents across the country just in 2015 [48••].
High rates of violence against physicians have been reported in Turkey (more than 70% of specialty physicians reported some form of violence) [49, 50]; in Palestine, on both West Bank and Gaza Strip EDs, especially among younger personnel [51]; in Israel [52]; in Jordan [53]; in Syria [54]; and in Iraq [55, 56]. Data showed however, as in other countries, that only 16.4% of assaulted workers reported the violent acts to the authority because of the perception of uselessness and fear of negative consequences [57].
Africa
In Africa, there is still paucity of research [58]. In a cross-sectional study carried out in 2020 in Sudan [59], 50% of doctors reported they had been victims of some form of violence (mainly verbal violence and, as for other countries, among younger doctors). A Nigerian study showed similar findings among physicians in general [60] while the prevalence of physical violence against mental health professionals showed that 33% of the doctors report physical assaults in the previous 12 months and 73% in their whole career [61]. Slightly lower but still significant figures were reported in Botswana [62].
Violence Against Physicians in the COVID-19 Period
After the WHO declaration of the novel coronavirus disease (COVID-19) secondary to the SARS-CoV2 virus, as an international public health emergency [63], data were presented regarding the high risk for frontline clinicians to be exposed to negative consequences, including infections, fatigue, anxiety, depression, emotional exhaustion, burnout, and also workplace violence [64, 65, 66•]. Although, initially, physicians and nurses were celebrated as heroes, there is evidence of increased violence against health professionals associated with the COVID-19 pandemic [67••].
This is not a new phenomenon, since historically, likelihood of attacks from patients and/or families becomes higher as clinicians need to implement unwelcome yet essential prevention and control measures, such as quarantining patients, or banning family visits, both of which may disrupt communications between staff and patients/families.
Studies carried out in different countries, such as Brazil [68], Egypt [69], Jordan [70], and Canada [71], showed that key correlates of COVID-19-related workplace violence (including being beaten, threatened, and verbally offended), along with working in MHUs or EDs, were the following: being involved in direct care of infected patients; having infected family/friends/colleagues; not having children or partners; and a high workload. Similar data were reported in India and Pakistan [72, 73] Iraq [74], Mexico [75], and Peru [76], but not in Israel where at least at the beginning of the pandemic, there was a general decrease of violent incidents in comparison to the corresponding period in 2019 [77].
According to the World Medical Association, the figures of violence vary from country to country. [78]. However, although violence rates are higher in conflict zones and developing countries, the phenomenon does not spare high-income, industrialized, and peaceful countries, such as France, the UK, Australia, and the USA [79, 80].
Causes and Consequences of Violence Against Physicians
The development and implementation of effective workplace violence prevention programs that account for risks faced by physicians require the assessment of causes, risk factors, and consequences of violence against physicians.
Causes
There are several causes and risk factors favoring violence against physicians. Some include professional and individual physician-related factors; some are related to communication and doctor-patient relationship issues; some are related to the patients and their personal or clinical condition; finally, a complex intertwining of social, organizational, and political factors should be also considered [81, 82•] (Table 3).Table 3 Factors increasing the risk of occupational violence in health care settings (from Kumari et al., modified)
Workplace and policy issues Patient factors Physicians factors Doctor-patient relationship Sociocultural issues
-Poor demarcation between staff-only area and patient area
-Overcrowded, uncomfortable, or noisy waiting rooms
-Poor access to exits, poor lighting, blind spots without surveillance
-Unsecured furnishings which could be used as weapons
-Poor policies and work practices including understaffing, cutting staff resources, no investment in staff training, or development of guidelines
-Increased bureaucracy
-Medical causes (e.g., delirium, intoxication, substance abuse)
-Psychosocial stressors
-Previous negative experiences with healthcare
-History of violence
-Psychiatric disturbances including personality disorders
-Interpersonal style of control or dominance
-Poor impulse control
-Family conflicts
-Lack of staff training in communication skills, treatment of conditions associated with violent behavior, de-escalation techniques
-Working alone
-Limited experience
-Poor control on one’s own emotions (e.g., anger, anxiety, frustration)
-Emotional exhaustion, burnout, or psychological distress symptoms
-Patients: frustration, perception of not being respected, not being listened to, or being treated unfairly
-Increased waiting times
-Physicians: detachment and interpersonal frictions
-Poor customer services
- Poverty, unemployment, and social marginalization
- Language barriers
- Cultural differences
- Reduced respect for authority, negative media messages
- Population density, especially in metropolitan areas
- Violence as a way to receive attention
Working with health authorities to implement policies and guidelines for the prevention and management of violence (e.g., education programs, specific procedures, and guidelines)
Among physician-related factors, lack of communication skills and empathy, unkindness, and negative or hurtful comments clearly predispose the patient’s reaction. Also, physician younger age, inexperience, gender (i.e., female), and not knowing how to de-escalate and when and how to escape are part of the problem. Regarding the interpersonal doctor-patient relationship, dissatisfaction with prescriptions and treatment methods, disagreement with doctors, and unawareness of own body language are frequent initiators of violence.
Among patient factors, personality traits predisposing to aggression, history of violence, the influence of drugs and alcohol, confusion states, high levels of stress, and accumulation of negative life events reduce the threshold to aggression. Also, long waiting time for diagnostic examinations, treatments, and physician consultation may increase dissatisfaction, with risk of irritability, aggressivity, and, eventually, escalation to violent behaviors.
As far as organizational factors are concerned, certain settings (as said, EDs, MHUs), lack of resources and of staffing (e.g., staff reduction for budget reasons), and poor workplace conditions (e.g., architectural design of location, uncomfortable physical conditions of the rooms) are correlated to an increased risk of violence [83]. Also, action on the hospital administration to solve several problems (e.g., lack of support for staff, reduced interest for the staff mental health, burnout, and other psychological job-related complications; lack of violence-prevention programs, staff empowerment, and shared governance) is important, although data about the efficacy of intervention are contrasting not showing [84•] or showing [85•] positive effects.
A special topic has to do with “political” factors that can create conflicts in the population through confusing messages. From one side, politicians (and political parties) tend to announce and proclaim the effectiveness and efficacy of the healthcare system and the powerful investment in public health; on the other side, there is no country where the funds to the healthcare system are cut by the governments, with reduction of resources and a negative influence of on the quality of care. This can cause the request by citizens to have a perfect service fixing all problems, while in reality, what they have to face is characterized by limits and the deficits of the system itself (e.g., long waiting lists to receive attention; burned-out staff).
A further political implication has been made evident in COVID time when the public policy stances taken by healthcare organizations, including physicians or groups of physicians, can place physicians into the crosshairs of highly charged, divisive, controversial political struggles and in turn can increase the likelihood of affective and predatory violence. Specific examples of “politically motivated” or “politically related” violence against physicians might include the following:Escalating confrontations over masks, vaccination status, arguments with patients demanding care using medications that lack scientific support for treatment of COVID, and refusal of acute care for COVID due to patient’s disbelief in COVID
Physician burnout in the face of dealing with the front-line consequences of these political divides, leading to diminished reserve and greater risk of arguments with patients resulting in affective violence
Increased targeting of physicians who perform abortions in the wake of increased controversy about legal changes related to abortion rights, as has recently occurred in the USA
Increased targeting of physicians working in gender reassignment in the context of increased controversy about gender identity issues
Lastly, culturally and socially, a negative attitude towards healthcare providers has to do with a social stance of unlimited expectations of health and longevity and the implicit belief that technologic progress has given to medical science a curative solution for all the health problems. As a result, both individual patients and society may present a splitting mechanism towards doctors, with an intense de-idealization when the reality is that invulnerability and immortality are not part of the world and that a number of health problems cannot be “fixed” (according to the often forgotten message that medicine, as the science of the human, cures sometimes, relieves often, but should comforts always) [86].
Consequences
Violence incidents bear consequences that go further than the moment of exposure, both on a physical and, sometimes most importantly, on a psychological and moral basis, with a potential influence not only on the individual but also on their family and social relationships (spillover effect). In addition, repeated episodes of aggression, or a major traumatic one, may dramatically erode patient-physician trust and lead to poorer health outcomes [87•, 88] (Fig. 1).
Fig. 1 Consequences of exposure to violence incidents
In this regard, loss of occupational performance has been reported as the most common result of exposure to violence, with changes in attitudes and decision making and increase of defensive medicine behavior (e.g., excessive prescriptions of drugs and investigations, inappropriate referrals, and consultation requests) [89, 90].
Solid data document that physicians who faced violence are at a great risk to develop psychological problems, such as depression, insomnia, post-traumatic stress symptoms with flashbacks, avoidance and hypervigilance, intense fear episodes, and anxiety, leading to deterioration of quality of life, absenteeism, and, in extreme circumstances, decision to quit the medical profession. Loss of self-esteem and feelings of shame, along with a sense of lowered safety and defeat, can occur as long-term sequelae of cumulative minor violence exposures or of a major dramatic episode. These consequences have been shown significantly higher in cases of physical violence or sexual harassment than in cases of verbal abuse and more in women than in men [91]. Both verbal and physical violence, however, impacted on subjective sleep quality, tobacco consumption, headache, eating disorders, and lifestyle deterioration, with psychological distress partially mediating the relationship between work-related violence and health damage [92, 93•, 94].
Burnout symptoms (i.e., emotional exhaustion, depersonalization and detachment in the relationship with patients, demotivation, and loss of job satisfaction) are statistically more frequent in physicians subjected to verbal and physical violence [95•, 96].
A further form of psychological reaction in physicians’ victims of violence could be embitterment and its psychopathological form, defined post-traumatic embitterment disorder (PTED) [97••]. Unlike PTSD, PTED is characterized by the fact that the event is experienced as unjust, as a personal insult, and as a violation of basic beliefs and values, with a predominant sense of embitterment and emotional arousal when reminded of the event, as the key core psychological components of the disorder. Also, mood impairment; downheartedness unspecific somatic complaints; phobic symptoms; social withdrawal; reduction of energy, motivation, and drive; and increase in irritability with possible aggression towards oneself and others can be present While studies have been conducted about these clinical conditions in the workplace, more data are needed regarding embitterment and PTED secondary to violence against physicians [98].
Prevention and Management of Workplace Violence in Healthcare Setting
Although most incidents can be avoidable, studies on violence against physicians have been designed to quantify the problem, and few have described experimental methods to prevent such violence. Three recent critical reviews of the literature documented moderate evidence that integrated prevention programs might decrease the risks of patient-initiated violence [99•]. These programs incorporate at least two different areas of intervention, based on the most frequent causes of violence, including both organizational and the healthcare staff levels (Table 4).Table 4 Possible interventions for physicians exposed to violence
Training and education on violence prevention (e.g., communication skills, non-violent crisis interventions, de-escalation techniques) and treatment (e.g., use of tranquilizers, restrains) tailored for physicians
Correct attention to high-risk patients (e.g., patients with drug or substance intoxication, confusion); correct interpretation of signs of aggressivity
Enhanced security measures in the clinical settings (e.g., help or panic buttons, direct telephone lines to security and police), safety environment (e.g., lighting, mirrors, cameras)
Development of safety standards within healthcare facilities (e.g., protocols, mandatory reporting of events, zero tolerance of violence (violence is not part of the job!)
Timely response after violence against physicians (e.g., debriefing, psychological assessment, and/or psychiatric treatment)
First Level of Intervention
A first level of action concerns the organizational area, from simple interventions to more specific ones. Excessive waiting time for care and clients’ disinformation about the reasons for waiting has been documented as a key factor kindling violence, along with lack of violence prevention measures, and lack of drugs or needed services is responsible for at least 15% episodes. Therefore, leaflets, clear information about the possible problems in the daily organization, and attention to comfort are part of these procedures. Hospitals which offer more comfortable physical conditions (e.g., private rooms with an air-conditioning system, or television) show lower rates of patient-initiated violence towards physicians (violence inhibitors). In facilities where procedures and culture of reporting are lacking, violence episodes occur significantly more frequently than in those where reporting systems are effective [100••, 101•]. The issue of under-reporting of violent incidents is extremely important and can be caused by several reasons which should be carefully examined in order to make the whole care system more aware. With respect to this, one reason for physicians and nurses to not report violence against them is because of previous experience of no action taken (e.g., administration tending to minimize these phenomena to protect the image of the institute; tendency to accept that violence is part nurses’ and physicians’ work environment); fear of the consequences (e.g., further threats or wish to revenge from the perpetrator if legal reports are made, worry or guilt feelings to have done mistakes in the relationship with the patient); and lack of management support (e.g., tendency to consider this as a problem of the single individual as a scapegoat, rather than a problem of the system) [15•]. Prompting healthcare professionals to always report incidents in a non-judgmental way and disseminating a policy based on protecting the staff (e.g., “violence is not part of the job” initiative in the USA) [6] increase the chances to make violence prevention programs more effective. More specific infrastructure changes, including high-security systems, cameras, alarm systems, improved lighting, rapid access to police, or hospital security, are a necessary part of the organization of healthcare systems, both in clinics and hospitals.
At organizational level, it is also necessary to consider changes in politics, culture, and society, in both the hospital and the media communication, which plays a key role in the public perception of the healthcare system. Political factors, such as the presence or absence of an effective public healthcare system, are another variable at stake: patients’ and families’ heightened anxiety about the disease and finance difficulties in sustaining the cost of healthcare are an important component of violence initiation.
Second Level of Intervention
A second area of intervention targets physician-related risk factors of violence and aims at promoting behavioral changes. In general, an inappropriate staff attitude can account for up to 20% of cumulative rates of physical and verbal aggression episodes in ICUs [102, 103]. Tailored training initiatives in self-awareness, communications skills, de-escalation techniques, improvement of staff-patient, and inter-staff relationships are necessary. After a violent aggression, debriefing, regular follow-up sessions, and psychological support should be performed with the aim to acknowledge and reassure not only the victim, but also the entire staff. A tense and unempathetic working atmosphere, dissatisfaction, and feelings of distress or fear, burnout, and low-quality interactions with colleagues are, in fact, further risk factors of violence exposure in physicians. The recognition of the difference between interventions to prevent or mitigate acts of affective violence versus predatory violence is also important.. Therefore, in the healthcare system, in the prevention of predatory violence, one of the most important concepts is that individuals who perpetrate such violence usually exhibit a variety of behavioral warning signs amidst numerous risk factors for targeted violence in the lead up to the ultimate act of violence. Thus, the capacity to recognize warning signs of violence and how to deal with should be part of training. Of course, an adequate number of staff personnel should be warranted, since not only the quality of training but also the necessary number of resources is important if a correct and complete response to patients’ needs is the aim of the system.
Last, as previously said, training and intervention should focus on the tendency of physicians not to report the aggression. Being victims of violence (therefore a bad act) when doing one’s own profession of care (therefore a good act) is a cause of sense of injustice that can cause not only burnout (i.e., emotional exhaustion, poor personal accomplishment, detachment towards patients) but demoralization and embitterment. Programs revolving around a “zero tolerance” policy that seeks to share a workplace culture of awareness and safety, “see something, say something – then we can do something [to stop something bad from happening]”, will help the doctor to work through his/her possible sense of guilt or shame and victimization.
Discussion
This review examined violence against physicians from a multifaceted point of view, highlighting the complexity of a phenomenon whose roots are to be found in intertwined cultural, societal, and organizational factors. A worldwide widespread and disturbing pattern of violence towards physicians is in fact occurring, and the medical profession, dignified and valued in the past, seems to be losing not only its sacredness, but even its reputation [104].
Finding a solution to invert this trend is not an easy task, and the few studies that have focused on interventions to reduce violence against physicians have highlighted the unlikelihood of finding a simple, one-size-fits-all approach, since the problem is multidimensional and extremely complex.
A major obstacle appears to be the underestimation of the phenomenon. As said, a number of doctors subjected to violence do not report the incident, and a high percentage of those who report decide to not proceed nor undertake any legal action, frequently considering any acting useless.
The absence of standard procedures for reporting violence and the lack of any encouragement to disclose violence events play a further deterrent role. Many hospitals have neither structured policies/procedure to prevent and manage violent abuses by patients, nor comprehensive training programs on workplace violence. In this light, physicians’ safety appears to assume a lower priority compared with patient safety, and even when present, programs are designed and conceived for patient safety—not worker safety—as the primary goal.
In the paucity of data that define effective steps to prevent violence against physicians, approaches to the problem should be undertaken at various levels. Legislators should consider effective consequences for violence against healthcare workers as a special class of offense; incident reporting procedures should be adopted, and physicians are encouraged to overcome fear of stigmatization and violence acceptance culture. Training programs and the implementation of cost-effective, evidence-based solutions should be ensured.
Our review shows some limitations. First, results presented here are drawn from a literature synthesis and therefore share the limits of the original research, such as the risk of under-reporting, the lack of homogeneity in the definition of violence, and the fact that nearly every study was based on voluntary retrospective surveys, an approach that risks both selection and recall bias. Another important element concerns the subjectivity of physicians, as the interpretation of the episodes of aggression can vary among the victims. A further consideration is that there is not a unique tool (e.g., questionnaire, interview, checklist) used to evaluate the phenomenon in the included studies. Thus, while the impact of violence can be measured by using traditional symptomatic scales to measure depression, anxiety, PTSD, or PTED, the development of specific measures for violence is necessary [105].
Conclusion
Violence against physicians is an urgent global problem with consequences on an individual and societal level. Physicians’ deterioration of quality of life, burnout, and traumatic stress are linked to physical and mental health problems, with poor job performance, low organizational commitment, and turnover intentions which, in a domino effect, fall on patients’ quality of care.
In this light, it appears crucial that, like all other workers, healthcare workers have the right to safety at work acknowledged. Further research should be particularly focused on detecting evidence-based strategies of the prevention and management of violence against physicians under a multidisciplinary perspective.
Acknowledgements
The editors would like to thank Dr. Philip Saragoza for taking the time to review this manuscript.
Declarations
Conflict of Interest
Rosangela Caruso, Tommaso Toffanin, Federica Folesani, Bruno Biancosino, Francesca Romagnolo, Michelle B. Riba, Daniel McFarland, Laura Palagini, Martino Belvederi Murri, Luigi Zerbinati, and Luigi Grassieach declare no potential conflicts of interest. Michelle Riba is an editor of Current Psychiatry Reports but did not review this article or make editorial suggestions regarding accepting or rejecting this article. She receives royalties or a stipend from Springer, Guilford, Cambridge, American Psychiatric Press Inc., and Current Psychiatry Reports. Luigi Grassi has received royalties for books from Springer and personal fees from EISAI and ANGELINI for scientific consultancy outside the submitted work.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
This article is part of the Topical Collection on Complex Medical-Psychiatric Issues
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
1. WHO Violence against health workers. 2020. https://www.who.int/violence_injury_prevention/violence/workplace/en/
2. World Medical Association. 73rd World Health Assembly, Agenda Item 3: Covid-19 pandemic response. (2020). Available online at: https://www.wma.net/wp-content/uploads/2020/05/WHA73-WMA-statement-on-Covid-19-pandemic-response-.pdf. Accessed 7 Jun 2020
3. Letter to the editor: assaults upon Medical Men’ JAMA 1892; 18:399–400.
4. •• Nowrouzi-Kia B, Chai E, Usuba K, Nowrouzi-Kia B CJ. Prevalence of type ii and type iii workplace violence against physicians: a systematic review and meta-analysis. Int J Occup Environ Med NIOC Heal Organ. 2019;10:99–110. This paper reviews and meta-analyzes prevalence of violence perpetrated against physicians.
5. • Phillips JP. Workplace violence against health care workers in the United States. N Engl J Med. 2016 Apr 28;374(17):1661–9. The review focuses on the current knowledge about workplace violence in various US health care settings, including the prevalence across professions, potential risk factors, and the use of metal detectors in preventing violence.
6. U.S. Department of Labor Occupational Safety and Health Administration. Guidelines for preventing workplace violence for healthcare and social service workers (osha.gov). OSHA 3148-06R 2016. https://www.osha.gov/sites/default/files/publications/osha3148.pdf
7. Odes R Chapman S Harrison R Ackerman S Hong O Frequency of violence towards healthcare workers in the United States' inpatient psychiatric hospitals: a systematic review of literature Int J Ment Health Nurs 2021 30 1 27 46 33150644
8. • Caruso R, Antenora F, Riba M, Murri MB, Biancosino B, Zerbinati L, et al. Aggressive behavior and psychiatric inpatients: a narrative review of the literature with a focus on the european experience. Curr Psychiatr Rep. 2021 Apr 7;23(5):29. This is a review specifically aiming at illustrating the risk of violence in the general hospital by patients with psychiatric disorder.
9. Weltens I, Bak M, Verhagen S, Vandenberk E, Domen P, van Amelsvoort T, Drukker M. Aggression on the psychiatric ward: prevalence and risk factors. A systematic review of the literature. PLoS One. 2021 Oct 8;16(10): e0258346.
10. US Bureau of Labor Statistics. Census of fatal occupational injuries (CFOI). current and revised data. Washington, DC. https://www.bls.gov/iif/oshcfoi1.htm
11. McEllistrem JE Affective and predatory violence: a bimodal classification system of human aggression and violence Aggress Violent Beh 2004 10 1 30 10.1016/j.avb.2003.06.002
12. Pariona-Cabrera P, Cavanagh J, Bartram T. Workplace violence against nurses in health care and the role of human resource management: a systematic review of the literature. J Adv Nurs. 2020. 11 2020 Jul;76(7):1581–1593
13. Duan X Ni X Shi L Zhang L Ye Y Mu H Li Z Liu X Fan L Wang Y The impact of workplace violence on job satisfaction, job burnout, and turnover intention: the mediating role of social support Health Qual Life Outcomes 2019 17 1 93 10.1186/s12955-019-1164-3 31146735
14. • Mento C, Silvestri MC, Bruno A, Muscatello MRA, Cedro C, Pandolfo G, Zoccali RA. Workplace violence against healthcare professionals: a systematic review. Aggress. Violent Behav. 2020;51:101381. This paper shows that workplace violence might lead to various negative impacts on health workers’ psychological and physical health, such as increase in stress and anxiety levels and feelings of anger, guilty, insecurity, and burnout.
15. • Chakraborty S, Mashreky SR, Dalal K. Violence against physicians and nurses: a systematic literature review. J Publ Health (Berl.). 2022;1–19. 10.1007/s10389-021-01689-6. The paper summarizes the findings of a large series of studies on violence showing that adequate policy making and implementation and operational research are required to further mitigate the episodes of violence especially in emergency departments.
16. Reddy IR, Ulkrani J, India V, Ulkrani V. Violence against doctors: a viral epidemic?. Ind J Psychiatr. 2019 Apr;61:S782–5.
17. Ghosh K Violence against doctors: a pandemic in the making Eur J Inter Med 2018 50 e9 e10 10.1016/j.ejim.2017.08.017
18. Pompeii L Dement J Schoenfisch A Perpetrator, worker and workplace characteristics associated with patient and visitor perpetrated violence (type II) on hospital workers: a review of the literature and existing occupational injury data J Safety Res 2013 44 57 64 10.1016/j.jsr.2012.09.004 23398706
19. Ashton RA Morris L Smith I A qualitative meta-synthesis of emergency department staff experiences of violence and aggression Int’l Emer Nurs 2018 39 13 19 10.1016/j.ienj.2017.12.004
20. Wuellner SE Bonauto DK Exploring the relationship between employer recordkeeping and underreporting in the BLS Survey of Occupational Injuries and Illnesses Am J Ind Med 2014 57 1133 1143 10.1002/ajim.22350 25099477
21. Kowalenko T, Gates D, Gillespie GL, Succop P, Mentzel TK. Prospective study of violence against ED workers. Am J Emerg Med. 2013;31:197–205.
22. Kumar NS Munta K Kumar JR Rao SM Dnyaneshwar M Harde Y A survey on workplace violence experienced by critical care physicians Indian J Crit Care Med 2019 23 7 295 301 10.5005/jp-journals-10071-23202 31406429
23. Rosenman KD Kalush A Reilly MJ Gardiner JC Reeves M Luo Z How much work-related injury and illness is missed by the current national surveillance system? J Occup Environ Med 2006 48 357 365 10.1097/01.jom.0000205864.81970.63 16607189
24. Kelly EL Subica AM Fulginiti A Brekke JS Novaco RW A cross-sectional survey of factors related to inpatient assault of staff in a forensic psychiatric hospital J Adv Nurs 2015 71 1110 1122 10.1111/jan.12609 25546118
25. Vorderwülbecke F Feistle M Mehring M Schneider A Linde K Aggression and violence against primary care physicians—a nationwide questionnaire survey Dtsch Arztebl Int 2015 112 10 159 165 25837740
26. Jatic Z Erkocevic H Trifunovic N Tatarevic E Keco A Sporisevic L Hasanovic E Frequency and forms of workplace violence in primary health care Med Arch 2019 73 1 6 10 10.5455/medarh.2019.73.6-10 31097851
27. Ferri P Silvestri M Artoni C Di Lorenzo R Workplace violence in different settings and among various health professionals in an Italian general hospital: a cross-sectional study Psychol Res Behav Manag 2016 9 263 275 10.2147/PRBM.S114870 27729818
28. Grassi L, Peron L, Marangoni C, Zanchi P, Vanni A. Characteristics of violent behaviour in acute psychiatric in-patients: a 5-year Italian study Acta Psychiatr Scand. 2001 Oct;104(4):273–9
29. Civilotti C Berlanda S Iozzino L Hospital-based healthcare workers victims of workplace violence in Italy: a scoping review Int J Environ Res Public Health 2021 18 11 5860 10.3390/ijerph18115860 34072551
30. • Lorettu L, Nivoli AMA, Daga I, Milia P, Depalmas C, Nivoli G, Bellizzi S. Six things to know about the homicides of doctors: a review of 30 years from Italy. BMC Publ Health. 2021 Jul 5;21(1):1318. The paper examines the risk of murder among physicians, with the conclusion that doctors should be aware that the risk of being killed is not limited to hospital settings and that their patients’ family members might also pose a threat to them.
31. National Observatory of Aggressions to Physicians (ONAM) Workgroup, General Council of Official Medical Associations of Spain (CGCOM). National report on aggressions to physicians in Spain 2010-2015: violence in the workplace-ecological study. BMC Res Notes. 2018 Jun 4;11(1):347.
32. Ranjan R, Singh M, Pal R, Das JK, Gupta S. Epidemiology of violence against medical practitioners in a developing country (2006–2017). 2018;5(3):153–160
33. Marwah A, Ranjan R, Singh M, Meeenakshi, Das JK, Pal R. Safety of doctors at their workplace in India perspectives and issues. Indian J Public Health Res Dev. 2018;9:139.
34. Ghosh K Violence against doctors: a wake-up call Indian J Med Res 2018 148 2 130 133 10.4103/ijmr.IJMR_1299_17 30381535
35. Singh M Intolerance and violence against doctors Indian J Pediatr 2017 84 10 768 773 10.1007/s12098-017-2435-9 28799002
36. Singh G Singh A Chaturvedi S Khan S Workplace violence against resident doctors: a multicentric study from government medical colleges of Uttar Pradesh Indian J Pub Health 2019 63 2 143 146 10.4103/ijph.IJPH_70_18 31219064
37. Sharma S, Gautam PL, Sharma S, Kaur A, Bhatia N, Singh G, Kaur P, Kumar A. Questionnaire-based evaluation of factors leading to patient-physician distrust and violence against healthcare workers. Indian J Crit Care Med. 2019 Jul;23:302–9.
38. Jain G Agarwal P Sharma D Agrawal V Yadav SK Workplace violence towards resident doctors in Indian teaching hospitals: a quantitative survey Trop Doct 2021 51 3 463 465 10.1177/00494755211010005 33906545
39. Ahmed F Khizar Memon M Memon S Violence against doctors, a serious concern for healthcare organizations to ponder about Ann Med Surg (Lond) 2017 15 25 3 5
40. •• Lu L, Dong M, Wang SB, Zhang L, Ng CH, Ungvari GS, Li J, Xiang YT. Prevalence of workplace violence against health-care professionals in China: a comprehensive meta-analysis of observational surveys. Trauma Violence Abuse. 2020 Jul;21(3):498–509. This is a comprehensive meta-analysis on prevalence of workplace violence among healthcare professionals in China included 47 studies and 81,771 professionals.
41. Zhang X, Li Y, Yang C, Jiang G. Trends in workplace violence involving health care professionals in China from 2000 to 2020: a review. Med Sci Monit. 2021 Jan 8;27:e928393
42. Li P Xing K Qiao H Fang H Ma H Jiao M Hao Y Li Y Liang L Gao L Kang Z Cui Y Sun H Wu Q Liu M Psychological violence against general practitioners and nurses in Chinese township hospitals: incidence and implications Health Qual Life Outcomes 2018 16 1 117 10.1186/s12955-018-0940-9 29871642
43. Hall BJ, Xiong P, Chang K, Yin M, Sui XR. Prevalence of medical workplace violence and the shortage of secondary and tertiary interventions among healthcare workers in China. J Epidemiol Community Health. 2018 Jun;72(6):516-518.
44. Fang H Wei L Mao J Jia H Li P Li Y Fu Y Zhao S Liu H Jiang K Jiao M Qiao H Wu Q Extent and risk factors of psychological violence towards physicians and Standardised Residency Training physicians: a Northern China experience Health Qual Life Outcomes 2020 18 1 330 10.1186/s12955-020-01574-y 33028344
45. Ma J Chen X Zheng Q Zhang Y Ming Z Wang D Wu H Ye H Zhou X Xu Y Li R Sheng X Fan F Yang Z Luo T Lu Y Deng Y Yang F Liu C Liu C Li X Serious workplace violence against healthcare providers in China between 2004 and 2018 Front Public Health 2021 15 8 574765 10.3389/fpubh.2020.574765
46. Tian Y Yue Y Wang J Luo T Li Y Zhou J Workplace violence against hospital healthcare workers in China: a national WeChat-based survey BMC Public Health 2020 20 1 582 10.1186/s12889-020-08708-3 32349727
47. Cheung T Lee PH Yip PSF Workplace violence toward physicians and nurses: prevalence and correlates in Macau Int J Environ Res Public Health 2017 14 8 879 10.3390/ijerph14080879 28777333
48. •• Liu T, Tan X. Troublemaking in hospitals: performed violence against the healthcare professions in China Health Sociol Rev. 2021 Jul;30(2):157–170. Very complete review of the phenomenon of Yi Nao in China as a form of violence against physicians and health workers in general.
49. Kaya S, Bilgin Demir İ, Karsavuran S, Ürek D, İlgün G. Violence against doctors and nurses in hospitals in Turkey. J Forensic Nurs. 2016 Jan-Mar;12(1):26-34.
50. Çevik M Gümüştakim RŞ Bilgili P Ayhan Başer D Doğaner A Saper SHK Violence in healthcare at a glance: the example of the Turkish physician Int J Health Plann Manage 2020 35 6 1559 1570 10.1002/hpm.3056 32996223
51. Hamdan M Abu HA Workplace violence towards workers in the emergency departments of Palestinian hospitals: a cross-sectional study Hum Resour Health 2015 7 13 28 10.1186/s12960-015-0018-2
52. Nevo T Peleg R Kaplan DM Freud T Manifestations of verbal and physical violence towards doctors: a comparison between hospital and community doctors BMC Health Serv Res 2019 19 1 888 10.1186/s12913-019-4700-2 31771580
53. Alhamad R, Suleiman A, Bsisu I, Santarisi A, Al Owaidat A, Sabri A, et al. Violence against physicians in Jordan: An analytical cross-sectional study. PLoS One. 2021;16(1): e0245192. 2
54. Mohamad O AlKhoury N Abdul-Baki MN Alsalkini M Shaaban R Workplace violence toward resident doctors in public hospitals of Syria: prevalence, psychological impact, and prevention strategies: a cross-sectional study Hum Resour Health 2021 19 1 8 10.1186/s12960-020-00548-x 33413470
55. Lafta RK Falah N Violence against health-care workers in a conflict affected city Med Confl Surviv 2019 35 1 65 79 10.1080/13623699.2018.1540095 30406677
56. Haar RJ, Read R, Fast L, Blanchet K, Rinaldi S, Taithe B, Wille C, Rubenstein LS. Violence against healthcare in conflict: a systematic review of the literature and agenda for future research. Confl Health. 2021 May 7;15(1):37.
57. Al Anazi RB AlQahtani SM Mohamad AE Hammad SM Khleif H Violence against health-care workers in governmental health facilities in Arar City, Saudi Arabia ScientificWorldJournal 2020 20 2020 6380281
58. Njaka S Edeogu OC Oko CC Goni MD Nkadi N Workplace violence (WPV) against healthcare workers in Africa: a systematic review Heliyon 2020 6 9 e04800 10.1016/j.heliyon.2020.e04800 32964153
59. Elamin MM, Hamza SB, Abbasher K, Idris KE, Abdallah YA, Muhmmed KAA, Alkabashi THM, Alhusseini RT Mohammed SES, Mustafa AAM. Workplace violence against doctors in Khartoum State, Sudan, 2020. Sudan J Med Sci. 2021;16(2):301–319.
60. Seun-Fadipe CT Akinsulore AA Oginni OA Workplace violence and risk for psychiatric morbidity among health workers in a tertiary health care setting in Nigeria: prevalence and correlates Psychiatry Res 2019 272 730 736 10.1016/j.psychres.2018.12.177 30832193
61. Akanni OO Osundina AF Olotu SO Agbonile IO Otakpor AN Fela-Thomas AL Prevalence, factors, and outcome of physical violence against mental health professionals at a Nigerian psychiatric hospital East Asian Arch Psychiatry 2019 29 1 15 19 10.12809/eaap1727 31237252
62. Olashore AA Akanni OO Ogundipe RM Physical violence against health staff by mentally ill patients at a psychiatric hospital in Botswana BMC Health Serv Res 2018 18 1 362 10.1186/s12913-018-3187-6 29751808
63. World Health Organization. The coronavirus disease (COVID-19) outbreak. 2020. Available online at: http://www.hoint.com. Accessed 30 Mar 2020
64. Lai J Ma S Wang Y Cai Z Hu J Wei N Factors associated with mental health outcomes among health care workers exposed to coronavirus disease 2019 JAMA Netw Open 2020 3 e203976 6
65. Zhao K Zhang G Feng R Wang W Xu D Liu Y Anxiety, depression and insomnia: a cross-sectional study of frontline staff fighting against COVID-19 in Wenzhou, China Psychiatry Res 2020 292 113304 4
66. • Devi S. COVID-19 exacerbates violence against health workers. Lancet. 2020;396(10252):658. 10.1016/S0140-6736(20)31858-4. This paper reports that hundreds of incidents of violence and harassment occurred in healthcare settings as a consequence of COVID-19.
67. •• Chew NWS, Lee GKH, Tan BYQ, et al. A multinational, multicentre study on the psychological outcomes and associated physical symptoms amongst healthcare workers during COVID-19 outbreak. Brain Behav Immun. 2020;S0889–1591:30523–30527. Study reporting the psychological and physical consequences of COVID-19 pandemic on healthcare workers.
68. Bitencourt MR, Alarcão ACJ, Silva LL, Dutra AdC, Caruzzo NM, Roszkowski I, et al. Predictors of violence against health professionals during the COVID-19 pandemic in Brazil: a crosssectional study. PLoS ONE 2021;16(6):e0253398.
69. Arafa A, Shehata A, Youssef M, Senosy S. Violence against healthcare workers during the COVID-19 pandemic: a cross-sectional study from Egypt. Arch Environ Occup Health. 2021 Sep 30:1–7. 0
70. Ghareeb NS El-Shafei DA Eladl AM Workplace violence among healthcare workers during COVID-19 pandemic in a Jordanian governmental hospital: the tip of the iceberg Environ Sci Pollut Res Int 2021 28 43 61441 61449 10.1007/s11356-021-15112-w 34173953
71. Cukier A Vogel L Escalating violence against health workers prompts calls for action CMAJ 2021 193 49 E1896 10.1503/cmaj.1095980 34903595
72. McKay D Heisler M Mishori R Catton H Kloiber O Attacks against health-care personnel must stop, especially as the world fights COVID-19 Lancet 2020 395 10239 1743 1745 10.1016/S0140-6736(20)31191-0 32445692
73. Bhatti OA Rauf H Aziz N Martins RS Khan JA Violence against healthcare workers during the COVID-19 pandemic: a review of incidents from a lower-middle-income country Ann Glob Health 2021 87 1 41 10.5334/aogh.3203 33977084
74. Lafta R Qusay N Mary M Burnham G Violence against doctors in Iraq during the time of COVID-19 PLoS ONE 2021 16 8 e0254401 10.1371/journal.pone.0254401 34358239
75. Rodríguez-Bolaños R Cartujano-Barrera F Cartujano B Flores YN Cupertino AP Gallegos-Carrillo K The urgent need to address violence against health workers during the COVID-19 pandemic Med Care 2020 58 7 663 10.1097/MLR.0000000000001365 32520840
76. Muñoz Del Carpio-Toia A, Begazo Muñoz Del Carpio L, Mayta-Tristan P, Alarcón-Yaquetto DE, Málaga G. Workplace violence against physicians treating COVID-19 patients in Peru: a cross-sectional study. Jt Comm J Qual Patient Saf. 2021 Oct;47(10):637-645
77. Basis F Moskovitz K Tzafrir S Did the events following the COVID-19 outbreak influence the incidents of violence against hospital staff? Isr J Health Policy Res 2021 10 1 36 10.1186/s13584-021-00471-z 34140034
78. World Medical Association Condemns Attacks on Health Care Professionals. 2020. Available online at: https://www.wma.net/news-post/world-medicalassociation-condemns-attacks-on-health-care-professionals/. Accessed 7 Jun 2020.
79. McKay D, Heisler M, Mishori R, Catton H, Kloiber O. Attacks against health-care personnel must stop, especially as the world fights COVID-19. Lancet. 2020;395:1743–5.
80. Alsuliman T Mouki A Mohamad O Prevalence of abuse against frontline health-care workers during the COVID-19 pandemic in low and middle-income countries East Mediterr Health J 2021 27 5 441 442 10.26719/2021.27.5.441 34080671
81. Shafran-Tikva S Chinitz D Stern Z Feder-Bubis P Violence against physicians and nurses in a hospital: how does it happen? A mixed-methods study Isr J Health Policy Res 2017 6 1 59 10.1186/s13584-017-0183-y 29089061
82. • Kumari A, Kaur T, Ranjan P, Chopra S, Sarkar S, Baitha U. Workplace violence against doctors: characteristics, risk factors, and mitigation strategies J Postgrad Med. 2020 Jul-Sep;66(3):149–154. This is a comprehensive review of workplace violence against doctors, discussing the prevalence, degree of violence, predictors, impact on physical and psychological health, and intervention strategies to devise practical actions against workplace violence.
83. Spelten E, Thomas B, O'Meara PF, Maguire BJ, FitzGerald D, Begg SJ. Organisational interventions for preventing and minimising aggression directed towards healthcare workers by patients and patient advocates. Cochrane Database Syst Rev. 2020 Apr 29;4(4):CD012662.
84. • Raveel A, Schoenmakers B. Interventions to prevent aggression against doctors: a systematic review. BMJ Open. 2019 Sep 17;9(9):e028465. This review failed to gather sufficient numerical data to perform a meta-analysis. A large-scale cohort study would add to a better understanding of the effectiveness of interventions.
85. • Wirth T, Peters C, Nienhaus A, Schablon A. Interventions for workplace violence prevention in emergency departments: a systematic review. Int J Environ Res Public Health. 2021 Aug 10;18(16):8459. The paper examines the role of intervention in reducing violence against physicians, and it shows a positive effect in the form of a reduction of violent incidents or an improvement in how prepared staff were to deal with violent situations
86. Gordon J. Medical humanities: to cure sometimes, to relieve often, to comfort always. MJA. 2005;182:5–8. 10.5694/j.1326-5377.2005.tb06543.x
87. • Kaur A, Ahamed F, Sengupta P, Majhi J, Ghosh T. Pattern of workplace violence against doctors practising modern medicine and the subsequent impact on patient care, in India. PLoS One. 2020 Sep 18;15(9):e0239193. The paper indicates that workplace violence has a significant effect on the psycho-social well-being of doctors, as well as on patient management.
88. Vento S, Cainelli F, Vallone A. Violence against healthcare workers: a worldwide phenomenon with serious consequences. Front Publ Health. 2020;8:570459.
89. Mittal S Garg S Violence against doctors - an overview J Evolution Med Dent Sci 2017 6 2948–2751 7
90. Morganstein JC, West JC, Ursano RJ. Work-Associated Trauma. In: Brower KJ, Riba MB (Eds) Physician mental health and well-being research and practice. Cham, Switzerland: Springer; 2017. p. 33–60.
91. Geoffrion S Goncalves J Marchand A Boyer R Marchand A Corbière M Guay S Post-traumatic reactions and their predictors among workers who experienced serious violent acts: are there sex differences? Ann Work Expo Health 2018 62 4 465 474 10.1093/annweh/wxy011 29548023
92. Lever I Dyball D Greenberg N Stevelink SAM Health consequences of bullying in the healthcare workplace: a systematic review J Adv Nurs 2019 75 12 3195 3209 10.1111/jan.13986 30816567
93. • Magnavita N, Di Stasio E, Capitanelli I, Lops EA, Chirico F, Garbarino S. Sleep problems and workplace violence: a systematic review and meta-analysis. Front Neurosci. 2019 Oct 1;13:997. This review indicates that occupational exposure to physical, verbal, or sexual violence is associated with sleep problems.
94. Cannavò M, La Torre F, Sestili C, La Torre G, Fioravanti M. Work related violence as a predictor of stress and correlated disorders in emergency department healthcare professionals. Clin Ter. 2019 Mar-Apr;170(2): e110-e123.
95. • Hacer TY, Ali A. Burnout in physicians who are exposed to workplace violence. J Forensic Leg Med. 2020 Jan;69:101874. Research study showing that doctors who were exposed to violence at work were exposed to verbal and psychological violence more than physical violence and especially psychological violence had a significant negative effect on burnout.
96. Shi J Wang S Zhou P Shi L Zhang Y Bai F Xue D Zhang X The frequency of patient-initiated violence and its psychological impact on physicians in China: a cross-sectional study PLoS ONE 2015 10 6 e0128394 10.1371/journal.pone.0128394 26030143
97. •• Linden M, Arnold CP. Embitterment and posttraumatic embitterment disorder (PTED): an old, frequent, and still underrecognized problem Psychother Psychosom. 2021;90(2):73-80. The paper analyzes the history of post-traumatic embitterment disorder as a consequence of traumatic events, including those happening in the workplace.
98. Linden M, Arnold CP. Embitterment in the Workplace. In: Grassi L, McFarland D, Riba MB, editors. Depression, burnout and suicide in physicians. Cham, Switzerland: Springer; 2021. p. 137–146.
99. • Lozano JM, Martínez Ramón JP, Morales Rodríguez FM. Doctors and nurses: a systematic review of the risk and protective factors in workplace violence and burnout. Int J Environ Res Public Health. 2021 Mar 22;18(6):3280. The review indicates that risk factors for workplace violence in healthcare sector are structural, organizational, and personal (e.g., age, gender) while protective factors are the quality of the working environment, mutual support networks, or coping strategies.
100. •• Aljohani B, Burkholder J, Tran QK, Chen C, Beisenova K, Pourmand A. Workplace violence in the emergency department: a systematic review and meta-analysis. Publ Health. 2021 Jul;196:186–197. This paper reviews and meta-analyzes ED workplace violence.
101. • Maguire BJ, O'Meara P, O'Neill BJ, Brightwell R. Violence against emergency medical services personnel: a systematic review of the literature. Am J Ind Med. 2018 Feb;61(2):167–180. This review underlines that emergency unit leaders and personnel should work together with researchers to design, implement, evaluate, and publish intervention studies designed to mitigate risks of violence to EMS personnel.
102. Udoji MA, Ifeanyi-Pillette IC, Miller TR, Lin DM. Workplace Violence against anesthesiologists: we are not immune to this patient safety threat. Int Anesthesiol Clin. 2019;57(3):123–137.
103. Harthi M Olayan M Abugad H Abdel WM Workplace violence among health-care workers in emergency departments of public hospitals in Dammam Saudi Arabia East Mediterr Health J 2020 26 12 1473 1481 10.26719/emhj.20.069 33355386
104. Grassi L, Daniel McFarland D, Riba MB. Medical professionalism and physician dignity: are we at risk of losing it? In: Grassi L, Daniel McFarland D, Riba MB, editors. Depression, Burnout and Suicide in Physicians. Berlin: Springer; 2022. p. 11–25.
105. Kumari A Singh A Ranjan P Sarkar S Kaur T Upadhyay AD Verma K Kappagantu V Mohan A Baitha U Development and validation of a questionnaire to evaluate workplace violence in healthcare settings Cureus 2021 13 11 e19959 34976540
| 36445636 | PMC9707179 | NO-CC CODE | 2022-12-01 23:21:38 | no | Curr Psychiatry Rep. 2022 Nov 29;:1-14 | utf-8 | Curr Psychiatry Rep | 2,022 | 10.1007/s11920-022-01398-1 | oa_other |
==== Front
Spat. Inf. Res.
Spatial Information Research
2366-3286
2366-3294
Springer Nature Singapore Singapore
499
10.1007/s41324-022-00499-6
Article
Rejuvenating impact of COVID-19 lockdown on major environmental parameters: an Indian perspective
Jha Deepak Kumar 1
http://orcid.org/0000-0001-9732-0251
Yashvardhini Niti [email protected]
2
Samiksha 2
Kumar Amit 3
1 Department of Zoology, S.M.P. Rajkiya Mahila Mahavidyalaya, Ballia, Uttar Pradesh 277401 India
2 grid.412457.1 0000 0001 1276 6626 Department of Microbiology, Patna Women’s College, Patna, Bihar 800001 India
3 grid.412457.1 0000 0001 1276 6626 Department of Botany, Patna University, Patna, Bihar 800005 India
29 11 2022
113
2 6 2022
10 11 2022
12 11 2022
© The Author(s), under exclusive licence to Korea Spatial Information Society 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The recent coronavirus outbreak caused severe impact on the life of people. Despite of several health and economic losses, COVID-19 pandemic induced lockdown proved to be a boon for the environment. This review highlights the positive impact of COVID-19 induced lockdown on the environment; enumerating its effect on air quality indices, water quality indices, wildlife and noise pollution, therefore, focussing on the brighter side of the effects of lockdown. Notably, in India, rivers like Ganga and Yamuna showed a drastic reduction in water pollutant levels. For the first time in a generation, the Himalayas were visible from various parts of India. The amount of waste generated also showed a decline during the lockdown, and wildlife was seen blooming. During the lockdown period temperature levels were also recorded low as compared to 2019 between March to June. Hence, this review emphasizes the beneficial impacts of lockdown on different pollution parameters as well as wildlife in India.
Keywords
COVID-19
Lockdown
Environment
Wildlife
Water quality indices
Air quality indices
==== Body
pmcIntroduction
Coronavirus disease (COVID-19) designated as 2019- nCoV was first reported in Wuhan, China on December 31, 2019 [1]. COVID-19 caused by SARS-CoV-2 belongs to the family Coronaviridae, and is an enclosed RNA virus having a crown-like structure [2, 3]. With the unprecedented rise in the number of COVID-19 infections across the globe, coronavirus outbreak has been declared as public health emergency of international concern by the World Health Organization (WHO) on 30th January 2020 (WHO, 2020) [4]. Nonetheless, 2019-nCoV is the third coronavirus that arouse in the human population, alarming the whole global public health institutions after the outbreak of SARS-CoV (Severe Acute Respiratory Syndrome Coronavirus) in 2002–2003 and MERS-CoV (Middle East Respiratory Syndrome Coronavirus) in the year 2012 [1].
To fight back and curtail the spread of infection, nationwide lockdown was introduced in India [5]. Lockdown is meant to curb the spread of infection from humans to humans and to establish the healthcare network adequately [6]. The outbreak of COVID-19 was catastrophic and the implementation of lockdown was undoubtedly a major step for the movement and control of infection which slowed down the transmission but led to the loss of economy, unemployment, loss of jobs and sources of livelihood; in turn causing devastation for the humans in an unpredictable way. COVID-19 lockdown brought anxiety and a sense of fear all around the globe [7]. The negative impacts of pandemic are tremendous and terrible but there are unexpected beneficial and positive impacts too. Global lockdown has drastically revitalized the environment in different ways [8]. Lockdown proved to be a reset button for the environment helping it to bounce back. Prior to COVID-19, pollution was the primary issue, and different organisations worked throughout the world to lower the pollution levels, particularly urban pollution in the form of carbon dioxide, nitrogen dioxide, sulphur dioxide, and particulate matter [4]. According to the report of WHO [20], almost 8% of total deaths in the world occurred due to air pollution. Profound impact on the environment has been seen right from the decline in air pollution, cleaner and healthier river water, decrease in carbon dioxide and aerosol level, dramatical drop in the level of nitrogen dioxide to clearer skies and improved climatic conditions. AOD (Aerosol Optical Length) values dropped over various regions of India and a significant reduction upto 50% was recorded over North-Central regions during the period of lockdown analyzed using MODIS satellite [5]. In crowded cities, noise pollution dropped below 60 db [9]. Even the generation of municipal solid waste (MSW) dropped considerably in India. Due to the reduction in pollution, the cases of jaundice, chikungunya, typhoid, respiratory disease, etc. also reduced in hospitals [10]. Restricted movement of human, vehicles and industrial production resulted in reduction in concentration of PM 10 and less man-made heat flux resulting from human movement caused declination in LST (Land Surface Temperature) [11]. Plants were seen growing better due to cleaner air and water and less human interference resulted in more coverage and oxygen.
As reported by Zambrano-Monserrate et al. (2020) [12], AQI (Air Quality Index) in North, South, East, Central and Western India showed sharp decline with 44, 33, 29, 15 and 32% respectively. The capital of India, Delhi saw the reduction of almost 70% in the levels of nitrogen dioxide and PM2.5 [13]. Rivers mainly Ganga and Yamuna in India were seen clearer and significant drop in the water pollution level [14]. The electricity usuage saw a steep decline due to lockdown in India; Maharashtra and Tamil Nadu showed a 20–30% declination in electricity demand [15]. The quantity of waste generated also came down during lockdown, the percentage of solid waste and paper waste went down by 42 and 62% respectively in major cities of India [16]. Due to lower human intrusion and pollution rates, sea turtles were seen laying eggs on the beaches [17]. The lockdown gave a rare opportunity to witness clean air, water, calm wildlife and liveable cities which created a very positive impact on the environment. Reports state that Dhauladhar range in Himachal Pradesh, India was visible from Jalandhar and Mt. Everest was visible from parts of Bihar during lockdown. As reported by Guha and Govil [18] values of LST (Land Surface Temperature), NDVI (Normalized Difference Vegetation Index) and LST- NDVI correlation in Raipur, Chhattisgarh (India) showed the environment of Raipur city during the lockdown compared to previous years. In this present review, we tried to highlight some positive impact of COVID-19 induced lockdown on the environment of India and suggest some strategies to reduce pollution.
Methods
In this review, we emphasized on the positive impacts of COVID-19 induced lockdown on three pollution parameters, air, water and noise as well as on the wildlife. This study was performed in context to India mainly, therefore, the first step was to extract articles on COVID-19 and environment. For this study, we searched in SCOPUS database with keywords COVID-19, lockdown, positive impact, water quality indices, and air quality indices. PubMed and Google Scholar databases were also used for searching articles related to impact of lockdown on environment of India. The articles since the onset of this disease (March 2020) till July 2021 were used as a reference in this study. The articles selected for this review as reference were studied thoroughly. Several lockdown policies and duration were introduced in India at regular intervals from time to time as shown in Table 1. These lockdown policies were imposed to reduce the rate of COVID-19 infection and in other way proved to be boon for the environment. The positive impact here signifies the significant reduction in air pollutants, water pollutants, solid waste and decrease in noise pollution as shown in Fig. 1. The data for satellite images of India for the concentration of different air pollutants like PM2.5, PM10, CO2, SO2 and CO were retrieved from https://earth.nullschool.net [19]. The AQI details were obtained from CPCB website for different cities of India before and after lockdown. The water quality parameters like BOD, pH and DO were retrieved from CPCB website.Table 1 Phases of lockdown in India, 2020
Phase Period Duration
Phase 1 25th March, 2020–14th April, 2020 21 days
Phase 2 15th April, 2020–3rd May, 2020 19 days
Phase 3 4th May, 2020–17th May, 2020 14 days
Phase 4 18th May, 2020–31st May, 2020 14 days
Unlock 1.0 1st June, 2020–30th June, 2020 30 days
Unlock 2.0 1st July, 2020–31st July, 2020 31 days
Unlock 3.0 1st August, 2020–31st August, 2020 31 days
Unlock 4.0 1st September, 2020–30th September, 2020 30 days
Unlock 5.0 1st October, 2020–31st October, 2020 31 days
Unlock 6.0 1st November, 2020–30th November, 2020 30 days
Fig. 1 COVID-19 lockdown and its impact
Review of literature
COVID-19 lockdown and its impact on air pollution
Air pollution is the largest environmental risk in the world causing 8% of total world deaths [20]. Air quality index (AQI) is the evaluation of air quality and lower AQI value is directly proportional to better quality of air [4]. Due to significant slowdown of industrialization, urbanization and social and economic activities, air quality was seen improving in different parts of the world [21]. According to the reports of Health Effects Institute (HEI), 2019, 95% of the earth’s population are inhaling polluted air [3].
Lockdown measures proved to have a positive effect on environment which lead to the reduction in gases like NO2, CO2, CO, PM (particulate matter) [22] (Fig. 2). Carbon emissions resulted in 25% reduction due to lockdown as estimated by the Energy and Clean Air Research Centre [23]. Decline in fossil fuel consumption, air pollution dropped in countries like China, Italy and India [24].Fig. 2 COVID-19 lockdown impact on air, water and sound pollution
India due to its large population, heavy traffics and number of industries lead to high AQI in all the major cities. The satellite images of India showing the pollutant concentration of PM2.5, PM10, SO2, CO and NO2 before and after the lockdown period is shown in figures 3, 4, 5, 6 and 7 [25]. In India 12.4 lakh deaths in the year 2017 occurred due to air pollution [26]. India has experienced fast industrialization during the past two decades but this development has come with a heavy price in terms of pollution [25]. However, due to lockdown India too witnessed reduction in various pollution. In the major cities of India such as Ahmedabad, Mumbai and Pune during the period of lockdown NO2 levels decreased between 40 and 50% [27]. Mahato et al. (2020) [28] have reported, three weeks of lockdown in Delhi, India from 24th March, 2020 showed a steep decline in air pollution. In Lucknow a significant decline of NO2 was observed from pre COVID lockdown to post COVID-19 lockdown period i.e., from 22–158 to 3–59 μg/m3 and also in New Delhi before lockdown NO2 concentration ranged from 4 to 158 μg/m3 while post COVID lockdown it was found to be 9–112 μg/m3 (Fig. 7) [29, 30]. The air AQI in Mumbai in March, 2020, was approximately 153 but in June, 2020 air quality was recorded to be around 66, similar effect was also seen in Karnataka (Bangalore) [31].Fig. 3 PM2.5 accumulation in India, before and after lockdown
Fig. 4 PM10 accumulation in India, before and after lockdown
Fig. 5 CO emission in India, before and after lockdown
Fig. 6 SO2 emission in India, before and after lockdown
Fig. 7 NO2 emission in India, before and after lockdown
Depletion of aerosol accumulation during COVID-19 lockdown was also observed in India which is the major cause of air pollution [32]. In New Delhi, whose AQI was around 300–400 due to smog and pollution lowered to 94 [25]. The highest improvement in Air Quality was recorded by West Bengal that was 64% while Orissa recorded the lowest improvement with a percentage of 10 with respect to PM 2.5 levels (Fig. 3); while the percentage reduction in PM2.5 was more than PM10 in Haryana, Bihar and Uttar Pradesh (Fig. 4) [33]. According to IQAir 2019 World Air Quality Report [34], Ghaziabad-the biggest city of Western Uttar Pradesh reported to be the most polluted city among 14 cities of India out of the world’s 20 most polluted cities with a PM2.5 pollution level of 110.2 μg/m3 in 2019 which is much more than the permissible limit of 60 μg/m3 (Table 2) [25, 34]. It was found that 85% reduction occurred in PM2.5 concentrations due to lockdown in Ghaziabad [25]. Tremendous decrement has been recorded in the pollutant concentration in the major metropolitan cities of India (Mumbai, Kolkata, Delhi and Chennai during the lockdown [35–37]. According to the Energy and Resources Institute (T.E.R.I.), 2021 (Delhi), a 61% reduction of NOx was observed during lockdown and 43% reduction in P.M2.5. The meteorological parameters such as wind speed, mixing height and ventilation coefficient were observed to be less during the lockdown period. Moreover, during pre- lockdown period ventilation coefficient showed more depression and reduced ambient concentration (Table 3). In the year 2021, during the second wave of COVID-19 in India, once again Himalayas were visible from Saharanpur (Uttar Pradesh) where the AQI was 85 as stated by Indian Forest Service official Ramesh Pandey [38].Table 2 AQI of different cities of India from 25th March to 25th April in 2019 and 2020
Cities of India AQI (2019)
From 25th March to 25th April AQI (2020)
From 25th March to 25th April % reduction
Delhi 200 100 50
Mumbai 98 72 27
Bangalore 116 57 51
Chennai 74 48 35
Kolkata 97 78 20
Hyderabad 95 65 32
Jalandhar (Punjab) 122 42 66
Jaipur 84 65 23
Gandhinagar 127 63 50
Vizag 131 63 52
Amritsar 89 65 27
Howrah 87 67 23
(Source: CPCB)
Table 3 Percentage change in PM 2.5, NOx, WS (Wind Speed), MH (Mixing Height) and VC (Ventilation coefficient) in 2020 with respect to 2019 in Delhi
(Source: T.E.R.I., 2021)
Parameters Pre-lockdown Lockdown Unlock 1
PM2.5 − 17 − 43 − 27
NOx − 12 − 61 − 43
WS (Wind Speed) − 16 − 15 − 34
MH (Mixing Height) + 61 − 12 − 27
VC (Ventilation Coefficient) + 34 − 26 − 52
COVID-19 lockdown and its impact on water pollution
Locking down in homes and halting factories, aviation, vehicles remarkably reduced the pollution problem improving the quality of water. Under the pandemic situation industrial effluents (one of the major cause of water pollution) entering the rivers and water bodies came to a halt improving the water quality [25]. Due to reduction in 90% turbidity in the canals, sunlight penetration reached the water beds, thereby providing benefits to the creatures in it [39]. As domestic sewage, industrial wastes, bathing of humans, and boat traffic were curtailed; quality of river water in India uplifted [4]. Ganga, the holiest river of India sadly has been the most polluted river in the world [40]. However, according to the reports of Dr. P.K. Mishra and real time water analysis of Central Pollution Control Board of India, water quality of Ganga was observed 40–50% improved (Table 4) [14, 35]. As per CPCB's data, Ganga water at Haridwar and Rishikesh was found suitable for drinking as sewage and industrial effluents decreased to 500% [4]. Suspended Particulate Matter (SPM) of Lake Vembanad, Kerala dropped by 15.9% which improved the surface water quality of the lake [41]. Singhal and Mato (2020), reported the improved quality of all major rivers of India due to temporary suspension of industries during lockdown. Aman et al. (2020), [42] reported, the average SPM of river Sabarmati, Ahmedabad (Gujarat) decreased to 16.79% as compared to pre-lockdown period with 36.48 and average SPM. Data of UPCB (Uttrakhand Pollution Control Board) demonstrated that total coliforms (40–90 MPN/100 mL), biochemical oxygen demand (0.6–1.2 mg/L), pH (7.4–7.8) and Dissolved oxygen (DO) (9.4–10.6 mg/L) of river Ganga met the standard surface water quality of India [20] (Fig. 2). River Yamuna also showed decrease in level of water pollutants like pH, BOD, COD and DO during the lockdown period as compared to the pre-lockdown period as shown in Table 5.Table 4 Assessment of water quality of Ganga as on March 28, 2020 at monitoring stations in different cities of India
(Source: SANDRP, 2020)
Monitoring stations in Kanpur pH DO (mg/L) BOD (mg/L) Ammonia (mg/L)
U/S of Ganga barrage 7.90 8 2.1 0.49
D/S of Ganga barrage 7.91 7.90 1.2 1.1
At Shuklagunj 7.68 8.51 2.1 0.79
Monitoring stations in Patna pH DO (mg/L) BOD (mg/L) Ammonia (mg/L)
Gandhighat, NIT Patna 8.05 8.5 1.8 0.22
Gulabi Ghat 8.10 8.5 1.8 0.21
D/S Patna, Gaighat 8.04 8.4 1.8 0.12
Monitoring stations in Allahabad pH DO (mg/L) BOD (mg/L) Ammonia (mg/L)
Kadaghat NIL 8.8 2.6 NIL
U/S Rasulabad ghat NIL 9.4 2.7 NIL
D/S Sangham NIL 8.6 2.7 NIL
Monitoring stations in Varanasi pH DO (mg/L) BOD (mg/L) Ammonia (mg/L)
U/S Varanasi NIL 8.5 3.6 NIL
D/S Varanasi NIL 8.4 3.7 NIL
D/S Tarighat NIL 8.9 3.5 NIL
Monitoring stations in Haridwar pH DO (mg/L) BOD (mg/L) Ammonia (mg/L)
Harki Pouri 7.6 10.2 0.6 NIL
U/S Bindhughat 7.8 9.8 1 NIL
D/S Balakumari Mandir 7.7 10 1 NIL
Monitoring stations in West Bengal pH DO (mg/L) BOD (mg/L) Ammonia (mg/L)
Howrah–Shivpuri 7.05 3.90 1.25 0.19
Uluberia 7.35 7.30 2.20 0.22
Table 5 Water quality of River Yamuna, at different stations before and during lockdown period, 2020
(Source: https://www.dpcc.delhigovt.nic.in/home/monthly_analysis_report#gsc.tab=0)
Location Phase pH COD
(mg/L) BOD
(mg/L) DO
(mg/L)
Palla Pre-lockdown
16/03/2020
7.8 8 2.5 8.4
Lockdown
06/04/2020
8 12 2.8 6.9
Surghat
(D/S of Wazirbad Barrage)
Pre-lockdown
16/03/2020
7.9 10 3 4.8
Lockdown
06/04/2020
8.06 16 3.8 6.8
Khajori Paltoon Pool
(D/S Najafgarh Drain
Pre-lockdown
16/03/2020
7.6 84 30 NIL
Lockdown
06/04/2020
7.24 116 33 NIL
Kudesia Ghat Pre-lockdown
16/03/2020
7.6 80 28 NIL
Lockdown
06/04/2020
7.37 60 25 NIL
ITO Bridge Pre-lockdown
16/03/2020
7.9 76 25 NIL
Lockdown
06/04/2020
7.63 32 22 2.3
Nizamudin Bridge Pre-lockdown
16/03/2020
7.7 76 24 0.8
Lockdown
06/04/2020
7.46 42 16 2.3
Agra Canal
(Okhla)
Pre-lockdown
16/03/2020
7.7 132 42 NIL
Lockdown
06/04/2020
7.46 42 16 4.8
After meeting Shahdara Drain
(D/S Okhla Barrage)
Pre-lockdown
16/03/2020
8.0 152 58 NIL
Lockdown
06/04/2020
7.63 76 23 NIL
Agra Canal
(Jaitpur)
Pre-lockdown
16/03/2020
7.6 92 32 NIL
Lockdown
06/04/2020
7.45 48 17 4.2
The Vice Chairman of Delhi Jal Board stated that situation of river Yamuna too witnessed improvement in water quality and appeared blue [14]. Water quality of tributaries like Hemavati, Shimsha, Lakshamanathirta and River Cauvery also reported significant improvement as compared to pre-lockdown period according to Karnataka State Pollution Control Board (Table 6) [43]. Faecal coliform level also showed significant reduction in river bodies. As reported by B.D. Joshi, an environmentalist, after a long period Ganga water became fit for ‘achaman' meaning ritual sipping. According to CPCB report of September 23, 2020 percentage increase in compliance to the bathing criteria in water quality of seven rivers were noticed during lockdown as shown in Table 7.Table 6 Compliance status of Rivers in April, 2020
(Source: CPCB, 2020)
Rivers Total no. of monitoring locations No. of complying locations with percentage No. of non-complying locations with percentage
Beas 22 21 (95%) 1 (5%)
Cauvery 33 32 (97%) 1 (3%)
Chambal 13 6 (46%) 7 (54%)
Ganga 54 25 (46%) 29 (54%)
Godavari 37 29 (78%) 8 (22%)
Krishna 18 16 (89%) 2 (11%)
Mahi 14 1 (7%) 13 (93%)
Sutlej 23 18 (78%) 5 (22%)
Tapi 8 7 (87%) 1 (13%)
Yamuna 12 8 (67%) 4 (33%)
Table 7 List of rivers which showed improvement in water quality for outdoor bathing during lockdown
(Source: CPCB, 2020)
Rivers % increment in compliance to the bathing criteria
Brahmani 85–100
Brahmaputra 87.5–96.97
Godavari 65.8–78.7
Cauvery 90.5–96.97
Krishna 84.6–94.4
Tapi 77.8–87.5
Yamuna 42.8–66.67
COVID-19 lockdown and its impact on fauna and wildlife
COVID-19 pandemic gave wildlife a period to bloom and to get out of their habituated places [3]. As humans were confined to their houses during lockdown, animals and birds came out in notice and were seen venturing the human-zones [4]. Due to peace and calm in urban areas, wildlife emerged to the residential areas [8]. Due to less noise and air pollution, resident birds were seen breeding much more than earlier [44]. Road—killing of both amphibians and reptiles reduced during lockdown [45].
Due to COVID-19 lockdown there were reduction in water pollution and improvement in water quality which imparted positive impacts on aquatic life too [40]. As reported on 25th April, 2020 by Times of India, Ganges Dolphin (a South Asian river dolphin) was back at Kolkata, India. Wildlife Institute of India (WII) using an app ‘Lockdown Wildlife Tracker’ tracked the comfortable movement of wildlife straying outside habitats which was a great initiative by WII [46]. Sambar deer was seen wandering on the roads of Chandigarh, a civet was spotted on a zebra crossing in Kerala, flamingos were seen congregating in Mumbai, a leopard was seen on the empty streets of Hyderabad. Flies, wasps, bees, butterflies were seen frequently in numbers on the crops and other plants indicating restoration of ecology and biodiversity [14].
COVID-19 lockdown and its impact on noise pollution
Noise pollution causes hearing loss or impairment and has several physiological and psychological effects [47, 48]. Globally around 360 million people are vulnerable to loss of hearing due to noise pollution [49]. In older people, noise pollution has been linked to high blood pressure and sensory impairment [50, 51]. Elevated levels of sound generated from construction works, machines, vehicles causes noise pollution leading to detrimental effects on human health [12, 52]. According to National Institute on Deafness and Other Communication Disorders, long or excessive exposure to sound at or above 85 decibels causes hearing impairment. As per WHO guidelines during the night for a peaceful sleep noise level should be 30 dB (Fig. 3). Man-made noises have negative impacts on wildlife and invertebrates too which causes imbalance in the ecosystem [20, 53].
The level of noise pollution decreased in most of the cities worldwide due to temporary suspension of human activities [12]. According to Central Pollution Control Board of India, noise level of residential areas of Delhi got reduced from 55 to 40 dB during daytime and during night it reduced from 45 to 30 dB [35]. In Delhi, April 2020, the average noise level was observed to be around 40–50 dB on an average while compared to last year (April, 2019) which was 100 dB and higher [54]. Chirping of birds can be heard more often due to reduction in noise pollution [54]. According to West Bengal Pollution Control Board, Kolkata’s noise pollution declined by 50–75% during the lockdown period. Underwater noise reduction in oceans during the lockdown proved to be good for whales and other sea mammals and for scientists, to study the effects of quieter oceans on marine wildlife [55].
Result and discussion
The review highlights the positive impact of COVID-19 induced lockdown on the environment of India. Table 1 shows the different lockdown policies introduced in India since the onset of this disease. The lockdown caused sharp decline in the Air quality indices with more than 50% decline in many states of India as shown in Table 2. Jalandhar showed highest percent reduction in Air quality parameters of 66% followed by Bangalore and Delhi (Table 2). The different parameters of air quality like PM2.5, PM10, NO2, SO2 and CO reported sharp decline in concentration as shown in the satellite images of India taken before and after the lockdown period (Figs. 3, 4, 5, 6 and 7). This decline was recorded in nearly all cities of India due to less use of automobiles and industries. The water quality parameters like pH, BOD, Ammonia, DO concentration decreased during the lockdown period as compared to the unlock period as shown in Table 4 in Ganga water at different monitoring locations of India. The pollutant level in water bodies declined due to shut down of industries and no human interventions. The water quality parameters of river Yamuna also witnessed improvement at different stations of India (Table 5). Water quality of many rivers like Chambal, Cauvery, Krishna, Godavari and many others showed a decline in all the water polluting parameters at different stations as shown in Tables 5, 6 and 7.
Due to rapid transmission of novel coronavirus disease (COVID-19), nationwide lockdown was adopted to restrict the human mobility to slow down the rate of ongoing infections. Undoubtedly, the imposition of lockdown was found to be advantageous for nature’s rejuvenation, as all kinds of commercial, social, industrial, economic and urbanization activities were suddenly put on hold. Consequently, nature restores itself and showed significant improvements in the air and water quality parameters, noise pollution as well as undisturbed and calm wildlife [30].
It is evident from the reports of several previous studies and other print and electronic media that water quality parameters of many rivers of India showed marked improvements including Ganga, Cauvery, Sutlej and Yamuna. The reason behind clean water in the rivers is that, very less amount of industrial effluents was reaching into the river body amid the lockdown period. In addition to this, several other factors have also been showing ameliorating effect in enhancing the water quality indices in the rivers such that less demand of irrigation water, above average rainfall, substantial reduction in the religious and cultural activities like pupa, cremations bathing and many other rituals which has been performed on the bank of rivers [56]. This resulted in increased DO of river Ganga from 6.5 to 8 ppm and reduced BOD from 4 to 3 ppm near Kanpur [57]. At some places due to the self-cleansing property of Ganga the water has reached drinking standards during the lockdown period. River Yamuna was also seen with blue and clear water with aquatic life moving around because of less detergents and chemicals due to lockdown. The clear water in river has significant effect on aquatic life as it flourishes in clean water as witnessed during lockdown.
The impact of COVID-19 lockdown on the improvement of air quality signifies the direct relation of air pollution with economic as well as energy generation activities. This calls for the requirement of green energy based system or use of renewable energy to reduce air pollution globally. COVID-19 lockdown reduced the use of fuel due to less traffic on road and airways, so public transport should be an alternative to reduce air pollution even after lockdown. The COVID-19 lockdown gave a major improvement in air quality in short span of time. This lockdown provides solution for nature’s rejuvenation from time to time along with sustainable development. The lockdown gave a positive hope that there can be reduction in environmental pollution with less human intervention.
Wildlife also flourished during the lockdown period. Many birds and animals were seen in cities which were lost earlier due to pollution and human interventions. This provides an insight to researchers to plan in a meticulous way how the coexistence of humans as well as animals is possible on earth. Noise pollution also declined due to reduced use of automobiles, aircraft and industries. COVID-19 pandemic reduced human activity to a great extent and hence 50% reduction in noise pollution was recorded in India [58]. Overall this few months lockdown is not sufficient to curb the pollution, but it hints the government policymakers to induce such restrictions regularly. There should proper removal of wastages as well as sewage treatment plants needs to be installed to reduce the pollution level. So, comprehensive pollutant study at different locations is required to formulate accurate policies to reduce the pollution level.
Conclusion
The outbreak of COVID-19 pandemic has shattered millions of lives globally and is still hitting several countries around the globe severely. The tragic impacts of coronavirus bring a lot of negativity but somehow the lockdown measures have proved beneficial to the environment. The effect of lockdown led to reduction in pollution and the Mother Earth was seen restoring itself. This review provides evidences of significant reduction in air, water and noise pollution and the positive impacts of lockdown on wildlife giving a better insight of the effects of lockdown in India. This calls for policy makers to amend the national policies for better environment in long run. To mitigate the pollution problem in future, government and the citizens should adopt measures and ensure lockdown restrictions regularly to curb environmental deterioration. Therefore, adopting a good lifestyle that reduces the greenhouse gases and carbon footprint in the atmosphere is urgently needed for positive impact on environment as well as to restore wildlife on Earth.
Declarations
Conflict of interest
All the authors declare they have no conflict of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Deepak Kumar Jha and Niti Yashvardhini have contributed equally.
==== Refs
References
1. Unhale SS Ansar QB Sanap S Thakhre S Wadatkar S Bairagi R Sagrule S Biyani KR A review on corona virus (Covid-19) International Journal of Pharmaceutical and Life Sciences 2020 6 4 109 115
2. Milibari AA Current situation of coronavirus disease: (COVID-19) review article Health Science Journal 2020 1 005
3. Rupani PF Nilashi M Abumalloh RA Asadi S Samad S Wang S Coronavirus pandemic (COVID-19) and its natural environmental impacts International Journal of Environmental Science and Technology: IJEST 2020 17 11 4655 4666 10.1007/s13762-020-02910-x 32904898
4. Arora S Bhaukhandi KD Mishra PK Coronavirus lockdown helped the environment to bounce back The Science of the Total Environment 2020 742 140573 10.1016/j.scitotenv.2020.140573 32619844
5. Soni P Effects of COVID-19 lockdown phases in India: An atmospheric perspective Environment, Development and Sustainability 2021 23 8 12044 12055 10.1007/s10668-020-01156-4 33424429
6. Miglani A Effect of lockdown during COVID-19: An Indian perspective International Journal of Science & Healthcare Research 2020 5 3 55 61
7. Singh S Roy D Sinha K Parveen S Sharma G Joshi G Impact of COVID-19 and lockdown on mental health of children and adolescents: A narrative review with recommendations Psychiatry Research 2020 293 113429 10.1016/j.psychres.2020.113429 32882598
8. Kumar A Malla MA Dubey A With corona outbreak: Nature started hitting the reset button globally Frontiers in Public Health 2020 8 569353 10.3389/fpubh.2020.569353 33072704
9. Bar H COVID-19 lockdown: Animal life, ecosystem and atmospheric environment Environment, Development and sustainability 2021 23 6 8161 8178 10.1007/s10668-020-01002-7 33020695
10. Jadhav RV Aher JS Bhagare AM Dhaygude AC COVID-19 era: What’s impact of the lockdown on India’s environment? Journal of Chemistry, Environmental Sciences and its Applications 2020 7 1 1 6 10.15415/jce.2020.71001
11. Muhammad S Long X Salman M COVID-19 pandemic and environmental pollution: A blessing in disguise? The Science of the Total Environment 2020 728 138820 10.1016/j.scitotenv.2020.138820 32334164
12. Zambrano-Monserrate MA Ruano MA Sanchez-Alcalde L Indirect effects of COVID-19 on the environment The Science of the Total Environment 2020 728 138813 10.1016/j.scitotenv.2020.138813 32334159
13. Thiessen, T. (2020). How clean air cities could outlast COVID-19 lockdowns. https://www.forbes.com/sites/tamarathiessen/2020/04/10/how-clean-air-cities-could-outlast-covid-19-lockdowns/?sh=3de864dc6bb5
14. Das SS Positive impact of Covid-19 lockdown on environment Gorteria Journal 2021 34 2 112 119
15. Hussain, N. (2020). Environmental impact of a lockdown. Centre for Development Policy and Practice.
16. Yadav, A. (2020). Analysis of impact on environment: Pre and post lockdown. SSRN; 2020. 10.2139/ssrn.3689299
17. Andhare R Shrivas Y Guru S Joshi A Indirect impact of Covid-19 lockdown on society and environment Medico-Legal Update 2020 20 4 1266 1269
18. Guha S Govil H COVID-19 lockdown effect on land surface temperature and normalized difference vegetation index Geomatics, Natural Hazards and Risk 2021 12 1 1082 1100 10.1080/19475705.2021.1914197
19. Beccario, C. (2016). Earth null school [Open source code]. https://github.com/cambecc/earth. Retrieved from https://earth.nullschool.net
20. World Health Organization. (2016). Ambient air pollution: a global assessment of exposure and burden of disease. World Health Organization. http://apps.who.int/iris/handle/10665/250141
21. Rume T Islam S Environmental effects of COVID-19 pandemic and potential strategies of sustainability Heliyon 2020 6 9 e04965 10.1016/j.heliyon.2020.e04965 32964165
22. Gupta N Tomar A Kumar N The effect of COVID-19 lockdown on the air environment in India Global Journal of Environmental Science and Management 2020 6 SI 31 40
23. Shakil MH Munim ZH Tasnia M Sarowar S COVID-19 and the environment: A critical review and research agenda The Science of the Total Environment 2020 745 141022 10.1016/j.scitotenv.2020.141022 32711074
24. Paital B Nurture to nature via COVID-19, a self-regenerating environmental strategy of environment in global context The Science of the Total Environment 2020 729 139088 10.1016/j.scitotenv.2020.139088 32388136
25. Lokhandwala S Gautam P Indirect impact of COVID-19 on environment: A brief study in Indian context Environmental Research 2020 188 109807 10.1016/j.envres.2020.109807 32574854
26 Balakrishnan K Dey S Gupta T The impact of air pollution on deaths, disease burden, and life expectancy across the states of India: The Global Burden of Disease Study 2017 Lancet Planet Health 2019 3 1 e26 e39 10.1016/S2542-5196(18)30261-4 30528905
27. Wright, R. (2020). The world’s largest coronavirus lockdown is having a dramatic impact on pollution in India.
28. Mahato S Pal S Ghosh KG Effect of lockdown amid COVID-19 pandemic on air quality of the megacity Delhi, India The Science of the Total Environment 2020 730 139086 10.1016/j.scitotenv.2020.139086 32375105
29. Srivastava S Kumar A Bauddh K Gautam AS Kumar S 21-day lockdown in India dramatically reduced air pollution indices in Lucknow and New Delhi, India Bulletin of environmental contamination and toxicology 2020 105 1 9 17 10.1007/s00128-020-02895-w 32495123
30. Yashvardhini N Kumar A Gaurav M Sayrav K Jha DK Positive impact of COVID-19 induced lockdown on the environment of India’s national capital, Delhi Spatial Information Research 2021 30 249 259 10.1007/s41324-021-00427-0
31. Kambalagere Y A study on air quality index (AQI) of Bengaluru, Karnataka during lockdown period to combat coronavirus disease (Covid-19): Air quality turns ‘better’ from ‘hazardous’ Studies in Indian Place Names 2020 40 69 2394 3114
32. Gautam S The influence of COVID-19 on air quality in India: A boon or inutile Bulletin of Environmental Contamination and Toxicology 2020 104 6 724 726 10.1007/s00128-020-02877-y 32394052
33. Sekar, A., Jasna, R. S., Binoy, B. V., Mohan, P., & Kuttiparichel Varghese, G. (2022). Air quality change and public perception during the COVID-19 lockdown in India. Gondwana Research: International Geoscience Journal. 10.1016/j.gr.2022.04.023. Advance online publication 10.1016/j.gr.2022.04.023.
34. IQAir. (2019). World air quality report (Region & City PM2.5 Ranking). IQAir Group.
35. Central Pollution Control Board, New Delhi. (2020). Assessment of Impact of lockdown on Water Quality of Major Rivers.
36. Sharma S Zhang M Anshika Gao J Zhang H Kota SH Effect of restricted emissions during COVID-19 on air quality in India The Science of the Total Environment 2020 728 138878 10.1016/j.scitotenv.2020.138878 32335409
37. Lau H Khosrawipour V Kocbach P Mikolajczyk A Schubert J Bania J Khosrawipour T The positive impact of lockdown in Wuhan on containing the COVID-19 outbreak in China Journal of Travel Medicine 2020 27 3 taaa037 10.1093/jtm/taaa037 32181488
38. Hindustan Times. (2021). https://www.hindustantimes.com/trending/himalayas-visible-from-up-s-saharanpur-again-breathtaking-pics-are-going-viral-101621592009535.html
39. Muralikrishnan, V. (2020). COVID-19 lockdown: Some good news for the environment. ScIU.
40. Khan I Shah D Shah SS COVID-19 pandemic and its positive impacts on environment: An updated review International Journal of Environmental Science and Technology: IJEST 2021 18 2 521 530 10.1007/s13762-020-03021-3 33224247
41. Yunus AP Masago Y Hijioka Y COVID-19 and surface water quality: Improved lake water quality during the lockdown The Science of the Total Environment 2020 731 139012 10.1016/j.scitotenv.2020.139012 32388159
42. Aman MA Salman MS Yunus AP COVID-19 and its impact on environment: Improved pollution levels during the lockdown period—A case from Ahmedabad, India Remote Sensing Applications: Society and Environment 2020 20 100382 10.1016/j.rsase.2020.100382 34173436
43. The Hindu. (2020). Cauvery, tributaries look cleaner as pandemic keeps pollution away.
44. India New England. (2020). India sees surge in the population of birds and butterflies amid lockdown. https://indianewengland.com/2020/04/india-sees-surge-in-the-population-of-birds-and-butterflies-amid-lockdown/
45. Manenti R Mori E Di Canio V Mercurio S Picone M Caffi M Brambilla M Ficetola GF Rubolini D The good, the bad and the ugly of COVID-19 lockdown effects on wildlife conservation: Insights from the first European locked down country Biological Conservation 2020 249 108728 10.1016/j.biocon.2020.108728 32863391
46. Wild Life Institute of India. (2020). Lockdown wildlife tracker. https://wii.gov.in/
47. Hinchcliffe R Review: Global perspective of noise-induced hearing loss as exemplified by Nigeria Journal of Audiological Medicine 2002 11 1 1 24
48. Oguntunde PE Okagbue HI Oguntunde OA Odetunmibi OO A study of noise pollution measurements and possible effects on public health in Ota Metropolis, Nigeria Open Access Macedonian Journal of Medical Sciences 2019 7 8 1391 1395 10.3889/oamjms.2019.234 31110591
49. Sims, J. (2020). Will the world be quieter after the pandemic?.
50. Ebare MN Omuemu VO Isah EC Assessment of noise levels generated by music shops in an urban city in Nigeria Public Health 2011 125 9 660 664 10.1016/j.puhe.2011.06.009 21875726
51. Ntui AI Noise sources and levels at the University of Calabar Library, Calabar, Nigeria African Journal of Library, Archives & Information Science 2009 19 1 53 63
52. Goines L Hagler L Noise pollution: A modem plague Southern Medical Journal 2007 100 3 287 294 10.1097/smj.0b013e3180318be5 17396733
53. Solan M Hauton C Godbold JA Wood CL Leighton TG White P Anthropogenic sources of underwater sound can modify how sediment-dwelling invertebrates mediate ecosystem properties Scientific Reports 2016 6 20540 10.1038/srep20540 26847483
54. Gandhiok, J., & Ibra, M. (2020). The Times of India.Covid-19: Noise pollution falls as lockdown rings in sound of silence. https://timesofindia.indiatimes.com/india/covid-19-noise-pollution-falls-as-lockdown-rings-in-sound-of-silence/articleshow/75309318.cms
55. The Guardian. (2020). Silence is golden for whales as lockdown reduces ocean noise.
56. SANDRP. (2020). Ganga-Yamuna-Cauvery flow cleaner in lockdown: What can we learn?. DRP News Bulletin, South Asia Network on Dams, Rivers and People (SANDRP). Retrieved from: https://sandrp.in/2020/04/06/drp-nb-6-april-2020-ganga-yamuna-cauvery-flow-cleaner-in-lockdown-what-can-we-learn/#more-34730
57. Pathak SS Mishra P A review of the Ganga river water pollution along major urban centres in the state of Uttar Pradesh, India International Ressearch Journal of Engineering and Technology 2020 7 3 1202 1210
58 Somala SN Seismic noise changes during COVID-19 pandemic: A case study of Shillong, India Natural Hazards 2020 103 1623 1628 10.1007/s11069-020-04045-1 32421076
| 0 | PMC9707182 | NO-CC CODE | 2022-12-01 23:20:21 | no | Spat. Inf. Res.. 2022 Nov 29;:1-13 | utf-8 | null | null | null | oa_other |
==== Front
Z Relig Ges Polit
Z Relig Ges Polit
Zeitschrift Fur Religion, Gesellschaft Und Politik
2510-1218
2510-1226
Springer Fachmedien Wiesbaden Wiesbaden
136
10.1007/s41682-022-00136-x
Artikel
Verschwörung als Ersatzreligion? Religiosität, Spiritualität und Verschwörungsaffinität in Zeiten gesellschaftlicher Krisen
Conspiracy as a surrogate for religion? Religiosity, spirituality, and affinity towards conspiracy myths in times of crisishttp://orcid.org/0000-0003-4920-7881
Schwaiger Lisa [email protected]
1
Schneider Jörg [email protected]
2
http://orcid.org/0000-0002-4964-2528
Eisenegger Mark [email protected]
12
Nchakga Camille [email protected]
1
1 grid.7400.3 0000 0004 1937 0650 Institut für Kommunikationswissenschaft und Medienforschung (IKMZ), Universität Zürich, Andreasstrasse 15, 8050 Zürich, Schweiz
2 grid.7400.3 0000 0004 1937 0650 Forschungszentrum Öffentlichkeit und Gesellschaft (fög), Universität Zürich, Andreasstrasse 15, 8050 Zürich, Schweiz
29 11 2022
122
29 4 2022
16 9 2022
18 10 2022
© The Author(s), under exclusive licence to Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2022, Springer Nature oder sein Lizenzgeber (z.B. eine Gesellschaft oder ein*e andere*r Vertragspartner*in) hält die ausschließlichen Nutzungsrechte an diesem Artikel kraft eines Verlagsvertrags mit dem/den Autor*in(nen) oder anderen Rechteinhaber*in(nen); die Selbstarchivierung der akzeptierten Manuskriptversion dieses Artikels durch Autor*in(nen) unterliegt ausschließlich den Bedingungen dieses Verlagsvertrags und dem geltenden Recht.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
In Krisenzeiten gewinnen alternative Weltdeutungen an Definitionsmacht, die Leitbilder der öffentlichen Sphäre in Frage stellen. Im Zuge der COVID-19-Pandemie verbreiten sich solche Narrative in Form von Verschwörungsmythen. Sie bieten Erklärungen für komplexe Sachverhalte, indem davon ausgegangen wird, dass gesellschaftliche Ereignisse von geheimen Eliten geplant und gesteuert werden. Darüber hinaus beruhen sie häufig auf pseudotranszendentalen Erklärungen, die sich einer Überprüfung entziehen, aber schicksalhafte Ereignisse mit Sinn füllen. In diesem Zusammenhang stellt sich die Frage, ob Verschwörungsmythen religiöse oder spirituelle Züge beinhalten. In dieser Studie untersuchen wir mit einer im April 2022 durchgeführten repräsentativen Bevölkerungsbefragung in der deutsch- und französischsprachigen Schweiz (n = 1221) den Zusammenhang zwischen Spiritualität, Religiosität und der Affinität gegenüber Verschwörungsmythen. Die Resultate zeigen zunächst, dass insbesondere die emotionale Betroffenheit während der COVID-19-Pandemie ein wichtiger Prädiktor im Zusammenhang mit Verschwörungsaffinität ist. Je negativer die emotionale Betroffenheit, umso höher ist die Affinität gegenüber Verschwörungsmythen. Umgekehrt wirken positive Emotionen während der Pandemie negativ auf die Verschwörungsaffinität. Während die Konfessionszugehörigkeit wie auch Religiosität kaum die Verschwörungsaffinität beeinflussen, erweist sich in unserer Studie Spiritualität als wichtigster Prädiktor. Wir schliessen daraus, dass in säkularisierten Gesellschaften Spiritualität ohne die Einbindung in traditionelle religiöse Gemeinschaften mit einer verstärkten Offenheit für Verschwörungsmythen als Ersatzreligion einhergeht.
In times of crisis, alternative interpretations of the world that challenge the public sphere gain impact. During the COVID-19 pandemic, such narratives spread in the form of conspiracy myths. They give explanations for complex questions by assuming that social events are planned and controlled by secret elites. Moreover, they are often based on pseudo-transcendental explanations that defy scrutiny but fill fateful events with meaning. In this context, the question arises whether conspiracy myths contain religious or spiritual elements. In this study, we use a representative population survey conducted in April 2022 in German- and French-speaking Switzerland (n = 1221) to examine the relationship between spirituality, religiosity, and affinity toward conspiracy myths. The results show that emotional concern during the COVID-19 pandemic is an important predictor related to conspiracy affinity. The more negative the emotional concern, the higher the affinity towards conspiracy myths. Conversely, positive emotions during the pandemic negatively affect conspiracy affinity. While religious affiliations and religiosity hardly influence conspiracy affinity, spirituality turns out to be the most important predictor in our study. We conclude that in secularized societies spirituality without involvement in traditional religious communities can result in an increased openness to conspiracy myths as a surrogate for religion.
Schlüsselwörter
Verschwörung
Religion
Spiritualität
Krise
Pandemie
Befragung
Keywords
Conspiracy
Religion
Spirituality
Crisis
Pandemic
Survey
==== Body
pmcEinleitung
Gesellschaftliche Krisenzeiten sind von Gefühlen der Unsicherheit und des Kontrollverlusts sowie von Existenzängsten in der Bevölkerung geprägt, was sich auch im Zuge der COVID-19-Pandemie deutlich zeigte und zeigt (Kirchler et al. 2020). In Zeiten der Krise suchen Individuen Antworten auf komplexe Fragen; sie suchen nach Lösungen, um die Welt, in der sie leben, und aktuelle Geschehnisse besser zu verstehen. Offizielle Statements der politischen und medialen Öffentlichkeit sind in diesen von Unsicherheit geprägten Zeiten jedoch für manche Bevölkerungsgruppen nicht immer zufriedenstellend, was zu einer Suche nach und Offenheit gegenüber alternativen Erklärungen führt (Konkes und Lester 2015). In diesem Beitrag möchten wir unterschiedliche Phänomene fokussieren, die für Menschen die Funktion einer Sinnstiftung erfüllen und insbesondere in Krisenzeiten besonders relevant sind. Einerseits sind es religiöse Zugehörigkeit, religiöser Glaube und Spiritualität, die seit jeher eine Möglichkeit bieten, schicksalhafte Ereignisse mit Sinn zu füllen und einem Gefühl des Kontrollverlusts entgegenzuwirken (Kay et al. 2010). Andererseits verdeutlichen zahlreiche Studien aus unterschiedlichen Disziplinen, dass der Glaube an Verschwörungsmythen in krisenhaften Zeiten, die von Unsicherheit geprägt sind, zunimmt (Douglas et al. 2017; van Prooijen und Douglas 2017). Beide angesprochen Phänomene – religiöser bzw. spiritueller Glaube wie auch Verschwörungsglaube – können demnach als Bewältigungsstrategien in schicksalhaften Phasen interpretiert werden, da sowohl religiöse und spirituelle wie auch verschwörungsmythische Glaubensüberzeugungen potenzielle Erklärungen für das Unerklärliche liefern, obgleich sie sich einer rationalen Verifikation entziehen. In diesem Kontext untersuchen – insbesondere theoretische – Studien mehr und mehr den Zusammenhang zwischen Religiosität, Spiritualität und Verschwörungsglauben (Robertson 2015, 2016; Dyrendal et al. 2018a; Wood und Douglas 2018). Es stellt sich die Frage, ob in modernen, säkularisierten Gesellschaften der Glaube an Verschwörungsmythen ein Ersatz für religiösen Glauben bieten kann (Robertson 2016). Im Kontext der COVID-19-Pandemie wird auch vermehrt empirisch zu dem Thema geforscht, wenn auch mit teils kontroversen Resultaten (z. B. Jasinskaja-Lahti und Jetten 2019; Gligoric et al. 2021; Ladini 2021; Frenken et al. 2022; Lowicki et al. 2022). Es erscheint insbesondere eine getrennte Betrachtung der beiden Konstrukte Religiosität und Spiritualität essenziell, um das Phänomen Verschwörungsglauben erklären zu können.
In dieser Studie arbeiten wir zunächst die Ähnlichkeiten und Unterschiede zwischen den Funktionen von Religiosität, Spiritualität und dem Glauben an Verschwörungsmythen in Zeiten gesellschaftlicher Krisen theoretisch heraus, um diese schliesslich empirisch zu untersuchen. Diese Studie widmet sich daher folgender Forschungsfrage: Wie hängen die individuelle Religiosität und Spiritualität mit der Affinität gegenüber Verschwörungsmythen zusammen? Die Studie basiert auf Daten einer repräsentativen Online-Bevölkerungsbefragung im April 2022 in der deutsch- und französischsprachigen Schweiz. Ziel des Beitrages ist es, empirische Erkenntnisse zum Zusammenhang zwischen Religiosität, Spiritualität und Verschwörungsaffinität zu liefern und die diesbezügliche Forschungslücke zu füllen, indem wir uns auf aktuelle Daten im Kontext der COVID-19-Pandemie beziehen.
Die Schweiz ist ein Beispiel für eine moderne, westliche und säkularisierte Demokratie, die – obwohl die Resilienz gegenüber Misinformationen (als Sammelbegriff für die Verbreitung sämtlicher Informationen, die nicht der Wahrheit entsprechen (Scheufele und Krause 2019)) in der Schweiz vergleichsweise hoch ist (Humprecht et al. 2020) – auch wie andere Länder während der COVID-19-Pandemie mit der Verbreitung von Verschwörungsmythen konfrontiert war (Vogler et al. 2021). Die religiöse Landschaft in der Schweiz zeichnet sich durch einen hohen Anteil an Katholikinnen und Katholiken (34,4 %) und Evangelisch-Reformierten (22,5 %) aus. 5,4 % der Bevölkerung gehören muslimischen und aus dem Islam hervorgegangenen Gemeinschaften an, 7,2 % anderen religiösen Gemeinschaften und 29,4 % der Schweizer Bevölkerung haben keine Religionszugehörigkeit (Bundesamt für Statistik 2022a). Obwohl der Anteil der Konfessionslosen zunimmt, waren laut Bundesamt für Statistik (2022a) im Jahr 2019 für 56 % der Bevölkerung Religion und Spiritualität „in schwierigen Momenten des Lebens“ eher oder sehr wichtig. Es kann davon ausgegangen werden, dass unabhängig von einer religiösen Zugehörigkeit Religiosität und Spiritualität weiterhin für die Mehrheit der Bevölkerung vor allem in Krisenzeiten einen hohen Stellenwert einnehmen.
Theoretischer Hintergrund
Verschwörungsmythen in Zeiten gesellschaftlicher Krisen
Die COVID-19-Pandemie ist geprägt von der Verbreitung von Verschwörungsmythen, die Antworten auf komplexe Fragen bieten sollen, während offizielle Stellungnahmen der Regierungen, Medien und von wissenschaftlichen Akteurinnen und Akteuren nicht mehr als zufriedenstellend wahrgenommen werden (Bruns et al. 2020). Generell fallen unter Verschwörungsmythen jene Annahmen, die davon ausgehen, dass gesellschaftliche Ereignisse im Geheimen von meist elitären Gruppen gesteuert werden; nichts passiere aus Zufall und unterschiedliche Ereignisse würden miteinander in Zusammenhang stehen (Clarke 2002; Castanho Silva et al. 2017; Douglas et al. 2019). Verschwörungsmythen stellen sich in den seltensten Fällen als wahr heraus, doch sind sie gleichzeitig in der Regel weder eindeutig falsifizierbar noch verifizierbar. Wir ziehen in diesem Beitrag daher den Begriff der ‚Verschwörungsmythen‘ jenem der ‚Verschwörungstheorien‘ vor, da letzterer eine potenzielle Falsifizier- oder Verifizierbarkeit implizieren würde (Spiegel et al. 2020). Zudem fasst der Mythenbegriff den Kern der Verschwörungserzählungen klarer: Es handelt sich vielmehr um Narrationen und alternative Weltdeutungen als um verifizierte Erklärungen von konkreten Ereignissen oder Phänomenen (Spiegel et al. 2020). Mythen zeugen von paradigmatischem Charakter, um so bestimmten Ereignissen Bedeutung zu verleihen (Dyrendal et al. 2018a) – wie es auch bei religiösen Erzählungen der Fall ist.
Die Diffusion alternativer Narrative – in diesem Fall von Verschwörungsmythen – ist kennzeichnend für Krisenzeiten (Imhof 1996; van Prooijen und Douglas 2017). Im Gegensatz zu strukturzentrierten Phasen, werden in Zeiten der Krise Leitbilder der öffentlichen Sphäre, also die öffentliche Meinung aus Politik, Medien und Ökonomie, stärker hinterfragt (Imhof 1996). Wir leiten den Krisenbegriff öffentlichkeitstheoretisch her, wonach Krise nach Imhof (1996, S. 211) als die „Erosion vormals selbstverständlicher, konsensuell geteilter Gegenwartsinterpretationen wie Zukunftsperspektiven und damit die Aufgabe vieler Konventionen, Denkgewohnheiten und Entwicklungspfade“ definiert wird. Es handelt sich demnach um ein unerwartetes Ereignis, das die Routine durchbricht und wodurch ein Rückgriff auf etablierte Entscheidungen und Normen nicht mehr möglich ist (Oevermann 2016). Die COVID-19-Pandemie ist demnach als traumatische Krise (Oevermann 2016) zu interpretieren, die in diesem Fall durch ein negatives Ereignis und einer ungewissen Zukunft geprägt ist. Alternative Gruppierungen erlangen in diesen Phasen eine stärkere Deutungsmacht, indem sie alternative Weltdeutungen in den Diskurs bringen, die jene Antworten bringen sollen, welche die politische und mediale Öffentlichkeit nicht bietet (Imhof 1996). Verschwörungsmythen erfüllen ebendiese Funktion, indem sie einerseits die öffentliche Meinung konkurrieren (Schwaiger 2022) und in Zeiten der Unsicherheit und des gefühlten Kontrollverlusts Erklärungen für komplexe gesellschaftliche Ereignisse bieten (Konkes und Lester 2015; van Prooijen und Douglas 2017). Aktuelle Studien verdeutlichen, dass im Zuge der COVID-19-Pandemie die Verbreitung von Verschwörungsmythen zugenommen hat (Bruns et al. 2020; Gemenis 2021; Pummerer et al. 2021; Srol et al. 2021). Die gestiegene Verbreitung von Verschwörungsmythen ist jedoch nicht nur in der aktuellen Krisenphase begründet, sondern steht darüber hinaus mit dem digitalen Wandel respektive der Etablierung von Digitalplattformen in Zusammenhang. Obwohl Verschwörungsmythen aus historischer Sicht schon immer existierten, erlauben es die Funktionen von digitalen Medien auch Laien, Inhalte selbstständig aufzubereiten und zu verbreiten, ohne dass diese auf ihren Wahrheitsgehalt hin geprüft oder Quellen transparent angegeben werden müssen, wie es bei professionellen journalistischen Inhalten der Fall ist (Neuberger 2018; Schwaiger 2022). Studien verdeutlichen, dass sich vor allem polarisierende, skandalisierende Inhalte auf sozialen Medien stärker verbreiten als wissenschaftlich fundierte Informationen (Konkes und Lester 2015; Zollo et al. 2015).
Insgesamt lassen sich sowohl sozialer als auch digitaler Wandel als Treiber für die Verbreitung von Verschwörungsmythen festhalten. Es sind insbesondere Krisenzeiten, in denen Personen anfälliger für den Glauben an Verschwörungsmythen werden, da sie hier mit Gefühlen der Unsicherheit und Unzufriedenheit mit politischen Entscheidungen konfrontiert sind (van Prooijen und Douglas 2017; Bode und Vraga 2018; Biddlestone et al. 2020). Die COVID-19-Pandemie ist dementsprechend als gesellschaftliche Krisenphase zu betrachten, die schliesslich auch zu persönlichen Krisen führen kann. Wir stellen daher folgende Hypothesen auf:
H1
Je höher die negative emotionale Betroffenheit während der COVID-19-Pandemie, desto grösser ist die Affinität gegenüber Verschwörungsmythen.
H2
Je ausgeprägter die positiv empfundenen Emotionen während der COVID-19-Pandemie, desto geringer ist die Affinität gegenüber Verschwörungsmythen.
Es stellt sich dabei die Frage, inwieweit Religiosität und Spiritualität im Vergleich zur emotionalen Betroffenheit als erklärende Faktoren hinsichtlich der Affinität gegenüber Verschwörungsmythen wirken. Fraglich ist zudem, ob Verschwörungsglaube auch einen Ersatz für religiösen Glauben darstellen kann, zumal in modernen, säkularisierten Gesellschaften das Vertrauen in göttliche Fügungen an Relevanz verliert (Robertson 2015). Im folgenden Abschnitt wird daher die Rolle von Religiosität und Spiritualität in Krisenzeiten erörtert, bevor der Zusammenhang zwischen Religiosität, Spiritualität und Verschwörungsglauben untersucht wird.
Zur Rolle von Religiosität und Spiritualität in Krisenzeiten
Kontingenzbewältigung im Sinne einer Unterstützung der Menschen bei existenziell bedrohlichen Situationen im Leben ist eine zentrale Funktion von Religion (Höllinger und Aschauer 2022). Ähnlich dem Glauben an Verschwörungsmythen, der in Krisenzeiten zunimmt, steigt auch religiöser Gaube bei bedrohenden und mit Unsicherheiten verbundenen Ereignissen (Wood und Douglas 2018). Die Relevanz von Religiosität in Krisenzeiten wie auch individueller Spiritualität – die in diesem Beitrag in Abgrenzung zur traditionellen, konfessionellen und institutionellen Religiosität betrachtet wird –, wird auch im Kontext der COVID-19-Pandemie durch aktuelle Studien bestätigt. Eine Bevölkerungsbefragung in Österreich während der Pandemie kam beispielsweise zu dem Befund, dass für jene Personen, die sich stark durch das Virus bedroht fühlten, der Grad an Religiosität wie auch Spiritualität ausgeprägter war als bei Personen, die diese Bedrohung geringer wahrgenommen haben (Höllinger und Aschauer 2022). Weiter deuten die Daten darauf hin, dass religiöse Personen ein höheres Institutionenvertrauen respektive Vertrauen in die Politik aufweisen, während spirituelle Personen der Krisenpolitik deutlich kritischer gegenüber stehen und beispielsweise eine geringere Bereitschaft zeigen, Massnahmen einzuhalten (Höllinger und Aschauer 2022). Auch international zeigen sich ähnliche Befunde, welche die Relevanz von Religion und Spiritualität in Pandemiezeiten verdeutlichen. Eine Befragungsstudie in Polen zeigt, dass etwa ein Fünftel der Befragten während der Pandemie mehr Zeit für religiöse Praktiken wie das Beten aufgebracht hat als vor der Krise (Boguszewski et al. 2020). Auch in Brasilien bestätigt sich das Resultat, religiöser und spiritueller Glauben soll einer Onlinebefragung zufolge eine Bewältigungsstrategie in Pandemiezeiten sein. Religiöse und spirituelle Aktivitäten korrelieren demnach mit geringerer Besorgnis und weniger Angst (Lucchetti et al. 2021). Dies deckt sich mit dem Befund von Roberto et al. (2020), die mit einer quantitativen und qualitativen Studie den positiven Zusammenhang zwischen Spiritualität und Resilienz während Krisenzeiten für Frauen bestätigen. Der Zusammenhang wird durch eine internationale Studie von Bentzen (2021) verdeutlicht. Die Autorin zeigt, dass im Frühjahr 2020 auch über die Plattform Google die Suche nach „Prayer“ und verwandten Begriffen laut Google Trends in 107 Ländern signifikant zugenommen hat.
Die aufgeführten Studien verdeutlichen, dass nicht immer eine eindeutige definitorische Trennung zwischen den Begriffen Religiosität und Spiritualität erfolgt. Wir möchten in dieser Studie konkret auf die unterschiedlichen Effekte der beiden Phänomene im Zusammenhang mit Verschwörungsglauben eingehen. Daher beziehen wir uns theoretisch wie auch methodisch auf zwei Definitionen, welche die beiden Konstrukte annähernd trennscharf erfassen. Zum einen beziehen wir uns auf den Religiositätsbegriff nach Huber (2008), der unterschiedliche Dimensionen von Religiosität definiert. Huber (2008) fasst unter diese Kerndimensionen den Intellekt, der sich in dem Interesse an religiöse Fragen (z. B. Sinnsuche) manifestiert, den Glauben oder die Ideologie (z. B. Haltung gegenüber Transzendenz), die öffentliche Praxis (z. B. Gottesdienstbesuch), die private Praxis (z. B. Gebet) und die Erfahrung (z. B. religiöse Gefühle). Zum anderen beziehen wir uns auf die Definition von Spiritualität nach Houtman und Tromp (2021). Die Autoren grenzen Spiritualität bewusst von einem christlich-geprägten Religiositätsbegriff ab, der beispielsweise durch einen Glauben an Gott und durch institutionelle Aspekte geprägt ist. Houtman und Tromp (2021) sprechen daher von einer post-christlichen Spiritualität, die als Weltanschauung zu deuten ist, geprägt von der Annahme eines ganzheitlichen und lebendigen Kosmos, einer Immanenz des Sakralen und dem Gefühl eines spirituellen Selbst, das losgelöst von bestimmten religiösen Traditionen existiert. Zusammengefasst bezieht sich Huber (2008) stärker auf eine institutionelle Form von Religiosität, während Houtman und Tromp (2021) ebendies als Abgrenzungsmerkmal betrachten. Gleichzeitig gibt es Überschneidungen in den Konzepten. So ist davon auszugehen, dass Spiritualität auch in religiösen, institutionalisierten Kontexten ein zentrales Phänomen ist.
Religiosität, Spiritualität und Verschwörungsmythen
Der Zusammenhang zwischen Religiosität, Spiritualität und dem Glauben an Verschwörungsmythen wird seit einigen Jahren insbesondere theoretisch bearbeitet (z. B. Johannsen 2010; Ward und Voas 2011; Franks et al. 2013; Robertson und Dyrendal 2018; Wood und Douglas 2018; Parmigiani 2021). Aber auch empirisch wird das Thema mehr und mehr untersucht (z. B. Jasinskaja-Lahti und Jetten 2019; Gligoric et al. 2021; Ladini 2021; Frenken et al. 2022; Lowicki et al. 2022) – insbesondere aufgrund der Zunahme an Verschwörungsmythen im Kontext der COVID-19-Pandemie –, wenn auch in diesem Bereich noch weiterer Forschungsbedarf besteht.
Der theoretische Diskurs befasst sich vorwiegend mit der Frage, wie der Glaube an Verschwörungsmythen in Relation zu religiösem Glauben steht respektive ob Verschwörungsglaube ein Substitut für Religiosität oder Spiritualität darstellen kann. Robertson et al. (2018) unterscheiden zunächst zwischen den verschiedenen möglichen Erscheinungsformen von Verschwörung im Kontext Religion: Denkbar seien den Autoren zufolge Verschwörungsmythen in Religionen, z. B., wenn religiöse Gruppierungen Mythen über bestimmte Ereignisse verbreiten. Weiter gäbe es Verschwörungsmythen über Religionen und schliesslich Verschwörungsmythen als Religion (Robertson et al. 2018). Wood und Douglas (2018) gehen in diesem letzteren Zusammenhang davon aus, dass Verschwörungsmythen ähnliche Bedürfnisse adressieren wie auch Religionen, indem sie Erklärungen bieten und Ordnung in der zunehmend komplexer werdenden Welt schaffen, insbesondere in Situationen, die gefühlte Unsicherheit hervorrufen. Glaube an Verschwörungsmythen ist wie auch religiöser Glaube in diesem Kontext sinn- und bedeutungsstiftend und bietet (pseudo‑)transzendentale, unüberprüfbare Erklärungen für Phänomene, die ansonsten zusammenhanglos und unerklärbar erscheinen würden (Wood und Douglas 2018). Auch aufgrund der Annahmen, dass Verschwörungsmythen ähnlich wie religiöse Erzählungen schwer zu falsifizieren sind und sie identitätsstiftend wirken, werden sie als möglicher Religionsersatz in säkularisierten Gesellschaften betrachtet (Johannsen 2010). Verschwörungsmythen weisen zudem Ähnlichkeiten zu religiösen Ritualen auf, indem gemeinsame Antworten auf komplexe Fragen im Sinne eines gemeinsamen Glaubensbekenntnisses gegeben werden, was zudem gruppenbildend wirkt (Franks et al. 2013) – Aspekte, die beispielsweise von Wood und Douglas (2018) nicht als gemeinsamer Nenner zwischen Religiosität und Verschwörungsdenken betrachtet werden. Zudem unterscheiden sich den Autoren zufolge zwischen religiösen und verschwörungsmythischen Gruppierungen die Vorstellungen von einer gerechten Welt.
Empirische Studien, die den Zusammenhang zwischen Religiosität und Verschwörungsglauben untersuchen, kommen zu unterschiedlichen Resultaten. Eine Metastudie mit Befragungsdaten aus Deutschland, Polen und den Vereinigten Staaten kommt zu dem Befund, dass Religiosität positiv mit Verschwörungsdenken korreliert (Frenken et al. 2022). Die Autoren sehen dies in den vergleichbaren Erklärungsstilen und Ideologien begründet, wonach sowohl Religiosität als auch Verschwörungsmythen implizieren, dass mächtige Kräfte Schicksale der Welt bestimmen. Eine weitere Studie aus Polen bestätigt, dass Personen, die während der Pandemie mehr Zeit mit religiösen Praktiken verbracht haben, eher Massnahmen der Regierungen nicht beachteten und eine höhere Affinität gegenüber dem Glauben an Verschwörungsmythen haben (Boguszewski et al. 2020). Zu einem differenzierteren Befund kommen Lowicki et al. (2022), wonach nur eine besonders starke Ausprägung fundamentalistischer Religiosität positiv mit Verschwörungsglauben korreliert, während die auch für die vorliegende Studie genutzte Zentralitätsskala von Religiosität (Huber 2008) nicht oder sogar negativ damit in Zusammenhang steht. Auch Ladini (2021) spricht sich für einen differenzierten Blick auf das Thema aus, da in einer empirischen Studie in Italien ausschliesslich alternative Religiosität, gemessen an dem Glauben an Reinkarnation, mit Verschwörungsglauben positiv korreliert.
Während Studien mit einem Fokus auf Religiosität zu unterschiedlichen Erkenntnissen gelangen, scheint vor allem Spiritualität ein Treiber für Verschwörungsglauben zu sein. So zeigt eine aktuelle Studie von Gligoric et al. (2021), dass neben den psychologischen Faktoren Narzissmus und einer geringeren Ausprägung analytischen Denkens vor allem Spiritualität den Glauben an Verschwörungsmythen im Zusammenhang mit COVID-19 beeinflusst. Die Autorinnen und Autoren verweisen in diesem Zusammenhang auf die „New Age Spirituality“ (Gligoric et al. 2021, S. 1172), die die Grundlage bietet für den von Ward und Voas (2011) geprägten Begriff der „Conspirituality“. Conspirituality umschreibt hybride Glaubenssysteme und Weltanschauungen wie auch eine politisch-spirituelle Philosophie, wonach davon ausgegangen wird, dass geheime Gruppen die politische und soziale Ordnung kontrollieren, wie auch steuern, und dass nichts zufällig passiert. Gleichzeitig – so die Autor:innen – gäbe es ein neues Paradigma des Bewusstseins, das sich in der New Age Spirituality manifestiert und eine problemlösende Rolle einnimmt (Ward und Voas 2011; Asprem und Dyrendal 2015). Ähnlichkeiten zwischen Spiritualität und Verschwörungsglauben scheinen daher gegeben, unter anderem aufgrund der unterstellten Annahme, dass nichts zufällig passiere respektive Ereignisse miteinander in Zusammenhang stehen und nichts so ist, wie es scheint (Gligoric et al. 2021).
Obwohl in einigen Studien Spiritualität und Religiosität im Zusammenhang mit Verschwörungsglauben als ein Konstrukt betrachtet werden (z. B. Marques et al. 2021), zeigen die unterschiedlichen Resultate im aktuellen Forschungsstand, dass eine Trennung von Spiritualität und Religiosität essenziell ist. Aufgrund der erörterten aktuellen Studienbefunde und den Ähnlichkeiten zwischen Religiosität und Verschwörungsmythen gehen wir von einem positiven Zusammenhang zwischen den beiden Variablen aus. Zudem postulieren wir, dass Spiritualität aufgrund der Popularität von New Age Movements und ihren esoterischen Zügen (Knoblauch 2000), die sich auch in Verschwörungsmythen wiederfinden (Schwaiger 2022), einen stärkeren Zusammenhang mit der Affinität gegenüber Verschwörungsmythen aufweist als Religiosität.
H3
Je höher die Religiosität, desto höher die Affinität gegenüber Verschwörungsmythen.
H4
Je höher die Spiritualität, desto höher die Affinität gegenüber Verschwörungsmythen.
H5
Spiritualität hängt stärker mit der Affinität gegenüber Verschwörungsmythen zusammen als Religiosität.
Die Mehrheit aktueller Studien in dem Themenbereich Religiosität, Spiritualität und Verschwörung betrachtet die von den Befragten selbst eingeschätzte Religiosität und Spiritualität als mögliche Prädiktoren. Die Konfessionszugehörigkeit wird dabei nur selten herangezogen (z. B. Lobato et al. 2014). Wir möchten daher untersuchen, ob es Unterschiede zwischen den Konfessionen hinsichtlich der Affinität gegenüber Verschwörungsmythen gibt. Dabei gehen wir davon aus, dass in modernen, säkularisierten Gesellschaften wie der Schweiz, in der die institutionelle Kirchenbindung abnimmt (Bundesamt für Statistik 2022a), die selbst eingeschätzte Religiosität und Spiritualität stärkere Effekte auf Verschwörungsglauben hat als die Konfessionszugehörigkeit.
H6
Religiosität hängt stärker mit der Affinität gegenüber Verschwörungsmythen zusammen als die Konfessionszugehörigkeit.
H7
Spiritualität hängt stärker mit der Affinität gegenüber Verschwörungsmythen zusammen als die Konfessionszugehörigkeit.
Methodik
Daten
Die Daten wurden mittels einer repräsentativen Bevölkerungsbefragung in der deutsch- und französischsprachigen Schweiz vom 31. März bis zum 19. April 2022 erhoben. Der Online-Fragebogen wurde auf Deutsch erstellt und auf Französisch übersetzt. Die Befragung wurde in Kooperation mit dem Marktforschungsinstitut Intervista (2022) durchgeführt, das aus seinem zertifizierten Online-Panel mit über 100.000 aktiven Panelisten Teilnehmerinnen und Teilnehmer aus der deutsch- und französischsprachigen Schweiz rekrutierte. Der Fragebogen beinhaltete Fragen zu Religiosität und Spiritualität im Allgemeinen, wie auch der COVID-19-Pandemie und damit in Zusammenhang stehende Fragen zur Verschwörungsaffinität und emotionalen Betroffenheit. Insgesamt nahmen 1221 Personen zwischen 18 und 74 Jahren an der Befragung teil. Die französischsprachige Region wurde überproportional abgedeckt und in der Gesamtstichprobe entsprechend der Anteile an der Wohnbevölkerung zurückgewichtet. Innerhalb von vier Grossregionen wurden proportional zur Bevölkerungsverteilung Quotenzellen interlocked nach Geschlecht und Alter gebildet, wodurch auch die schweizweite Bildungsverteilung gut abgebildet wird (Bundesamt für Statistik 2022b). Der für die Analysen gewichtete Datensatz besteht somit aus 50 % männlichen und 49,5 % weiblichen Befragten; 0,5 % der Befragten haben eine nonbinäre Geschlechtszugehörigkeit angegeben. Die gewichtete Alters- und Bildungsverteilung wird in Tab. 1 zusammengefasst.Alter
18–29 Jahre 19,8 %
30–44 Jahre 29,0 %
45–59 Jahre 30,0 %
60–74 Jahre 21,2 %
Höchster Bildungsabschluss
Obligatorische Schule 17,3 %
Sekundarstufe 2 45,6 %
Tertiärstufe 37,1 %
Operationalisierung
Religionszugehörigkeit
Die Religionszugehörigkeit wurde als nominale Variable abgefragt. Wir unterscheiden in dieser Studie zwischen (1) römisch-katholisch; (2) evangelisch-reformiert; (3) freikirchlich; (4) sonstige religiöse Zugehörigkeit (z. B. orthodox, muslimisch, jüdisch, hinduistisch, buddhistisch oder sonstige) und (5) keine Religionszugehörigkeit.
Religiosität
Religiosität wurde unabhängig von der Religionszugehörigkeit abgefragt. Es handelt sich um fünf Items, die auf einer fünfstufigen Likert-Skala abgefragt wurden (z. B. „Wie oft kommt es vor, dass sie über religiöse Themen nachdenken?“). Die Items wurden von der Skala zur Messung von Religiosität von Huber (2008) übernommen (siehe Anhang) (α = 0,89; M = 2,43).
Spiritualität
Die Spiritualitätsskala besteht aus sieben Items, die ebenfalls auf einer fünfstufigen Likert-Skala abgefragt wurden (z. B. „Die persönliche Spiritualität ist wichtiger als die Zugehörigkeit zu einer religiösen Tradition“). Die Skala wurde aus dem Englischen von Houtman und Tromp (2021) übernommen und ins Deutsche übersetzt (siehe Anhang) (α = 0,88; M = 3,29).
Die Skalen Religiosität und Spiritualität wurden zudem mittels Hauptkomponentenanalyse mit Varimax-Rotation auf ihre Trennschärfe hin geprüft. Die resultierende Lösung mit zwei Komponenten, die einen Eigenwert grösser als eins aufweisen, entspricht den beiden Skalen (vgl. Tab. 2). Somit konnte die Trennschärfe der beiden Konstrukte (und Relevanz dieser) empirisch bestätigt werden. Die Lösung zeigt zudem, dass zwei Items der Spiritualitätsskala („Das gesamte Universum entspringt der einen universellen spirituellen Energie.“ und „Es gibt eine Art Geist oder Lebenskraft, die alles Leben durchdringt.“) zwar stärker auf die Komponente Spiritualität rotieren, aber auch auf der Komponente Religiosität schwache bis mittlere Ladungen erzielen. Daraus kann geschlossen werden, dass Religiosität auch spirituelle Einstellungen integriert, während sich dies umgekehrt nicht zeigt.Items Komponenten
Religiosität Spiritualität
Wie oft kommt es vor, dass Sie beten? 0,860 –
Wie oft kommt es vor, dass Sie Situationen erleben, in denen Sie das Gefühl haben, dass Gott oder etwas Göttliches in Ihr Leben eingreift? 0,827 –
Wie stark glauben Sie daran, dass es Gott oder etwas Göttliches gibt? 0,819 –
Wie oft kommt es vor, dass Sie über religiöse Themen nachdenken? 0,776 –
Wie oft kommt es vor, dass Sie an Gottesdiensten teilnehmen? 0,762 –
Jeder Mensch hat ein höheres spirituelles Selbst, das erweckt und erleuchtet werden kann – 0,787
Das Göttliche hat seinen Ursprung nicht ausserhalb, sondern in jedem Menschen – 0,729
Der Kosmos ist eine lebendige Einheit – 0,706
Die persönliche Spiritualität ist wichtiger als die Zugehörigkeit zu einer religiösen Tradition – 0,693
Die eine und einzig wahre Religion gibt es nicht, aber es gibt Wahrheiten, die man in allen Religionen der Welt finden kann – 0,684
Das gesamte Universum entspringt der einen universellen spirituellen Energie 0,366 0,666
Es gibt eine Art Geist oder Lebenskraft, die alles Leben durchdringt 0,473 0,662
Extraktionsmethode: Hauptkomponentenanalyse
Rotationsmethode: Varimax mit Kaiser-Normalisierung
Es sind nur Komponentenladungen grösser 0,3 dargestellt
Emotionale Betroffenheit während der Krise
Die Befragten mussten auf einer fünfstufigen Likert-Skala angeben, wie häufig sie unterschiedliche Emotionen während der COVID-19-Pandemie empfunden haben (z. B. Einsamkeit, Angst, Kontrollverlust). Die Items wurden übernommen von der „Studie zu Religion und gesellschaftlichem Zusammenhalt in Zeiten der Corona-Pandemie“ der Universität Münster (Yendell et al. 2021). Aus den acht Items wurden zwei Skalen gebildet, um positive Emotionen (3 Items; α = 0,68; M = 3,16) und negative Emotionen (5 Items; α = 0,86; M = 2,43) abzubilden (siehe Anhang).
Affinität gegenüber Verschwörungsmythen
Die Skala zur Messung der Affinität gegenüber Verschwörungsmythen (Schwaiger et al. 2022) besteht aus zwölf Items, die jeweils auf einer fünfstufigen Likert-Skala beantwortet wurden (siehe Anhang). Die Skala misst Affinität gegenüber Verschwörungsmythen in unterschiedlichen Dimensionen mit jeweils zwei Items, darunter Anti-Elitismus (z. B. „Man muss davon ausgehen, dass es geheime Eliten gibt, die die Welt kontrollieren“), Populismus (z. B. „Die Bürger:innen sind sich oft einig, aber die Politiker:innen verfolgen ganz andere Ziele“), Wissenschaftsskepsis (z. B. „Die sogenannte Wissenschaft verleugnet viele Wahrheiten, die nicht in ihr Konzept passen“), Anti-System-Einstellungen (z. B. „Das gesamte System ist im Kern verrottet“), Links-Extremismus (z. B. „Die Mächtigen sind daran interessiert, dass soziale Ungleichheit und Armut bestehen bleibt“), und Rechts-Extremismus (z. B. „Uns werden Veränderungen aufgezwungen, die unsere Kultur gefährden“). Es wurden bewusst keine konkreten Verschwörungsmythen abgefragt, sondern breitere Einstellungsmuster als Prädisposition für Verschwörungsglauben. Aus allen Items wurde ein Index gebildet (α = 0,94; M = 2,77).
Soziodemographische Variablen
Sämtliche Analysen wurden kontrolliert für Geschlecht (männlich, weiblich, divers), Alter (metrisch) und Bildung (niedrig, mittel, hoch).
Resultate
Die Verteilung nach Religionszugehörigkeiten ist vergleichbar mit der Schweizer Bevölkerungsverteilung (Bundesamt für Statistik 2022a), mit einem resultierenden Anteil von 30,4 % Katholikinnen und Katholiken, 25,3 % Evangelisch-Reformierten, 3,5 % Freikirchenmitgliedern, 5,5 % mit einer anderen Religionszugehörigkeit und 35,2 % ohne Religionszugehörigkeit.
Zur Testung der Hypothesen wurden insgesamt vier Regressionsmodelle durchgeführt, um Veränderungen in der Erklärungskraft durch Zunahme von Variablenblöcken zu veranschaulichen. Modell 1 beinhaltet neben den soziodemographischen Variablen Alter, Geschlecht und Bildung die beiden Indices zur emotionalen Betroffenheit während der Krise (positive Emotionen; negative Emotionen). In Modell 2 wurden die religiösen Konfessionen hinzugenommen (römisch-katholisch, evangelisch-reformiert, freikirchlich, sonstige religiöse Zugehörigkeit in Referenz zu keiner Zugehörigkeit). Modell 3 beinhaltet zusätzlich den Religiositätsindex und das finale Modell 4 wurde durch den Spiritualitätsindex ergänzt (vgl. Tab. 3). Durch das schrittweise Vorgehen wird deutlich, dass die Erklärungskraft der Modelle durch Hinzunahme der Konfessionszugehörigkeit in Modell 2 um 1,6 Prozentpunkte steigt; Religiosität als zusätzlicher Prädiktor erhöht die Erklärung in Modell 3 um weitere 1,2 Prozentpunkte. Der grösste Anstieg an Erklärungskraft erfolgt in Modell 4, durch Hinzunahme von Spiritualität als unabhängige Variable, mit einem resultierenden korrigierten R2 von 15,8 %. Modell 1 Modell 2 Modell 3 Modell 4
Beta (β) p Beta (β) p Beta (β) p Beta (β) p
Männlicha 0,198 0,322 0,227 0,253 0,229 0,246 0,147 0,442
Weiblicha 0,213 0,285 0,236 0,234 0,228 0,247 0,117 0,540
Alter 0,161 0,000 0,174 0,000 0,146 0,000 0,103 0,000
Bildung −0,181 0,000 −0,178 0,000 −0,177 0,000 −0,162 0,000
Positive Emotionen −0,020 0,484 −0,028 0,321 −0,048 0,091 −0,085 0,003
Negative Emotionen 0,150 0,000 0,148 0,000 0,130 0,000 0,101 0,001
Römisch-katholischb – – −0,017 0,595 −0,056 0,097 −0,034 0,294
Evangel.-reformiertb – – −0,046 0,147 −0,076 0,020 −0,056 0,082
Freikircheb – – 0,035 0,226 −0,017 0,596 0,036 0,246
Sonstige Religionb – – 0,116 0,000 0,080 0,010 0,101 0,001
Religiosität – – – – 0,133 0,000 0,021 0,549
Spiritualität – – – – – – 0,283 0,000
Korr. R2 0,071 0,087 0,099 0,158
N = 1221
aReferenz: divers
bReferenz: keine Religionszugehörigkeit
Die in Modell 1 signifikanten Prädiktoren Alter, Bildung und negative emotionale Betroffenheit während der COVID-19-Pandemie bleiben auch durch Zunahme weiterer Prädiktoren in den anderen Modellen hochsignifikant. So ist im finalen Modell 4 die Affinität gegenüber Verschwörungsmythen höher, je älter die Befragten sind (β = 0,103; p = 0,000) und je niedriger der Bildungsgrad ist (β = −0,162; p = 0,000). Das Geschlecht hingegen spielt keine signifikante Rolle in diesem Zusammenhang. Emotionale Betroffenheit hängt sowohl positiv wie auch negativ höchst signifikant mit Verschwörungsaffinität zusammen: Wenn positive Emotionen (z. B. Fürsorge und Liebe von anderen) während bzw. trotz der COVID-19-Pandemie gefühlt wurden, sinkt die Affinität gegenüber Verschwörungsmythen (β = −0,085; p = 0,003). Noch stärker ist der umgekehrte Effekt: Je ausgeprägter die negative emotionale Betroffenheit (z. B. Kontrollverlust) während der Krise, desto höher ist die Verschwörungsaffinität (β = 0,101; p = 0,001). Die Hypothesen H1 und H2 sind somit bestätigt.
Ein interessantes Bild zeigt sich hinsichtlich der unterschiedlichen Religionszugehörigkeiten. Im Vergleich zu Personen ohne religiöses Bekenntnis, haben nur sonstige Religionszugehörigkeiten einen positiven Effekt auf Verschwörungsaffinität (Modell 4: β = 0,101; p = 0,001). Demnach kann von einer Zugehörigkeit zu den beiden dominierenden christlichen Kirchen der Schweiz, der römisch-katholischen Kirche und der evangelisch-reformierten Kirche, oder von der Mitgliedschaft in Freikirchen nicht auf eine höhere oder niedrigere Affinität gegenüber Verschwörungsmythen geschlossen werden.
Von besonderem Interesse in dieser Studie ist die Frage, inwiefern Religiosität und Spiritualität auf Verschwörungsaffinität wirken. Hier lohnt sich ein Vergleich zwischen dem Regressionsmodell 3 und Modell 4. Während durch Zunahme des Religiositätsindex in Modell 3 die Erklärungskraft des Modells nur minimal steigt (Korrigiertes R2 = 0,099), obwohl ein hochsignifikanter positiver Effekt von Religiosität auf Verschwörungsaffinität feststellbar ist (β = 0,133; p = 0,000), verschwindet dieser signifikante Effekt in Modell 4 durch Hinzunahme des Spiritualitätsindex als Prädiktor. Zudem steigt die Erklärungskraft von Modell 4 um 5,9 Prozentpunkte auf 15,8 %. Die Resultate des finalen Modells zeigen somit, dass Verschwörungsaffinität durch Spiritualität vorhergesagt wird: Je spiritueller eine Person, desto höher ist die Affinität gegenüber Verschwörungsmythen (β = 0,283; p = 0,000). Im Vergleich zu den weiteren Prädiktoren handelt es sich um den stärksten Zusammenhang. Höhere Religiosität hingegen wirkt im finalen Modell nicht positiv verstärkend auf Verschwörungsaffinität.
Bezugnehmend auf unsere aufgestellten Hypothesen lässt sich somit festhalten, dass nur Spiritualität einen positiven signifikanten Effekt auf Verschwörungsaffinität hat, wodurch Hypothese H4 bestätigt ist. H3, die davon ausgeht, dass Religiosität ebenso eine beeinflussende Variable ist, kann nicht bestätigt werden. Die Annahme, dass Spiritualität einen stärkeren Effekt auf die Affinität gegenüber Verschwörungsmythen hat als Religiosität (H5) wird jedoch bestätigt. H6, die davon ausgeht, dass Religiosität stärker mit Verschwörungsaffinität zusammenhängt als die Religionszugehörigkeit, konnte nicht belegt werden, da Religiosität als Prädiktor im finalen Modell keinen signifikanten Effekt aufzeigt. Hingegen hat die Zugehörigkeit zu sonstigen Religionen (z. B. muslimisch, hinduistisch, buddhistisch) einen positiven Effekt auf Verschwörungsaffinität. Schliesslich wird H7 angenommen: Spiritualität hat einen stärkeren Effekt auf die Affinität gegenüber Verschwörungsmythen als die Religionszugehörigkeit. Wie diese Resultate zu interpretieren sind, wird in der abschliessend Diskussion zusammengefasst.
Diskussion
Die zunehmende Verbreitung von Verschwörungsmythen während der COVID-19-Pandemie verstärkte die bereits breite Auseinandersetzung mit der Thematik im wissenschaftlichen Diskurs. Neben den digitalen Möglichkeiten, die eine Verbreitung von nicht verifizierten Inhalten ermöglichen und erleichtern (Mahl et al. 2022), bestätigte sich während der Pandemie die Relevanz von krisenhaften Ereignissen für die Aufnahmebereitschaft von alternativen Weltdeutungen wie Verschwörungsmythen (Imhof 1996, 2011; van Prooijen und Douglas 2017). Der Zusammenhang zwischen Religiosität, Spiritualität und Verschwörungsglauben, der bislang vorwiegend theoretisch untersucht wurde (z. B. Franks et al. 2013; Robertson 2015; Dyrendal et al. 2018b), erlangte im Kontext der Pandemie neue Aufmerksamkeit für empirische Forschungen. Unter anderem werden die Fragen gestellt, wie Religiosität, Spiritualität und Verschwörungsglaube zusammenhängen; ob und inwiefern Religiosität respektive Spiritualität eine Art Ersatzreligion darstellen; oder ob Religiosität und Spiritualität die Affinität gegenüber Verschwörungsmythen in Zeiten der COVID-19-Pandemie erhöhen. Befeuert wurden diese Fragen durch die Annahme, dass insbesondere in krisenhaften Phasen, die von Unsicherheiten und Kontrollverlust geprägt sind, sowohl religiöser Glaube der Kontingenzbewältigung dienlich sein kann (Höllinger und Aschauer 2022), aber auch Verschwörungsglaube (Wood und Douglas 2018). Theoretische Studien sind sich zudem mehrheitlich einig, dass in Bezug auf sinnstiftende Erklärungen für Unerklärliches und gruppenbildende und identitätsstiftende Funktionen Ähnlichkeiten zwischen religiösem und spirituellem Glauben einerseits sowie Verschwörungsglauben andererseits gegeben sind (Johannsen 2010; Franks et al. 2013).
Insbesondere hinsichtlich der Rolle von Religiosität zeigt sich in bisherigen empirischen Studien ein kontroverses Bild. Während manche Studien von einem positiven Zusammenhang zwischen Religiosität und Verschwörungsglauben berichten (Boguszewski et al. 2020; Frenken et al. 2022), sprechen sich andere Autorinnen und Autoren für eine differenzierte Betrachtung aus, wonach beispielsweise nur bestimmte Ausprägungen von Religiosität mit Verschwörungsaffinität korrelieren, oder sogar negative Zusammenhänge festgestellt werden (Ladini 2021; Lowicki et al. 2022). Insbesondere die Abgrenzung zwischen Religiosität und alternativer Religiosität (Ladini 2021) respektive Spiritualität scheint im Kontext Verschwörungsaffinität von Bedeutung zu sein. So gibt es Befunde, dass Spiritualität den Glauben an Verschwörungsmythen positiv beeinflusst, gestützt durch die Annahme eines neuen Glaubenssystems der „Conspirituality“ (Ward und Voas 2011; Asprem und Dyrendal 2015; Gligoric et al. 2021).
Aufgrund der teils kontroversen bisherigen Befunde war es Ziel der vorliegenden Studie, den Zusammenhang zwischen Religiosität, Spiritualität und Verschwörungsaffinität empirisch zu überprüfen. Dabei war es uns insbesondere ein Anliegen, Religiosität und Spiritualität getrennt voneinander als Treiber für Verschwörungsaffinität zu betrachten. Dabei stützten wir uns auf etablierte Skalen der Religiosität (Huber 2008) und Spiritualität (Houtman und Tromp 2021), die im Rahmen dieser Studie mittels Hauptkomponentenanalyse auf ihre Trennbarkeit hin untersucht wurden. Die Analyse zeigt deutlich, dass die zwei theoretisch hergeleiteten Konstrukte Religiosität und Spiritualität empirisch getrennt voneinander betrachtet werden müssen, wenn auch einzelne Items der Spiritualität auch auf der Religiositätsskala höhere Ladungen erzielen. Diese Trennung von Religiosität und Spiritualität hilft schliesslich bei der Interpretation der Ergebnisse. An dieser Stelle muss jedoch angemerkt werden, dass die Resultate und deren Interpretation durch ebendiese Definitionen von Religiosität (Huber 2008) und Spiritualität (Houtman und Tromp 2021) klar bedingt sind. Die Wahl anderer theoretischer Definitionen und Operationalisierungen, die beispielsweise beide Konstrukte ähnlicher definieren, hätten das Ergebnis der Hauptkomponentenanalyse potenziell verändert.
Durch die insgesamt vier Regressionsmodelle zur Vorhersage von Affinität gegenüber Verschwörungsmythen konnte deutlich gezeigt werden, durch welche Prädiktoren die Erklärungskraft der Modelle gestärkt wird. So konnten zunächst die Hypothesen bestätigt werden, wonach emotionale Betroffenheit in Krisenphasen die Affinität gegenüber Verschwörungsmythen erhöht, sofern es sich um negative Emotionen handelt, und schliesslich senkt, wenn positive Emotionen gefühlt wurden. Dieser Befund bestätigt bisherige theoretische Annahmen, wonach Verschwörungsaffinität in belastendenden, krisenhaften Situationen zunimmt (z. B. van Prooijen und Douglas 2017). Schliesslich handelt es sich bei der COVID-19-Pandemie um eine traumatische Krise (Oevermann 2016), die gesellschaftliche und persönliche Routinen plötzlich und unvorhergesehen durchbrach und für deren Bewältigung es an gesellschaftlichen Erfahrungen und Normen mangelte. Die Suche nach klaren Antworten, die schliesslich durch verschwörungsmythische Narrative versprochen wurden, gewann dadurch an Relevanz. Die noch unzureichend erforschte Rolle von religiöser Konfessionszugehörigkeit im Zusammenhang mit Verschwörungsaffinität stellt sich als gering heraus. So zeigt sich, dass die Zugehörigkeit zu den zwei dominierenden christlichen Landeskirchen der Schweiz, der römisch-katholischen und der evangelisch-reformierten Kirche, sowie eine Mitgliedschaft in Freikirchen nicht signifikant Verschwörungsaffinität beeinflusst. Ausschliesslich die Zugehörigkeit zu sonstigen Religionen stellt sich als signifikanter Treiber für Verschwörungsaffinität im Vergleich zu keiner Konfessionszugehörigkeit heraus. Wir interpretieren diesen statistisch positiven Effekt als Zeichen dafür, dass Personen, die sich zu religiösen Minderheiten in der Schweiz zählen, eine geringere institutionelle und soziale Einbindung in der Schweiz aufweisen und daher ein höheres Risiko haben, sich durch den Glauben an Verschwörungsmythen einen entsprechenden Ersatz in krisenhaften Zeiten zu suchen. Eine eindeutige Interpretation ist allerdings schwierig, da in der Stichprobe dieser Studie nur 5,5 % der Befragten in diese Gruppe fallen und darunter unterschiedliche Religionen, die in der Schweiz zu den Minderheiten zählen, gefasst werden. Hier besteht weiterer Forschungsbedarf. Durch die Hinzunahme des Religiositätsindex in das Regressionsmodell wird darüber hinaus auch dieser Effekt kleiner. Besonders interessant, auch im Vergleich zu bisheriger empirischer Forschung, ist der Befund im finalen Regressionsmodell. So zeigt sich, dass insbesondere Spiritualität ein Treiber für Verschwörungsaffinität ist, der nicht nur die Erklärungskraft des Regressionsmodells deutlich erhöht, sondern auch offenbart, dass Religiosität an sich in Summe keinen signifikanten Einfluss auf die Affinität gegenüber Verschwörungsmythen hat. Spiritualität lässt sich zudem insgesamt als stärkster Effekt identifizieren, auch gegenüber üblichen signifikanten Prädiktoren wie Alter und Bildung. Es ist davon auszugehen, dass Spiritualität den Religiositätseffekt vermindert, zumal auch religiöse Personen in gewisser Weise spirituell sind, allerdings andere Faktoren der Spiritualität stärker mit Verschwörungsglauben in Einklang zu bringen sind. Mit Verweis auf die Hauptkomponentenanalyse wird diese Interpretation deutlich: So rotieren auch einzelne Items der Spiritualitätsskala auf die Komponente Religiosität, umgekehrt zeigt sich dieser Befund jedoch nicht. Weiter lässt sich das Resultat auch theoretisch, unter Rückbezug auf die Definitionen von Religiosität, Spiritualität und Verschwörungsmythen, interpretieren. So operationalisieren wir in dieser Studie Religiosität nach dem Religiositätskonzept von Huber (2008), wonach Religiosität stärker an institutionellen Indikatoren gemessen wird, wie beispielsweise mittels der Häufigkeit des Gottesdienstbesuches oder des Betens. Es ist davon auszugehen, dass eine stärkere institutionell geprägte Religiosität die Verschwörungsaffinität nicht signifikant erhöht, da diese gesellschaftlichen und institutionell etablierten Strukturen bereits Halt in Krisenphasen geben. Spiritualität hingegen ist individueller geprägt und losgelöst von konkreten religiösen Traditionen (Houtman und Tromp 2021), womit eine geringere soziale Einbindung einhergehen kann. Spiritualität als Konzept distanziert sich demnach stärker von gesellschaftlich etablierten Strukturen, einem Narrativ, dem sich auch Verschwörungsmythen mitunter bedienen, indem sie sich gegen die öffentliche Meinung richten.
In Summe möchten wir durch dieses Resultat verdeutlichen, dass eine Trennung von Religiosität und Spiritualität in dem sensiblen Kontext der Erforschung von Verschwörungsglauben essenziell ist, um Fehlschlüsse zu vermeiden. Das stufenweise Vorgehen im Zuge der Regressionsanalysen zeigt, dass eine Analyse ohne den Prädiktor Spiritualität die Annahme bestärkt hätte, dass Religiosität ein Treiber für Verschwörungsaffinität ist. Der Effekt verflüchtigt sich allerdings, wenn zusätzlich Spiritualität betrachtet wird. Somit konnten unsere Hypothesen teilweise bestätigt werden: Je höher die Spiritualität, desto grösser ist die Affinität gegenüber Verschwörungsmythen. Der Effekt ist stärker als die religiöse Zugehörigkeit. Nicht bestätigt werden konnte die Hypothese, dass auch Religiosität ein positiver Treiber für Verschwörungsaffinität ist.
Jede Studie hat Limitationen, die auch hier mitberücksichtigt werden müssen. Wir haben in dieser Studie neben soziodemographischen Kontrollvariablen insbesondere auf emotionale Betroffenheit während der COVID-19-Pandemie, religiöse Zugehörigkeit und schliesslich Religiosität und Spiritualität als Prädiktoren für Affinität gegenüber Verschwörungsmythen fokussiert. Selbsterklärend gibt es weitere bereits häufig erforschte Variablen, die Verschwörungsaffinität signifikant beeinflussen, wie beispielsweise politische Einstellungen (z. B. Castanho Silva et al. 2017; Hollander 2018; Wood und Gray 2019), Misstrauen gegenüber der Regierung oder Autoritäten (z. B. Darwin et al. 2011; Bode und Vraga 2018) oder Mediennutzung (z. B. Stecula und Pickup 2021). Wir haben in dieser Studie bewusst auf Prädiktoren wie Religiosität und Spiritualität fokussiert, wenn auch weitere Variablen die Erklärungskraft der Regressionsmodelle erhöhen würden. In Hinblick auf bereits publizierte empirische Studien, die sich mit dem Zusammenhang zwischen Religiosität und Verschwörungsdenken auseinandersetzen, gibt es weitere zu berücksichtigende Aspekte. Insbesondere definitorische Unterschiede respektive unterschiedliche Operationalisierungen können zu variierenden Resultaten führen. Es macht beispielsweise einen Unterschied, ob Religiosität anhand der religiösen Konfession operationalisiert wird, oder die beiden Indikatoren getrennt voneinander abgefragt werden, wie in der vorliegenden Studie. Ebenso verhält es sich, wenn Religiosität und Spiritualität nicht getrennt voneinander abgefragt werden. Daraus können Fehlschlüsse resultieren, die wir in dieser Studie versucht haben zu vermeiden. Die Schwierigkeit, einheitliche Definitionen für Religiosität wie auch Spiritualität zu entwickeln, ist im wissenschaftlichen Diskurs bekannt – und verhält sich im Kontext des Verschwörungsbegriffs nicht anders. So sprechen Dyrendal et al. (2018a, S. 21) im Zusammenhang mit den Begriffen ‚Religion‘ und ‚Verschwörungstheorien‘ von „two fuzzy terms“. Insbesondere die Operationalisierung von Verschwörungsglauben respektive -affinität variiert. Es existieren beispielsweise viele Studien, die konkrete Verschwörungsmythen und deren Glaubwürdigkeit abfragen (Bruns et al. 2020; Balta et al. 2021; Eberl et al. 2021). In der vorliegenden Studie haben wir uns dazu entschieden, nicht konkrete Ereignisse mit verschwörungsmythischem Charakter in die Befragung mit aufzunehmen, sondern breitere Einstellungsmuster, die Affinitäten im Sinne von Prädispositionen für Verschwörungsglauben abbilden.
Abschliessend stellt sich die Frage, inwiefern Verschwörungsglaube als Ersatz für religiösen Glauben gedeutet werden kann. Den Resultaten zufolge erweist sich Religiosität per se nicht als signifikanter Treiber für Verschwörungsaffinität, weshalb postuliert werden kann, dass religiöser Glaube seine Funktion zur Kontingenzbewältigung in krisenhaften Phasen erfüllt. Vielmehr befördert Spiritualität die Affinität gegenüber Verschwörungsmythen. Wir schliessen daraus, dass in säkularisierten Gesellschaften Spiritualität ohne die Einbindung in traditionelle religiöse Gemeinschaften mit einer verstärkten Offenheit für Verschwörungsmythen als Ersatzreligion einhergeht.
Anhang
Operationalisierungen
Religiosität (Huber 2008)
Wie oft kommt es vor, dass Sie … (1 = gar nicht; 5 = sehr oft)… über religiöse Themen nachdenken?
… an Gottesdiensten teilnehmen?
… beten?
… Situationen erleben, in denen Sie das Gefühl haben, dass Gott oder etwas Göttliches in Ihr Leben eingreift?
Wie stark glauben Sie daran, dass es Gott oder etwas Göttliches gibt? (1 = überhaupt nicht; 5 = sehr stark)
Spiritualität (übersetzt aus dem Englischen; Houtman und Tromp 2021)
Wie stark stimmen Sie den folgenden Aussagen zu? (1 = stimme gar nicht zu; 5 = stimme voll und ganz zu)Die persönliche Spiritualität ist wichtiger als die Zugehörigkeit zu einer religiösen Tradition.
Jeder Mensch hat ein höheres spirituelles Selbst, das erweckt und erleuchtet werden kann.
Es gibt eine Art Geist oder Lebenskraft, die alles Leben durchdringt.
Das Göttliche hat seinen Ursprung nicht ausserhalb, sondern in jedem Menschen.
Die eine und einzig wahre Religion gibt es nicht, aber es gibt Wahrheiten, die man in allen Religionen der Welt finden kann.
Der Kosmos ist eine lebendige Einheit.
Das gesamte Universum entspringt der einen universellen spirituellen Energie.
Emotionen während der COVID-19 Pandemie (Yendell et al. 2021)
Wie oft haben Sie in Zeiten der Corona-Pandemie folgende Emotionen empfunden? (1 = nie; 5 = sehr oft)Energie für Neues (Skala: Positive Emotionen)
Einsamkeit (Skala: Negative Emotionen)
Hilflosigkeit (Skala: Negative Emotionen)
Fürsorge und Liebe von anderen (Skala: Positive Emotionen)
Angst (Skala: Negative Emotionen)
Optimismus (Skala: Positive Emotionen)
Unsicherheit (Skala: Negative Emotionen)
Kontrollverlust (Skala: Negative Emotionen)
Affinität gegenüber Verschwörungsmythen (Schwaiger et al. 2022)
Wie stark stimmen Sie den folgenden Aussagen über die heutige Gesellschaft in der Schweiz zu? (1 = stimme überhaupt nicht zu; 5 = stimme voll und ganz zu)Die Bürger:innen sind sich oft einig, aber die Politiker:innen verfolgen ganz andere Ziele.
Was man in der Politik „Kompromiss“ nennt, ist in Wirklichkeit nichts Anderes als ein Verrat der eigenen Prinzipien.
Man muss davon ausgehen, dass es geheime Eliten gibt, die die Welt kontrollieren.
Vieles was geschieht, wird von Personen und Organisationen kontrolliert, die im Verborgenen agieren.
Die Wissenschaft wird durch wirtschaftliche und politische Interessen gesteuert.
Die sogenannte Wissenschaft verleugnet viele Wahrheiten, die nicht in ihr Konzept passen.
Heutzutage bekommt man immer mehr vorgeschrieben, was man zu denken hat.
Uns werden Veränderungen aufgezwungen, die unsere Kultur gefährden.
Die Mächtigen sind daran interessiert, dass soziale Ungleichheit und Armut bestehen bleibt.
Grosskonzerne und die Wirtschaft insgesamt sorgen dafür, dass sich die Machtverhältnisse nicht ändern.
Das gesamte System ist im Kern verrottet.
Das politische System müsste grundlegend verändert werden.
==== Refs
Literatur
Asprem E Dyrendal A Conspirituality reconsidered: how surprising and how new is the confluence of spirituality and conspiracy theory? Journal of Contemporary Religion 2015 30 3 367 382 10.1080/13537903.2015.1081339
Balta E Rovira Kaltwasser C Yagci AH Populist attitudes and conspiratorial thinking Party Politics 2021 10.1177/13540688211003304
Bentzen Sinding J In crisis, we pray: religiosity and the COVID-19 pandemic Journal of Economic Behavior and Organization 2021 192 541 583 10.1016/j.jebo.2021.10.014 34744223
Biddlestone M Green R Douglas KM Cultural orientation, power, belief in conspiracy theories, and intentions to reduce the spread of COVID-19 British Journal of Social Psychology 2020 59 3 663 673 10.1111/bjso.12397 32592420
Bode L Vraga EK See something, say something: correction of global health misinformation on social media Health Communication 2018 33 9 1131 1140 10.1080/10410236.2017.1331312 28622038
Boguszewski R Makowska M Bożewicz M Podkowińska M The COVID-19 pandemic’s impact on religiosity in Poland Religions 2020 11 12 1 14 10.3390/rel11120646
Bruns A Harrington S Hurcombe E ‘Corona? 5G? or both?’: the dynamics of COVID-19/5G conspiracy theories on Facebook Media International Australia 2020 177 1 12 29 10.1177/1329878x20946113
Bundesamt für Statistik. 2022a. Religionen. https://www.bfs.admin.ch/bfs/de/home/statistiken/bevoelkerung/sprachen-religionen/religionen.html. Zugegriffen: 6. Apr. 2022.
Bundesamt für Statistik. 2022b. Stand und Entwicklung. https://www.bfs.admin.ch/bfs/de/home/statistiken/bevoelkerung/stand-entwicklung.html. Zugegriffen: 28. Apr. 2022.
Castanho Silva B Vegetti F Littvay L The elite is up to something: exploring the relation between populism and belief in conspiracy theories Swiss Political Science Review 2017 23 4 423 443 10.1111/spsr.12270
Clarke S Conspiracy theories and conspiracy theorizing Philosophy of the Social Sciences 2002 32 2 131 150 10.1177/004931032002001
Darwin H Neave N Holmes J Belief in conspiracy theories. The role of paranormal belief, paranoid ideation and schizotypy Personality and Individual Differences 2011 50 8 1289 1293 10.1016/j.paid.2011.02.027
Douglas KM Sutton RM Cichocka A The psychology of conspiracy theories Current Directions in Psychological Science 2017 26 6 538 542 10.1177/0963721417718261 29276345
Douglas KM Uscinski JE Sutton RM Cichocka A Nefes T Siang CA Understanding conspiracy theories Political Psychology 2019 40 S1 3 35 10.1111/pops.12568
Dyrendal A Asprem E Robertson DG Dyrendal ADGR Asprem E Conspiracy theories and the study of religion(s): what we are talking about, and why it is important Handbook of Conspiracy Theory and Contemporary Religion 2018 Leiden Brill 21 47
Dyrendal ADGR Asprem E Handbook of conspiracy theory and contemporary religion 2018
Eberl J-M Huber RA Greussing E From populism to the “plandemic”: why populists believe in COVID-19 conspiracies Journal of Elections, Public Opinion and Parties 2021 31 sup1 272 284 10.1080/17457289.2021.1924730
Franks B Bangerter A Bauer MW Conspiracy theories as quasi-religious mentality: an integrated account from cognitive science, social representations theory, and frame theory Frontiers in Psychology 2013 4 1 12 10.3389/fpsyg.2013.00424 23382719
Frenken M Bilewicz M Imhoff R On the relation between religiosity and the endorsement of conspiracy theories: the role of political orientation Political Psychology 2022 10.1111/pops.12822
Gemenis K Explaining conspiracy beliefs and scepticism around the COVID-19 pandemic Swiss Political Science Review 2021 27 2 229 242 10.1111/spsr.12467 35923362
Gligoric V Moreira da Silva M Eker S van Hoek N Nieuwenhuijzen E Popova U The usual suspects: How psychological motives and thinking styles predict the endorsement of well-known and COVID-19 conspiracy beliefs Applied Cognitive Psychology 2021 35 1171 1181 10.1002/acp.3844 34177101
Hollander BA Partisanship, individual differences, and news media exposure as predictors of conspiracy beliefs Journalism & Mass Communication Quarterly 2018 95 3 619 713 10.1177/1077699017728919
Höllinger F Aschauer W Aschauer W Glatz C Prandner D Die Bedeutung von Religiosität und Spiritualität in Krisenzeiten Die österreichische Gesellschaft während der Corona-Pandemie. Ergebnisse aus sozialwissenschaftlichen Umfragen 2022 Wiesbaden Springer VS 121 153
Houtman D Tromp P Ai AL Wink P Paloutzian RF Harris KA The post-Christian spirituality scale (PCSS): misconceptions, obstacler, prospects Assessing spirituality in a diverse world 2021 Cham Springer 35 57
Huber S Kerndimensionen, Zentralität und Inhalt. Ein interdisziplinäres Modell der Religiosität Journal für Psychologie 2008 16 3 1 17 10.7892/boris.73105
Humprecht E Esser F van Aelst P Resilience to online disinformation: a framework for cross-national comparative research The International Journal of Press/Politics 2020 25 3 493 516 10.1177/1940161219900126
Imhof K Imhof K Romano G Intersubjektivität und Moderne Die Diskontinuität der Moderne. Zur Theorie sozialen Wandels 1996 Frankfurt am Main Campus 200 292
Imhof K Die Krise der Öffentlichkeit. Kommunikation und Medien als Faktoren des sozialen Wandels 2011 Frankfurt am Main Campus
intervista. 2022. https://www.intervista.ch. Zugegriffen: 6. Apr. 2022.
Jasinskaja-Lahti I Jetten J Unpacking the relationship between religiosity and conspiracy beliefs in Australia British Journal of Social Psychology 2019 58 4 938 954 10.1111/bjso.12314 30706498
Johannsen K Fathi S Verschwörungstheorien erkennen und analysieren Komplotte, Ketzer und Konspirationen. Zur Logik des Verschwörungsdenkens – Beispiele aus dem Nahen Osten 2010 Bielefeld transcript 33 42
Kay AC Gaucher D McGregor I Nash K Religious belief as compensatory control Personality and Social Psychology Review 2010 14 1 37 48 10.1177/1088868309353750 20040614
Kirchler E Pitters J Kastlunger B Kirchler E Pitters J Kastlunger B Beginn der Krise Psychologie in Zeiten der Krise. Eine wirtschaftspsychologische Analyse der Coronavirus-Pandemie 2020 Wiesbaden Springer 7 20
Knoblauch H Populäre Religion. Markt, Medien und Popularisierung von Religion Zeitschrift für Religionswissenschaft 2000 8 143 161 10.1515/zfr.2000.8.2.143
Konkes C Lester L Incomplete knowledge, rumour and truth seeking Journalism Studies 2015 18 7 826 844 10.1080/1461670x.2015.1089182
Ladini R Religious and conspiracist? An analysis of the relationship between the dimensions of individual religiosity and belief in a big pharma conspiracy theory Italian Political Science Review/Rivista Italiana di Scienza Politica 2021 52 1 33 50 10.1017/ipo.2021.15
Lobato E Mendoza J Sims V Chin M Examining the relationship between conspiracy theories, paranormal beliefs, and pseudoscience acceptance among a university population Applied Cognitive Psychology 2014 28 5 617 625 10.1002/acp.3042
Lowicki P Marchlewska M Molenda Z Karakula A Szczepanska D Does religion predict coronavirus conspiracy beliefs? Centrality of religiosity, religious fundamentalism, and COVID-19 conspiracy beliefs Personality and Individual Differences 2022 187 111413 10.1016/j.paid.2021.111413 34866726
Lucchetti Giancarlo Goes Leonardo G Amaral Stefanie G Ganadjian Gabriela T Andrade Isabelle Almeida Paulo OA Spirituality, religiosity and the mental health consequences of social isolation during Covid-19 pandemic International Journal of Social Psychiatry 2021 67 6 672 679 10.1177/0020764020970996 33135559
Mahl D Schäfer MS Zeng J Conspiracy theories in online environments: an interdisciplinary literature review and agenda for future research new media & society 2022 10.1177/14614448221075759
Marques MD Ling MMNW Kerr JR McLennan J Australasian public awareness and belief in conspiracy theories: motivational correlates Political Psychology 2021 43 1 177 198 10.1111/pops.12746
Neuberger C Nuernbergk C Neuberger C Journalismus in der Netzwerköffentlichkeit. Zum Verhältnis zwischen Profession, Partizipation und Technik Journalismus im Internet 2018 Wiesbaden Springer 11 80
Oevermann U Becker-Lenz R Franzmann A Jansen A Jung M „Krise und Routine“ als analytisches Paradigma in den Sozialwissenschaften Die Methodenschule der Objektiven Hermeneutik 2016 Wiesbaden Springer VS 43 114
Parmigiani G Magic and politics: conspirituality and COVID-19 Jornal of the American Academy of Religion 2021 89 2 506 529 10.1093/jaarel/lfab053
van Prooijen Jan-Willem Douglas Karen M Conspiracy theories as part of history: the role of societal crisis situations Memory Studies 2017 10 3 323 333 10.1177/1750698017701615 29081831
Pummerer L Böhm R Lilleholt L Winter K Zettler I Sassenberg K Conspiracy theories and their societal effects during the COVID-19 pandemic Social Psychological and Personality Science 2021 10.1177/19485506211000217
Roberto A Sellon ASTC Josalin H-J Winslow H Impact of spirituality on resilience and coping during the COVID-19 crisis: A mixed-method approach investigating the impact on women Health Care for Women International 2020 41 11–12 1313 1334 10.1080/07399332.2020.1832097 33085579
Robertson DG Conspiracy theories and the study of alternative and emergent religions Nova Religio 2015 19 2 5 16 10.1525/nr.2015.19.2.5
Robertson DG UFOs, conspiracy theories and the new Age. Millennial conspiracism 2016 London Bloomsbury Publishing (Bloomsbury Advances in Religious Studies)
Robertson DG Dyrendal A Uscinski JE Conspiracy theories and religion Conspiracy theories and the people who believe them 2018
Robertson DG Asprem E Dyrendal A Dyrendal ADGR Asprem E Introducing the field: conspiracy theory in, about, and as religion Handbook of conspiracy theory and contemporary religion 2018 Leiden Brill 1 18
Scheufele Dietram A Krause Nicole M Science audiences, misinformation, and fake news Proceedings of the National Academy of Sciences of the United States of America 2019 116 16 7662 7669 10.1073/pnas.1805871115 30642953
Schwaiger L Gegen die Öffentlichkeit. Alternative Nachrichtenmedien im deutschsprachigen Raum 2022 Bielefeld transcript
Schwaiger L Schneider J Rauchfleisch A Eisenegger M Mindsets of conspiracy: a typology of affinities towards conspiracy myths in digital environments Convergence: The International Journal of Research into New Media Technologies 2022 10.1177/13548565221106427
Spiegel S Anton A Amlinger C Pause J Nitzke S Verschwörungstheorien als narratives Phänomen Zeitschrift für Fantastikforschung 2020 8 1 1 49 10.16995/zff.3415
Srol Jakub Ballova Mikuskova Eva B Cavojova Vladimira When we are worried, what are we thinking? Anxiety, lack of control, and conspiracy beliefs amidst the COVID-19 pandemic Appl Cogn Psychol 2021 10.1002/acp.3798
Stecula DA Pickup M How populism and conservative media fuel conspiracy beliefs about COVID-19 and what it means for COVID-19 behaviors Research and Politics 2021 10.1177/2053168021993979
Vogler D Schwaiger L Rauchfleisch A Marschlich S Siegen D Udris L fög – Forschungszentrum Öffentlichkeit und Gesellschaft Wahrnehmung von Desinformation in der Schweiz Jahrbuch Qualität der Medien. Schweiz – Suisse – Svizzera 2021 Basel Schwabe 25 36
Ward C Voas D The emergence of conspirituality Journal of Contemporary Religion 2011 26 1 103 121 10.1080/13537903.2011.539846
Wood M Douglas K Dyrendal ADGR Asprem E Are conspiracy theories a surrogate for god? Handbook of conspiracy theory and contemporary religion 2018 Leiden Brill 87 105
Wood MJ Gray D Right-wing authoritarianism as a predictor of pro-establishment versus anti-establishment conspiracy theories Personality and Individual Differences 2019 138 163 166 10.1016/j.paid.2018.09.036
Yendell A Hidalgo O Hillenbrand C Die Rolle von religiösen Akteuren in der COVID-19-Pandemie: Eine theoriegeleitete empirische Analyse mit politischen Handungsempfehlungen 2021 Stuttgart ifa
Zollo Fabiana Novak Petra K Del Vicario Michaela Bessi Alessandro Mozetic Igor Scala Antonio Emotional dynamics in the age of misinformation PLoS One 2015 10 9 e0138740 10.1371/journal.pone.0138740 26422473
| 36465327 | PMC9707184 | NO-CC CODE | 2022-12-01 23:20:21 | no | Z Relig Ges Polit. 2022 Nov 29;:1-22 | utf-8 | Z Relig Ges Polit | 2,022 | 10.1007/s41682-022-00136-x | oa_other |
==== Front
Nano Res
Nano Res
Nano Research
1998-0124
1998-0000
Tsinghua University Press Beijing
5160
10.1007/s12274-022-5160-2
Research Article
Anion-tuned nickel chalcogenides electrocatalysts for efficient 2e− ORR towards H2O2 production in acidic media
Sun Qingjia 1
Xu Guanxing 1
Xiong Bingyan 1
Chen Lisong [email protected]
12
Shi Jianlin [email protected]
3
1 grid.22069.3f 0000 0004 0369 6365 Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai, 200062 China
2 grid.22069.3f 0000 0004 0369 6365 Institute of Eco-Chongming, Shanghai, 202162 China
3 grid.9227.e 0000000119573309 State Key Laboratory of High Performance Ceramics and Superfine Microstructures, Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai, 200050 China
29 11 2022
17
27 8 2022
28 9 2022
7 10 2022
© Tsinghua University Press 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Electrocatalytic 2e− oxygen reduction reaction (2e− ORR) is a promising approach to producing H2O2 at ambient temperature and pressure especially in acidic media, which, however, is hindered by the high cost of precious metal-based electrocatalysts. Hence, the development of efficient earth-abundant electrocatalysts and reaction mechanism exploration for H2O2 production by 2e− ORR in acidic solution are critically important but remain challenging at present. In this work, NiSe2 has been developed as a novel and high-performance 2e− ORR electrocatalyst in acidic media, moreover, using nickel chalcogenides as the models, the influence of different anion species (Se22−, S22−, and O2−) on 2e− ORR electrocatalytic performance of the catalysts has been investigated. The synthesized NiSe2 exhibits outstanding 2e− ORR performance of high selectivity (90%) and long-term durability (12 h). The maximum H2O2 concentration of NiSe2 reaches 988 ppm, which is the highest among all the reported transition metal chalcogenides. This work demonstrates a novel point of view in anion tuning for designing high-efficiency transition-metal-based electrocatalysts for 2e− ORR.
Electronic Supplementary Material
Supplementary material (additional experimental procedures, characterizations, and computational details) is available in the online version of this article at 10.1007/s12274-022-5160-2.
Keywords
hydrogen peroxide
oxygen reduction reaction
electrosynthesis
anions
==== Body
pmcElectronic supplementary material
Anion-tuned nickel chalcogenides electrocatalysts for efficient 2e− ORR towards H2O2 production in acidic media
Appendix
Acknowledgements
This work was supported by Shanghai Science and Technology Committee Rising-Star Program (No. 22QA1403400), the Natural Science Foundation of Shanghai (No. 21ZR1418700), and the National Natural Science Foundation of China (No. 52201271). The authors would like to thank ECNU Multifunctional Platform for Innovation for support of SEM and TEM characterizations (004).
==== Refs
References
[1] Tang J Y Zhao T S Solanki D Miao X B Zhou W G Hu S Selective hydrogen peroxide conversion tailored by surface, interface, and device engineering Joule 2021 5 1432 1461 10.1016/j.joule.2021.04.012
[2] Sun Y Y Han L Strasser P A comparative perspective of electrochemical and photochemical approaches for catalytic H2O2 production Chem. Soc. Rev. 2020 49 6605 6631 10.1039/D0CS00458H 32760937
[3] Jiang K Zhao J J Wang H T Catalyst design for electrochemical oxygen reduction toward hydrogen peroxide Adv. Funct. Mater. 2020 30 2003321 10.1002/adfm.202003321
[4] Han G H Lee S H Hwang S Y Lee K Y Advanced development strategy of nano catalyst and DFT calculations for direct synthesis of hydrogen peroxide Adv. Energy Mater. 2021 11 2003121 10.1002/aenm.202003121
[5] Pang Y Y Xie H Sun Y Titirici M M Chai G L Electrochemical oxygen reduction for H2O2 production: Catalysts, pH effects and mechanisms J. Mater. Chem. A 2020 8 24996 25016 10.1039/D0TA09122G
[6] Xia C Xia Y Zhu P Fan L Wang H T Direct electrosynthesis of pure aqueous H2O2 solutions up to 20% by weight using a solid electrolyte Science 2019 366 226 231 10.1126/science.aay1844 31601767
[7] Zhang X D Zeng Y M Shi W Y Tao Z Liao J J Ai C Z Si H W Wang Z P Fisher A C Lin S W S-scheme heterojunction of core—shell biphase (1T-2H)-MoSe2/TiO2 nanorod arrays for enhanced photoelectrocatalytic production of hydrogen peroxide Chem. Eng. J. 2022 429 131312 10.1016/j.cej.2021.131312
[8] Wang Y L Waterhouse G I N Shang L Zhang T R Electrocatalytic oxygen reduction to hydrogen peroxide: From homogeneous to heterogeneous electrocatalysis Adv. Energy Mater. 2021 11 2003323 10.1002/aenm.202003323
[9] Liu Y M Quan X Fan X F Wang H Chen S High-yield electrosynthesis of hydrogen peroxide from oxygen reduction by hierarchically porous carbon Angew. Chem., Int. Ed. 2015 54 6837 6841 10.1002/anie.201502396
[10] Kim H W Ross M B Kornienko N Zhang L Guo J H Yang P D McCloskey B D Efficient hydrogen peroxide generation using reduced graphene oxide-based oxygen reduction electrocatalysts Nat. Catal. 2018 1 282 290 10.1038/s41929-018-0044-2
[11] Wang Z Li Q K Zhang C H Cheng Z H Chen W Y McHugh E A Carter R A Yakobson B I Tour J M Hydrogen peroxide generation with 100% faradaic efficiency on metal-free carbon black ACS Catal. 2021 11 2454 2459 10.1021/acscatal.0c04735
[12] Jung E Shin H Lee B H Efremov V Lee S Lee H S Kim J Antink W H Park S Lee K S Atomic-level tuning of Co-N-C catalyst for high-performance electrochemical H2O2 production Nat. Mater. 2020 19 436 442 10.1038/s41563-019-0571-5 31932671
[13] Yang Q H Xu W W Gong S Zheng G K Tian Z Q Wen Y J Peng L M Zhang L J Lu Z Y Chen L Atomically dispersed Lewis acid sites boost 2-electron oxygen reduction activity of carbon-based catalysts Nat. Commun. 2020 11 5478 10.1038/s41467-020-19309-4 33127912
[14] Zhang F F Zhu Y L Tang C Chen Y Qian B B Hu Z W Chang Y C Pao C W Lin Q Kazemi S A High-efficiency electrosynthesis of hydrogen peroxide from oxygen reduction enabled by a tungsten single atom catalyst with unique terdentate N1O2 coordination Adv. Funct. Mater. 2022 32 2110224 10.1002/adfm.202110224
[15] Miner E M Fukushima T Sheberla D Sun L Surendranath Y Dinca M Electrochemical oxygen reduction catalysed by Ni3(hexaiminotriphenylene)2 Nat. Commun. 2016 7 10942 10.1038/ncomms10942 26952523
[16] Wang M J Dong X Meng Z D Hu Z W Lin Y G Peng C K Wang H S Pao C W Ding S Y Li Y Y An efficient interfacial synthesis of two-dimensional metal-organic framework nanosheets for electrochemical hydrogen peroxide production Angew. Chem., Int. Ed. 2021 60 11190 11195 10.1002/anie.202100897
[17] Liu C Li H Chen J S Yu Z X Ru Q Li S Z Henkelman G Wei L Chen Y 3d transition-metal-mediated columbite nanocatalysts for decentralized electrosynthesis of hydrogen peroxide Small 2021 17 2007249 10.1002/smll.202007249
[18] Gao R J Pan L Li Z W Shi C X Yao Y D Zhang X W Zou J J Engineering facets and oxygen vacancies over hematite single crystal for intensified electrocatalytic H2O2 production Adv. Funct. Mater. 2020 30 1910539 10.1002/adfm.201910539
[19] Gao J J Liu B Progress of electrochemical hydrogen peroxide synthesis over single atom catalysts ACS Materials Lett. 2020 2 1008 1024 10.1021/acsmaterialslett.0c00189
[20] Zhang J Y Xia C Wang H F Tang C Recent advances in electrocatalytic oxygen reduction for on-site hydrogen peroxide synthesis in acidic media J. Energy Chem. 2022 67 432 450 10.1016/j.jechem.2021.10.013
[21] Jirkovsky J S Panas I Ahlberg E Halasa M Romani S Schiffrin D J Single atom hot-spots at Au-Pd nanoalloys for electrocatalytic H2O2 production J. Am. Chem. Soc. 2011 133 19432 19441 10.1021/ja206477z 22023652
[22] Siahrostami S Verdaguer-Casadevall A Karamad M Deiana D Malacrida P Wickman B Escudero-Escribano M Paoli E A Frydendal R Hansen T W Enabling direct H2O2 production through rational electrocatalyst design Nat. Mater. 2013 12 1137 1143 10.1038/nmat3795 24240242
[23] Li H Wen P Itanze D S Hood Z D Adhikari S Lu C Ma X Dun C C Jiang L Carroll D L Scalable neutral H2O2 electrosynthesis by platinum diphosphide nanocrystals by regulating oxygen reduction reaction pathways Nat. Commun. 2020 11 3928 10.1038/s41467-020-17584-9 32764644
[24] Gao J J Yang H B Huang X Hung S F Cai W Z Jia C M Miao S Chen H M Yang X F Huang Y Q Enabling direct H2O2 production in acidic media through rational design of transition metal single atom catalyst Chem 2020 6 658 674 10.1016/j.chempr.2019.12.008
[25] Tang C Chen L Li H J Li L Q Jiao Y Zheng Y Xu H L Davey K Qiao S Z Tailoring acidic oxygen reduction selectivity on single-atom catalysts via modification of first and second coordination spheres J. Am. Chem. Soc. 2021 143 7819 7827 10.1021/jacs.1c03135 33983725
[26] Xu J Y Wei X K Costa J D Lado J L Owens-Baird B Goncalves L P L Fernandes S P S Heggen M Petrovykh D Y Dunin-Borkowski R E Interface engineering in nanostructured nickel phosphide catalyst for efficient and stable water oxidation ACS Catal. 2017 7 5450 5455 10.1021/acscatal.7b01954
[27] Masa J Barwe S Andronescu C Sinev I Ruff A Jayaramulu K Elumeeva K Konkena B Cuenya B R Schuhmann W Low overpotential water splitting using cobalt-cobalt phosphide nanoparticles supported on nickel foam ACS Energy Lett. 2016 1 1192 1198 10.1021/acsenergylett.6b00532
[28] Zhou Z Kong Y Tan H Huang Q W Wang C Pei Z X Wang H Z Liu Y Y Wang Y H Li S Cation-vacancy-enriched nickel phosphide for efficient electrosynthesis of hydrogen peroxides Adv. Mater. 2022 34 2106541 10.1002/adma.202106541
[29] Song F Z Li W Yang J Q Han G Q Yan T Liu X Rao Y Liao P L Cao Z Sun Y J Interfacial sites between cobalt nitride and cobalt act as bifunctional catalysts for hydrogen electrochemistry ACS Energy Lett. 2019 4 1594 1601 10.1021/acsenergylett.9b00738
[30] Dai Z F Geng H B Wang J Luo Y B Li B Zong Y Yang J Guo Y Y Zheng Y Wang X Hexagonal-phase cobalt monophosphosulfide for highly efficient overall water splitting ACS Nano 2017 11 11031 11040 10.1021/acsnano.7b05050 29077385
[31] Sun Y Wu J Zhang Z Liao Q L Zhang S C Wang X Xie Y Ma K K Kang Z Zhang Y Phase reconfiguration of multivalent nickel sulfides in hydrogen evolution Energy Environ. Sci. 2022 15 633 644 10.1039/D1EE02985A
[32] Fang Z W Peng L L Qian Y M Zhang X Xie Y J Cha J J Yu G H Dual tuning of Ni-Co-A (A = P, Se, O) nanosheets by anion substitution and holey engineering for efficient hydrogen evolution J. Am. Chem. Soc. 2018 140 5241 5247 10.1021/jacs.8b01548 29608305
[33] Long X Li G X Wang Z L Zhu H Y Zhang T Xiao S Guo W Y Yang S H Metallic iron-nickel sulfide ultrathin nanosheets as a highly active electrocatalyst for hydrogen evolution reaction in acidic media J. Am. Chem. Soc. 2015 137 11900 11903 10.1021/jacs.5b07728 26338434
[34] Zhang X L Hu S J Zheng Y R Wu R Gao F Y Yang P P Niu Z Z Gu C Yu X X Zheng X S Polymorphic cobalt diselenide as extremely stable electrocatalyst in acidic media via a phase-mixing strategy Nat. Commun. 2019 10 5338 10.1038/s41467-019-12992-y 31767845
[35] Liu M Q Wang J A Klysubun W Wang G G Sattayaporn S Li F Cai Y W Zhang F C Yu J Yang Y Interfacial electronic structure engineering on molybdenum sulfide for robust dual-pH hydrogen evolution Nat. Commun. 2021 12 5260 10.1038/s41467-021-25647-8 34489450
[36] Wang M J Zhang N Feng Y G Hu Z W Shao Q Huang X Q Partially pyrolyzed binary metal-organic framework nanosheets for efficient electrochemical hydrogen peroxide synthesis Angew. Chem., Int. Ed. 2020 59 14373 14377 10.1002/anie.202006422
[37] Wang Y L Shi R Shang L Waterhouse G I N Zhao J Q Zhang Q H Gu L Zhang T R High-efficiency oxygen reduction to hydrogen peroxide catalyzed by nickel single-atom catalysts with tetradentate N2O2 coordination in a three-phase flow cell Angew. Chem., Int. Ed. 2020 59 13057 13062 10.1002/anie.202004841
[38] Liang J Wang Y Y Liu Q Luo Y L Li T S Zhao H T Lu S Y Zhang F Asiri A M Liu F G Electrocatalytic hydrogen peroxide production in acidic media enabled by NiS2 nanosheets J. Mater. Chem. A 2021 9 6117 6122 10.1039/D0TA12008A
[39] Ross R D Sheng H Y Parihar A Huang J Z Jin S Compositionally tuned trimetallic thiospinel catalysts for enhanced electrosynthesis of hydrogen peroxide and built-in hydroxyl radical generation ACS Catal. 2021 11 12643 12650 10.1021/acscatal.1c03349
[40] Xia F Li B M Liu Y Q Liu Y Z Gao S Y Lu K Kaelin J Wang R Y Marks T J Cheng Y W Carbon free and noble metal free Ni2Mo6S8 electrocatalyst for selective electrosynthesis of H2O2 Adv. Funct. Mater. 2021 31 2104716 10.1002/adfm.202104716
[41] Ji Y Liu Y D Zhang B W Xu Z F Qi X Xu X Ren L Du Y Zhong J X Dou S X Morphology engineering of atomic layer defect-rich CoSe2 nanosheets for highly selective electrosynthesis of hydrogen peroxide J. Mater. Chem. A 2021 9 21340 21346 10.1039/D1TA05731F
[42] Sheng H Y Hermes E D Yang X H Ying D W Janes A N Li W J Schmidt J R Jin S Electrocatalytic production of H2O2 by selective oxygen reduction using earth-abundant cobalt pyrite (CoS2) ACS Catal. 2019 9 8433 8442 10.1021/acscatal.9b02546
[43] Sheng H Y Janes A N Ross R D Kaiman D Huang J Z Song B Schmidt J R Jin S Stable and selective electrosynthesis of hydrogen peroxide and the electro-Fenton process on CoSe2 polymorph catalysts Energy Environ. Sci. 2020 13 4189 4203 10.1039/D0EE01925A
[44] Zhang X L Su X Z Zheng Y R Hu S J Shi L Gao F Y Yang P P Niu Z Z Wu Z Z Qin S Strongly coupled cobalt diselenide monolayers for selective electrocatalytic oxygen reduction to H2O2 under acidic conditions Angew. Chem., Int. Ed. 2021 60 26922 26931 10.1002/anie.202111075
[45] Zhang L C Liang J Yue L C Dong K Xu Z Q Li T S Liu Q Luo Y L Liu Y Gao S Y CoTe nanoparticle-embedded N-doped hollow carbon polyhedron: An efficient catalyst for H2O2 electrosynthesis in acidic media J. Mater. Chem. A 2021 9 21703 21707 10.1039/D1TA06313H
[46] Zhang L C Liang J Yue L C Xu Z Q Dong K Liu Q Luo Y L Li T S Cheng X H Cui G W N-doped carbon nanotubes supported CoSe2 nanoparticles: A highly efficient and stable catalyst for H2O2 electrosynthesis in acidic media Nano Res 2022 15 304 309 10.1007/s12274-021-3474-0 33936561
[47] Zhou H Q Yu F Liu Y Y Sun J Y Zhu Z He R Bao J M Goddard W A Chen S Ren Z F Outstanding hydrogen evolution reaction catalyzed by porous nickel diselenide electrocatalysts Energy Environ. Sci. 2017 10 1487 1492 10.1039/C7EE00802C
[48] Gill T M Zheng X L Comparing methods for quantifying electrochemically accumulated H2O2 Chem. Mater. 2020 32 6285 6294 10.1021/acs.chemmater.0c02010
[49] Liu C C Gong T Zhang J Zheng X R Mao J Liu H Li Y Hao Q Y Engineering Ni2P-NiSe2 heterostructure interface for highly efficient alkaline hydrogen evolution Appl. Catal., B: Environ. 2020 262 118245 10.1016/j.apcatb.2019.118245
[50] Wang T T Guo X S Zhang J Y Xiao W Xi P X Peng S L Gao D Q Electronic structure modulation of NiS2 by transition metal doping for accelerating the hydrogen evolution reaction J. Mater. Chem. A 2019 7 4971 4976 10.1039/C8TA11286J
[51] Liu Z Zhang C Z Liu H Feng L G Efficient synergism of NiSe2 nanoparticle/NiO nanosheet for energy-relevant water and urea electrocatalysis Appl. Catal., B: Environ. 2020 276 119165 10.1016/j.apcatb.2020.119165
[52] Boppella R Tan J Yun J Manorama S V Moon J Anion-mediated transition metal electrocatalysts for efficient water electrolysis: Recent advances and future perspectives Coord. Chem. Rev. 2021 427 213552 10.1016/j.ccr.2020.213552
[53] Back S Na J Ulissi Z W Efficient discovery of active, selective, and stable catalysts for electrochemical H2O2 synthesis through active motif screening ACS Catal. 2021 11 2483 2491 10.1021/acscatal.0c05494
[54] Choi C H Kwon H C Yook S Shin H Kim H Choi M Hydrogen peroxide synthesis via enhanced two-electron oxygen reduction pathway on carbon-coated Pt surface J. Phys. Chem. C 2014 118 30063 30070 10.1021/jp5113894
| 36465524 | PMC9707185 | NO-CC CODE | 2022-12-01 23:20:21 | no | Nano Res. 2022 Nov 29;:1-7 | utf-8 | Nano Res | 2,022 | 10.1007/s12274-022-5160-2 | oa_other |
==== Front
SN Compr Clin Med
SN Compr Clin Med
Sn Comprehensive Clinical Medicine
2523-8973
Springer International Publishing Cham
1353
10.1007/s42399-022-01353-4
Review
The Effects of Conservative Interventions for treating Diastasis Recti Abdominis in Postpartum Women: a Review with Meta-analysis
Weingerl Iva 1
Kozinc Žiga 12
http://orcid.org/0000-0003-0747-3735
Šarabon Nejc [email protected]
134
1 grid.412740.4 0000 0001 0688 0879 Faculty of Health Sciences, University of Primorska, Polje 42, 6310 Izola, Slovenia
2 grid.412740.4 0000 0001 0688 0879 University of Primorska, Andrej Marušič Institute, Muzejski Trg 2, 6000 Koper, Slovenia
3 Human Health Department, InnoRenew CoE, Livade 6, 6310 Izola, Slovenia
4 grid.513315.0 S2P, Science to Practice, Laboratory for Motor Control and Motor Behavior, LtdTehnološki Park 19, 1000 Ljubljana, Slovenia
29 11 2022
2023
5 1 1018 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Diastasis recti abdominis (DRA) is a common occurrence in postpartum women, and it is unclear what types of nonsurgical interventions are most effective in preventing and/or reducing it. The aim of this review with meta-analysis was to investigate which conservative treatment approaches are the most effective for the management of postpartum DRA. After a thorough search of the PubMed and Scopus databases, we reviewed 14 articles. The literature suggests that abdominal exercise programs are generally effective in treating DRA at various postpartum periods. There is preliminary but promising evidence of the efficacy of electrical stimulation in combination with exercise. In addition, abdominal kinesiotaping can be used in conjunction with other interventions. Limitations of previous research include (a) the use of different measurement methods (palpation, calipers, ultrasound) and sites, (b) the evaluation of treatment effects in different time periods, and (c) the use of a wider range of exercise combinations. Although abdominal exercise is a cornerstone of DRA treatment, the optimal exercise combination is currently unknown due to these limitations.
Keywords
Kinesiotaping
Electrical stimulation
Abdominal exercise
Exercise intervention
Inter-recti distance
issue-copyright-statement© Springer Nature Switzerland AG 2023
==== Body
pmcIntroduction
During pregnancy, the two muscle bellies of the rectus abdominis muscle, which are connected by the linea alba, elongate and curve round as the abdominal wall expands to make room for the growing fetus [1, 2]. Diastasis of the rectus abdominis (DRA) occurs as a result of hormonal changes leading to elastic changes in the connective tissue, mechanical stresses on the abdominal wall from the growing fetus, and displacement of the abdominal organs [3, 4]. In most pregnant women, the inter-recti distance (IRD) increases due to stretching and thinning of the linea alba [5, 6]. This is a natural process and happens to some degree in most pregnancies. However, in some cases, the abdominal muscle changes may become excessive and persist after delivery [7]. DRA is defined as an excessive separation (i.e., increased IRD) between the bellies of the recti abdominis muscles. DRA can occur anywhere along the linea alba, from the xiphoid process to the public bone, and is quantified by the inter-recti distance [8, 9].
Studies have shown that IRD increases around the 14th week of gestation and continues to increase until delivery. Clinically significant DRA usually occurs early in the last trimester of pregnancy and its peak incidence is immediately after and in the first weeks following childbirth. The prevalence of DRA ranges from 32 to 95% in late pregnancy and 30–68% in the postpartum period [7]. After childbirth, the changes in the abdominal musculature, especially in the linea alba and rectus abdominis sheath, usually return to normal and have no significant impact on health [10]. Recovery is most marked between 1 day and 8 weeks after delivery, followed by a plateau phase, and if no interventions are performed, IRD may remain elevated throughout life [11].
The abdominal wall plays an important role in posture and postural control, trunk and pelvic stability, respiration, trunk movement, and support of the abdominal viscera [1]. Persistent DRA can potentially lead to a range of symptoms and dysfunctions [10, 12]. These can manifest as altered trunk mechanics, impaired pelvic stability, and altered posture, which can make the lumbar spine and pelvis more susceptible to injury. Previous research has also found that DRA is related to low back pain [13], possibly due to associated decrements in trunk muscle strength, lumbopelvic instability, and pelvic floor weakness [14, 15]. A variety of factors have been shown to increase the risk of DRA, including multiparity, maternal age, body mass index before pregnancy, and at 6 months postpartum, weight gain during pregnancy, Beighton’s hypermobility score, baby weight at birth, abdominal circumference at gestational week 35, exercise training level before, during and after pregnancy, and type of delivery [16].
Women with DRA are usually referred to physical therapists for conservative treatment [17]. However, little is known about the optimal training methods for treating DRA. A 2014 review acknowledged that prenatal exercise can reduce the incidence of DRA by 35% [18]. Anecdotally, regular exercise before pregnancy and during the prenatal period appears to decrease the risk of developing DRA and reduce IRD [18, 19]. Abdominal exercises targeting the trunk flexors are commonly prescribed for women with postpartum DRA. Other regularly used nonsurgical interventions include postural and back training, external support, and aerobic exercise [20–22]. However, it remains unclear which types of nonsurgical interventions are most effective in preventing and/or reducing DRA. Therefore, the aim of this review was to determine which conservative treatment approaches may be most effective in reducing IRD in postpartum women.
Methods
Inclusion Criteria
Study inclusion criteria were structured according to PICOS tool [23]:Population (P): postpartum females
Intervention (I): exercise or other non-pharmacological interventions intended for treating DRA
Comparisons (C): control groups, receiving no intervention or placebo intervention; or comparisons of two different interventions
Outcomes (O): Inter-rectus distance
Study design (S): randomized controlled trials and non-randomized clinical trials
Search Strategy
PubMed and Scopus databases were searched in January 2022 without regard to the date of publication. We used the following combination of search key words: (diastasis recti OR stress urinary incontinence OR abdominal separation OR urinary incontinence) AND (exercise OR therapy OR intervention) AND (pregnancy OR pregnant OR postpartum) AND (months OR weeks OR follow-up). Additionally, reference lists of several review articles describing interventions for older adults were carefully scrutinized. Finally, we performed a backward and forward citation search; specifically, we reviewed reference lists of all articles that were already retrieved through the database, and we also reviewed all papers that have cited them, using the Google Scholar platform. Potentially relevant articles were screened in full text, followed by additional screening for their eligibility by the additional reviewers.
Data Extraction
The extracted data included (a) baseline and post-intervention means and standard deviations for IRD; (b) baseline demographics of participants (gender, age, body height, body mass, body mass index); (c) intervention characteristics, such as duration of the intervention, number of sessions per week, volume (number of exercises, sets, and repetitions), breaks between exercises and sets, supervision, and progression of exercise difficulty); (d) IRD measurement characteristics (method, inclusion of contraction, measurement location, number of trials). Regarding the latter, previous studies have differed in regard to IRD measurement method (ultrasound, caliper, manual measurements), location (4.5 cm above and below the umbilicus being the most common) and inclusion of contraction (either measuring IRD during abdominal contraction or in relaxed condition). Data were carefully entered into Microsoft Excel 2016 (Microsoft, Redmond, WA, USA). If the data were presented in a graphical rather than tabular form, we used Adobe Illustrator Software (version CS5, Adobe Inc., San Jose, CA, USA) to accurately determine the means and standard deviations. In case of missing data, the corresponding author of the respective articles was contacted by e-mail. If no response was received after 21 days, the author was contacted again. If the author did not reply to the second inquiry, the data was considered irretrievable.
Data Analysis and Synthesis
The main data analyses were carried out in Review Manager (Version 5.3, Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014, London, UK). Before the results were entered into the meta-analytical model, the pre-post differences and pooled standard deviations were calculated according to the following formula SD = √[(SDpre2 + SDpost2) − (2 × r × SDpre × SDpost). The correction value (r), which represents the pre-test–post-test correlation of outcome measures, was conservatively set at 0.75. It should be noted that a change in the correction value in the range between 0.5 and 0.9 had little effect on the pooled SD and would not change the outcomes of the meta-analyses. For the meta-analyses, the inverse variance method for continuous outcomes with a random-effects model was used. The effect sizes were expressed as standardized mean difference (SMD). The respective 95% confidence intervals were also calculated and reported. Even when only one study was considered, we calculated the SMD to enable the comparison of effect sizes across studies. We conducted separate analyses depending on the location of measurement and contraction (i.e., measurement with and without muscle contraction), but not the measurement method. Statistical heterogeneity among studies was determined by calculating the I2 statistics. According to Cochrane guidelines, the I2 statistics of 0% to 40% might not be important, 30% to 60% may represent moderate heterogeneity, 50% to 90% may represent substantial heterogeneity, and 75% to 100% indicates considerable heterogeneity [24]. The threshold for statistical significance was set at p ≤ 0.05 for the main effect size and the subgroup difference tests.
Results
In total, 975 records from the PubMed database and 423 from the Scopus database were retrieved (total n = 1398). After the removal of the duplicates, 1112 records remained for screening. Based on the examination of the titles, 215 papers were left for the examination of the abstract, after which 22 papers were chosen for full-text evaluation. After reading the full texts, 13 papers were included in the systematic review. One additional paper was identified in the reference list of previous papers, for a total of 14 papers reviewed. The procedure is shown in Fig. 1 in detail. The overview of the included studies is provided in Table 1.Fig. 1 PRISMA flowchart of article retrieval
Table 1 Overview of included studies
Study Participants Measurement methods Intervention
Awad et al., 2021 n = 50; 3–6 months postpartum; mean age: 27.1 ± 3.5 years IRD measured 4.5 cm below and above the umbilicus, both during rest and contraction, measured with an ultrasound Plank exercise, performed for 8 weeks, 3 times per week, combined with dietary advice. Difficulty progressed with unilateral variation and using the Swiss ball
Bobowik & Dąbek, 2018 n = 40; 0–3 days postpartum; mean age: 32.3 ± 5.9 years IRD measured by palpation, undefined location and presence/absence of contraction Physiotherapeutic program, performed daily for 6 weeks. Consisting of 20-min posture position accelerating the involution of the uterus and exercises aimed at consolidating the correct correlation between the long exhalation and the movement. Advice for prevention of DRA and kinesiotaping were also included
El-Mekawy et al., 2013 n = 30; 2 days postpartum; aged 25–35 years IRD measured with a dial up caliper; location and contraction undefined Abdominal exercise program (static abdominal contraction, posterior pelvic tilt, reverse sit-up exercise, trunk twist, and reverse trunk twist exercise), performed 3 times per week for 6 weeks, and a daily abdominal exercise at home
Gluppe et al., 2018 n = 175, 6 weeks postpartum; mean age: 29.8 ± 4.1 years The prevalence of women diagnosed with DRA, determined with palpation Abdominal exercise program lasting for 4 months, conspiring of 1 session of guided exercise per week and daily home exercises (hollowing exercise, half-plank, side-plank, oblique sit-up, and straight sit-up)
Kamel et al., 2017 n = 60; 2 months postpartum; mean age: 29.3 ± 2.9 years IRD measured with ultrasound on an undefined location, in relaxed position Neuromuscular electrical stimulation (frequency 80 pulses/min, with pulse width 0.1–0.5 ms, and as on:off ratio of 5 s:10 s for the total stimulation time of 30 min) combined with exercise (sit-ups, reverse sit-ups, reverse trunk twists, and U-seat exercises), performed 3 times a week for 8 weeks
Keshwani et al., 2021 n = 32, 22 ± 2 days postpartum; mean age 31 ± 2 years IRD measured and averaged at 5 locations with an ultrasound Exercise therapy lasting for 12 weeks, conspiring of 1 session of guided exercise per week and daily home exercises (isolated activation of the transversus abdominis muscles, bent knee leg lifts in crook lying, eccentric trunk flexion and side planks with progressive variations). Depending on group allocation, therapy was done alone or in combination with abdominal binding
Kim et al., 2022 n = 52, 6–12 months postpartum; mean age: 31.68 ± 3.92 IRD measured 2.5 cm above umbilicus with an ultrasound, in a relaxed position Core stabilization exercise program, performed 2 times per week for 6 weeks. Hollowing technique used for stabilization. One group performed the exercises on site; for the other group, the program was conducted via ZOOM. It was also recommended that the exercise programs are implemented as often as possible
Laframboise et al., 2021 n = 8, 6–24 months postpartum; mean age: 35.6 ± 3.2 years IRD measured 4.5 cm below and above the umbilicus, both during rest and contraction, using a nylon digital caliper Online program including a variety of exercises, mostly isometric, targeting core function and strength, performed for 12 weeks, 3 times per week. Exercise examples: glute bridges, lying supine with feet planted and knees bent at a 90-degree angle, lifting glutes while shoulders stay planted; planks, isometric holding at the “top” of a pushup position; leg raises/marches while in a glute bridge
Ptaszkowska et al., 2021 n = 24; 6–12 months postpartum; aged 18–38 years IRD measured at the umbilicus and 4.5 cm above and below it, using a digital caliper Application of kinesiotapes for 48 h
Ramírez-Jiménez et al., 2021 n = 12, no control group, 5.7 ± 3.0 months postpartum, mean age 39.0 ± 3.0 years IRD measured 4.5 cm above the umbilicus during contraction, using a digital 150 mm hardened stainless steel caliper Hypopressive exercise program (hypopressive breathing), performed 3 times a week for 4 weeks, in five postures. Hypopressive maneuvers included: slow, deep diaphragmatic breaths, exhaling the air to the reserve volume, Expiratory apnea for 10 to 25 s with the focus on expanding the chest by moving the ribs in an upward and outward
Saleem et al., 2021 n = 40; 2–4 months postpartum; mean age: 29.8 ± 4.1 years IRD measured at 4.5 above and below the umbilicus during contraction, using both palpation and digital caliper Two exercise programs, performed 3 times per week for 6 weeks. The first program included abdominal crunches, static glutei, kegels, and isometric back strengthening exercises while in the other, the crunches were replaced by double straight leg raise
Thabet & Alshehri, 2019 n = 15; postpartum, no control group, aged between 23 and 33 years IRD measured 4.5 cm below and above the umbilicus during contraction, using digital nylon calipers Deep core stability-strengthening program (abdominal bracing, diaphragmatic breathing, pelvic floor contraction, plank, and isometric abdominal contraction), performed 3 times a week for 8 weeks
Vaishnavi et al., 2018 n = 15; aged between 25 and 30 years, no control group IRD measured 4.5 cm below and above the umbilicus during contraction, using a caliper Abdominal exercise program (pelvic tilt, reverse sit-up, reverse trunk twist, U-seat, transverse exercise), performed for 6 weeks every day
Wei et al., 2021 32 women who had DRA for 6 months, mean age: 33.47 ± 0.3 years IRD measured 2.0 cm below and above the umbilicus in a relaxed position Electrical stimulation of abdominal muscles with strengthening exercises of internal and external oblique muscles, once a day for 6 weeks
IRD, inter-recti distance; DRA, diastasis recti abdominis
The Effect of Exercise on IRD At Rest
Only two studies with exercise-based interventions and control groups (58 participants in total) assessed IRD above the umbilicus (4.5 cm in both) at rest [25, 26]. The pooled SMD was 2.39, but was not statistically significant (CI: − 0.67, 5.42; p = 0.13; I2 = 92%). However, one of the studies [26] was done on 8 participants only, which included women at 6–24 months postpartum. The other study (n = 25 per group; 3–6 months postpartum) showed a large effect of plank exercise (SMD = 3.91, CI = 2.93–4.88, p < 0.011). The same two studies also assessed IRD 4.5 cm below the umbilicus at rest. The pooled effect was again not statistically significant (SMD = 1.27; CI: − 1.34 to 3.88; p = 0.340; I2 = 91%). Again, the first study [25] showed a significant reduction in IRD in the plank exercise group (SMD = 2.54, CI: 1.78–3.30, p < 0.001).
Effect of Exercise on IRD During Contraction
Three studies (98 participants in total) assessed IRD above the umbilicus during voluntary contraction [25–27]. The meta-analysis indicated large reductions of IRD in the experimental groups (SMD = 2.85; CI: 0.74–4.97; p = 0.008; I2 = 90%). The removal of the small study with 8 participants only [26] resulted in even larger and more consistent effect (SMD = 3.89; CI = 1.92–5.86; p < 0.001; I2 = 85%). Two studies [25, 26] also assessed IRD below the umbilicus during rectus abdominis contraction, and showed a completely opposite effects (SMD = 2.31 and − 1.16, respectively), making the pooled effect small and non-significant (SMD = 0.65; CI: -2.74 to 4.14). The former study [25] with a larger sample size showed a very large beneficial effect of exercise (SMD = 2.31; CI: 1.58–3.03; p < 0.001).
One study [28] without a control group investigated the effect of hypopressive exercise on IRD in 12 women at 2 or more months postpartum. The authors reported statistically significant reductions in IRD (p < 0.05). Assuming a null effect in a potential control group, the effect would translate into large SMD (3.22; CI: 1.94–4.50). An additional study (including 15 postpartum women) with exercise-based intervention (abdominal exercise) and no control group was conducted [29]. Assuming a null effect in a potential control group, the effect would translate into large SMD for the measurement above (2.04; CI: 1.02–3.05) and below the umbilicus (2.09; CI: 1.06–3.12) during contraction.
One study did not measure the exact IRD, but reported only the prevalence of DRA [30]. The participants (n = 157; women 6 weeks postpartum) were randomized in experimental and control group. The participants in the experimental group received supervised pelvic floor exercises and were encouraged to also perform the program at home, while the control group only received 1 session with exercise instructions. The prevalence of DRA was reduced similarly in both groups (from 54–55 to 30–41%), indicating no differences between the groups.
Effect of Exercise Interventions with Undefined/Unclear IRD Measurement
Two studies [31, 32] assessed the effect of exercise interventions (56 participants in total, women 0–3 weeks postpartum), but did not specify the details regarding their IRD assessments. The pooled effect was not statistically significant (SMD = 0.94; CI: − 1.64, 3.94; p = 0.5), as the studies showed the opposite effect. In the first study [31], exercise induced reductions in IRD statistically significantly larger than in the passive control group (SMD = 2.32; CI: 1.50–3.15; p < 0.001), however, there was no difference between exercise and control groups in the second study [32] (SMD = − 0.47; − 1.47 to 0.53; p = 0.360). In addition, abdominal crunch exercise was very effective to improve IRD in postpartum women (SMD = 4.47; CI: 3.26–5.57) in another study [33], who used caliper measurements, but did not specify if the IRD was assessed at rest or during contraction.
Other Interventions
One study [34] investigated the immediate effect of kinesiotaping on IRD in 12 women 6–12 weeks postpartum. The IRD reduction was statistically significant for the measurement above the umbilicus (SMD = 1.12; CI = 0.25–2.00; p = 0.01), but not for the measurement below the umbilicus (SMD = 0.62; CI = − 0.20–1.45; p = 0.140).
One study compared neuromuscular electrical stimulation and abdominal exercise, to abdominal exercise alone, with 30 participants (women at least 2 months postpartum) per group [35]. The parameters for electrical stimulation were: a frequency 80 pulses/min, with pulse width 0.1–0.5 ms, and as on:off ratio of 5 s:10 s for the total stimulation time of 30 min. The analysis showed that the combination of neuromuscular electrical stimulation and abdominal exercise decreased IRD statistically significantly more than exercise alone (SMD = 2.88; CI: 2.14–3.61; p < 0.001). Another study assessed the effect of electrical stimulation followed by exercise in comparison to no intervention [36]. The IRD decreased in the intervention group only (from 3.1 ± 0.7 cm to 2.4 ± 0.7 above the umbilicus, and 2.0 ± 0.6 cm to 1.7 ± 0.6 cm below the umbilicus) [36].
Two studies (46 participants in total; 0–3 weeks postpartum) compared wearing abdominal belt or abdominal binding with exercise-based interventions [32, 37]. The first study strongly favored the exercise interventions (SMD = 2.51; CI: 1.52–3.50; p < 0.001), while there were no statistically significant differences between groups in the second study (SMD = − 0.41; CI: − 1.49 to 0.51; p = 0.340), although the direction of the effect favored the use of abdominal binding. The pooled effect was not statistically significant (SMD = 1.01; CI: − 1.93 to 3.95; p = 0.501; I2 = 94%). In the latter study, the effect of combined exercise and abdominal binding was not superior to abdominal binding alone (SMD = 0.12; p = 0.822).
One study compared the effectiveness of online (e-platform based) and offline core stabilization exercise protocols for reducing IRD. During the 6 weeks of intervention, the two groups performed the program twice for 40 min. For the online group, the exercise sessions were conducted through Zoom. In both groups, it was recommended that the exercise programs are implemented as often as possible. The IRD decreased from 1.99 ± 0.26 cm to 1.37 ± 0.40 cm (p < 0.001) in the online group and from 1.92 ± 0.30 cm to 1.18 ± 0.30 cm in the offline group (p < 0.001), with no differences observed between the groups (p > 0.05) [38].
Discussion
The aim of this review was to determine the scope of existing interventions for DRA in postpartum women and to compare their effectiveness in reducing IRD. The literature indicates that DRA is very common in pregnant and postpartum women. Yet, there are no clear and effective exercise-based approaches that can be recommended to women with DRA. With this in mind, we decided to conduct a review of previously studied interventions. We found only a handful of intervention studies, with different approaches and different measurement methods, making it difficult to compare results. Data from the studies included in this paper suggest that abdominal exercise plays a central role in reducing IRD and associated symptoms. Preliminary evidence suggests the potential of electrical stimulation as a complementary tool in the rehabilitation of IRD. In addition, online-based exercise programs appear to be as effective as on-site programs, which is important in light of potential limitations of group exercise programs, such as during the COVID -19 pandemic.
The major drawback of the current literature is the discrepancy in IRD measurement methods, as we found studies in which DRA was measured at rest or during contraction, as well as studies with undefined protocols. Some studies lacked control groups [28] and one study reported only the prevalence of DRA [30]. In addition to exercise-based interventions, one study examined the efficacy of kinesiotaping [34] and two studies examined the effect of neuromuscular electrical stimulation in combination with exercise [35, 36]. Two studies compared abdominal binding with exercise-based intervention [32, 37]. Because these studies had very different designs and interventions, quantitative statistical analysis within study groups (IRD at rest/IRD during contraction/undefined IRD measurement/other interventions) was limited.
In the group of studies that examined IRD at rest, no statistically significant pooled effects were observed for IRD either above or below the umbilicus. However, a large reduction in IRD was achieved in one study that used progressive prone plank exercise [25]. The group of interventions in which IRD was measured during voluntary contraction showed a large reduction in IRD above the umbilicus but not below the umbilicus. Because of these discrepancies, the pooled effect was small and not significant. Intervention studies that used other interventions, methods such as kinesiotaping, neuromuscular electrical stimulation, and/or abdominal exercise and wearing an abdominal belt with/without exercise found different effects. Kinesiotaping showed improvement above the umbilicus but not bellow [34], neuromuscular electrical stimulation was effective in combination with abdominal exercise [35] and the abdominal binding showed inconsistent effects when combined with exercise [32, 37].
Regardless of whether IRD was measured at rest or during contraction, the studies showed a decrease in IRD in several intervention groups. Therapeutic exercises are designed to activate both fast twitch and slow twitch fibers of the skeletal muscles, and they affect the metabolic demand needed to produce a given muscle force, which leads to increase in muscular power and endurance [39]. Recently, it has been shown that individuals with increased IRD have lower muscle stiffness in the rectus abdominis and transversus abdominis muscles, which is another indication of the importance of strengthening the abdominal muscles [40]. Some authors suggested that an increased mind-muscle connection (i.e., meaningful muscle activation) may contribute to the observed improvement [26]. Studies indicated the importance of activating the transversus abdominis muscle, as it brings together the bellies of the rectus abdominus muscle, improves the integrity of the linea alba, and increases fascial tension, thereby reducing DRA [41]. Interestingly, several studies showed that the changes in IRD were not uniform along the linea alba. The improvements limited to a specific region along the linea alba demonstrate the potential for rehabilitation, but also require further research to optimize exercise interventions.
Ideally, carefully designed rehabilitation exercises have the potential to positively impact DRA and related outcomes in the postpartum period, but this area appears to be under-researched. The studies we included in this meta-analysis also had a number of shortcomings that make conclusions difficult. This may be due, in part, to the different methods used to measure IRD. The studies we examined used different points along the linea alba for IRD measurement. Caution should be used when interpreting results measured at different points, because IRD magnitudes can vary considerably. A recent systematic review showed that the average IRD distance of the nullipara was 8.77 mm, the distance in the epigastric region was 7.22 mm, and that in the infraumbilical region was 4.09 mm [42]. Although IRD measurements are generally reliable and valid [43], the methods used to assess IRD were different in the various studies and were performed either at rest or during contraction (or nonspecifically), which may have affected the validity of our comparisons. Some, but not all, of these variables could be accounted for in the meta-analytic models. Some research also suggests that measuring the integrity (i.e., depth) of the linea alba in conjunction with IRD may be a more comprehensive approach to determining the severity of pregnancy-related DRA [44]. In recent years, ultrasonography-based measurements are gaining recognition as a reliable tool for measuring IRD [45]. Small and very different sample sizes in the studies is one of the major limitations despite good study designs. In addition, the comparison between the studies is confounded by the natural recovery, which is particularly evident in early postpartum period [7].
Conclusions
Given the prevalence of DRA, more research is needed on potential effective interventions. From existing publications, abdominal exercises are at the forefront of effective protocols to reduce IRD. Future interventions should consider the use of more uniform IRD measurement devices, preferably using ultrasound or calipers, which are more reliable than manual methods. Because of the impact that DRA can have on postpartum women’s quality of life, strategies to assess the severity of DRA and the integrity of the linea alba should be incorporated into postpartum health care. This would allow timely determination of whether DRA is present so that women can receive appropriate referral/treatment.
Author Contribution
IW and NS conceptualized the paper. IW and ZK performed the literature review. ZK performed the data analysis. IW and NS wrote the first manuscript draft. All authors worked on finalizing the paper.
Declarations
Competing Interests
The authors declare no competing interests.
This article is part of the Topical Collection on Medicine
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Gilleard WL Brown JMM Structure and function of the abdominal muscles in primigravid subjects during pregnancy and the immediate postbirth period Phys Ther 1996 76 750 762 10.1093/ptj/76.7.750 8677279
2. Axer H Keyserlingk DGV Prescher A Collagen fibers in linea alba and rectus sheaths: I. General scheme and morphological aspects J Surg Res 2001 96 127 34 10.1006/jsre.2000.6070 11181006
3. Blaschak MJ Boissonnault JS Incidence of diastasis recti abdominis during the childbearing year Phys Ther 1988 68 1082 1087 10.1093/ptj/68.7.1082 2968609
4. Rath AM Attali P Dumas JL Goldlust D Zhang J Chevrel JP The abdominal linea alba: an anatomo-radiologic and biomechanical study Surg Radiol Anat 1996 18 281 288 10.1007/BF01627606 8983107
5. Beer GM Schuster A Seifert B Manestar M Mihic-Probst D Weber SA The normal width of the linea alba in nulliparous women Clin Anat 2009 22 706 711 10.1002/ca.20836 19637295
6. Barbosa S De Sá RAM Coca Velarde LG Diastasis of rectus abdominis in the immediate puerperium: correlation between imaging diagnosis and clinical examination Arch Gynecol Obstet 2013 288 299 303 10.1007/s00404-013-2725-z 23435798
7. Rett M Braga M Bernardes N Andrade S Prevalence of diastasis of the rectus abdominis muscles immediately postpartum : comparison between primiparae and multiparae Rev Bras Fisioter 2009 13 275 280 10.1590/S1413-35552009005000037
8. da Mota PGF Pascoal AGBA Carita AIAD Bø K Prevalence and risk factors of diastasis recti abdominis from late pregnancy to 6 months postpartum, and relationship with lumbo-pelvic pain Man Ther 2015 20 200 205 10.1016/j.math.2014.09.002 25282439
9. Sperstad JB Tennfjord MK Hilde G Ellström-Engh M Bø K Diastasis recti abdominis during pregnancy and 12 months after childbirth: prevalence, risk factors and report of lumbopelvic pain Br J Sports Med 2016 50 1092 1096 10.1136/bjsports-2016-096065 27324871
10. Hills NF Graham RB McLean L Comparison of trunk muscle function between women with and without diastasis recti abdominis at 1 year postpartum Phys Ther 2018 98 891 901 10.1093/ptj/pzy083 30011041
11. Liaw L-J Hsu M-J Liao C-F Liu M-F Hsu A-T The relationships between inter-recti distance measured by ultrasound imaging and abdominal muscle function in postpartum women: a 6-month follow-up study J Orthop Sports Phys Ther 2011 41 435 443 10.2519/jospt.2011.3507 21289454
12. Zappile-Lucis M Quality of life measurements and physical therapy management of a female diagnosed with diastasis recti abdominis J Womens Health Phys Therap 2009 33 22 10.1097/01274882-200933010-00023
13. Puri J Sharma S Samuel AJ Chahal A Investigate correlation between diastasis of rectus abdominis muscle and low back pain in obese women J Lifestyle Med 2021 11 38 42 10.15280/jlm.2021.11.1.38 33763341
14. Yuan S Wang H Zhou J Prevalence and risk factors of low back and pelvic pain in women with rectus abdominis diastasis: a multicenter retrospective cohort study Korean J Pain 2022 35 86 96 10.3344/kjp.2022.35.1.86 34966015
15. Gluppe S Ellström Engh M Kari B Women with diastasis recti abdominis might have weaker abdominal muscles and more abdominal pain, but no higher prevalence of pelvic floor disorders, low back and pelvic girdle pain than women without diastasis recti abdominis Physiother (United Kingdom) 2021 111 57 65
16. Cavalli M Aiolfi A Bruni PG Manfredini L Lombardo F Bonfanti MT Prevalence and risk factors for diastasis recti abdominis: a review and proposal of a new anatomical variation Hernia 2021 25 883 890 10.1007/s10029-021-02468-8 34363190
17. Keeler J Albrecht M Eberhardt L Horn L Donnelly C Lowe D Diastasis recti abdominis: a survey of women’s health specialists for current physical therapy clinical practice for postpartum women J Women’s Heal 2012 36 131 142
18. Benjamin DR, van de Water ATM, Peiris CL Effects of exercise on diastasis of the rectus abdominis muscle in the antenatal and postnatal periods: a systematic review Physiother (United Kingdom) 2014 100 1 8
19. Chiarello CM Falzone LA McCaslin KE Patel MN Ulery KR The effects of an exercise program on diastasis recti abdominis in pregnant women J Womens Health Phys Therap 2005 29 11 16 10.1097/01274882-200529010-00003
20. Sheppard S Part I: management of postpartum gross divarication recti J Assoc Chart Physiother Womens Heal 1996 79 22 24
21. Thornton SL Thornton SJ Management of gross divarication of the recti abdominis in pregnancy and labour Physiother (United Kingdom) 1993 79 457 458
22. Pascoal AG Dionisio S Cordeiro F Mota P Inter-rectus distance in postpartum women can be reduced by isometric contraction of the abdominal muscles: a preliminary case-control study Physiother (United Kingdom) 2014 100 344 348
23. Methley AM, Campbell S, Chew-Graham C, McNally R, Cheraghi-Sohi S. PICO, PICOS and SPIDER: a comparison study of specificity and sensitivity in three search tools for qualitative systematic reviews. BMC Health Serv Res. 2014;14:579.
24. Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al. Cochrane handbook for systematic reviews of interventions. Cochrane Handb Syst Rev Interv. John Wiley & Sons, 2019.
25. Awad E Mobark A Zidan AA Hamada HA Shousha T Effect of progressive prone plank exercise program on diastasis of rectus abdominis muscle in postpartum women: a randomized controlled trial J Hum Sport Exerc 2021 16 395 403
26. Laframboise FC Schlaff RA Baruth M Postpartum exercise intervention targeting diastasis recti abdominis Int J Exerc Sci 2021 14 400 409 34055160
27. Thabet AA Alshehri MA Efficacy of deep core stability exercise program in postpartum women with diastasis recti abdominis: a randomised controlled trial J Musculoskelet Neuronal Interact 2019 19 62 68 30839304
28. Ramírez-Jiménez M, Alburquerque-Sendín F, Garrido-Castro JL, Rodrigues-de-Souza D. Effects of hypopressive exercises on postpartum abdominal diastasis, trunk circumference, and mechanical properties of abdominopelvic tissues: a case series. Physiother Theory Pract. 2021;Ahead of print.
29. Vaishnavi G Mohan Kumar G Jayson CJ Kirupa K Tharani G Kamatchi K Effectiveness of exercise in treating rectus abdominis diastasis Biomed 2018 38 583 587
30. Gluppe SL Hilde G Tennfjord MK Engh ME Bø K Effect of a postpartum training program on the prevalence of diastasis recti abdominis in postpartum primiparous women: a randomized controlled trial Phys Ther 2018 98 260 268 10.1093/ptj/pzy008 29351646
31. Bobowik PŻ Dąbek A Physiotherapy in women with diastasis of the rectus abdominis muscles Adv Rehabil 2018 32 11 17 10.5114/areh.2018.80964
32. Keshwani N Mathur S McLean L The impact of exercise therapy and abdominal binding in the management of diastasis recti abdominis in the early postpartum period: a pilot randomized controlled trial Physiother Theory Pract 2021 37 1018 1033 10.1080/09593985.2019.1675207 31642725
33. Saleem Z Khan AA Farooqui SI Yasmeen R Rizvi J Effect of exercise on inter-recti distance and associated low back pain among postpartum females; a randomized controlled trial J Fam Reprod Heal 2021 15 202 209
34. Ptaszkowska L, Gorecka J, Paprocka‐borowicz M, Walewicz K, Jarzab S, Majewska‐pulsakowska M, et al. Immediate effects of kinesio taping on rectus abdominis diastasis in postpartum women—preliminary report. J Clin Med. 2021;10:5043.
35. Kamel DM Yousif AM Neuromuscular electrical stimulation and strength recovery of postnatal diastasis recti abdominis muscles Ann Rehabil Med 2017 41 465 474 10.5535/arm.2017.41.3.465 28758085
36. Wei R Yu F Ju H Jiang Q Effect of electrical stimulation followed by exercises in postnatal diastasis recti abdominis via MMP2 gene expression Cell Mol Biol 2021 67 82 88 10.14715/cmb/2021.67.5.12
37. El-Mekawy H Eldeeb A El-Lythy M El-Begawy A Effect of abdominal exercises versus abdominal supporting belt on postpartum abdominal efficiency and rectus separation Int Sci Index [Internet] 2013 7 44 48
38. Kim S Yi D Yim J The effect of core exercise using online videoconferencing platform and offline-based intervention in postpartum woman with diastasis recti abdominis Int J Environ Res Public Health 2022 19 7031 10.3390/ijerph19127031 35742279
39. Reiman MP Lorenz DS Integration of strength and conditioning principles into a rehabilitation program Int J Sports Phys Ther [Internet] 2011 6 241 253 21904701
40. He K, Zhou X, Zhu Y, Wang B, Fu X, Yao Q, et al. Muscle elasticity is different in individuals with diastasis recti abdominis than healthy volunteers. Insights Imaging. 2021;12:1–11.
41. Mota P Pascoal AG Carita AI Bø K The immediate effects on inter-rectus distance of abdominal crunch and drawing-in exercises during pregnancy and the postpartum period J Orthop Sports Phys Ther. United States 2015 45 781 8 10.2519/jospt.2015.5459
42. Wang Y, Wang H. Systematic review and meta-analysis of the inter-recti distance on ultrasound measurement in nulliparas. J Plast Surg Hand Surg. 2022.
43. Benjamin DR, Frawley HC, Shields N, Georgiou C, Taylor NF. Establishing measurement properties in the assessment of inter-recti distance of the abdominal muscles in a postnatal women. Musculoskelet Sci Pract. 2020;49:102202.
44. Dufour S, Hurd A, Lis E, Speckley J, Stotesbury A, Wright C. Pregnancy-related diastasis rectus abdominis: impact of a multi-component group-based intervention. Obstet Gynecol Int J. 2019;10:87–93.
45. Qu E Wu J Zhang M Wu L Zhang T Xu J The ultrasound diagnostic criteria for diastasis recti and its correlation with pelvic floor dysfunction in early postpartum women Quant Imaging Med Surg 2021 11 706 713 10.21037/qims-20-596 33532270
| 36466120 | PMC9707186 | NO-CC CODE | 2022-12-01 23:20:21 | no | SN Compr Clin Med. 2023 Nov 29; 5(1):10 | utf-8 | SN Compr Clin Med | 2,022 | 10.1007/s42399-022-01353-4 | oa_other |
==== Front
Inflammopharmacology
Inflammopharmacology
Inflammopharmacology
0925-4692
1568-5608
Springer International Publishing Cham
1105
10.1007/s10787-022-01105-9
Original Article
Efficacy and safety of granulocyte–macrophage colony-stimulating factor (GM-CSF) antibodies in COVID-19 patients: a meta-analysis
Xi An-ran 1
Luo Yi-jun 1
Guan Jin-Tao 2
Wang Wei-Jie 3
http://orcid.org/0000-0003-0033-525X
Xu Zheng-Hao [email protected]
14
1 grid.268505.c 0000 0000 8744 8924 Laboratory of Rheumatology and Institute of TCM Clinical Basic Medicine, College of Basic Medical Science, Zhejiang Chinese Medical University, Binwen Road 548, Hangzhou, 310053 Zhejiang China
2 grid.469601.c First People’s Hospital of Taizhou, Taizhou, 318020 Zhejiang China
3 grid.268505.c 0000 0000 8744 8924 The Second Affiliated Hospital of Zhejiang, Chinese Medical University, Hangzhou, 310053 Zhejiang China
4 grid.268505.c 0000 0000 8744 8924 Key Laboratory of Neuropharmacology and Translational Medicine of Zhejiang Province, Zhejiang Chinese Medical University, Hangzhou, China
29 11 2022
111
20 9 2022
17 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Objective
This study aims to determine the efficacy and safety of granulocyte–macrophage colony-stimulating factor (GM-CSF) antibodies in COVID-19 patients.
Methods
We searched Cochrane Library, PubMed, Embase, and ClinicalTrials.gov databases until July 27, 2022. Both randomized control trials (RCTs) and cohort studies were included and analyzed separately. The outcomes included mortality, incidence of invasive mechanical ventilation (IMV), ventilation improvement rate (need oxygen therapy to without oxygen therapy), secondary infection, and adverse events (AEs). The odds ratio (OR) with a 95% confidence interval (CI) was calculated by a random-effects meta-analysis model.
Results
Five RCTs and 2 cohort studies with 1726 COVID-19 patients were recruited (n = 866 in the GM-CSF antibody group and n = 891 in the control group). GM-CSF antibodies treatment reduced the incidence of IMV, which was supported by two cohort studies (OR 0.16; 95% CI 0.03, 0.74) and three RCTs (OR 0.62; 95% CI 0.41, 0.94). GM-CSF antibodies resulted in slight but not significant reductions in mortality (based on two cohort studies and five RCTs) and ventilation improvement (based on one cohort study and two RCTs). The sensitive analysis further showed the results of mortality and ventilation improvement rate became statistically significant when one included study was removed. Besides, GM-CSF antibodies did not increase the risks of the second infection (based on one cohort study and five RCTs) and AEs (based on five RCTs).
Conclusion
GM-CSF antibody treatments may be an efficacious and well-tolerant way for the treatment of COVID-19. Further clinical evidence is still warranted.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10787-022-01105-9.
Keywords
Meta-analysis
Granulocyte–macrophage colony-stimulating factor
Antibody
COVID-19
Natural Science Foundation of ZhejiangNo. LY22H280007 Xu Zheng-Hao http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 82174005 Xu Zheng-Hao Foundation of Zhejiang Chinese Medical UniversityNo. Q2019Y02 Xu Zheng-Hao
==== Body
pmcIntroduction
The COVID-19 pandemic is still threatening public health with its serious outcomes and strong infectivity (Chen et al. 2021). By 30th August 2022, confirmed COVID-19 cases surpassed 601 million, and more than 6.48 million deaths have been reported worldwide. The most common symptoms of COVID-19 include fever, dry cough, shortness of breath, and severe complications include pneumonia and even acute respiratory distress syndrome (ARDS) (Wiersinga et al. 2020). Vaccinations still could not give immediate or 100% protection and may be resistant by a mutant virus. Thus, it is still urgent to develop and evaluate new therapeutic strategies for COVID-19.
Granulocyte–macrophage colony-stimulating factor (GM-CSF) is an important myelopoietic growth factor and pro-inflammatory cytokine which had been identified to play a key role in lung inflammatory and immunological diseases (Mehta et al. 2020). It had been reported that GM-CSF expression is increased in the leukocytes of patients with COVID-19 (Zhou et al. 2020). Because of its immune hyperactivity, GM-CSF may play a significant role in the pathogenesis of COVID-19. GM-CSF antibodies inhibiting the GM-CSF signaling axis by targeting GM-CSF, such as Gimsilumab, lenzilumab, namilumab, and otilimab, or GM-CSF receptors (GM-CSFR), such as mavrilimumab. Several clinical trials of GM-CSF antibodies for the treatment of COVID-19 have been conducted. Antagonizing the GM-CSF receptor or directly binding circulating GM-CSF may be beneficial for treating hyperinflammation-related lung injury (Bonaventura et al. 2020; Lang et al. 2020; Sterner et al. 2019). However, the efficacy and safety of GM-CSF antibodies in the treatment of COVID-19 have not been clarified (Khan et al. 2020; Rizk et al. 2020). Therefore, the present meta-analysis was conducted to investigate the effects of GM-CSF antibody treatment in COVID-19 patients.
Methods
This study was conducted according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) (Liberati et al. 2009) and registered in PROSPERO (CRD42020221450; https://www.crd.york.ac.uk/prospero/).
Inclusion and exclusion criteria and data collection
Inclusion criteria were: (1) studies concerning GM-CSF antibody treatment alone or in combination with other specific treatments for adult COVID-19 patients; (2) studies were designed as randomized control trials (RCTs) or cohort studies.
Exclusion criteria were: (1) studies without controls or only self-control; (2) data were not available or were repeated; (3) reviews, editorials, letters, commentaries.
Search strategy
The search strategy was conducted by A.X. and Y.L. for the Cochrane Library, Embase, PubMed, and ClinicalTrials.gov. References for each piece of literature were manually reviewed. Any disagreement was solved by ZX. In detail, relevant literature published from 1 December 2019 and until 27 July 2022 was searched with the strategies shown in Supplemental Tables 1.
Study selection and data extraction
Literature selection was undertaken by A.X. and Y.L. The duplicate literature was excluded using EndNote 20; the eligible literature with full text were reviewed and selected according to the inclusion and exclusion criteria. J.G. and W.W. gathered data from relevant literature, including therapeutic strategies, the number of recruited patients, ventilation conditions, death number, severe case number, ICU admission risk, risks of secondary infection, and severe events (e.g., sepsis, acute kidney injury, and cardiac injury).
Assessment of study quality
For cohort studies, the Newcastle–Ottawa Scale (NOS) (Wells et al. 2014) was used to assess study quality (Supplemental Table 2). For RCTs, Cochrane risk-of-bias tool was used (Supplemental Fig. 1).
Statistical analysis
Forest plots were used to display the pooled estimates, which were reported as odds ratios (ORs) and 95% confidence intervals (CIs). As the studies included in the analysis are not functionally identical (such as the use of different kinds of drugs and different standard of care methods), a random-effects model (Mantel–Haenszel method) was employed for the meta-analysis. Heterogeneity among studies was evaluated by χ2, I2, df, and Tau2. I2 values selected zero. I2 values of 25%, 50%, and 75% were, respectively, indicated low, moderate, and high heterogeneity. Funnel plots and Egger’s asymmetry test were used to assess publication bias. Revman 5.3 was used for statistical processing. In addition, to evaluate the strength and stability of the meta-analysis, sensitivity analysis was conducted by omitting the individual studies one by one.
Results
Selection of eligible literatures
According to the Fig. 1, 1480 pieces of literatures were initially identified based on their titles and abstracts; 212 literatures were removed for duplicate; 1260 literatures were removed by the inclusion and exclusion criteria. Finally, the meta-analysis included seven pieces of relevant literatures (Cremer et al. 2021; Criner et al. 2022; De Luca et al. 2020; Fisher et al. 2022; Patel et al. 2022; Temesgen et al. 2020, 2021). Of note, one preprint of Temesgen et al. 2021 was included, because it is the results of a registered clinical trial from ClinicalTrials.gov (number: NCT04351152).Fig. 1 Flow diagram of selection of eligible literatures
Characteristics of included literatures
The GM-CSF antibody group (n = 866) and the control group (n = 891) were from seven studies (Cremer et al. 2021; Criner et al. 2022; De Luca et al. 2020; Fisher et al. 2022; Patel et al. 2022; Temesgen et al. 2020, 2021). Two of included studies were high-quality cohort studies (De Luca et al. 2020; Temesgen et al. 2020) and the other five were RCTs (Cremer et al. 2021; Criner et al. 2022; Fisher et al. 2022; Patel et al. 2022; Temesgen et al. 2021); two of included studies were used anti-GM-CSFR antibody (Cremer et al. 2021; De Luca et al. 2020), and the other five were used anti-GM-CSF antibodies (Criner et al. 2022; Fisher et al. 2022; Patel et al. 2022; Temesgen et al. 2020, 2021). Characteristics of the included studies are summarized in Table 1, including the study design (type of study), treatment methods, sample sizes, the standard of care, primary endpoints, and secondary endpoints. The NOS scores of included non-RCTs are shown in Supplementary Table 2 and the risk of bias of included RCTs was shown in Supplementary Fig. 1.Table 1 Characteristics of the studies included in the meta-analysis
Reference Type of study Dosage of GM-CSF antibodies Control Outcomes Standard care
N Method N Method
Nonrandomized studies (cohort studies)
Luca et al. 2020 Single-center prospective cohort study 13 Mavrilimumab iv 6 mg/kg and standard care 26 Standard care Clinical improvement endpoint Oral hydroxychloroquine (200 mg twice a day), intravenous azithromycin (500 mg, once daily until patient tested negative for urine antigen for Legionella pneumophila), oral lopinavir–ritonavir (400 and 100 mg, respectively, twice a day), and respiratory support with supplemental oxygen or non-invasive ventilation with continuous positive airway pressure (with a positive end expiratory pressure of 10 cm of water)
Temesgen et al. 2020 Single-center retrospective cohort study 12 ①lenzilumab 600 mg iv and standard care 27 Standard care Incidence of clinical improvement Remdesivir and/or dexamethasone and US FDA guidelines
RCTs
Patel et al. 2022 Large, global, double-blind, randomized, placebo-controlled study 397 90 mg infusion of otilimab and standard care 396 Standard care The primary endpoint: Proportion of patients alive and free of respiratory failure
The secondary endpoints: time to recovery from respiratory failure; time to all-cause mortality; time to last dependence on supplementary oxygen; time to final ICU discharge; time to first discharge from investigator site; time to final hospital discharge
US FDA guidelines
Fisher et al. 2022 Randomized, multi-arm, parallel-group, open-label, trial 55 Namilumab (150 mg) iv once and standard care 54 Standard care The primary endpoint: CRP concentration
The secondary endpoints: WHO clinical progression scale
UK guidelines
Cremer et al. 2021 Multicenter, double-blind, randomized trial 21 Mavrilimumab iv 6 mg/kg and standard care 19 Standard care The primary endpoint: patients alive and off supplemental oxygen
The secondary endpoint: mortality; the proportion of patients alive and without respiratory failure
US FDA guidelines
Temesgen et al. 2021 Randomized, double-blind, placebo-controlled trial 255 Lenzilumab iv tid (600 mg, each) and standard care 257 Standard care The primary endpoint: survival without ventilation
The secondary endpoint: time to recovery; the proportion of patients with the composite of IMV, ECMO, or death; ventilator-free days; duration of ICU stay; and mortality
Remdesivir and/or dexamethasone and US FDA guidelines
Criner et al. 2022 Randomized, double-blind, placebo-controlled trial 113 Gimsilumab 400 mg iv on Day 1 and 200 mg iv on Day 8, and standard care 112 Placebo and standard care The primary endpoint: mortality, the secondary endpoint: proportion of subjects who survived with mechanical ventilation-free; mechanical ventilation-free days, and time to hospital discharge US FDA guidelines
iv intravenous, tid three times a day, ICU intensive care unit, WHO World Health Organization, CRP C-reactive protein, IMV invasive mechanical ventilation, ECMO extracorporeal membrane oxygenation, FDA Food and Drug Administration
Mortality
1724 COVID-19 patients from 7 studies contribute to mortality data. GM-CSF antibodies were used in 847/1724 patients. As shown in Fig. 2, the risk of mortality in the GM-CSF antibody group was slightly lower than that in the control group, which was supported by two cohort studies (OR 0.24; 95% CI 0.04,1.42) and five RCTs (OR 0.84; 95% CI 0.63, 1.10). Heterogeneity was low for both cohort studies and RCTs. Of note, the sensitive analysis further showed that the results of mortality based on RCTs become statistically significant (OR 0.73, 95% CI 0.55, 0.98, p = 0.04, Supplemental Table 3) when the study of Criner et al. 2022 was removed. Fig. 2 Risk of mortality between GM-CSF antibodies and control groups
Incidence of invasive mechanical ventilation (IMV)
To estimate the risk of incidence of IMV, we analyzed the data of 822 COVID-19 patients from 5 studies, including 395 in the GM-CSF antibody group and 427 in the control group. As shown in Fig. 3, incidence of IMV in GM-CSF antibodies group was lower than that in control group, which was supported by two cohort studies (OR 0.16; 95% CI 0.03, 0.74) and three RCTs (OR 0.62; 95% CI 0.41, 0.94). Heterogeneity was low for both cohort studies and RCTs. The results are similar when excluding each one of these included studies (Supplemental Table 3).Fig. 3 The risk of IMV between the GM-CSF antibodies treatment group and the control group in COVID-19 patients
Ventilation improvement rate
The rate of ventilation improvement (need oxygen therapy to without oxygen therapy) from 3 controlled clinical trial, involving 304 COVID-19 patients (147 in the GM-CSF antibodies treatment group and 157 in the control group), were analyzed. A total of 99/147 in the GM-CSF antibodies treatment group and 92/157 patients in control group had improved ventilation, respectively. As shown in Fig. 4, the treatment of GM-CSF antibodies did not significantly improve ventilation in patients, which was supported by one cohort study of De Luca et al. (2020) (OR 5.50; 95% CI 1.01, 29.85) and two RCTs (OR 2.02, 95% CI 0.34, 12.12). Besides, when excluded the study of Criner et al. 2022, the results of ventilation improvement rate based on the study of Criner et al. 2022 become statistically significant (OR 6.07, 95% CI 1.58, 23.33, p = 0.009, Supplemental Table 3).Fig. 4 Ventilation improvement rate between GM-CSF antibodies and control groups COVID-19 patients
Secondary infection
The data of secondary infection were from 6 studies involving 1718 patients. Among them, 854 patients were in the GM-CSF antibody group and 864 patients were in control group. As presented in Fig. 5, secondary infection was not found to differ significantly across groups, which was supported by both one cohort study (OR 0.25, 95% CI 0.01, 5.19) and five RCTs (OR 1.00, 95% CI 0.71, 1.41). Heterogeneity was low for both cohort studies and RCTs. The results are similar when excluding each one of these included studies (Supplemental Table 3).Fig. 5 The risk of secondary infection between the GM-CSF antibodies treatment group and the control group in COVID-19 patients
Adverse events
Among the included trials, five RCTs reported covering nine kinds of adverse events (AEs), four RCTs reported the total number of treatment patients with AEs, and four RCTs reported serious adverse events (SAEs). As shown in Table 2 and Supplementary Fig. 2, compared with control group, GM-CSF antibodies treatment did not increase the AE rate (OR 0.95, 95% CI 0.86, 1.05, P = 0.31) and SAE rate (OR 0.85, 95% CI 0.66, 1.08, P = 0.18). Of note, GM-CSF antibodies treatment was almost associated with the lower rate of pulmonary AEs based on 2 trials involving 1305 COVID-19 patients (OR 0.99, 95% CI 0.59, 1.00, P = 0.05). Heterogeneity was low for both cohort studies and RCTs. The results are similar when excluding each one of these included studies (Supplemental Table 3).Table 2 Adverse events between the GM-CSF antibodies treatment group with control group
Author Patel 2022 Fisher 2022 Cremer 2021 Temesgen 2021 Criner 2022
Group A B A B A B A B A B
Number of patients 397 396 55 54 21 19 255 257 113 112
Patients with AEs 274 265 30 29 68 84 84 77
Patients with SAEs 124 147 10 10 5 4 47 45
Gastrointestinal AEs 54 53
Diarrhea 15 18
Constipation 39 35
Liver function events 6 5 5 3 11 7 8 10
ALT elevation 5 3 5 3
GGT elevation 6 4
Hepatocellular injury 6 5
Abnormal liver enzyme test 8 10
Lung function event 69 85 66 80
Respiratory failure 19 21 24 31
Pulmonary embolism 13 25 5 3
Hypoxemia 10 13 15 15
Acute respiratory failure 10 11 18 22
Pneumothorax 17 15 0 6
Acute respiratory distress syndrome 4 3
Cardiac disorders event 12 18 11 11 3 3
Atrial fibrillation 12 18 3 3
Cardiac arrest 8 4
Cardiorespiratory arrest 3 4
Acute myocardial infarction 0 3
Renal function event 23 25 4 3 5 8 5 9
Renal failure 2 0
Acute kidney injury 23 25 4 3 5 8 3 9
Infection 81 71 6 6 2 1 7 14 15 9
Pneumonia 43 29 4 5 2 1 3 3
Urinary tract infection 13 14
Septic shock 18 16 5 9 7 6
Sepsis 7 12 2 1 2 5 5 0
Blood system 18 22 6 6
Anemia 18 22 6 6
MODS 12 16 1 3 3 6
Other adverse events 137 107 2 6 2 5 2 0
Pyrexia 20 15
Hypernatremia 20 10 1 1
Hypokalemia 15 16
Hyperkalemia 17 13 1 5
Hypotension 14 16 2 5
Hypertension 17 10
Delirium 17 17
Decubitus ulcer 16 9
Infusion-related reaction 1 1 2 0
A GM-CSF antibodies treatment group, B control group
Discussion
The COVID-19 pandemic remains an urgent problem around the globe, with the global outbreak still not fully contained and cases rising. Prior research indicates that patients with COVID-19 infection have increased levels of GM-CSF in their blood, especially in the later stages of the disease (Leavis et al. 2022). Previous studies revealed that GM-CSF can enhance the expression of many pro-inflammatory cytokines and chemokines, such as interleukin-1 (IL-1), interleukin-6 (IL-6), and tumor necrosis factor (Becher et al. 2016). The anti-GM-CSF monoclonal antibodies might have broader effects than other immunomodulatory approaches on the systemic pro-inflammatory responses accompanying the cytokine release syndrome in COVID-19. In addition, the anti-GM-CSF monoclonal antibodies have been approved by the FDA for emergency compassionate use in COVID-19 patients (Humanigen, Inc. 2020).
Based on the considerable potential of GM-CSF antibodies to treat COVID-19, this article intends to evaluate the clinical efficacy and safety of GM-CSF antibodies for COVID-19. Our meta-analysis involved seven studies that examined GM-CSF antibodies as a treatment for COVID-19 patients, including two studies used anti-GM-CSFR antibodies and the other five studies used anti-GM-CSF antibodies. We found that: (1) treatment with GM-CSF antibodies decreased the incidence of IMV; (2) GM-CSF antibodies resulted in a slight but not significant decrease in mortality and the ventilation improvement rate, while sensitive analysis showed that the results of mortality and ventilation improvement rate became statistically significant when one included study were removed; (3) GM-CSF antibodies did not increase risks of secondary infection and AEs in COVID-19 patients. Taken together, our study suggests that GM-CSF antibodies is well tolerated and may have substantial effects in reducing the incidence of IMV among COVID-19 patients. Further studies on the beneficial effects of GM-CSF antibodies in COVID-19 patients remain needed, especially for mortality and the ventilation improvement rate.
GM-CSF antibodies inhibit the GM-CSF signaling axis by targeting GM-CSF, or GM-CSFR. Both strategies blocked the same interaction regardless of whether they targeted GM-CSF or GM-CSFR in theory (Zhou et al. 2020). In our meta-analysis, the treatment with GM-CSF antibodies (pooled the data of both strategies together) mainly decreased the incidence of IMV but only slightly affected mortality and ventilation improvement rate. Two of our included articles used mavrilimumab, the monoclonal antibody against GM-CSFR, which has a similar effect to GM-CSF target antibodies for decreasing the incidence of IMV the treatment of COVID-19. However, results of mortality and ventilation improvement rate became statistically significant after removing Criner et al. 2022, one of the RCTs used mavrilimumab. These results indicate the type of GM-CSF antibodies remains the potential factor contributing to the heterogeneities and might influence the efficacy of GM-CSF antibodies treatment for COVID-19 patients.
Secondary infection often indicates severe severity and results in mortality in COVID-19 patients (Guo et al. 2021). Secondary infection is also common side effects or AEs of many anti-inflammatory therapies, such as glucocorticoid, Janus kinase inhibitors, and IL-6 antibodies (Russell et al. 2020; Shah et al. 2022; Winthrop et al. 2018). Kimmig et al. (2020) and Quartuccio et al. (2020) revealed that COVID-19 patients used IL-6 antibodies may have a higher rate of secondary infection; Roumier et al. (2021) also reported that early inhibition of IL-6 may trigger many AEs in COVID-19 patients. Different from IL-6 antibodies, here we found that GM-CSF inhibition treatment did not increase the risks of secondary infection and other AEs in COVID-19 patients. Besides, GM-CSF antibodies treatment was almost significantly reduced the rate of pulmonary AEs in COVID-19 patients. Thus, GM-CSF inhibition treatment might be safer than IL-6 antibodies and other anti-inflammatory therapies.
Several limitations in our meta-analysis should be noted as follows: (1) Funnel-plot analysis and Egger’s test were not performed because of the small number of included studies. (2) One study is a pre-publication, which can lead to publication bias or other potential risks. Two of the seven studies are nonrandomized, which may reduce the credibility of the combined results (though we analyzed the nonrandomized studies and RCTs separately). (3) The different standards of care used in these studies prevent us from conducting an adequate comparison and analysis. As an example, oxygen therapy and nursing measures may influence mortality in COVID-19 patients. (4) The study population is different for some studies, which implies a risk of selection bias. (5) Moreover, there are some other factors influencing the results in this study, such as other drugs, ventilator availability, and vaccines. The synergistic effect or other interactions of these factors (such as dexamethasone (Vohra et al. 2021)) remain unclear.
Conclusions
Our meta-analysis shows treatment with GM-CSF antibodies decreased the incidence of IMV and does not increase the risks of second infection and AEs. It is likely that GM-CSF antibodies reduced mortality and the ventilation improvement rate. Further RCTs, cohort studies, and multicenter clinical studies are still needed.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 8443 KB)
Author contributions
The study was designed by JG and ZX; AX and YL ran the search strategies and undertook the search. JG and WW extracted data and analysis. AX and YL wrote the manuscript. AX, YL, JG and WW contribute to editing the manuscript and checking all the extracted data under the supervision of ZX.
Funding
This work was funded by Natural Science Foundation of Zhejiang (No. LY22H280007), National Natural Science Foundation of China (82174005), and Foundation of Zhejiang Chinese Medical University (No. Q2019Y02).
Data availability
All data are presented in the text.
Declarations
Conflict of interest
The authors declare that there is no conflict of interest.
Ethical approval
Consent statement and ethical approval are not required as the current study was based on published data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
An-ran Xi, Yi-jun Luo and Jin-Tao Guan have contributed equally to this work and should be considered the co-first authors.
==== Refs
References
Becher B Tugues S Greter M GM-CSF: from growth factor to central mediator of tissue inflammation Immunity 2016 45 963 973 10.1016/j.immuni.2016.10.026 27851925
Bonaventura A Vecchié A Wang TS Lee E Cremer PC Carey B Rajendram P Hudock KM Korbee L Van Tassell BW Targeting GM-CSF in COVID-19 pneumonia: rationale and strategies Front Immunol 2020 11 1625 10.3389/fimmu.2020.01625 32719685
Chen P Nirula A Heller B Gottlieb RL Boscia J Morris J Huhn G Cardona J Mocherla B Stosor V SARS-CoV-2 neutralizing antibody LY-CoV555 in outpatients with COVID-19 New Engl J Med 2021 384 229 237 10.1056/NEJMoa2029849 33113295
Cremer PC Abbate A Hudock K McWilliams C Mehta J Chang SY Sheng CC Van Tassell B Bonaventura A Vecchié A Mavrilimumab in patients with severe COVID-19 pneumonia and systemic hyperinflammation (MASH-COVID): an investigator initiated, multicentre, double-blind, randomised, placebo-controlled trial Lancet Rheumatol 2021 3 e410 e418 10.1016/S2665-9913(21)00070-9 33754144
Criner GJ Lang FM Gottlieb RL Mathews KS Wang TS Rice TW Madduri D Bellam S Jeanfreau R Case AH Anti-granulocyte-macrophage colony-stimulating factor monoclonal antibody gimsilumab for COVID-19 pneumonia: a randomized, double-blind, placebo-controlled trial Am J Respir Crit Care Med 2022 205 1290 1299 10.1164/rccm.202108-1859OC 35290169
De Luca G Cavalli G Campochiaro C Della-Torre E Angelillo P Tomelleri A Boffini N Tentori S Mette F Farina N GM-CSF blockade with mavrilimumab in severe COVID-19 pneumonia and systemic hyperinflammation: a single-centre, prospective cohort study Lancet Rheumatol 2020 2 e465 e473 10.1016/S2665-9913(20)30170-3 32835256
Fisher BA Veenith T Slade D Gaskell C Rowland M Whitehouse T Scriven J Parekh D Balasubramaniam MS Cooke G Namilumab or infliximab compared to standard of care in hospitalised patients with COVID-19 (CATALYST): a phase 2 randomised adaptive trial Lancet Respir Med 2022 10 3 255 266 10.1016/S2213-2600(21)00460-4 34922649
Guo M Gao M Gao J Zhang T Jin X Fan J Wang Q Li X Chen J Zhu Z Identifying risk factors for secondary infection post-SARS-CoV-2 infection in patients with severe and critical COVID-19 Front Immunol 2021 12 715023 10.3389/fimmu.2021.715023 34659204
Humanigen, Inc. (2020) FDA Approves Emergency IND Use of Humanigen's Lenzilumab for Compassionate Use in COVID-19 Patients. Available online at: https://www.accesswire.com/583610/FDA-Approves-Emergency-IND-Use-of-Humanigens-Lenzilumab-For-Compassionate-Use-In-COVID-19-Patients. Accessed 2 Apr 2020
Khan AR Soneja M Tirlangi PK Wig N Mavrilimumab for severe COVID-19 Lancet Rheumatol 2020 2 e661 e662 10.1016/S2665-9913(20)30306-4 33521661
Kimmig LM Wu D Gold M Pettit NN Pitrak D Mueller J Husain AN Mutlu EA Mutlu GM IL-6 inhibition in critically Ill COVID-19 patients is associated with increased secondary infections Front Med (lausanne) 2020 7 583897 10.3389/fmed.2020.583897 33195334
Lang FM Lee KM Teijaro JR Becher B Hamilton JA GM-CSF-based treatments in COVID-19: reconciling opposing therapeutic approaches Nat Rev Immunol 2020 20 507 514 10.1038/s41577-020-0357-7 32576980
Leavis HL van de Veerdonk FL Murthy S Stimulating severe COVID-19: the potential role of GM-CSF antagonism Lancet Respir Med 2022 10 223 224 10.1016/S2213-2600(21)00539-7 34863335
Liberati A Altman DG Tetzlaff J Mulrow C Gøtzsche PC Ioannidis JP Clarke M Devereaux PJ Kleijnen J Moher D The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration Ann Intern Med 2009 151 W65 94 10.7326/0003-4819-151-4-200908180-00136 19622512
Mehta P Porter JC Manson JJ Isaacs JD Openshaw PJM McInnes IB Summers C Chambers RC Therapeutic blockade of granulocyte macrophage colony-stimulating factor in COVID-19-associated hyperinflammation: challenges and opportunities Lancet Respir Med 2020 8 822 830 10.1016/S2213-2600(20)30267-8 32559419
Patel J Beishuizen A Ruiz XB Boughanmi H Cahn A Criner GJ Davy K de-Miguel-Díez J Fernandes S François B A randomized trial of otilimab in severe COVID-19 pneumonia (OSCAR) Eur Respir J 2022 10.1183/13993003.01870-2021 36229048
Quartuccio L Sonaglia A McGonagle D Fabris M Peghin M Pecori D De Monte A Bove T Curcio F Bassi F Profiling COVID-19 pneumonia progressing into the cytokine storm syndrome: results from a single Italian centre study on tocilizumab versus standard of care J Clin Virol 2020 129 104444 10.1016/j.jcv.2020.104444 32570043
Rizk JG Kalantar-Zadeh K Mehra MR Lavie CJ Rizk Y Forthal DN Pharmaco-immunomodulatory therapy in COVID-19 Drugs 2020 80 1267 1292 10.1007/s40265-020-01367-z 32696108
Roumier M Paule R Vallée A Rohmer J Ballester M Brun AL Cerf C Chabi ML Chinet T Colombier MA Tocilizumab for severe worsening COVID-19 pneumonia: a propensity score analysis J Clin Immunol 2021 41 303 314 10.1007/s10875-020-00911-6 33188624
Russell CD Millar JE Baillie JK Clinical evidence does not support corticosteroid treatment for 2019-nCoV lung injury Lancet 2020 395 473 475 10.1016/S0140-6736(20)30317-2 32043983
Shah R Shah J Gohil J Revathi G Surani S Secondary infections in patients with COVID-19 pneumonia treated with tocilizumab compared to those not treated with tocilizumab: a retrospective study at a tertiary hospital in kenya Int J Gen Med 2022 15 2415 2425 10.2147/IJGM.S356547 35264878
Sterner RM Sakemura R Cox MJ Yang N Khadka RH Forsman CL Hansen MJ Jin F Ayasoufi K Hefazi M GM-CSF inhibition reduces cytokine release syndrome and neuroinflammation but enhances CAR-T cell function in xenografts Blood 2019 133 697 709 10.1182/blood-2018-10-881722 30463995
Temesgen Z Assi M Shweta FNU Vergidis P Rizza SA Bauer PR Pickering BW Razonable RR Libertin CR Burger CD GM-CSF neutralization with lenzilumab in severe COVID-19 pneumonia: a case-cohort study Mayo Clin Proc 2020 95 2382 2394 10.1016/j.mayocp.2020.08.038 33153629
Temesgen Z Burger CD Baker J Polk C Libertin C Kelley C Marconi VC Orenstein R Durrant C Chappell D Lenzilumab efficacy and safety in newly hospitalized COVID-19 subjects: results from the live-air phase 3 randomized double-blind placebo-controlled trial medRxiv 2021 8 29
Vohra M Sharma AR Satyamoorthy K Rai PS Pharmacogenomic considerations for repurposing of dexamethasone as a potential drug against SARS-CoV-2 infection Per Med 2021 18 389 398 10.2217/pme-2020-0183 34086487
Wells GA, Wells G, Shea B, Shea B, O’Connell D, Peterson J, Welch LM, Tugwell P, Ga SW, Zello GA, Petersen JA (2014) The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Available online at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp
Wiersinga WJ Rhodes A Cheng AC Peacock SJ Prescott HC Pathophysiology, transmission, diagnosis, and treatment of coronavirus disease 2019 (COVID-19): a review JAMA 2020 324 782 793 10.1001/jama.2020.12839 32648899
Winthrop KL Melmed GY Vermeire S Long MD Chan G Pedersen RD Lawendy N Thorpe AJ Nduaka CI Su C Herpes zoster infection in patients with ulcerative colitis receiving tofacitinib Inflamm Bowel Dis 2018 24 2258 2265 10.1093/ibd/izy131 29850873
Zhou Y Fu B Zheng X Wang D Zhao C Qi Y Sun R Tian Z Xu X Wei H Pathogenic T-cells and inflammatory monocytes incite inflammatory storms in severe COVID-19 patients Natl Sci Rev 2020 7 998 1002 10.1093/nsr/nwaa041 34676125
| 36445552 | PMC9707187 | NO-CC CODE | 2022-12-01 23:20:21 | no | Inflammopharmacology. 2022 Nov 29;:1-11 | utf-8 | Inflammopharmacology | 2,022 | 10.1007/s10787-022-01105-9 | oa_other |
==== Front
Stud Russ Econ Dev
Stud Russ Econ Dev
Studies on Russian Economic Development
1075-7007
1531-8664
Pleiades Publishing Moscow
7343
10.1134/S1075700722060077
Science and Technology
Russian Science and Technology: Rise or Progressive Lag (Part I)
Klepach A. N. 1
Vodovatov L. B. 2
Dmitrieva E. A. [email protected]
2
1 grid.473260.4 0000 0004 0562 8681 Institute of Economic Forecasting, Russian Academy of Sciences, Moscow, Russia
2 Research and Expert Review Institute, Vnesheconombank of the Russian Federation, Moscow, Russia
28 11 2022
2022
33 6 631644
6 6 2022
15 6 2022
20 6 2022
© Pleiades Publishing, Ltd. 2022, ISSN 1075-7007, Studies on Russian Economic Development, 2022, Vol. 33, No. 6, pp. 631–644. © Pleiades Publishing, Ltd., 2022.Russian Text © The Author(s), 2022.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The article proposes an approach to a comprehensive assessment of the level of scientific and technological activity in Russia in comparison with leading foreign countries. An analysis has been made of the current imbalances in the development of the scientific and technological sphere in Russia and the reasons for the failure to fulfill many strategic goals. A forecast of scientific and technological development in the context of the existing management system has been made and an assessment has been taken of the potential effect and cost of measures aimed at increasing the technological sovereignty of Russia and the formation of an advanced knowledge economy.
Keywords:
science
technology
innovation
science financing
technological sovereignty
public administration
issue-copyright-statement© Pleiades Publishing, Ltd. 2022
==== Body
pmcIntroduction. Almost all countries and experts consider science and technology to be the main driver of modern development. In Russia, science employs almost 0.8% of all those working in the national economy and creates 1.4% of GDP. This is not much, but the overall contribution of science and technology to the development of the Russian economy is many times higher. However, despite certain positive results, in general, the sphere of technological development and especially science is, in our opinion, the Achilles’ heel of the modern Russian economy. In fact, the main achievements in this area include the following:
— In the field of innovation, a system of technological and venture development institutions has been created: Rosnano, NTI, Skolkovo, RVC, the Foundation for Assistance to the Development of Small Forms of Enterprises in the Scientific and Technological Sphere (FSI) and a number of other funds.
— Funding for megascience class installations (national project Science and Universities) was increased, a grant system to support scientists and projects was created.
— In 2013–2020, there was an increase in the level of wages in the scientific sector in relation to the corporate sector, an influx of young scientists and researchers increased.
— There are separate breakthrough technological results (nanotubes, vaccines, composites, supercomputers and works on artificial intelligence, nuclear technologies and lasers, hypersound).
— Publication activity of scientists increased.
At the same time, Russia is lagging behind developed countries in terms of R&D funding [1], patent and publication activity, and the number of research personnel is declining. Most target indicators of the Decrees of the President of the Russian Federation of 2012 and 20181 for the development of the sphere of science and technology, except for the task of raising wages, have not been fulfilled (Table 1).
Table 1. Results of the implementation of strategic documents for the development of science and technology
Strategic documents Indicators Forecast value (estimate) Year
of achievement Russia 2020 (fact)
Concept of Long-Term Socio-Economic Development of Russia until 2020
and
Strategy for Innovative Development of Russia 2020
Research and development expenses, in % of GDP 2.5–3 2020 1.1
Share of industrial enterprises implementing technological innovations, % 40–50 21.5*
Share of innovative products, % 25–30 5.7
* In 2020, the share of organizations implementing technological innovations increased to 23% (up to 21.5% in industry). This growth is associated with a change in international recommendations on the statistical measurement of innovation, implemented by the OECD together with Eurostat (Oslo Manual). The value of the indicator for the Russian Federation for 2017, calculated according to the criteria of the 3rd edition of the Oslo Manual, amounted to 7.5%, when recalculated according to the criteria of the 4th edition of the Oslo Manual, it increased to 20.8%. The difference in the calculation is due to the use of three criteria for classifying an organization as innovative instead of one.
In the Concept of Long-Term Socio-Economic Development until 2020 (CLTD)2 adopted by the Government of the Russian Federation in 2008, the task of Russia’s transition to the innovative development of the economy was set. In 2011, the Innovative Development Strategy was adopted, the purpose of which was to increase the innovative activity of business and the efficiency of the transformation of scientific ideas into technologies and market innovations, a state program for the development of science and technology was initiated, as well as a number of other decisions.
However, most of the targets of the CLTD-2020 and the Innovation Development Strategy were not achieved. In particular, domestic expenditures on R&D were planned in 2020 at the level of 3% of GDP, but in fact amounted to only 1.1%. Actually, the indicator of R&D expenses relative to GDP (1%) has been stagnating for almost 13 years. The problems that had accumulated over the years sharply escalated under the conditions of the hybrid war launched by the West against Russia, including actions to isolate the Russian scientific community from world science and a technological blockade.
What resources does Russia have in this war of minds and technologies, and how should the management of the scientific and technological complex be restructured in order to win and lift the country?
Comparative potential and effectiveness of the scientific and technological complex of Russia and other countries: relative and absolute dimensions. If in 2008 we were approximately on the same level with China in terms of the relative level of R&D expenses, now China has increased spending to 2.4% of GDP, despite the fact that its GDP is 5.5 times higher than in Russia (Table 2). In the United States, R&D expenses are 3.4% of GDP, in South Korea 4.8% of GDP. Although Russia now ranks 9th in the world in terms of R&D expenses (in terms of purchasing power parity), it is 12.1 times behind China and 15 times behind the United States, and there are all the prerequisites for this gap to widen [2]. In the context of fierce technological rivalry and the hybrid war unleashed against Russia by the collective West, such an imbalance in power becomes threatening.
Table 2. Comparative global dynamics of R&D expenses
Country R&D expenditures based on PPP, billion dollars R&D expenses, % of GDP Including government R&D expenses, % of GDP
2008 2020 2008 2020 2008 2020
Israel 8.7 19.8 4.3 5.44 0.5 0.5
Republic of Korea 43.9 112.9 3.0 4.81 0.8 1.1
USA 407.2 720.8 2.8 3.45 0.8 0.7
Japan 148.7 174.1 3.3 3.3 0.5 0.5
Germany 81.2 143.4 2.6 3.14 0.7 0.9
China 145.1 582.8 1.4 2.4 0.3 0.5
France 46.6 74.6 2.1 2.35 0.8 0.7
Great Britain 36.5 56.9 1.6 1.7 0.5 0.5
Czech Republic 3.6 8.9 1.2 1.9 0.6 0.7
Russia 30.1 47.9 0.97 1.1 0.6 0.7
Source: OECD, UK (PPP R&D expenses) 2019, Israel, Germany, France (Public R&D expenses, % of GDP) 2019.
If we evaluate R&D expenses taking into account the number of researchers, then the situation in Russia will be even less optimistic: in 2019, 1 researcher (in full-time equivalent) accounted for 3.5 times less research and development expenses than in the United States, and three times less than in Germany. According to this indicator, Russia occupies only 44th place3 (Table 3).
Table 3. Place of Russia among the leading countries of the world
Country R&D employees
by country: 2020 (thousand person-years, full-time equivalent) Country Number of researchers by country: 2020* (thousand person-years, full-time equivalent) Country Domestic R&D expenses by country: 2020 billion dollars in terms of PPP
1 China 4800.8 China 2109.5 USA 657.5
2 USA 1554.9 USA 1554.9 China 525.7
3 Japan 903.4 Japan 681.8 Japan 173.3
4 Russia 748.7 Germany 450.7 Germany 148.1
5 Germany 735.6 Republic of Korea 430.7 Republic of Korea 102.5
6 India 553.0 Russia 397.2 France 73.3
7 Republic of Korea 525.7 India 341.8 India 58.7
8 Great Britain 486.1 Great Britain 317.5 Great Britain 56.9
9 France 463.7 France 314.1 Russia 45.4
10 Italy 355.9 Brazil 180.0 Taiwan 44.0
11 Brazil 316.5 Canada 167.4 Italy 39.3
12 Taiwan 271.6 Italy 160.8 Brazil 36.3
13 Canada 238.1 Taiwan 159.2 Canada 31.0
* OECD data. Number of researchers: UK–2007, 2018, China—2012, Germany—2007.
Source: OECD, ANO VEB Institute.
The national project provided for Russia to maintain the 5th place in terms of the number of researchers in full-time equivalent among the leading countries of the world (according to the OECD) for the period from 2018 to 2021. However, according to the OECD, in 2020 the Republic of Korea overtook Russia in this indicator, shifting it to the 6th position in the rating. Thus, in 2020 Russia is ahead of China (the number of researchers in full-time equivalent is estimated at 2109.5 thousand people), the United States (1554.9 thousand people), Japan (681.8 thousand people), Germany (450.7 thousand people), Republic of Korea (430.7 thousand people). In Russia, this figure in 2020 decreased to 397.2 thousand people versus 400.6 thousand people in 2019. In the leading countries, the number of researchers is growing, while in Russia it has been declining for more than 20 years in a row.
The lag behind developed countries in the field of science and technology is quite large, which creates the preconditions for a “brain drain.” In Germany and the Czech Republic, the salary of scientific workers is 1.9 and 1.3 times higher than the corresponding Russian indicator (the gap for researchers is even higher).4 However, the issue of brain drain is not so much a matter of wages, but rather of the opportunity to realize one’s potential, the status of a scientist and engineer in society, as well as the access to the global scientific and technological community.
The capital–labor ratio of the scientific sector in Russia today is comparable to that of the scientific sector in the Czech Republic (Table 4).
Table 4. Resources of science: position of the Russian scientific complex
No. Country R&D,
billion dollars
in terms of PPP Capital-labor ratio, billion dollars Number
of researchers, thousand people Researcher’s salary in 2021
in dollars according
to glassdoor.com in terms of PPP * Patents**, thousand items Web of science publications, thousand items***
2008 2020 2008 2020**** 2008 2020***** 2021 2008 2020 2008 2020
1 USA 407.2 720.8 N/A N/A 10 242 429 496 327 606
2 China 145.1 582.8 N/A N/A 2069.7 4393 204 1441 108 614
3 Japan 148.7 174.1 N/A N/A 890.7 951.0 6074 510 423 77 117.3
4 Germany 81.2 143.4 N/A N/A 437.8 667.4 6501 172 168 83 162.5
5 Republic of Korea 43.9 112.9 N/A N/A 300.1 558.0 4808 173 261 34 86.1
6 France 46.6 74.6 163.0 225.3 289.0 430.0 5436 62 64 62 105.4
7 Great Britain 407.2 56.9 59.6 61.8 377.2 535.5 5780 51 53 86 194.9
8 Russia 30.1 47.9 11.3 18.0 375.8 346.4 3581 31 30 28 84.9
9 Israel 8.7 19.8 N/A N/A 4845 11 16 12 22.3
10 Czech Republic 3.6 8.9 12.8 12.3 44.2 65.1 4512 2 2 8 22.3
* Salary is estimated based on machine learning on the basis of millions of salaries from Glassdoor and latest government data sources.
** Patent statistics: https://www3.wipo.int/ipstats/IpsStatsResultvalue. Cited June 5, 2022.
*** Only web of science publications. The dynamics of the number of publications is calculated based on the data of the analytical system InCites (Clarivate Analytics) for Web of Science as of October 31, 2021. A publication means three types of documents indexed in Web of Science: an article, a review and a proceedings paper.
**** France’s capital-labor ratio—2019 data.
***** Number of researchers in China, Germany, France, UK in 2019.
Source: OECD, Rosstat, National Research University Higher School of Economics, ANO VEB Institute.
Naturally, not only the total capital–labor ratio is important, but also the technical equipment of researchers, since the potential capabilities of organizations in obtaining world-class scientific results and their competitive prospects largely depend on the availability of modern scientific equipment. The Decree of the President of the Russian Federation No. 204 of May 7, 2018 set the strategic task of updating at least 50% of the instrumentation base of the leading research and development organizations by 2024. In 2020, less than half (39%) of technical equipment was new equipment under the age of five years. However, most of the research organizations, especially those of an applied nature, remained outside the national project and, accordingly, without incentive measures to update the experimental and testing base.
Traditionally, the effectiveness of scientific activity is considered through publication and patent activity. In 2020, Russia ranked 14th in terms of the number of publications in Web of Science and 8th in Scopus; 10th place in terms of the number of applications for a patent for inventions (Table 5). Not the highest result, but its level relative to R&D expenses is very decent. With a significantly lower share of spending on science in GDP, the publication activity of scientists in Russia corresponds to and even exceeds similar values for other developed countries. Accordingly, the relative “cost” of one scientific publication in Russia is lower, which is confirmed by the RSF estimate of the “cost” of publications at the expense of scientific grants allocated by the Foundation, about 2 million rubles for an article in a top-rated journal.
Table 5. Comparative efficiency of scientific activity
No. Country R&D expenses,
% of GDP Capital–labor ratio
per researcher,
million dollars* Patents per researcher** Publications per researcher
2008 2020 2008 2020 2008 2020 2008 2020
1 Israel 4.3 5.4 N/A N/A N/A N/A N/A N/A
2 Rep. Korea 3.0 4.8 N/A N/A 0.6 0.5 0.1 0.2
3 Japan 3.3 3.3 N/A N/A 0.6 0.4 0.1 0.1
4 Germany 2.6 3.14 N/A N/A 0.4 0.3 0.2 0.2
5 USA 2.8 3.45 N/A N/A 0.4 0.3 0.3 0.4
6 China 1.4 2.4 N/A N/A 0.1 0.7 0.1 0.3
7 France 2.1 2.35 0.6 0.5 0.2 0.1 0.2 0.2
8 Great Britain 1.6 1.7 0.2 0.1 0.1 0.1 0.2 0.4
9 Czech Republic 1.2 1.9 0.3 0.2 0.03 0.03 0.1 0.1
10 Russia 0.97 1.1 0.03 0.1 0.08 0.09 0.1 0.2
* France—capital–labor ratio per researcher, 2018.
** USA (2008, 2020), China (2008)—patents per researcher (thousand person-years, full-time equivalent).
Source: OECD, Rosstat, ANO VEB Institute.
This indirectly testifies to the good performance of the domestic scientific sector, at least in the field of fundamental science, in contrast to the opinion often expressed in the expert community and in government about the inefficiency of Russian science.
The low number of patents (in 2020, Russia was not among the top ten countries), as well as the extremely low share of high-tech exports (the ratio of R&D and high-tech exports) really indicate significant problems with the development of the applied science sector responsible for broadcasting the results of fundamental science in pilot and serial production with the involvement of business funds represented by public and private companies and corporations (Fig. 1). It should be taken into account that such exported high-tech products as nuclear fuel and reactors, according to the international classification, do not belong to high-tech exports. Nevertheless, this does not change the overall low rating of Russian high-tech exports.
Fig. 1. Comparison of the level of development of the country, R&D expenses and technological exports, 2020 (the size of the “bubble” is the share of high-tech exports in exports).
For all the importance of the science-intensive export factor, it is necessary to take into account the significant structural difference between the Russian scientific and technological complex and other developed countries. Historically, it was focused not on exports, which were mainly raw materials, but on solving the internal problems of the state, including those related to defense capability. The structure of scientific publications, as well as patent activity in Russia, is largely concentrated in the fields of mathematics, physics and engineering, in contrast to medicine and information technology, which are a priority in the West.
We have yet to create a truly realistic comprehensive system for assessing the achieved scientific and technological potential instead of fragmentary indicators borrowed from Western experience. We need new criteria (except for publications and patents) and a new science assessment system based on a qualified expert assessment, indicators of work with the industry, promotion of R&D results by levels of technological readiness. According to the President of the Russian Academy of Sciences A. M. Sergeev, “the main result will not be an article, but an expert assessment of specialists and the final product”.5
In various coordinate systems of scientific and technological activity, Russia occupies from 6th to 12th place (Table 6). According to a comprehensive assessment of ANO VEB Institute,6 in 2019 we have an honorable 7th place in the world table of scientific ranks.
Table 6. Comprehensive assessment of the place of the Russian scientific complex in the world
Country Resources Results Top 500 universities Final place
DRDE Researchers Publications Patents
USA 1 2 1 2 1 1
China 2 1 2 1 5 2
Japan 3 3 6 3 6 3
Germany 4 4 4 5 3 4
South Korea 5 5 13 4 8 6
France 6 8 7 26 7 10
Great Britain 7 7 3 6 2 5
Russia 8 6 12 11 7 7
Italy 9 9 8 10 10 8
Canada 10 10 9 13 6 9
Spain 11 11 11 22 10 12
Netherlands 12 13 15 8 9 11
Switzerland 13 18 19 7 13 13
Sweden 14 14 20 12 13 14
Belgium 15 15 22 16 14 15
Poland 16 12 17 29 19 17
Austria 17 16 26 15 16 16
Singapore 18 23 33 25 18 20
Denmark 19 20 24 17 16 18
Czech Republic 20 19 27 34 18 21
Finland 21 21 36 20 13 19
Norway 22 24 32 28 17 22
Ireland 23 25 44 27 16 23
Portugal 24 17 25 74 17 25
Hungary 25 22 48 39 20 24
Slovenia 26 27 56 56 20 28
Slovakia 27 26 49 54 20 26
Luxembourg 28 28 76 31 20 27
Source: OECD, ANO VEB Institute.
In general, in terms of the level of scientific and technological activity, Russia is approximately in the same place as in terms of GDP in terms of purchasing power parity. Such positioning is more characteristic of an economy that is trying to gain a foothold on what has been achieved, with the risk of losing its occupied place, than for an economy that is breaking through upwards.
Science and technology policy priorities in Russia: controversial searches. What are the reasons for this huge discrepancy between ambitious goals and modest results? The year of science has ended, but there has been no visible scientific and technological upsurge.
The first reason is chronic underfunding, usually explained by insufficient efficiency, and by reference to the high share of public funding compared to Western countries. True, in terms of absolute volumes, both state and, even more so, private funding per employee or per key area of research and development is extremely small. At the same time, Russian private business with large incomes, unlike Western countries, is concentrated mainly in the fuel and energy and raw materials sectors, where the relative level of R&D expenses in the West is also low. On the whole, in terms of the relative (to revenue) level of R&D expenses, but not in absolute terms, our leading industrial companies are not inferior to Western ones.
Despite the implementation of the Presidential Decree to increase the salaries of scientists included in the target categories, the prestige of scientific activity is not increasing, while the inflationary surge in the scientific and educational sphere in 2022 has so far remained without adequate compensation. All this contributes to the continuation of the trend towards a reduction in the number of scientists and researchers. There is still a shortage of modern scientific equipment, especially of domestic design, with a limited scale of implementation of megascience projects in Russia, despite the PIK and SKIF projects.
The second reason, perhaps more important, is the lack of a systematic, consistent policy for the development of science and technology and a “lost aim of priorities and principles.” We are persistently trying to develop the scientific and educational sphere along the American path, relying on the leading role of universities in the development of science and the formation of a venture capital market as the basis for innovative technologies and projects. However, the Russian tradition is closer to the German model and its advantage, rather than disadvantage, is the presence of powerful academic institutions and industry science, including in the form of state research centers (SRCs) [3].
Despite repeated attempts to reform the field of science and create a modern innovation system, one can speak of a crisis in the management system of the country’s scientific and technological complex.
The list of priority areas and critical technologies has been updated long time ago, in 2002, 2006, 2011 and 2015.7 Despite the fact that the number of positions in the list of critical technologies has consistently decreased, none of the lists was accompanied by an indication of additional funding for the priorities included in the list, as a result of which it did not become a real tool for highlighting the most important areas, and is mainly used for ritual references in the preparation of various scientific programs and funding applications. In the Strategy for Scientific and Technological Development of the Russian Federation (SNTD), a list of seven areas appeared, which actually began to be used as a priority. They were based on an analysis of the “big challenges” facing the country, and they were supposed to be specified at the next stage, which was never implemented.
In the new version of the state program “Scientific and technological development of the Russian Federation” (SP STD),8 an attempt was made to formally align projects and financing instruments under the logic of priorities set in the STD Strategy. To do this, the research work that was previously carried out by line ministries is rather schematically combined into large blocks to correspond to the various priority areas of the Strategy, which in essence is a matter of classification, but not the allocation of real priorities for the purpose of their priority funding. The old system of determining the priorities of technological development by the Presidential Decree does not work, and a new integral system has not been formed. Under these conditions, the plans for scientific research (government order) do not meet the breakthrough tasks and global challenges facing the country, and are largely guided by the principle “by what has been achieved.”
A partial selection of priorities based on forecasting new markets was also carried out during the formation of the National Technology Initiative (NTI) [4]. The Presidium of the Council under the President of the Russian Federation for the Modernization of the Economy and Innovative Development of Russia approved seven roadmaps: Autonet, Aeronet, Marinet, Neuronet, Technet, Healthnet, Energynet. To date, the number of NTI directions has increased to 13, Foodnet, Safenet, Edunet, Sportnet, Homenet, Wearnet have been added. However, the fashion and e-sports markets can hardly claim the role of a priority area for research and development.
In 2022, at a meeting of the Council for Science and Education under the President of the Russian Federation, three new most important innovative projects of national importance were announced, which can also be considered as a choice of priorities:
— Russian scientific and technological platform for rapid response to infectious diseases.
— Creation of the Unified National Monitoring System for Climatically Active Substances.
— Low-carbon closed-loop energy.
Despite the relevance of these topics, taking into account the challenges of the COVID-19 epidemic [5] and the tasks of adaptation to climate change, in the context of a hybrid war, their priority is relatively reduced.
The now adopted new SP STD (funding is approved only until 2024) for 2022 provides for funding in the amount of more than a trillion rubles, but this is ensured mainly by including in the SP STD the amount of funding for research and development previously ordered by line ministries—the Ministry of Industry and Trade of Russia, the Ministry of Health of Russia and other departments. In other words, the apparent increase in funding for the new state program is associated with the merging into one program of all projects and major events, in the name of which there was the word “scientific …” from the rest of the state programs, while the coordination of all the former industry R&D of the Ministry of Education and Science has not been worked out yet. Such an association does not imply real coordination of projects. In nominal terms, until 2024, an average annual growth of funding by 0.9% per year is planned, which means a decrease in funding in real terms (in the old program for 2020, an increase of 2.7% per year was planned) (Table 7).
Table 7. State program for the development of science and technology (old and new), billion rubles
Indicator 2022 2023 2024
old new old new old new
SP STD 838.5 1075.5 881.8 1138.9 957.2 1173.5
FP 119.1 251.2 140.7 282.3 N/A 262.4
Departmental project 0.02 75.9 0.02 64.4 N/A 61.3
OM 719.4 741.0 N/A
FTP 0.6 0.6 N/A 0.6
Complex of process measures 744.8 788.6 N/A 846.2
Source: ANO VEB Institute.
In our opinion, in the new SP STD, taking into account the experience of NTI and road maps, it is necessary to single out specific scientific and technological areas, ensuring their priority funding [6]. A significant part of the scientific community agrees that these are artificial intelligence, microelectronic, photonic and quantum technologies, new materials and additive manufacturing, the Internet of things and 5/6G communications, medical technology and pharmacology, genetic and biotechnologies.9
Management dramas or barriers in the circulation of ideas and innovations. The development of science is the responsibility of the Ministry of Science and Education, while technological areas are the responsibility of line ministries, from the Ministry of Industry and Trade to the Ministry of Defense. The Academy of Sciences, although it did not turn into a club of scientists, lost the status of a scientific organization after the reform of 2013 and has rather an informal influence on the management of the scientific and technological process.
By Decree of the President of the Russian Federation No. 143 of March 15, 2021, On Measures to Improve the Efficiency of the State Science and Technology Policy, the functions of determining the strategic goals, objectives and priorities of the scientific and technological development of the Russian Federation are assigned to the Council under the President of the Russian Federation for Science and the permanent Commission on Scientific and Technological Development of the Russian Federation under the Government of the Russian Federation. However, the new functions and the creation of the commission did not change the nature of the management of the scientific and technological complex. The plurality and diversity of priorities at various levels indicate the absence of a coordinated system for determining the scientific and technological goals of these priorities, which then inevitably manifests itself both in their resource provision and in the concentration of managerial efforts in accordance with the designated goals. The expert potential of the Academy of Sciences is used to a small extent when considering key strategic decisions and large-scale scientific, technological and spatial projects in managing the development of science and technology.
The Ministry of Education and Science, currently mainly focused on higher education issues and the scientific agenda, especially applied science, is on the sidelines, while the importance of developing scientific and technological groundwork in the activities of sectoral departments is declining under the influence of a wave of current sectoral problems. In contrast to Soviet times, the coordination of scientific and technological developments in the civil and military spheres is also at a low level. For all the relativity of international university rankings, despite additional funding in the amount of 80 billion rubles, none of the universities participating in the 5/100 project has made it into the top 100 world rankings.
An analysis of the volume and structure of financing of the state program “Scientific and technological development of the Russian Federation” until 2024 also indicates that its main expenses go to the implementation of the subprogram “Ensuring the global competitiveness of Russian higher education”–ensuring current expenses for the implementation of educational programs and activities of higher education organizations within the framework of the Priority 2030 program and SECs. The share of the educational block in 2021–2023 accounts for 66.5% of the state program. Financing of directly scientific and scientific and technological activities is carried out according to the residual principle [7].
The process of creating scientific and educational centers (SECs) has not yet led to the formation of new scientific and educational consortiums capable of solving large-scale scientific problems. The educational agenda in them dominates over the scientific one. SECs do contribute to the involvement of scientists in the development of universities, but do not create conditions for the joint work of educational institutions with academic institutions as integral entities.
Prospects for the development of fundamental science until 2024 are largely determined by the implementation of the program of fundamental research and the participation of RAS institutes in the activities of the national project “Science and universities” through the activities of world-class scientific centers, centers for genomic and mathematical research, centers of the NTI competencies, programs for the creation of research facilities of the megascience class. The development of the system of grants from science foundations has given the support of basic science an important flexibility and individuality. At the same time, the loss of scientific status by the Academy of Sciences and the lack of coordination of its actions with the Russian Science Foundation, after its actual merger with the Russian Foundation for Basic Research, further increases disunity in the scientific community and its management system [8].
The attempts made to build an integral system from fundamental research through exploratory scientific and technological work to applied innovative developments, or an innovative lift, have not yet been successful. An example is the initiative of development, which is almost four years old, of complex scientific and technological programs of the full innovation cycle (CSTP), which provided for large-scale world-class scientific and technological projects, including in such important areas as new substances and materials, specialized robotics, baby food. For four years there has been discussion and adjustment, but not a single program has been launched. According to the idea, the state planned to provide support for exploratory R&D, and business was supposed to finance applied research and bring innovative products to the market, which ensured the unity of fundamental and applied science. So far, there is progress only in terms of business financing, in particular through Rosatom, but without state support. After the reformatting of the national project “Science” into “Science and Universities” in 2020, the development of CSTP actually became a nonpriority. In the approved SP STD, the planned financing of CSTP was reduced to two billion rubles per year (10 times less than the original passport of the national project “Science”), which does not allow us to consider even the already approved CSTP as powerful driver programs for the full scientific and technological cycle.
Currently, along with projects in industry state programs, a really significant state tool for the development of new scientific and technological areas in terms of applied and corporate science are the roadmaps of state companies launched in 2019 for the development of a number of high-tech areas, as well as “beacon projects” approved by the Government as part of strategic initiatives in 2021 as a continuation of the National Technology Initiative (NTI). It is assumed that the “beacon projects” of technology development should have a high multiplier effect for the development of the economy. In our opinion, beacon projects are more likely to be targeted innovative projects and do not have a large macroeconomic effect in the medium term, with the exception of the electric propulsion project. Moreover, they do not create a need for in-depth fundamental and exploratory research, nor do they entail any significant increase in country R&D expenses. These projects are, in essence, a continuation of the priorities adopted at the launch of the National Technology Initiative, a kind of NTI 2.0 [9].
To a large extent, there is no synergy of beacon projects with research and projects carried out within the framework of the roadmaps of state-owned companies and the activities of the SP STD, except for the areas of ICT and artificial intelligence technologies.
In the world practice of managing and supporting new areas of scientific and technological development by the state, the emphasis is on numerous research programs and partnerships created specifically to organize and support research in the field of new technologies, targeted funding is also provided through grant funds.
For example, in the field of quantum technologies at the EU level, the Quantum Flagship program (2018–2027) [10] is operating, and Germany is implementing the National Quantum Program (2021–2028). In the United States, in the field of new technologies in the electric power industry, there is the DOE Grid Modernization Initiative, a large program for the modernization of the US electric networks, which involves 17 national Laboratories operating under the auspices of the US Department of Energy [11].
These are quite long-term programs, while in Russia budget financing of roadmaps and programs is provided only until 2024. In particular, in the program “Digital economy of the Russian Federation”: for the development of quantum computing (13.3 billion rubles), artificial intelligence (24.6 billion rubles), 5G mobile communication networks (21.463 billion rubles), quantum communications (10.2 billion rubles). All together, these roadmaps cost about 1 billion dollars. State support abroad for similar areas is much larger: in the direction of quantum computing, in the United States—1.2 billion dollars, programs of the EU and individual European countries–a total of more than 5 billion euros, India—1.12 billion dollars; in the direction of artificial intelligence, China— 8 billion dollars, the United States — 6 billion dollars, programs of the EU and individual European countries—a total of more than 7 billion euros; in the direction of quantum communications, China—more than 15 billion dollars, Germany—more than 2 billion euros. At present, monitoring of the implementation of roadmaps by the state is entrusted to the Ministry of Economic Development of Russia.
The overall structure of science expenses in Russia is close to the similar structure in foreign countries. However, as is known, it is the sphere of applied research and development that is the most financially and capital intensive, while in Russian conditions it is here that the main deficit of investments and equipment is concentrated (Table 8) [12].
Table 8. Structure of internal current research and development expenses by type of work in 2020, %
Country Fundamental research Applied research Development
USA 16.4 19.0 64.5
China 6.0 11.3 82.7
Japan 13.0 19.4 67.6
Republic of Korea 14.7 22.5 62.8
France 22.7 41.4 36.0
Great Britain 18.3 42.1 39.7
Russia 18.8 20.0 61.2
Israel 10.0 10.1 79.9
Czech Republic 26.2 41.1 32.3
Source: OECD, Rosstat.
At present, the weakest link in the structure of the Russian scientific and technological complex is the link that ensures the transition from the stage of research and laboratory samples to pilot plants and small-scale production (TRL 4-7),10 refining and scaling new technologies.
The main potential of applied and engineering research in Russia is concentrated in the system of state research centers of the Russian Federation and in the field of corporate science, concentrated mainly in the largest state corporations and companies with state participation.
The system of the public sector of applied science is the most important component of the national innovation system and unites 44 scientific organizations with the status of state scientific centers, the activities of which are aimed at creating and developing technologies, promoting the results of search, applied research and experimental development, including their own production of high-tech goods.
Today, the system of state research centers of the Russian Federation in terms of its functionality and variety of work performed is comparable to the world’s largest associations that carry out applied problem-oriented research and development, such as the Fraunhofer Society (Germany) and the network of Carnot Institutes (France).
The fundamental difference between the state research centers of the Russian Federation and academic and university science is the predominance of expenses on applied research and experimental development. Despite the fact that all state research centers of the Russian Federation are only 1% of the country’s organizations performing research and development, they account for 20% of the country’s expenses for applied research as part of DRDE. At the same time, the share of state research centers of the Russian Federation in DRDE across the country in 2020 reached 7.7% (91.1 billion rubles). At the same time, the share of R&D expenses in the DRDE structure at the expense of nonbudgetary sources in the system of state research centers of the Russian Federation exceeds 50%.
At the present stage, the problem of further development of the system of state research centers of the Russian Federation, as well as of all applied science in the Russian Federation, is, among other things, insufficient legal support for such activities. It is advisable to update the normative and managerial categories “applied scientific research,” “exploratory scientific research,” “experimental developments,” “scientific and technological groundwork,” to consolidate the concept “innovative projects of a full life cycle.” Groundbreaking work does not fit into the procurement system according to Federal Law No. 44-FZ of April 5, 2013, On the Contract System in the Field of Procurement of Goods, Works, Services to Meet State and Municipal Needs, and Federal Law No. 223 of July 18, 2011-FZ, On the Procurement of Goods, Works, Services by Certain Types of Legal Entities, and they need a special management system and their own special regulatory framework [13].
Despite the significant potential, the state’s attention to the development of the system of state research centers of the Russian Federation is almost absent. It should also be noted that the functions of managing applied research are not provided for in the regulations of any federal executive body. As a result, today hardly anyone is directly responsible for supporting applied science in the country.
It is advisable to create a special section (target item of expenses): “Research and development carried out by state research centers of the Russian Federation,” which implies targeted budgetary financing of research and development carried out by the state research centers of the Russian Federation according to agreed development programs, including for organizations that have an organizational and legal form of commercial organization.
The status of state research centers should be clarified, both taking into account the American experience of “national laboratories” and the experience of the Kurchatov Institute and the Zhukovsky Center for Scientific Research, for which special legal acts were adopted.
It should be noted that more than 40% of the total national funding for the physical and technological sciences is concentrated in the system of US national laboratories, while federal funding accounts for up to 70% of all R&D expenditures, primarily for the operation of unique scientific facilities used by universities and industry. Many national laboratories operate in a government-owned, contractor-operated format, a model that enables breakthrough research in promising areas based on the use of large state-funded scientific facilities and equipment and the private initiative of the contractor. At the same time, network interactions formed for specific tasks allow national laboratories to solve interdisciplinary problems in a wide range of areas [14].
In general, the formation of intersectoral, interdisciplinary national research centers of applied science on the basis of the leading state research centers of the Russian Federation and the scientific research centers, following the example of the Kurchatov Institute and the Zhukovsky Center for Scientific Research, will allow planning and implementing complex scientific and technological projects and full-cycle programs that meet the challenges and priorities of the Strategy for Scientific and Technological Development of the Russian Federation.
The system of innovation development institutions that has developed in Russia is mainly focused on a variety of startup support mechanisms: the Innovation Promotion Fund, the Skolkovo Foundation, Rosnano, NTI, RVC, etc. The activities of development institutions, for all their importance for the development of innovations, are characterized by limited scientific, especially the fundamental, component. Domestic startups in the overwhelming majority do not develop, but use technologies of varying degrees of readiness for the commercialization of products based on them. The few exceptions are the most successful projects of Rosnano (for example, the company Oscial in the field of nanotubes). The bottleneck in the scientific and innovation cycle is the stage of pilot development and scaling, which in the current system of development institutions can be mainly handled by Rosnano only (to a lesser extent, by FRP and FPI). Currently, VEB.RF is in charge of coordinating development institutions, and a new model of interaction (“seamless integration”) and the so-called “innovation lift” still needs to be developed.
Applied science in Russia is concentrated mainly both in state research centers and in the largest state corporations and companies with state participation, which are obliged to implement innovative development programs (IDP) since 2011 [15] (Table 9). At present, the list of state-owned companies implementing IDP includes 57 state corporations, joint-stock companies and federal state unitary enterprises. In 2020, the total expenses of state-owned companies for the implementation of IDP amounted to about 1.4 trillion rubles, total R&D expenses 552 billion rubles. At the same time, state-owned companies’ R&D expenses reached 232 billion rubles, which is more than 60% of applied science expenses for the Russian Federation as a whole.
Table 9. R&D expenses of the largest state-owned companies by industry
Sectors of the economy R&D spending 2017 2018 2019 2020
Space sector pace 104.2 92.8 235.5
% of revenue 54.3 61.9 57.6 51.5
Aircraft industry pace 92.0 146.7 67.0
% of revenue 10.0 10.2 18.6 10.2
Shipbuilding, automated control systems and marine engineering pace 78.9 86.7 94.9
% of revenue 13.4 10.4 8.2 7.3
Chemistry and pharmaceuticals pace 86.5 67.0 94.9
% of revenue 22.7 18.3 28.8 23.2
Extraction and processing of raw materials pace 109.5 106.5 91.7
% of revenue 0.2 0.2 0.3 0.3
Energy pace 108.1 124.7 167.5
% of revenue 0.5 0.6 0.6 1.1
Transport and infrastructure pace 199.3 109.7 95.7
% of revenue 0.3 0.7 0.7 0.7
Communications and telecommunications pace 150.4 143.4 102.8
% of revenue 1.2 1.1 1.5 1.4
Source: Ministry of Economic Development of Russia, ANO VEB Institute.
The range of technologies developed by state-owned companies is very wide, and, as can be seen from Fig. 2, in a number of traditional areas in general, according to the companies themselves, it is not inferior to the level of development in the leading foreign peer companies. However, in terms of microelectronics technologies, space and energy technologies, there is a significant lag behind the world level.
Fig. 2. Comparative level of technological development of state-owned companies. The size of the “bubble” corresponds to the number of technologies analyzed.
At the same time, state-owned companies note that there is a shortage of breakthrough promising research that cannot be overcome by corporate research centers and state research centers, and therefore special approaches and mechanisms are needed to support the formation of scientific and technological groundwork and breakthrough risky developments. At present, business mainly plans its activities in the short and medium term and is not ready to set fundamental tasks for scientists that require serious exploratory research.
Despite calls for an outpacing increase in private R&D funding, state-owned companies have not increased R&D expenses in recent years, and somewhere have lowered its relative (to revenue) level.11 The state, through its representatives on the boards of directors, does not set them the task of increasing these expenses. In many ways, this is not just the result of inconsistency with the budget plans of corporations, but the lack of long-term sustainable priorities for technological and innovative development on the part of the state and companies.
At present, all FEED projects are financed in a general manner within the framework of investment programs of state-owned companies, and priority is usually given to low-risk projects with a high share of mastered imported technologies. In the new times that have come, the requirement to ensure technological sovereignty forces the creation of high-risk projects with a significant innovative and breakthrough component.
Attention to FEED in recent years has been pushed aside by the demand for the implementation of specific KPIs, mainly of a volumetric financial focus. It is advisable not to abolish, but to reformat the innovative development programs of state-owned companies (FEED 2.0) and their financing mechanisms, not excluding their transformation into subprograms of long-term programs for the development of state corporations. The need for stimulation, and for “forcing innovation” has not yet disappeared. The following necessary institutional innovations can be identified:
— Highlighting, as part of innovative development programs (DPR subprograms), of activities that are part of the roadmaps for the implementation of agreements between companies and the state on the development of advanced technologies; combining innovative programs with corporate digitalization programs and programs to reduce greenhouse gas emissions in order to avoid the multiplication of organizational structures within companies and the dispersion of efforts following every trendy agenda.
— Providing public-corporate innovators with the right to take risks when conducting research in the early stages and transition to managing a portfolio of innovative projects instead of waiting for the economic efficiency of each project.
— Initiating the creation of new ways for financing innovative projects at the stages of R&D and development through specialized corporate innovation support programs and corporate venture funds, or industry-specific R&D funds with a deduction of 1.5% of profits under the current legislation.
— Encouraging corporate science to start research in breakthrough technological areas (for a 10–15 year perspective), which require fundamental/exploratory work to be carried out jointly with external partners.
Thus, based on the results of the comparison of the level and trends in the development of the scientific and technological sphere in Russia with the leading foreign countries, carried out in this article, we can conclude that there is a potential in the field of fundamental science and high-tech big business that is sufficient to maintain technological parity. At the same time, for a technological breakthrough with the aim of Russia’s entry into the top five countries—world technological leaders—it is necessary to solve a set of problems related to the multiplicity of scientific and technological priorities, restrictions on the financing of science and the inconsistency of policy in relation to state programs of scientific and technological development, as well as insufficient support for the applied and corporate science sector.
In this regard, in the second part of the article, a set of measures will be proposed to accelerate scientific and technological development, including taking into account the need to level significant restrictions in which the domestic science sector found itself in conditions of technological blockade.
CONFLICT OF INTERESTS
The authors declare that they have no conflicts of interest.
1 Decree of the President of the Russian Federation No. 204 of May 7, 2018, On the National Goals and Strategic Objectives of the Development of the Russian Federation for the Period until 2024; Decree of the President of the Russian Federation No. 474 of July 21, 2020, On the National Development Goals of the Russian Federation for the Period until 2030.
2 Concept of Long-Term Socio-Economic Development of the Russian Federation for the period until 2020, approved by the Order of the Government of the Russian Federation No. 1662-r of November 17, 2008.
3 According to the materials of Institute for Statistical Studies and Economics of Knowledge, Higher School of Economics. https://issek.hse.ru/news/482453668.html. Cited June 5, 2022.
4 Source: Federal Office of Statistics of Germany, Bureau of Statistics of the Czech Republic.
5 A. M. Sergeev, How can we do science under sanctions? https://rg.ru/2022/05/31/1-iiunia-sostoiatsia-vybory-novyh-chlenov-rossijskoj-akademii-nauk.html. Cited June 5, 2022.
6 The assessment is based on a modified methodology of the Russian Ministry of Education and Science for calculating the indicator “Place of the Russian Federation in terms of research and development, including through the creation of an effective system of higher education.” PRF = (PRFOESD in terms of the number of researchers in full-time equivalent among the world’s leading countries × 0.3) + (PRFDRDE in terms of research and development expenses × 0.3) + (PRF A in terms of share in the total number of articles indexed in international databases data × 0.15) + (PRF P in terms of share in the total number of patent applications × 0.2) + PRF TOP500 in terms of the presence of top 500 universities in the QS ranking × 0.05). The indicator is expressed in whole units, and the weight coefficients are based on an assessment of the impact of the components of the statistical indicator on the development of the scientific and technological complex of the Russian Federation. The data are divided into main areas: resource potential (based on the indicators of staffing and the state’s financial expenses on research and development), the effectiveness of scientific activity (based on indicators of publication and patent activity), as well as the effectiveness of the higher education system based on the position of Russian universities in the international QS ranking.
7 Program of fundamental scientific research in the Russian Federation for the long term (2021–2030), approved by the Order of the Government of the Russian Federation No. 3684-r of December 31, 2020.
8 State Program “Scientific and technological development of the Russian Federation,” approved by the Decree of the Government of the Russian Federation of March 29, 2019 No. 377 (as amended by the Decree of the Government of the Russian Federation No. 1814of October 22, 2021).
9 A. M. Sergeev, How can we do science under sanctions? https://rg.ru/2022/05/31/1-iiunia-sostoiatsia-vybory-novyh-chlenov-rossijskoj-akademii-nauk.html. Cited June 5, 2022; A. N. Klepach, Socio-technological challenges of the Russian economy, Moscow Economic Forum MAEF. https://cyberleninka.ru/article/n/sotsialnyei-tehnologicheskie-vyzovy-rossiyskoy-ekonomiki/viewer. Cited June 5, 2022.
10 TRL. There are nine levels of technology readiness. From the first to the sixth levels, this is the development of technologies, which is carried out within the framework of research and development. From the seventh level and above, development engineering begins, or a demonstration of the operability of technologies on real devices under development. TRL 1—approval and publication of the basic principles of technology, TRL 2—formulation of the concept of technology and evaluation of the scope, TRL 3—beginning of research and development. Validation of characteristics, TRL 4—verification of the main technological components in the laboratory, TRL 5—verification of the main technological components in real conditions, TRL 6— testing a model or prototype in real conditions, TRL 7—demonstration of a prototype (trial model) in operation, TRL 8—completion of development and testing of the system in operation, TRL 9—demonstration of the technology in its final form during flight tests of the sample
11 FEED is a comprehensive tool for the development of innovations in companies, their structure includes activities in the following areas: development and implementation of innovative projects, improvement of innovation management mechanisms in companies, including in the field of intellectual property, development of an ecosystem of “open innovations” through interaction with small and medium-sized companies, organizations of science, higher education and objects of innovation infrastructure (innovation clusters and technology platforms), development of mechanisms for financing and investment in the innovation sector (including venture funds).
Translated by S. Avodkova
==== Refs
REFERENCES
1 Post-Crisis Recovery of the Economy and the Main Directions of the Forecast of the Socio-Economic Development of Russia for the Period up to 2035: Scientific Report, Ed. by A.A. Shirov (Nauka, Miscow, 2020) [in Russian]. 10.47711/sr1-2020
2 I. E. Frolov, “Russian high-technology complex under low inflation and government support limitation: The condition, capacity and tendencies for development,” Stud. Russ. Econ. Dev. 30 (4), 365–375 (2019).10.1134/S1075700719040051
3 Yu. A. Sytnyak and I. V. Tsvetkova, “Analysis of the activity of state scientific centers of the Russian Federation,” Izv. Inst. Inzhener. Fiz., No. 3 (33), 85–88 (2014).
4 E. V. Sibirskaya and L. V. Oveshnikova, “NTI as a strategic direction of Russia’s technological development,” Statist. Ekon., No. 1, 34–41 (2018). 10.21686/2500-3925-2018-1-34-41
5 Ganichev N. A. Koshovets O. B. Forcing the digital economy: How will the structure of digital markets change as a result of the COVID-19 pandemic? Stud. Russ. Econ. Dev. 2021 32 11 22 10.1134/S1075700721010056 33642845
6 Klepach A. N. “Social and technological challenges of the Russian economy,” Nauch. Tr. Vol’n. Ekon O-va Rossii 2021 230 103 112 10.38197/2072-2060-2021-230-4-103-112
7 Report on the Achieved Results of the 1st Stage and Plans for Participation in the Implementation of the 2nd Stage of the Strategy for Scientific and Technological Development of the Russian Federation, in the Framework of the Activities of State Scientific Centers of the Russian Federation, Ed. by E. N. Kablov (Assots. “Nauka,” Moscow, 2021) [in Russian].
8 “On the Development of Competition in the Field of Science,” in Materials of the Meeting of the NCC on the Legal, Psychological and Socio-Economic Problems of Society of the Department of Social Sciences of the Russian Academy of Sciences; Association of Russian Banks; National Research Institute for Trust, Dignity and Law, 2021, Ed. by G. A. Tosunyan (OOO “Novye Pechatnye Tekhnol., Moscow, 2022) [in Russian].
9 Klepach A. N. “Social and innovative turn of the Russian economy: plans and reality,” Nauch. Tr. Vol’n. Ekon O-va Rossii 2021 227 30 91 10.38197/2072-2060-2021-227-1-30-91
10 M. Riedel, M. Kovac, P. Zoller, J. Mlynek, T. Calarco, “Europe’s Quantum Flagship initiative,” Quantum Sci. Technol., No. 4 (2), 020501 (2019). 10.1088/2058-9565/ab042d
11 Ellis A. “DOE Grid Modernization Initiative and Sandia R&D,” Sandia National Lab. (SNL-NM) 2019 Albuquerque, NM (United States) No. SAN
12 Shepelev G. V. On the management of Russian science Upravl. Naukoi: Teor. Prakt. 2020 2 65 92 10.19181/smtp.2020.2.2.3
13 Klepach A. N. “Scientific and technological complex of Russia: Problems and prospects,” Nauch. Tr. Vol’n. Ekon O-va Rossii 2021 232 117 132 10.38197/2072-2060-2021-232-6-117-132
14 D. Yu. Faykov and D. Yu. Baidarov, “Features of the organization of production of civil products in the US national laboratories,” Ross. Vneshneekon. Vestn., No. 8, 40–62 (2020).
15 Gershman M. A. Programs of innovative development of companies with state participation: first results Forsait 2013 7 28 43
| 36466731 | PMC9707188 | NO-CC CODE | 2022-12-01 23:20:21 | no | Stud Russ Econ Dev. 2022 Nov 28; 33(6):631-644 | utf-8 | Stud Russ Econ Dev | 2,022 | 10.1134/S1075700722060077 | oa_other |
==== Front
Arch Womens Ment Health
Arch Womens Ment Health
Archives of Women's Mental Health
1434-1816
1435-1102
Springer Vienna Vienna
36443483
1277
10.1007/s00737-022-01277-x
Original Article
Pregnancy during the COVID-19 pandemic: a qualitative examination of ways of coping
Reynolds Kristin A. [email protected]
12
Pankratz Lily 1
Cameron Emily E. 1
Roos Leslie E. 1
Giesbrecht Gerald F. 3
Lebel Catherine 4
Tomfohr-Madsen Lianne M. 5
1 grid.21613.37 0000 0004 1936 9609 Department of Psychology, University of Manitoba, Winnipeg, Manitoba Canada
2 grid.21613.37 0000 0004 1936 9609 Department of Psychiatry, University of Manitoba, Winnipeg, Manitoba Canada
3 grid.22072.35 0000 0004 1936 7697 Departments of Pediatrics and Community Health Sciences, University of Calgary, Calgary, Alberta Canada
4 grid.22072.35 0000 0004 1936 7697 Department of Radiology, University of Calgary, Calgary, Alberta Canada
5 grid.22072.35 0000 0004 1936 7697 Department of Psychology, University of Calgary, Calgary, Alberta Canada
29 11 2022
112
7 4 2022
7 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The COVID-19 pandemic and related public health restrictions have impacted the mental health and coping strategies of many population groups, including people who are pregnant. Our study sought to explore the ways that pregnant people described coping with stressors associated with the pandemic. N = 5879 pregnant individuals completed the pan-Canadian Pregnancy During the COVID-19 Pandemic Survey between April and December 2020. We used descriptive statistics to quantify sociodemographic characteristics and thematic analysis (Braun & Clarke, 2006, 2019) to analyze n = 3316 open-ended text responses to the question “Can you tell us what things you are doing to cope with the COVID-19 pandemic?” The average age of participants was 32 years (SD = 4.4), with the majority identifying as White (83.6%), female (99.7%), married (61.5%), having completed post-secondary education (90.0%), and working full-time (75.4%). We categorized participant responses into two overarching thematic dimensions: (1) ways of coping and (2) coping challenges. Ways of coping included the following main themes: (1) taking care of oneself, (2) connecting socially, (3) engaging in pandemic-specific coping strategies, (4) keeping busy, (5) taking care of others, (6) creating a sense of normalcy, (7) changing perspectives, and (8) practicing spirituality. Coping challenges included the following: (1) the perception of coping poorly, (2) loss of coping methods, (3) managing frontline or essential work, and (4) worries about the future. Findings highlight important implications for targeted prenatal supports delivered remotely, including opportunities for social support, prenatal care, and mental health strategies.
Keyword
Prenatal
COVID-19
Coping
Qualitative
Research ManitobaThe Alberta Children's Hospital Research Institute and the Owerko CentreSocial Sciences and Humanities Research Council Postdoctoral Fellowship
==== Body
pmcPrenatal mental health
Pregnancy can be a joyous time in life, while also bringing hormonal, physical, emotional, and lifestyle challenges. Emotional distress, including anxiety and depression, is common in pregnancy and is elevated in pregnant people compared to non-pregnant people of similar ages (Dadi et al. 2020; Dennis et al. 2017; Fairbrother et al. 2016; Fawcett et al. 2020; Leach et al. 2017; Loughnan et al. 2018; Underwood et al. 2016; Zender and Olshansky 2009). Substantially elevated rates of perinatal distress are evident during the COVID-19 pandemic, with estimates of depressive and anxiety symptoms ranging from 40 to 57% and 60 to 81%, respectively (Khoury et al. 2021; Fan et al. 2021; Hessami et al. 2020; Yan et al. 2020). These pandemic-related prenatal symptom exacerbations are concerning for several reasons, including the detrimental impact of untreated mental health symptoms on pregnancy, postpartum, and fetal and infant development (Sun et al. 2021; Tomfohr-Madsen et al. 2021).
Impact of COVID-19 on prenatal mental health
Though integral to slowing the spread of the virus, pandemic restrictions including physical distancing come with a significant cost for mental health and well-being, both in the short and long term (Brooks et al. 2020). This is problematic for many pregnant people who have experienced changes to important activities such as prenatal support classes and groups, in-person physical and emotional supports from friends and family, access to health-related resources, and medical visitation (Burgess et al. 2021). Additionally, women are disproportionately impacted by restrictions such as school and daycare closures, loss of employment, increased demand for unpaid labor, and increased domestic violence (Boserup et al. 2020; Scott 2020; van der Linden 2020).
Prenatal mental health and coping strategies
With increased stressors and demands being placed on prenatal people throughout the pandemic, in the absence of typical supports, it is important to consider the ways in which they have been coping. Coping refers to “constantly changing cognitive and behavioral efforts to manage specific external and/or internal demands that are appraised as taxing or exceeding the resources of the person” (Lazarus and Folkman 1984, p. 141). Seeking and receiving social support to assist with coping is well-known to mitigate the effects of distress experienced during the perinatal period (Ibrahim et al. 2019; Khatri et al. 2019; Rowther et al. 2020). Social support network size, satisfaction with close relationships, and degree of social involvement can all act as buffers against stress and its negative impacts (Goletzke et al. 2017; Taubman-Ben-Ari 2019) and are identified as a key resource to reduce stress during pregnancy (Bloom et al. 2012; Rowther et al. 2020). An important facet of social support is the degree to which one’s preferred type of social support is accessible at the time it is needed most (Leahy-Warren et al. 2018). Seeking social support has been made more complicated due to COVID-19-related restrictions (Cigãran et al. 2021).
A limited number of studies have focused on the ways that pregnant people have been coping with the unprecedented changes associated with the COVID-19 pandemic. One recent study (Khoury et al. 2021) revealed that pregnant people who reported a higher negative impact of COVID-19 were more likely to use negative coping methods, such as denial, self-distraction, and substance use, which in turn was associated with increased risk of clinically significant levels of depressive and anxiety symptoms, and self-reported rates of stress, worry, and sleep difficulties. A subsequent qualitative study (Aydin and Aktaş 2021) reported that pregnant people used various ways of coping, such as focusing on the baby and finding indoor activities to participate in, to overcome their pandemic-related stress.
Summary and objectives
There remains a dearth of information highlighting the ways that pregnant people have coped with stressors associated with the COVID-19 pandemic and challenges that they have experienced in their coping. Considering the dramatic increase in estimates of anxiety and depression seen in pregnant people during the COVID-19 pandemic, paired with limitations in prenatal supports and resources, more research is needed to better understand the ways of coping applied by pregnant people during the pandemic. Using qualitative methods, our group sought to understand the unique perspectives of a large sample of Canadian pregnant people, concerning their expressed ways of coping with the COVID-19 pandemic.
Method
Participants and recruitment
We used data from the Pregnancy During the COVID-19 Pandemic Survey (Giesbrecht et al. 2021). Initial data collection began on April 5, 2020, with recruitment strategies including social media platforms (Facebook, Instagram) using advertisements that directed potentially eligible participants to the study website (www.pregnancyduringthepandemic.com). Participants were eligible to participate in the study if they were ≥ 17 years of age, ≤ 35 weeks gestation at enrollment, residing in Canada, and able to read and write in English or French. All participants provided informed consent prior to data collection. The current study utilized data from the baseline survey collected before July 8, 2020. The study received research ethics approval through the University of Calgary Conjoint Health Research Ethics Board (REB20-0500), and approval for use of secondary data was obtained from the University of Manitoba Research Ethics Board. We examined the following open-ended text-response item (English responses): “Can you tell us what things you are doing to cope with the COVID-19 pandemic?”.
Measures
Sociodemographic characteristics
Information was collected at baseline regarding sociodemographic variables (e.g., age, income, education).
Depressive symptoms
The Edinburgh Postnatal Depression Scale (EPDS; Cox et al. 1987) was used to measure the frequency and severity of depressive symptoms in the last week. The 10-item self-report scale is summed, with a possible range of 0–30; higher scores indicate greater severity of symptoms (Cox et al. 1987). Clinical cut-off scores of ≥ 13 are associated with major depressive disorder (Cox et al. 1987). Internal consistency was good in the current study (α = 0.88).
Anxiety symptoms
The Patient-Reported Outcomes Measurement Information System (PROMIS) Anxiety Adult 7-item short form was used to assess general anxiety symptoms (Pilkonis et al. 2011). Raw scores were converted to T-scores using US general population norms as per standard procedure; the possible score range was from 36.3 to 82.7 (M = 50.0, SD = 10.0). T-scores between 60.0 and 69.9 represent moderately elevated anxiety while scores ≥ 70.0 indicate severely elevated anxiety (Cella et al. 2010). Internal consistency was good in the current study (α = 0.88).
Analytic approach
We used descriptive statistics to quantify sociodemographic and mental health-related characteristics. We analyzed open-ended text responses following thematic analysis (Braun and Clarke 2006, 2019) using NVivo Qualitative Research Software to assist with data organization and quantification of representativeness or frequencies of main and sub-themes. Thematic analysis involves six interconnected stages, which the analysts move back and forth between data familiarization, generating initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the report (Braun and Clarke, 2006). The rigor of our approach was assured by a clear documentation of the coding process and thematic framework development (audit trail), including two qualitative analysts (KR and LP) coding the data and arriving at the thematic framework based on independent initial coding (negotiated validity) and a review of hierarchy charts and maps in NVivo to assess thoroughness and representativeness of themes and sub-themes in the data.
Results
Sample characteristics
Pregnant participants (n = 3316) provided responses to the open-ended text-based question exploring ways of coping with the COVID-19-pandemic and related stressors, leading to a total of 7073 responses within all our themes. A single participant’s response often fits into multiple themes; therefore, all participant numbers (denoted by n) described below reflect the number of total responses rather than the number of participants. Participants were, on average, 32.5 years old (range: 17.9–49.0 years old) and identified as White (83.1%) and female (99.7%). On average, participants had one child (range: 0–6) and were married (76.4%) or cohabitating (20.6%). Most participants had a bachelor’s degree (42.0%), completed trade school or community college (21.3%), or had a master’s degree (20.1%). Most participants were working full-time (75.4%). Nearly 30% of our sample had clinically elevated levels of depressive symptoms (27.8%; EPDS scores ≥ 13), and 51.4% of our sample had clinically elevated rates of anxiety symptoms (PROMIS scores ≥ 60).
Thematic analysis
Thematic analysis revealed two overarching thematic categories: (1) ways of coping and (2) coping challenges, each of which include separate main and sub-themes (see Tables 1 and 2 for a presentation of quotes associated with themes and sub-themes; see Figs. 1 and 2 for frequency-related diagramming of themes and sub-themes). The main and sub-themes within the ways of coping theme include the following: (1) taking care of oneself, (2) social connection (sub-themes: family time, virtual, telephone, and distanced visits), (3) pandemic-specific coping strategies (sub-themes: limiting social contact, managing news, mask wearing, and physical distancing), (4) keeping busy, (5) taking care of others, (6) creating a sense of normalcy, (7) changing perspectives, and (8) practicing spirituality. The main themes within the thematic category of coping challenges include the following: (1) the perception of coping poorly or “not coping,” (2) loss of coping methods, (3) managing pregnancy and frontline or essential work, and (4) worries about the future.Table 1 Qualitative responses—ways of coping
Theme Direct quote
Taking care of oneself I have worked very hard to ensure my athleticism and fitness goals have remained on track. Fitness and rigorous exercise are my work and life therapy. So far, while some days the energy is a battle, I have been able stick to my regular exercise regimen during this pregnancy
I reached out to the mental health supports available to physicians. I had a couple telephone visits with a psychiatrist, and I've been doing regular video visits with a psychologist. I started exercising. I've been learning and practicing mindfulness. I attended a mental health webinar for physicians that was very helpful. I've been very proactive
I’ve been taking things easy. Resting when I feel overwhelmed and not pushing myself too hard mentally or physically
I'm focusing on myself more. Listening to what my body needs, trying to focus on learning about my baby. It's been a time for personal growth
Connecting socially Ensuring that I keep in touch with friends and family via social media, virtual meetings, and socially distanced visits (e.g., drive-by greetings). Keeping social and maintaining relationships is very important to my mental health
Family time Enjoying the extra time with my dog and my husband, enjoying not having tons of commitments that usually make life busy
Focusing on my family, enjoying time spent with my daughter
My 14-month-old really helps me cope with everything that is going on. He brings so much joy and makes me laugh so much! I also have a very loving and supportive husband which makes life easier
Virtually At the beginning of the pandemic it was important to keep my social connections via FaceTime and zoom to help me cope
Face-Timing my Mom almost daily, keeping in touch with friends in a group Facebook chat
Telephone Long phone dates with friends. I don’t normally talk on the phone
Talking to people over the phone more often
Staying in touch with family and friends via phone calls
Forging new friendships through phone calls
Physically distanced visits My parents passing by our home (but remaining outside and keeping 2 m minimum distance)
I started having socially distanced hangouts when I was not coping well
Engaging in pandemic-specific coping strategies
Limiting social contact We are staying away from people. Friends. Crowded public places
We are not seeing family or friends, keeping ourselves distanced from them □.
I am making sure we stay at home and only go out for essential items. When I do go into a store, I make sure to hand sanitize as much as possible and keep distance from others. The hardest part of this is not being able to see people or visit in person. I used to love shopping but would trade that in for a girl’s night any day
Managing news intake I've cut back on my news consumption as that was triggering more stress and worry
I am trying to change my social media platforms, so I am more exposed to quality information from verifiable and trusted sources. I'm trying to limit my exposure to sensationalism and/or unqualified opinions on social media
I was reading the news a lot in the beginning and looking at death toll numbers. I try to only look at the news once a day now
I am 'Unplugging' from the news every now and again when it becomes too much
At first, the pandemic was very stressful and overwhelming. I did a lot of research to understand some of the details of what was going on. Knowing more information helps me see the big picture and helps to not feel out of control
I find myself googling a lot about COVID and pregnancy, as not much is known about its effects in the second trimester, especially on baby long term
Mask wearing I ensure that I wear mask in public places
Physical distancing Social distancing to promote health
Keeping physical distance when I am at work
Physically distancing to help reduce the spread of COVID-19
Keeping busy Distracting myself with art projects
Keeping busy, starting a side hustle
Playing phone games to disconnect from the current situation
Escaping by watching tv and too much time on social media
Spending lots of time distracting myself with social media and TV shows
Cleaning and changing my house arrangement just to keep myself busy and keep me from thinking
Taking care of others Feeling my baby’s movements inside of me make me extremely happy and hopeful
I’m focusing on my children and their happiness during the day
I'm trying to create a safe and peaceful environment in our household for my family despite feeling anxious sometimes
Creating a sense of normalcy Ensuring I keep a routine while working from home. This means changing out of my pajamas, working in my living room (not my bed) and eating proper meals. I've been doing video calls with friends weekly and seeing my Mom (while socially distancing) for walks outside a few times a week
I am trying to keep our lives as close to normal as possible. Making goals, working on projects, spending quality time with close family and friends as much as possible
Changing perspectives I am putting things into perspective. It could be much worse, others are really struggling right now or putting themselves on the front line
I am changing my perspective to help cope. For example, not being able to take my partner in the ultrasound to see if I had lost the pregnancy. I thought, imagine the people who have had COVID who cannot see their family members
I’m practicing gratitude—lots to be thankful for during this time. Working from home due to class suspension for students has allowed me to slow down and not take on too much. This has also allowed me to delve into work-related learning material I wouldn't have had time to previously—I enjoy learning
I’m taking things one day at a time. Just focusing on what I need to do each day of the week with my daughter and partner
Just taking it day by day and trying to look forward to brighter days ahead
Staying calm, keeping things very simple, basic, natural. Taking the time to get back to basic needs, reset habits and routine
I take one day at a time, accept the uncertainty as the new normal, try and go with the flow more, find ways to reduce stress, find joy in the things we can do
I try to not worry about things I have no control over. Having fear of the unknown is detrimental so I choose to live life like always with adaptations to keep me and my family safe
Practicing Spirituality Prayerfully, we pray that everything is under God's control, and our family and our friends will be able to cope well, and stay healthy, stay strong throughout this Covid-19 pandemic. And we pray that this Covid-19 pandemic will be finished
I am just praying a lot and talking to God to protect my baby
Table 2 Qualitative responses—coping challenges
Theme Direct quote
The perception of not coping poorly or “not coping” Not coping
No time to for anything else. Not exactly coping
Not coping well
I don't feel that I am coping well. Mostly, I just feel that I'm waiting for things to return to normal. It's difficult to accept that it will probably be a very long time before any real return to normalcy is possible. By the time that return does happen, I will be taking care of a small child, so my life will also be irreversibly changed by that
I’m trying to cope with sadness and loneliness throughout this pregnancy by keeping it to myself almost entirely. Have not even announced at 24 weeks. That way, I cannot feel depressed about no one offering support or reaching out. COVID-19 has stolen much of the joy of pregnancy for me
Only leaving my house to go for a walk around the block with my family or play in our backyard has taken a huge toll on my mental health. I often have days where I feel extreme anger coursing through my body, and I have no explanation why. I don't know if I am going crazy from not being able to get away from my home to even do something as trivial as grocery shopping or if it has to do with pregnancy hormones. I hide my feelings well from my partner and children because I don't want them to think they have done anything wrong
Loss of coping methods I have no good coping strategies. I feel disconnected from friends, a loss of independence, and nausea and tiredness from pregnancy have meant I'm not working out at all whereas I used to do so 5 days a week. No exercise, no alcohol, no social life, and no sex plus more work stress means I'm coping poorly
I have placenta previa so not being able to exercise or have intercourse have taken a toll on my emotional state, as I've lost my two biggest outlets
I’m currently on pelvic rest and modified bed rest and have been feeling quite isolated not being able to leave the house or go for walks as normal
I have very little coping mechanisms I can use during the pandemic. Close support people are not front line and do not understand the fear and frustration of feeling exposed and have poor advice. Cannot do usual activities with friend and family as normal or to cope. Cannot plan or prepare which is a coping mechanism for change
It's been impossible to cope with the expectation that I work 7 h/day while looking after my 2-year-old by myself. My husband still has to go in to work, and since we don't want to risk our parents' health, I have had no help with childcare. Due to my pregnancy, I am not able to work while my toddler naps because I am so exhausted from not sleeping and looking after. Then I have to stay up late to get work done. Also, I have had to turn my toddler into a TV zombie, which is something I never wanted to do. So that has caused me a lot of stress and guilt
Managing pregnancy and frontline or essential work I am not coping well during the pandemic. My work life has been completely flipped around. I went from working in the operating room Monday to Friday weekends off, to being thrown into floor nursing working nights and 10 days in a row and every second weekend. My work life in the pandemic has caused me a great amount of stress and anxiety and being pregnant I am concerned by the work environment and schedule I have been thrown into. All the work stress I am experiencing is disturbing my relationship and effecting my abilities to enjoy my early stages of pregnancy. It has forced my attention away from the joy of being pregnant and has put all my attention on my negative work environment
Working full time on the front line. Overwhelmed with stress and exhaustion. Very understaffed and over worked. Long hours. A lot of fear of the unknown of how this virus is going to affect baby and I when I'm hands on with positive COVID patients. Gives me a lot of anxiety and stress and a lot of tears. Feel very unsupported by my employer
The major thing I've done is stop working. As a registered midwife I am a frontline healthcare provider, and though the risk to pregnant people from COVID appears to be low, I could not isolate, I worried how I would manage if I got infected and put my family at risk. The stress of the early days of the pandemic in terms of uncertainty, lack of PPE, and the physical and emotional stress of midwifery- now all with my toddler at home instead of being in daycare, meant I was not doing either job (being a midwife or a mom) very well. I started my leave at 28 weeks, 8 weeks earlier than I had originally planned. This will be a financial burden for us, but I think it was the right choice. Doing it all was too hard
Worries about the future I have never done a home birth before and the closer I get to my due date has me feeling rather nervous. If COVID-19 was not an issue I would never have chosen to give birth at home. But having all my family with me supporting me is what I want and need
This is my first baby, and I pictured my pregnancy differently. Also, this is not asked but many women on Facebook including me have expressed anxiety over fear of having to give birth alone or without partner. If they have a cold, they will be considered symptomatic and not allowed in to the most stressful and scary thing I will ever do (first time mom). It leaves me so frightened and stressed anytime it crosses my mind, and this is not good for a growing baby
I am nervous for when things open back up – like going back to teaching at school in the fall. I am not currently struggling to cope, as I can remain in my own safe bubble. I am more worried for the future then what is happening currently
I’m trying really hard to not worry about money as we are quickly going through our savings
I'm very concerned about being able to get some of the things I need to have a baby (car seat, stroller, bassinet) and haven't been able to get pregnancy pillows I need for sleeping because so much is out of stock. I'm also concerned my mom will never see me pregnant and may not get to meet the baby—because of COVID
Fig. 1 Ways of coping
Fig. 2 Coping challenges
Ways of Coping
Taking care of oneself
The most frequently reported main theme (n = 2390) within the ways of coping category was taking care of oneself. Participants described many important strategies employed to focus on their health and well-being, including accessing supports, spending time outdoors, engaging in self-care, exercising, returning to previously enjoyed hobbies, and practicing self-compassion.
Connecting socially
Social connection was another commonly reported way of coping (n = 1483), with participants noting connecting through family time (n = 699), virtually (n = 568), by telephone (n = 153), and through physically distanced visits (n = 63).
Connecting through family time
Participants described that public health restrictions limiting the number of close contacts, paired with activity cancelations and closures, allowed for more time and slower time spent with immediate family.
Connecting virtually
Participants described the vital importance of virtual connection, particularly early on during the pandemic and during times when public health restrictions were more stringent.
Connecting by telephone
Many participants described talking on the telephone with family and friends more often during the pandemic, in comparison to pre-pandemic connections.
Connecting through physically distanced visits
With the rise of COVID-19 cases and public health restrictions during the time data was collected, participants navigated when and how to incorporate physically distanced visits into their social connections. Within the main theme of connecting socially, connecting through physically distanced visits included the smallest number of participant responses (n = 63).
Engaging in pandemic-specific coping strategies
Participants reported ways of coping specific to the COVID-19 pandemic, including limiting social contact (n = 319), managing news intake (n = 311), mask wearing (n = 81), and physical distancing (n = 49) when in public.
Limiting social contact
Although some participants noted their engagement in socially distanced visits with friends and family, many participants reported restricting in-person contact, particularly during times when COVID-19 rates were elevated.
Managing news intake
Participants described limiting news intake as a way of coping with COVID-19 related stress and anxiety, including news viewed through television, radio, the Internet, and social media. Participants noted an ebb and flow to their media consumption throughout the pandemic, balancing staying informed with pandemic-related information with over-consumption of news. Participants also described ways in which managing news intake was helpful to them. Specific to pregnancy, participants noted increased use of the Internet, including scanning online news sources for information about the impact of COVID-19 on pregnancy.
Mask wearing
Participants reported use of non-medical face masks to cope with stress associated with the pandemic.
Physical distancing
Physical distancing was the final pandemic-specific way of coping reported by participants.
Keeping busy
Participants described several ways that they “kept busy” through participation in distracting activities to cope with stress associated with the COVID-19 pandemic (n = 612). Any activities mentioned with words of distracted or avoidance were coded within this sub-theme, including household projects, immersion in work, watching television, and scanning social media.
Taking care of others
Participants also described taking care of their growing fetus, as well as taking care of their immediate families (n = 560). Participants described coping with pandemic-related stress by focusing on engaging in health-promoting behavior (for themself and baby), learning about infant rearing and development, and preparing one’s home and finances for baby’s arrival.
Creating a sense of normalcy
Participants also described the ways in which they tried to maintain pre-pandemic experiences, schedules, and routines with the hope of finding ways to create a “normal” life (n = 307).
Changing perspectives
Participants described engaging in several cognitive strategies to cope with pandemic-related stress, including social or global comparisons, practicing gratitude, focusing on the present moment, and accepting the things we cannot change or control (n = 413). Taking things 1 day at a time was a frequently noted statement within this sub-theme of changing perspectives.
Practicing spirituality
In our sample, n = 65 participants noted coping with stress by engaging in spiritual practices such as prayer, meditation, and trusting in their prescribed faith.
Coping challenges
Many participant responses (n = 483) included challenges they were facing. We categorized these reported challenges into four main themes: the perception of coping poorly or “not coping,” loss of coping methods, managing pregnancy and frontline or essential work, and worries about the future.
The perception of coping poorly or “not coping”
Participants (n = 214) described their struggle to cope with stressors experienced during the COVID-19 pandemic, noting that they were coping poorly or “not coping.” They identified social isolation and loneliness, challenges regulating emotions, and reduced practice of positive health habits as contributing to difficulties coping with pandemic-related stress.
Loss of coping methods
Some participants (n = 157) described that they were not able to engage in the ways that they used to cope with stressful situations before the COVID-19 pandemic. Participants also described a compounding impact of stress in terms of managing both pregnancy-related challenges in addition to the pandemic. The loss of emotional and instrumental support from friends, family members, and childcare centers impacted the ways in which participants were able to cope with pandemic-related stress.
Managing pregnancy and frontline or essential work
Several participant responses (n = 76) indicated their roles in frontline or essential work and the specific challenges they faced in their coping.
Worries about the future
Participants (n = 36) described anticipatory worry of an uncertain future. Upcoming labor and delivery, return to work, financial instability, and availability of supplies were some of the worries noted by participants.
Discussion
Our findings demonstrated that pregnant people reported a variety of ways in which they were coping during the pandemic. Participants reported coping by taking care of themselves, others close to them, or their growing fetus; finding social connection; keeping busy; practicing spirituality; changing perspective; focusing on the present moment; and employing pandemic specific strategies such as managing news intake and following public health orders. We also saw a small but notable portion of participants sharing challenges associated with coping. Participants outlined challenges such as loss of regular coping strategies due to the pandemic and related restrictions, being a frontline worker and managing specific challenges associated with this line of work, and worrying about what the future would hold in terms of labor and delivery and postpartum supports.
A promising finding that emerged from our data was that the most frequently coded theme was “taking care of oneself” as a way of coping. This included a range of activities such as exercise, accessing supports, and creating boundaries for oneself. Despite the negative impact we have seen COVID-19 have on pregnant people, many participants were trying to find ways to prioritize their self-care. Finding social connection was a frequently coded source of coping for our participants. Participants found unique ways to stay connected whether that be making more phone calls or driving by loved ones’ homes for a short outside visit, even while Canadian restrictions at the time forced families to remain within their household or in small bubbles (Assaly 2021). Encouraging pregnant people to take advantage of less traditional social connection approaches may be valuable even after the pandemic, given the importance of social support to mental health during the perinatal period and beyond (Milgrom et al. 2019). The loss of coping methods reported by many participants in this research (including loss of support from family and friends, childcare, and prenatal services) aligns with the exacerbated impact that COVID-19 has had on women and mothers (Aydin and Atkaş 2021; Khoury et al. 2021). Our findings did not highlight substance use as a prominent way of coping, with only a few participants using substances (e.g., alcohol, cannabis) to cope. These findings align with prior work in an overlapping sample who found rates of substance use among pregnant individuals were similar or lower rates of use compared to pre-pandemic rates (Kar et al. 2021).
Using a large Canadian sample of pregnant people, our findings extend prior research exploring ways of coping during the COVID-19 pandemic. Previous research has used quantitative methods (Khoury et al. 2021) or small sample sizes and limited discussion of coping (Aydin and Aktaş 2021). Furthermore, our findings are in line with research prior to COVID-19 which demonstrated common coping themes within this population were positive coping strategies, spirituality, and social support (Ibrahim et al. 2019; Shamsaei et al. 2019). Findings highlight the particular need for the provision of accessible social support provided through remote means (Mu et al. 2021).
Findings of this research should be considered within the context of its limitations. First, our participants were predominantly White (83.1%). Second, as each Canadian province experienced different case numbers and related restrictions, it was not possible to decipher what restrictions people were faced with at the time they completed the survey. Future research should continue to examine coping methods and unmet socioemotional needs among pregnant people during the pandemic with the goal of informing the development and dissemination of prenatal public health initiatives including targeted remote-accessed social support, prenatal care, and mental health strategies.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Assaly R (2021) How we got here: A timeline of Alberta’s response to the COVID-19 pandemic. Toronto Star. Accessed on February 10, 2022: https://www.thestar.com/news/canada/2021/09/16/how-we-got-here-a-timeline-of-albertas-response-to-the-covid-19-pandemic.html
Aydin R, Aktaş S (2021). An investigation of women’s pregnancy experiences during the COVID-19 pandemic: A qualitative study. Int J Clin Pract, 1-1110.1111/ijcp.14418
Bloom TL Bullock LFC Parsons L Rural pregnant women’s stressors and priorities for stress reduction Issues in Ment Health Nurs 2012 33 12 813 819 10.3109/01612840.2012.712087
Boserup B McKenney M Elkbuli A Alarming trends in US domestic violence during the COVID-19 pandemic Am J Emerg Med 2020 38 2753 2755 10.1016/j.ajem.2020.04.077 32402499
Braun V Clarke V Using thematic analysis in psychology Qual Res Psychol 2006 3 2 77 101 10.1191/1478088706qp063oa
Braun V Clarke V Reflecting on reflexive thematic analysis Qual Res Sport Exerc Health 2019 11 4 589 597 10.1080/2159676X.2019.1628806
Brooks SK Webster RK Smith LE Woodland L Wessely S Greenberg N Rubin GJ The psychological impact of quarantine and how to reduce it: rapid review of the evidence Lancet 2020 395 10227 912 920 10.1016/S0140-6736(20)30460-8 32112714
Burgess A Breman R Bradley D Dada S Burcher P Pregnant Women’s Reports of the Impact of COVID-19 on Pregnancy, Prenatal Care, and Infant Feeding Plans MCN Am J Matern Child Nurs 2021 46 1 21 29 10.1097/NMC.0000000000000673 33009009
Buss C Davis EP Shahbaba B Pruessner JC Head K Sandman CA Maternal cortisol over the course of pregnancy and subsequent child amygdala and hippocampus volumes and affective problems Proc Natl Acad Sci 2012 109 20 E1312 E1319 10.1073/pnas.1201295109 22529357
Cameron EE Joyce KM Delaquais CP Reynolds K Protudjer J Roos LE Maternal psychological distress & mental health service use during the COVID-19 pandemic J Affect Disord 2020 276 765 774 10.1016/j.jad.2020.07.081 32736186
Canada (2021). Coronavirus disease (COVID-19): Prevention and risks. Government of Canada. Accessed on August 23rd 2021 https://www.canada.ca/en/public-health/services/diseases/2019-novel-coronavirus-infection/prevention-risks.html
CDC (2022). How to Protect Yourself & Others. Your Health. Accessed on February 1st, 2022. https://www.cdc.gov/coronavirus/2019-ncov/prevent-getting-sick/prevention.html
CDC (2021). Pregnant people. Centers for Disease Control and Prevention. Accessed on January 8th 2022. https://www.cdc.gov/coronavirus/2019-ncov/need-extra-precautions/pregnant-people.html
Cella D Riley W Stone A Rothrock N Reeve B Yount S PROMIS Cooperative Group The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008 J Clin Epidemiol 2010 63 11 1179 1194 10.1016/j.jclinepi.2010.04.011 20685078
Cigăran RG Botezatu R Mînecan EM Gică C Panaitescu AM Peltecu G Gică N The psychological impact of the COVID-19 pandemic on pregnant women Healthcare 2021 9 6 725 10.3390/healthcare9060725 34204712
Cooper CL Marshall J Understanding executive stress 1977 New York PBI
Cox JL Holden JM Sagovsky R Detection of postnatal depression: Development of the 10-item Edinburgh Postnatal Depression Scale Br J Psychiatry 1987 150 782 786 10.1192/bjp.150.6.782 3651732
Dadi AF, Miller ER, Bisetegn TA, Mwanri L (2020). Global burden of antenatal depression and its association with adverse birth outcomes: An umbrella review. BMC Public Health, 20(1). 10.1186/s12889-020-8293-9
Dennis CL Chung-Lee L Postpartum depression help-seeking barriers and maternal treatment preferences: A qualitative systemic review Birth 2006 33 323 331 10.1111/j.1523-536X.2006.00130.x 17150072
Dennis CL Falah-Hassani K Shiri R Prevalence of antenatal and postnatal anxiety: Systematic review and meta-analysis Br J Psychiatry 2017 210 5 315 323 10.1192/bjp.bp.116.187179 28302701
DiPietro JA Kivlighan KT Costigan KA Rubin SE Shiffler DE Henderson JL Pillon JP Prenatal antecedents of newborn neurological maturation Child Dev 2010 81 1 115 130 10.1111/j.1467-8624.2009.01384.x 20331657
Evans K Spiby H Morrell JC Non-pharmacological interventions to reduce the symptoms of mild to moderate anxiety in pregnant women A systematic review and narrative synthesis of women’s views on the acceptability of and satisfaction with interventions Arch Womens Ment Health 2020 23 1 11 28 10.1007/s00737-018-0936-9 30613846
Fairbrother N Janssen P Anthony MM Tucker E Young AH Perinatal anxiety disorder prevalence and incidence J Affect Disord 2016 200 148 155 10.1016/j.jad.2015.12.082 27131505
Falah-Hassani K Shiri R Dennis CL The prevalence of antenatal and postnatal co-morbid anxiety and depression: A meta-analysis Psychol Med 2017 47 12 2041 2053 10.1017/S0033291717000617 28414017
Fan S, Guan J, Cao L, Wang M, Zhao H, Chen L, Yan L (2021). Psychological effects caused by COVID-19 pandemic on pregnant women: A systemic review with meta-analysis. Asian J Psychiatr, 56. 10.1016/j.ajp.2020.102533
Fawcett EJ, Fairbrother N, Cox L, White IR, Fawcett JM (2020). The prevalence of anxiety disorders during pregnancy and the postpartum period: A multivariate Bayesian meta-analysis. J Clin Psychiatry, 80(4). 10.4088/JCP.18r12527.
Field T Diego M Hernandez-Reif M Figueiredo B Deeds O Ascencio A Schanberg S Kuhn C Comorbid depression and anxiety effects on pregnancy and neonatal outcome Infant Behav Dev 2010 33 1 23 10.1016/j.infbeh.2009.10.004 19945170
Fonseca A Gorayeb R Canavarro MC Women’s help-seeking behaviours for depressive symptoms during the perinatal period: Socio-demographic and clinical correlated and perceived barriers to seeking professional help Midwifery 2015 31 1177 1185 10.1016/j.midw.2015.09.002 26433622
Furer P Alcolado G Reynolds K Hebert EA Group cognitive-behavioural therapy for perinatal anxiety disorders: Treatment development, content, and pilot results J Affect Disord Rep 2021 6 100195 10.1016/j.jadr.2021.100195
Giesbrecht G Bagshawe M van Sloten M MacKinnon A Dhillon A van de Wouw M Vaghef-Mehrabany E Rojas L Cattani D Lebel C Tomfohr-Madsen L Protocol for the Pregnancy During the COVID-19 Pandemic (PdP) study: A longitudinal cohort study of mental health among pregnant Canadians during the COVID-19 pandemic and developmental outcomes in their children MIR Research Protocols 2021 10 4 e25407 e25407 10.2196/25407
Goletzke J Kocalevent RD Hansen G Rose M Becher H Hecher K Arck PC Diemert A Prenatal stress perception and coping strategies: Insights from a longitudinal prospective pregnancy cohort J Psychosom Res 2017 102 8 14 10.1016/j.jpsychores.2017.09.002 28992901
Harris PA Taylor R Thielke R Payne J Gonzalez N Conde JG Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support J Biomed Inform 2009 42 2 377 381 10.1016/j.jbi.2008.08.010 18929686
Hessami K Romanelli C Chiurazzi M Cozzolino M COVID-19 pandemic and maternal mental health: a systematic review and meta-analysis J Matern Fetal Neonatal Med 2020 0 0 1 8 10.1080/14767058.2020.1843155
Hutchens BF Kearney J Risk factors for postpartum depression: An umbrella review J Midwifery Womens Health 2020 65 1 96 108 10.1111/jmwh.13067 31970924
Ibrahim SM Nicoloro-SantaBarbara J Auerbach MV Rosenthal L Kocis C Busso CE Lobel M Pregnancy-specific coping and changes in emotional distress from mid- to late pregnancy J Reprod Infant Psychol 2019 37 4 397 412 10.1080/02646838.2019.1578871 30773900
ICES (2022). ICES COVID-19 Dashboard. Data Discovery Better Health. Accessed on January 8th 2022. https://www.ices.on.ca/DAS/AHRQ/COVID-19-Dashboard
Johnson RC Slade P Obstetric complications and anxiety during pregnancy: is there a relationship? J Psychosom Obstet Gynaecol 2003 24 1 1 14 10.3109/01674820309042796 12685335
Kar P Tomfohr-Madsen L Giesbrecht G Bagshawe M Lebel C Alcohol and substance use in pregnancy during the COVID-19 pandemic Drug Alcohol Depend 2021 225 108760 10.1016/j.drugalcdep.2021.108760 34102507
Khatri GK Tran TD Fisher J Prevalence and determinants of symptoms of antenatal common mental disorders among women who had recently experienced an earthquake: A systematic review BMC Psychiatry 2019 19 1 1 17 10.1186/s12888-018-1986-2 30606141
Khoury JE Atkinson L Bennett T Jack SM Gonzalez A Coping strategies mediate the associations between COVID-19 experiences and mental health outcomes in pregnancy Arch Womens Ment Health 2021 24 6 1007 1017 10.1007/s00737-021-01135-2 34145499
Lau Y Wang Y Kwong DHK Wang Y Testing direct and moderating effects of coping styles on the relationship between perceived stress and antenatal anxiety symptoms J Psychosom Obstet Gynecol 2015 36 1 29 35 10.3109/0167482X.2014.992410
Lazarus RS, Folkman S (1984). Stress, appraisal, and coping. Springer Pub. Co
Leach LS Poyser C Fairweather-Schmidt K Maternal perinatal anxiety: A review of prevalence and correlates Clin Psychol 2017 21 1 4 19 10.1111/cp.12058
Leahy-Warren P Newham J Alderdice F Perinatal social support: panacea or a pitfall J Reprod Infant Psychol 2018 36 3 219 221 10.1080/02646838.2018.1477242 29916279
Loughnan SA Wallace M Joubert AE Haskelberg H Andrews G Newby JM A systematic review of psychological treatments for clinical anxiety during the perinatal period Arch Womens Ment Health 2018 21 481 490 10.1007/s00737-018-0812-7 29368048
Milgrom J Hirshler Y Reece J Holt C Gemmill AW Social support—a protective factor for depressed perinatal women? Int J Environ Res Public Health 2019 16 8 1426 10.3390/ijerph16081426 31010090
Mu TY Li YH Xu RX Chen J Wang YY Shen CZ Internet-based interventions for postpartum depression: A systematic review and meta-analysis Nurs Open 2021 8 3 1125 1134 10.1002/nop2.724 33373101
O’Mahen HA Flynn HA Preferences and perceived barriers to treatment for depression during the perinatal period J Womens Health 2008 17 1301 1309 10.1089/jwh.2007.0631
Pilkonis PA Choi SW Reise SP Stover AM Riley WT Cella D PROMIS Cooperative Group Item banks for measuring emotional distress from the Patient-Reported Outcomes Measurement Information System (PROMIS®): depression, anxiety, and anger Assessment 2011 18 3 263 283 10.1177/1073191111411667 21697139
Ponti L Smorti M Mediating role of labor on the relationship between prenatal psychopathologic symptoms and symptoms of postpartum depression in women who give birth vaginally J Obstet Gynecol Neonatal Nurs 2019 48 6 627 634 10.1016/j.jogn.2019.09.003 31626779
Reynolds K (2021). UM News Today. UM psychologists create free tool to help new parents grapple with anxiety during COVID-19: https://news.umanitoba.ca/um-psychologists-create-free-tool-to-help-new-parents-grapple-with-anxiety-during-covid-19/
Ritchie H, Mathieu E, Rodés-Guirao L, Appel C, Giattino C, Ortiz-Ospina E, Hasell J, MacDonald B, Beltekian D, Roser M (2021) "Coronavirus Pandemic (COVID-19)". Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/coronavirus' [Online Resource].
Rowther AA Kazi AK Nazir H Atiq M Atif N Rauf N Malik A Surkan PJ “A woman is a puppet” Women’s disempowerment and prenatal anxiety in Pakistan: A qualitative study of sources, mitigators, and coping strategies for anxiety in pregnancy Int J Environ Res Public Health 2020 17 14 1 19 10.3390/ijerph17144926
Scott K (2020). Women bearing the brunt of economic losses: One in five has been laid off or had hours cut. The Monitor. Accessed on August 23rd 2021. https://monitormag.ca/articles/women-bearing-the-brunt-of-economic-losses-one-in-five-has-been-laid-off-or-had-hours-cut/
Shamsaei F Maleki A Shobeiri F Soltani F Ahmadi F Roshanaei G The relationship between general health and coping style with perceived stress in primigravida healthy pregnant women: Using the PATH model Women Health 2019 59 1 41 54 10.1080/03630242.2018.1434587 29400638
Sun F Zhu J Tao H Ma Y Jin W A systematic review involving 11,187 participants evaluating the impact of COVID-19 on anxiety and depression in pregnant women J Psychosom Obstet Gynecol 2021 42 2 91 99 10.1080/0167482X.2020.1857360
Taubman-Ben-Ari O Pathways and barriers to parenthood: Existential concerns regarding fertility, pregnancy, and early parenthood Springer International Publishing 2019 10.1007/978-3-030-24864-2
Teixeria JMA Glover V Fisk NM Acute cerebral redistribution in response to invasive procedures in the human fetus AM J Obstet Gynecol MFM 1999 181 4 1018 1025 10.1016/S0002-9378(99)70340-6
Tomfohr-Madsen LM Racine N Giesbrecht GF Lebel C Madigan S Depression and anxiety in pregnancy during COVID-19: A rapid review and meta-analysis, Psychiatry Res, 300 ISSN 2021 113912 0165 1781 10.1016/j.psychres.2021.113912
Underwood L Waldie K D’Souza S Peterson ER Morton S A review of longitudinal studies on antenatal and postnatal depression Arch Womens Ment Health 2016 19 5 711 720 10.1007/s00737-016-0629-1 27085795
van de Loo KFE Vlenterie R Nikkels SJ Merkus PJFM Roukema J Verhaak CM Roeleveld N van Gelder MMHJ Depression and anxiety during pregnancy: The influence of maternal characteristics Birth 2018 45 4 478 489 10.1111/birt.12343 29517137
van der Linden C (2020). The moms are not alright: How coronavirus pandemic policies penalize mothers. The Conversation. Accessed on August 23, 2021. https://theconversation.com/the-moms-are-not-alright-how-coronavirus-pandemic-policies-penalize-mothers-144713
Woody CA Ferrari AJ Siskind DJ Whiteford HA Harris MG A systematic review and meta-regression of the prevalence and incidence of perinatal depression J Affect Disord 2017 219 86 92 10.1016/j.jad.2017.05.003 28531848
Yan H Ding Y Guo W Mental health of pregnant and postpartum women during the Coronavirus disease 2019 pandemic: A systematic review and meta-analysis Front Psychol 2020 11 1 12 10.3389/fpsyg.2020.617001 32038435
Zender R Olshansky E Women’s mental health: Depression and anxiety Nurs Clin North Am 2009 44 3 355 364 10.1016/j.cnur.2009.06.002 19683096
| 36443483 | PMC9707189 | NO-CC CODE | 2022-12-10 23:15:15 | no | Arch Womens Ment Health. 2022 Nov 29; 25(6):1137-1148 | utf-8 | Arch Womens Ment Health | 2,022 | 10.1007/s00737-022-01277-x | oa_other |
==== Front
J Digit Imaging
J Digit Imaging
Journal of Digital Imaging
0897-1889
1618-727X
Springer International Publishing Cham
36443529
720
10.1007/s10278-022-00720-w
Article
Bilateral Weighted Relative Total Variation for Low-Dose CT Reconstruction
He Yuanwei 12Yuanwei He
received the B.S. degree in Information and Computing Science from Chongqing University, Chongqing, China, in 2017. She is currently a doctoral student with the College of Mathematics and Statistics, Chongqing University, China. Her interest is in image processing.
http://orcid.org/0000-0001-6109-9583
Zeng Li [email protected]
12Li Zeng
received his B.S. degree in 1986 from Sichuan Normal University, M.S. degree in 1989 from Dalian University of Technology and Ph.D. degree (equivalently) in 1997 from Chongqing University. He was a postdoctor at the University of Basel, Switzerland, from 2000 to 2001. Now, he is a professor and PhD supervisor in Chongqing University. His research interests include industrial CT and digital image processing.
Chen Wei 3Chen Wei
received the Ph.D. degree from Third Military Medical University, Chongqing, China, in 2001. He is currently a Professor in Southwest Hospital, Army Medical University, Chongqing. His research interests include medical image processing and radiomics.
Gong Changcheng 4Changcheng Gong
received his B.S. degree in Information and Computing Science from Chongqing University in 2014, Ph.D. degree in the College of Optoelectronic Engineering from Chongqing University in 2019. He is currently an associate professor in Chongqing Technology and Business University, China. His interest is in image processing.
Shen Zhaoqiang 12Zhaoqiang Shen
received the B.S. degree in applied mathematics from Chongqing University, Chongqing, China, in 2016, where he is currently pursuing the Ph.D. degree with the College of Mathematics and Statistics. His research interest includes image reconstruction.
1 grid.190737.b 0000 0001 0154 0904 College of Mathematics and Statistics, Chongqing University, Chongqing, 401331 China
2 grid.190737.b 0000 0001 0154 0904 Engineering Research Center of Industrial Computed Tomography, Nondestructive Testing of the Education Ministry of China, Chongqing University, Chongqing, 400044 China
3 grid.416208.9 0000 0004 1757 2259 Department of Radiology, Southwest Hospital of AMU, Chongqing, Chongqing, 400038 China
4 grid.411578.e 0000 0000 9802 6540 College of Mathematics and Statistics, Chongqing Technology and Business University, Chongqing, 400067 China
28 11 2022
110
30 1 2022
14 10 2022
17 10 2022
© The Author(s) under exclusive licence to Society for Imaging Informatics in Medicine 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Low-dose computed tomography (LDCT) has been widely used for various clinic applications to reduce the X-ray dose absorbed by patients. However, LDCT is usually degraded by severe noise over the image space. The image quality of LDCT has attracted aroused attentions of scholars. In this study, we propose the bilateral weighted relative total variation (BRTV) used for image restoration to simultaneously maintain edges and further reduce noise, then propose the BRTV-regularized projections onto convex sets (POCS-BRTV) model for LDCT reconstruction. Referring to the spacial closeness and the similarity of gray value between two pixels in a local rectangle, POCS-BRTV can adaptively extract sharp edges and minor details during the iterative reconstruction process. Evaluation indexes and visual effects are used to measure the performances among different algorithms. Experimental results indicate that the proposed POCS-BRTV model can achieve superior image quality than the compared algorithms in terms of the structure and texture preservation.
Keywords
Low-dose computed tomography (LDCT)
Image reconstruction
Relative total variation
Structure preservation
http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 61771003 Zeng Li Graduate Research and Innovation Foundation of Chongqing, ChinaCYB21044 He Yuanwei General project of Chongqing Natural Science Foundationcstc2021jcyj-msxmX0679 Gong Changcheng Science and Technology Research Program of Chongqing Education Commission of ChinaKJQN202000808 Gong Changcheng
==== Body
pmcIntroduction
Computed tomography (CT) is a technique based on X-ray attenuation to reconstruct tomographic images of an object [1]. As one of the greatest scientific and technological achievements at the end of twentieth century, CT have had a revolutionary impact on clinical screening, diagnosis, image-guided surgery, image-guided radiotherapy, and other aspects [2, 3]. However, the widespread use of CT has aroused people’s concern about causing cancer or genetic abnormalities [4, 5]. Therefore, strategies were proposed to lower X-ray dose in CT scanning [6], such as reducing tube current or peak voltage, decreasing the number of projections, and improving the hardware of CT system. All of the above methods can reduce the X-ray dose absorbed by patients, while this paper mainly studies low-dose CT acquired by reducing the radiation tube current or tube voltage. CT imaging artifacts brought by these methods are not too serious. However, due to the low signal-to-noise ratio of the obtained projection data, images reconstructed by traditional methods are usually severely degraded by noise, which might affect radiologists’ diagnosis [7]. In order to reconstruct high-quality images from low-dose computed tomography (LDCT) projection data, methods including projection domain preprocessing, iterative reconstruction, and post-processing have been proposed.
Operations on sinogram, called projection preprocessing method, aim to reduce noise in raw data. Projection data could be considered a noisy image. Denoising filters are usually designed according to the noise characteristics or structure information of the sinogram. Compared with other methods, projection preprocessing takes the statistical properties of noise as prior information. In recent years, models like penalized likelihood [8] and sparsity-based sinogram denoising [9] have been proposed to suppress noise and avoid edge blurring in sinogram. Liu et al. developed a well-designed composite dictionary with discriminative features in sinogram domain [10]. Xie et al. made full use of the prior knowledge and statistical properties of LDCT sinogram to formulate the preprocessing as a standard maximum a posteriori estimation [11]. Karimi et al. proposed simultaneous sparse representation (SSR) method [12], which is similar to sparse representation-based dictionary learning [13, 14] and non-local means algorithm [15, 16]. Hu et al. developed an improved Wasserstein generative adversarial network (WGAN) framework for sinogram to enhance the reconstructed image quality [17].
Post-processing method is not limited by CT supplier. Some outstanding filters and networks have been presented to tackle LDCT restoration, such as BM3D [18, 19], conveying path-based convolutional encoder-decoder network (CPCE) [20], and joint bilateral filter net (JBFnet) [21]. Post-processing methods can be directly applied to the reconstructed CT images and easily integrated into the current CT workflow. Based on the statistical property of CT images, Choi et al. proposed a deep learning approach to statistical image restoration for LDCT [22]. Shiri et al. applied a residual convolutional neural network to cope with ultralow-dose CT images of COVID-19 patients [23]. Gholizadeh-Ansari et al. used dilated convolutions with different dilation rates to capture more contextual information in fewer layers for LDCT [24].
Iterative reconstruction methods can also be used to enhance image quality of LDCT by taking advantages of the measured data information and some prior knowledge. Gao et al. proposed a prior-based machine learning model for Bayesian reconstruction of ultralow-dose CT images [25]. He et al. developed a noise suppression-guided image filtering reconstruction (NSGIFR) algorithm [26] for LDCT reconstruction. They introduced guided image filtering (GIF) [27] and block-matching and 3D filtering (BM3D) [18, 19] to achieve an outstanding performance on noise suppression and texture preservation. As a commonly used regularization method, total variation (TV) [28, 29] uses L1-norm of the image gradient to decrease noise in LDCT image reconstruction. However, TV-based reconstruction results are weak in retaining image structure, at the same time introducing blocking artifacts, which affects the clinical diagnosis. Therefore, Yu et al. considered L0-norm-based LDCT image reconstruction model, only calculating the number of image gradients whose amplitude is non-zero while ignoring penalty for the large ones [30]. Recently, relative total variation (RTV) [31] was proposed to extract main structures from complicated texture patterns. Unlike L0-norm and L1-norm, RTV is defined in terms of the differences between the partial derivatives of textures and structures in an image. Window inherent variation (WIV) and window total variation (WTV) are, respectively, defined as denominator and numerator of RTV. The main structures in images stand out according to the definition of WIV, which could easily distinguish structures and the textures. For WIV, the partial derivatives of textures and noise in a local window tend to cancel each other out. However, both textures and noise contribute to the value of WTV. As a consequence, the textures and noise can be removed by penalizing RTV. In [32], Gong et al. proposed RTV-regularized projections onto convex set (POCS-RTV) reconstruction model. Based on this model, they used L-curve method to set parameters in POCS-RTV, called adaptive POCS-RTV (POCS-ARTV). This model is able to reduce noise by penalizing RTV term.
In this work, we aim to take advantages of prior knowledge in gradient domain to enhance the quality of LDCT. Even though experimental results in [32] show a satisfactory performance on low-intensity CT reconstruction, we observed that POCS-RTV tends to cause fuzzy edges and details in reconstructed results especially for extremely LDCT. We consider this phenomenon is related to the weighting function of RTV, because the weight is only based on distance between pixels, ignoring the importance of image gray-scale values. We draw lessons from the thought of bilateral filtering [33, 34] to propose the bilateral weighted relative total variation (BRTV). Spatial proximity and pixel value similarity are utilized as the weighting function of WIV. Two close and similar pixels make more contributions to the partial derivative in WIV so that a smaller BRTV could be acquired, which could suppress noise while preserve sharp edges and fine details at the same time. Structures, textures, and noise contribute to the value of WTV. Therefore, weights of WTV are set as the original ones in RTV.
Inspired by [33, 34], we first propose BRTV to simultaneously maintain edges and further reduce noise for image restoration, then propose the BRTV-regularized projections onto convex set (POCS-BRTV) model for LDCT reconstruction. Being different from POCS-RTV, the weights of WIV in POCS-BRTV rely on both the closeness to vicinity in the domain and similarity to vicinity in the range, which could better characterize the distance and difference between two pixels. Especially for LDCT, textures and edges are severely degraded by extremely high level noise. It is easy to confuse textures and noise when reconstruction models based on image sparsity in the gradient domain are used to tackle the LDCT images. POCS-BRTV shows outstanding performance on getting sharper edges as well as distinguishing fine details and noise for LDCT reconstruction. The main contributions of this work are threefold. Firstly, we develop an image restoration model, BRTV, based on proximity and similarity between two pixels in a local rectangular. Secondly, BRTV is utilized as the regularization term for LDCT reconstruction model. Penalizing BRTV term can suppress noise and preserve main structures. Finally, we develop an iterative algorithm to solve POCS-BRTV model.
This paper is structurally organized as follows. In the next section, we briefly introduce LDCT imaging model and RTV. In the third section, the proposed algorithm is presented, after which the experimental results of the simulated data are described. The conclusion of this study is given in the final section.
Methods
Low-Dose CT Imaging Model
In LDCT, relationship between pixels and projection data can be expressed as.1 g=Af+e.
Column vector g denotes the N-dimension projection data degraded by noise e; column vector f is M-dimension object image; A represents the projection coefficient matrix.
The task of CT imaging is to reconstruct the object image f from projection data g. In ideal circumstances, CT projection data is noise-free so that the reconstructed images can be obtained according to the inverse of projection matrix if there is enough computer memory available. As a matter of fact, LDCT cannot be directly reconstructed due to the limitation of the computer memory for problem dimension and the degradation by noise. To approximately find f from g, the least squares method of optimal objective function is usually adopted to solve the problem in Eq. (1), which is represented in the following form.2 f∗=argminf⩾0Af-g22+λRf.
The first term in Eq. (2) is a data fidelity term to constrain the reconstructed image and projection data. The second term is a penalty term, usually used for noise reduction and edge preservation. λ is a penalty parameter to balance the two terms just mentioned.
Relative Total Variation
Relative total variation (RTV) [31] can be treated as a special weighting total variation, which enforces the structure part to be sparse in gradient domain. The formulation of RTV can be expressed as.3 RTVp=DxpLxp+ε+DypLyp+ε,
where ε is a positive number to avoid denominators equaling to zero. Dxp and Dyp are windowed total variations (WTV) along x and y directions at the pixel p of image f, using for reflecting the absolute differences between neighboring pixels within a rectangular region Rp centered at pixel p. Structures, textures, and noise in region Rp contribute to WTV, which has no dependence on the sign of each partial derivative. Lxp and Lyp denote the windowed inherent variations (WIV) along x and y directions at pixel p, and the WIV helps to distinguish structures from image f. WTV and WIV are defined as.4 Dxp=∑q∈Rpkp,q·∂xfq,Dyp=∑q∈Rpkp,q·∂yfq,
5 Lxp=∑q∈Rpkp,q·∂xfq,Lyp=∑q∈Rpkp,q·∂yfq,
where ∂x and ∂y are the partial derivatives in two directions. kp,q is a weighting function defined based on the spatial affinity referring to a rectangular region centered at pixel p, setting as6 kp,q∝exp-xp-xq2+yp-yq22σ2,
where xp and yp are the horizontal and vertical coordinates at pixel p, and σ relates to the scale of window Rp.
The Proposed Reconstruction Algorithm
LDCT naturally contains severe noise, and this noise will corrupt the image quality of the complicated texture within the LDCT. RTV tends to blur the distinction between textures and noise so that it is less than satisfactory for LDCT image denoising. In [32], Gong et al. proposed relative total variation-regularized projections onto convex set (POCS-RTV) model. Experiments show a satisfactory performance on low-intensity CT reconstruction, while we observed that this model tends to cause fuzzy edges and textures in reconstructed results especially for extremely LDCT. We consider that this phenomenon is related to the weighting function of RTV, because kp,q is only based on distance between pixels p and q, ignoring the importance of image gray-scale values. We draw lessons from the thought of bilateral filtering [33, 34] to utilize spatial proximity and pixel value similarity as the weighting function of WIV. Therefore, bilateral weighted relative total variation (BRTV) is proposed to simultaneously maintain edges and further reduce noise. Based on the characteristics of BRTV and LDCT, we utilize BRTV as the regularization term of LDCT reconstruction model and propose the POCS-BRTV model.
Bilateral Weighted Relative Total Variation
BRTV considers both closeness to vicinity in the domain and similarity to vicinity in the range of WIV. Considering that WTV would extract edges, textures, and noise of an image at the same time, BRTV takes no account of the modifications on WTV. BRTV is set as7 BRTVp=DxpL~xp+ε+DypL~yp+ε,
where Dxp and Dyp are same as the ones in RTV. L~xp and L~yp utilize a new weighting function hp,q. They are defined as.8 Dxp=∑q∈Rpkp,q·∂xfq,Dyp=∑q∈Rpkp,q·∂yfq,
9 L~xp=∑q∈Rphp,q·∂xfq,L~yp=∑q∈Rphp,q·∂yfq,
where kp,q is the original weighting function in RTV. It was defined based on the closeness between pixel p and pixel q shown in Eq. (6). hp,q, as a composite similarity measure, is proportional to the product of gray similarity and position proximity between pixel p and pixel q:10 hp,q∝exp-xp-xq2+yp-yq22σ2·exp-fp-fq22σ2,
where xp and yp are the horizontal and vertical coordinates at pixel p. fp is the gray-scale value at pixel p. A large hp,q represents a greater closeness between pixel p and pixel q. As a result of that, the partial derivatives make more contributions for WIV to reach a smaller BRTV.
Lxp, L~xp, Lyp, and L~yp of a noisy image are displayed in Fig. 1 to verify the performance of hp,q. It can be observed that L~xp and L~yp show clearer structures with less blurry edges than Lxp and Lyp.Fig. 1 The noisy phantom image and corresponding Lxp, L~xp, Lyp, and L~yp images. The display window is [0, 0. 5]
POCS-BRTV Reconstruction Model
The proposed POCS-BRTV reconstruction model is set as11 minf⩾0Af-g22+λ·∑p∈fBRTVp.
This model is to reconstruct image f from projection data g. A represents the projection coefficient matrix. The first term is data fidelity term, which maintains the constancy between the reconstructed image and the measurements. The second term, BRTV, is applied to remove noise and preserve clearer edges during LDCT reconstruction process. λ is a weight to balance data fidelity and regularization term.
As an iterative algorithm is adopted to solve this problem, we linearize Af-g22 at point fk. Problem (11) could be expressed as12 minf⩾0Afk-g22+ATAfk-g,f-fk+γ/2·f-fk22+λ·∑p∈fBRTVp,
where γ is a positive parameter, and fk is the intermediate image at the kth iteration during reconstruction process. · represents inner product of the two vectors. Optimization problem (12) can be reformulated as the following concise form:13 minf⩾0f-fk-1/γ·ATAfk-g22+2λ/γ·∑p∈fBRTVp.
The term fk-1/γ·ATAfk-g∗ can be approximately implemented by simultaneous algebraic reconstruction technique (SART). The combination of SART and nonnegative constraints, called POCS, is to acquire the intermediate image fk+1/2. Let λ1 be equal to 2λ/γ. Therefore, problem (11) can be solved by a two-step iterative reconstruction algorithm, including POCS step [35] and BRTV step:14 fk+1/2=POCSfk,g,
15 fk+1=argminf⩾0f-fk+1/222+λ1·∑p∈fBRTVp.
To solve Eq. (15), we first discuss the x-direction measure of penalty term. It can be written as16 ∑pDx(p)L~x(p)+ε=∑p∑q∈Rpkp,q·∂xfq∑q∈Rphp,q·∂xfq+ε=∑q∑p∈Rqkp,q∑q∈Rphp,q·∂xfq+ε∂xfq≈∑q∑p∈Rqkp,qL~xp+ε1∂xfq+εg∂xfq2=∑qu~xqwxq∂xfq2
where εg is introduced for numerical stability. u~xq and wxq are set as.17 u~xq=∑p∈Rqkp,qL~xp+ε,wxq=1∂xfq+εg.
A similar process can be used to the y-direction measure in BRTV. Then, Eq. (15) can be written in a matrix form:18 vf-vfk+1/2Tvf-vfk+1/2+λ1vfTCxTU~xWxCxvf+vfTCyTU~yWyCyvf
where vf and vfk+1/2 represent the vector of f and fk+1/2, respectively. Cx and Cy are the matrices of forward difference operator. U~x, U~y, Wx, and Wy are the diagonal matrices consisting of u~xi, u~yi, wxi, and wyi, respectively. The optimization problem of Eq. (18) can be solved using the following equation:19 I+λ1CxTU~xtWxtCx+CyTU~ytWytCy·vft+1=vfk+1/2,
where I is an identity matrix. t represents the number of iterations of BRTV.
The complete workflow of the proposed POCS-BRTV model is summarized as Table 1.Table 1 LDCT reconstruction by the proposed POCS-BRTV model
Experiments and Analysis
In this section, experiments on digital photons are used for demonstrating performance of the proposed model. All experiments ran on a 3.20 GHz Intel(R) Core(TM) i7-8700 CPU with windows 10 64-bit system environments, and all the reconstruction algorithms were implemented by Matlab 2016b combining with Visual Studio 2015. All reconstructed images use root mean square error (RMSE), peak signal-to-noise ratio (PSNR), and structure similarity (SSIM) for quantitative analyses. The size of each phantom is 256×256 for all of the experiments in this paper. Projections from these phantoms are acquired via fan-beam CT, and the detector units are arranged in line. Poisson noise is added to projection data in order to simulate LDCT. The level of Poisson noise is negatively correlated with photon number I0. Specifically, a large I0 corresponds to a higher SNR. We set I0=1.0×104 and 1.0×105 for all digital photon experiments. The geometry scanning parameters in this work are set as Table 2.Table 2 Geometry scanning parameters of the simulated experiments
System parameter Parameter value
The distance from X-ray source to center of rotation 500.0 mm
The distance from center of rotation to detector 0 mm
Interval angle between two views 1
The number of detector units 372
The length of the detector 372 mm
Image size 256 × 256
Pixel size 1.0 × 1.0 mm2
In all of the experiments, the qualities of reconstructed images are highly affected by parameters, which are selected according to the evaluation indexes and visual effect. In this work, we fixed the number of max iterative reconstruction NiterMax=1000. A stopping criteria referring to errors between two successive steps is set for an earlier termination. The relaxation parameters αSART are empirically fixed as 0.15 and 0.25 corresponding to photon number I0 at 1.0 × 104 and 1.0 × 105, respectively. The value of the regularization parameter λ1 can control the smoothness of results. A larger one represents a greater smoothness. Making BRTV iteration number NBRTV larger could preserve the edge sharpness well. Because BRTV is implemented in each iteration during the reconstruction process, NBRTV set as 2 or 3 is enough. σ is a spatial parameter to control the window size for computing the windowed variations. A large σ refers to a bigger window size, as a result of which the images would be strongly smoothed. A bit larger ε helps preserve smoothly varying structures. In other words, a smaller one helps to preserve sharper edges and more fine textures. Besides, we found that the value of ε depends on image complexities, simple images corresponding to a small value of ε. It is worth noting that all of the parameters of POCS-RTV are set the same as those of POCS-BRTV for a fair comparison.
Shepp-Logan Phantom Experiment
The first simulation experiment uses a Shepp-Logan phantom to simulate LDCT case. In this experiment, the photon number is set as I0=1.0×104 and 1.0×105. Parameters for POCS-BRTV model in different corresponding cases are set as follows: (1) λ1=0.0007, NBRTV=2, σ=0.6, ε=0.000001; (2) λ1=0.00045, NBRTV=2, σ=0.5, ε=0.000001.
In Fig. 2, CT images are reconstructed using LDCT projection data via SART, POCS-TV, POCS-RTV, and POCS-BRTV. On the far left one is the noise-free phantom, in which a ROI is marked by a red rectangle and zoomed-in at the corner. It can be observed that images reconstructed by SART are severely degraded by noise. POCS-TV reconstruction can effectively reduce noise, while the results from POCS-TV are blurred by staircase effect (indicated by red arrows). Adjusting parameters could mitigate this problem, while at the same time details might be blurred during the reconstruction process. From the reconstructed images of POCS-RTV, edges indicated by red arrows are blurry, while these artifacts disappear in the results of POCS-BRTV. Clear edges can be seen from the results of POCS-BRTV. Our model represents better noise suppression and edge protection than other three algorithms.Fig. 2 The tomographic results of Shepp-Logan phantom and a zoomed-in region reconstructed by different algorithms with different noise levels. The first column is the reference image, the rest of columns are results reconstructed via SART, POCS-TV, POCS-RTV, and POCS-BRTV. Images from top to bottom are the results reconstructed with different photon numbers: 1.0 × 104 and 1.0 × 10.5. The display window is [0, 0. 5]
Table 3, listed RMSE, SSIM, and PSNR corresponding to Fig. 2, is to quantitatively analyze different reconstruction algorithms under different noise levels. The bold ones are the optimal indexes. POCS-BRTV outdistances the other three algorithms according to the indexes.Table 3 Global RMSEs, SSIMs, and PSNRs of reconstructed Shepp-Logan phantom with different photon numbers
Photon number Method RMSE SSIM PSNR
1 × 104 SART 0.0522 0.6791 25.6493
POCS-TV 0.0117 0.8248 38.6347
POCS-RTV 0.0066 0.9745 43.5711
POCS-BRTV 0.0050 0.9891 46.1031
1 × 105 SART 0.0193 0.8469 34.2898
POCS-TV 0.0041 0.9937 47.8366
POCS-RTV 0.0020 0.9962 53.8810
POCS-BRTV 0.0011 0.9985 59.0495
FORBILD Head Phantom Experiment
The second experiment is performed on the projection data of the FORBILD head phantom. Parameters for our model corresponding to I0=1.0×104 and 1.0×105 are set as follows: (1) λ1=0.01, NBRTV=3, σ=1, ε=0.0002; (2) λ1=0.004, NBRTV=3, σ=0.9, ε=0.00001.
The restored images from LDCT data are presented in Fig. 3, and the reference image is on the far left. Two ROIs are marked with red rectangles. The rest of columns are images processed by SART, POCS-TV, POCS-RTV, and POCS-BRTV, respectively. The zoomed-in ROIs corresponding to Fig. 3 are displayed in Fig. 4.Fig. 3 The tomographic results of FORBILD head phantom reconstructed by different algorithms with different noise levels. The first column is the reference image, the rest of columns are results reconstructed via SART, POCS-TV, POCS-RTV, and POCS-BRTV. Images from top to bottom are the results reconstructed with different photon numbers: 1.0 × 104 and 1.0 × 10.5. The display window is [0.5, 1.5]
Fig. 4 ROIs of the FORBILD head phantom and the reconstructed results in Fig. 3. The images in (a) are extracted from the reference FORBILD head phantom image. Images in (b) and (c) are extracted from results reconstructed with different photon numbers: 1.0 × 104 and 1.0 × 105, respectively. From top to bottom, images are ROI 1 and ROI 2. From left to right in (b) and (c), images are, respectively, reconstructed by SART, POCS-TV, POCS-RTV, and POCS-BRTV. The display window is [0.5, 1.5]
From the zoomed-in ROIs, it can be observed that all of the results from SART are severely disrupted by noise so that three minor dots in images cannot be identified even to the naked eye. Images reconstructed via POCS-TV tend to cause staircase effects, which are not smooth enough. POCS-RTV leads to over-smoothing results. Almost all of the three minor dots are smoothed by POCS-RTV. These problems are improved according to the results of POCS-BRTV. However, there is something that needs to be improved. When the noise in projection data intensifies, even though the POCS-BRTV model can significantly suppress noise, two extremely minor dots cannot be preserved in ROI1. Table 4 presents the quantitative evaluation indexes corresponding to the results in Fig. 3. The bold ones are optimal indexes. The results show that POCS-BRTV model presents the best performance at the two presented cases.Table 4 Global RMSEs, SSIMs, and PSNRs of reconstructed FORBILD head phantom with different photon numbers
Photon number Method RMSE SSIM PSNR
1 × 104 SART 0.1927 0.4384 14.2608
POCS-TV 0.0647 0.7113 23.7786
POCS-RTV 0.0294 0.8334 30.6457
POCS-BRTV 0.0274 0.8499 31.2318
1 × 105 SART 0.0725 0.6243 22.7593
POCS-TV 0.0240 0.9344 33.3779
POCS-RTV 0.0097 0.9671 40.2613
POCS-BRTV 0.0058 0.9867 44.7963
In Fig. 5, the horizontal profiles (177th row) of the reconstructed images from the incident X-ray intensity at I0=1.0×104 are provided to verify the advantages of POCS-BRTV, that it generates the nearest image, especially at edges, to the reference one. From the RIOs in Fig. 5, we can see that POCS-BRTV is able to preserve the sharp edges, while POCS-RTV tends to cause jitter on edges.Fig. 5 Horizontal profiles (177th row) of the reconstructed images shown in Fig. 3 by different algorithms with the incident X-ray intensity of 1 × 104
Discussion and Conclusion
In this paper, we proposed a LDCT reconstruction model based on the prior knowledge in gradient domain. This work extended the thinking of Gong et al. that they utilized the differences between partial derivatives generated by noise and structures, and developed POCS-RTV for low-intensity CT reconstruction [32]. POCS-RTV is defined based on WIV and WTV in the gradient domain. Structures and noise can be adaptively distinguished by WIV. However, the results from POCS-RTV tend to cause blur on edges especially under high-level noise. It might result from the weights of WIV that only depend on the distance between two pixels in a local rectangular. In fact, both nearby spatial location and similar gray values make contributions to characterize the closeness between two pixels. Therefore, we further integrated the similarity measurement with the weighting function of WIV, proposing POCS-BRTV to reconstruct LDCT. POCS-BRTV considers both position closeness and gray similarity to vicinity in the range of WIV, which helps preserve more fine details and sharp edges compared with POCS-RTV.
In summary, we propose a bilateral weighted relative total variation (POCS-BRTV) model for LDCT reconstruction in this work, integrating the thinking of bilateral filtering with POCS-RTV. Experimental results obtained at different noise levels show that the proposed POCS-BRTV model leads to significant improvements on LDCT images and especially achieves better results on preservation of edges and details during the reconstruction procedure.
Acknowledgements
This work was supported by the National Natural Science Foundation of China under Grant 61771003; Graduate Research and Innovation Foundation of Chongqing, China, under Grant CYB21044; General project of Chongqing Natural Science Foundation under Grant cstc2021jcyj-msxmX0679; and Science and Technology Research Program of Chongqing Education Commission of China under Grant KJQN202000808.
Declarations
Conflict of Interest
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Goldman, and W. L., “Principles of CT and CT technology”, J Nucl Med Technol, vol. 35, no. 3, pp. 115–128, 2007.
2. J. H. Siewerdsen, “Cone-Beam CT with a Flat-Panel Detector: From Image Science to Image-Guided Surgery”, Nucl Instrum Methods Phys Res A, vol. 648, no. supp-S1, pp. S241-S250, 2011.
3. Schubert T Jacob AL Takes M Menter T Gutzeit A Kos S CT-Guided Percutaneous Biopsy of a Mass Lesion in the Upper Presacral Space: A Sacral Transneuroforaminal Approach Cardiovascular & Interventional Radiology 2012 35 5 1255 1257 10.1007/s00270-011-0232-9 21789696
4. Frush DP Donnelly LF Rosen NS Computed Tomography and Radiation Risks: What Pediatric Health Care Providers Should Know Pediatrics 2003 112 4 951 957 10.1542/peds.112.4.951 14523191
5. Brenner DJ Hall E Computed tomography—an increasing source of radiation exposure New England Journal of Medicine 2007 357 22 2277 2284 10.1056/NEJMra072149 18046031
6. Naidich DP Marshall CH Gribbin C Arams RS Mccauley DI Low-dose CT of the lungs: Preliminary observations Radiology 1990 175 3 729 731 10.1148/radiology.175.3.2343122 2343122
7. Dhawan AP Rangayyan RM Gordon R Image restoration by Wiener deconvolution in limited-view computed tomography Applied Optics 1985 24 23 4013 10.1364/AO.24.004013 18224156
8. Li T Li X Wang J Wen J Lu H Hsieh J Liang Z Nonlinear sinogram smoothing for low-dose X-ray CT IEEE Transactions on Nuclear Science 2004 51 5 2505 2513 10.1109/TNS.2004.834824
9. J. Shtok, M. Elad, and M. Zibulevsky, “Sparsity-based sinogram denoising for low-dose computed tomography”, in IEEE ICASSP, 2011, pp. 569–572.
10. Liu J Ma J Zhang Y Chen Y Yang J Shu H Luo L Coatrieux G Yang W Feng Q Discriminative Feature Representation to Improve Projection Data Inconsistency for Low Dose CT Imaging IEEE Transactions on Medical Imaging 2018 36 12 2499 2509 10.1109/TMI.2017.2739841
11. Q. Xie, D. Zeng, Q. Zhao, D. Meng, Z. Xu, Z. Liang, and J. Ma, “Robust Low-Dose CT Sinogram Preprocessing via Exploiting Noise-Generating Mechanism”, IEEE Trans Med Imaging, vol. PP, no. 12, pp. 2487–2498, 2017.
12. Karimi D Ward RK Sinogram denoising via simultaneous sparse representation in learned dictionaries Physics in Medicine & Biology 2016 61 9 3536 10.1088/0031-9155/61/9/3536 27055224
13. Lu Y Zhao J Wang G Few-view image reconstruction with dual dictionaries Physics in Medicine & Biology 2011 57 1 173 189 10.1088/0031-9155/57/1/173
14. Q. Xu, H. Y. Yu, X. Q. Mou, L. Zhang, J. Hsieh, and G. Wang, “Low-Dose X-ray CT Reconstruction via Dictionary Learning”, IEEE Transactions on Medical Imaging, 2012.
15. Chun SY Dewaraja YK Fessler JA Alternating Direction Method of Multiplier for Tomography With Nonlocal Regularizers IEEE Transactions on Medical Imaging 2014 33 10 1960 1968 10.1109/TMI.2014.2328660 25291351
16. Buades A Coll B Morel JM Nonlocal Image and Movie Denoising International Journal of Computer Vision 2008 76 2 123 139 10.1007/s11263-007-0052-1
17. Hu Z Xue H Zhang Q Gao J Zhang N Zou S Teng Y Liu X Yang Y Liang D Zhu X Zheng H DPIR-Net: Direct PET Image Reconstruction Based on the Wasserstein Generative Adversarial Network IEEE Transactions on Radiation and Plasma Medical Sciences 2021 5 1 35 43 10.1109/TRPMS.2020.2995717
18. K. Dabov, A. Foi, V. Katkovnik, and K. Egiazarian, “Image denoising with block-matching and 3D filtering”, Proc. SPIE-IS&T Electronic Imaging, pp. 354–365, 2006.
19. Dabov K Foi A Katkovnik V Egiazarian K Image denoising by sparse 3-D transform-domain collaborative filtering IEEE Transactions on Medical Imaging 2007 16 8 2080 2095 10.1109/TIP.2007.901238
20. Kalra LS Cong W Wang G 3D Convolutional Encoder-Decoder Network for Low-Dose CT via Transfer Learning from a 2D Trained Network IEEE Transactions on Medical Imaging 2018 3 7
21. M. Patwari, R. Gutjahr, R. Raupach, and A. Maier, “JBFnet -- Low Dose CT Denoising by Trainable Joint Bilateral Filtering”, 2020.
22. K. Choi, J. S. Lim, and S. K. Kim, “StatNet: Statistical Image Restoration for Low-Dose CT using Deep Learning”, no. 99, pp. 1–1, 2020.
23. Shiri I Akhavanallaf A Sanaat A Salimi Y Askari D Mansouri Z Shayesteh SP Hasanian M Rezaei-Kalantari K Salahshour A Sandoughdaran S Abdollahi H Arabi H Zaidi H Ultra-low-dose chest CT imaging of COVID-19 patients using a deep residual neural network Eur Radiol 2021 31 3 1420 1431 10.1007/s00330-020-07225-6 32879987
24. Gholizadeh-Ansari M Alirezaie J Babyn P Deep Learning for Low-Dose CT Denoising Using Perceptual Loss and Edge Detection Layer Journal of Digital Imaging 2020 33 2 504 515 10.1007/s10278-019-00274-4 31515756
25. Gao Y Tan J Shi Y Lu S Gupta A Li H Liang Z Constructing a tissue-specific texture prior by machine learning from previous full-dose scan for Bayesian reconstruction of current ultralow-dose CT images J Med Imaging (Bellingham) 2020 7 3 032502 32118093
26. He Y Zeng L Yu W Gong C Noise suppression-guided image filtering for low-SNR CT reconstruction Medical & biological engineering & computing 2020 58 11 2621 2629 10.1007/s11517-020-02246-1 32839918
27. K. He, J. Sun, X. Tang, and m. intelligence, “Guided image filtering”, IEEE transactions on pattern analysis, vol. 35, no. 6, pp. 1397–1409, 2012.
28. E. Y. Sidky, Y. Duchin, X. Pan, and C. Ullberg, “A constrained, total-variation minimization algorithm for low-intensity x-ray CT”, Medical Physics, vol. 38 Suppl 1, no. S1, 2011.
29. Tian Z Jia X Yuan K Pan T Jiang SB Low-dose CT reconstruction via edge-preserving total variation regularization Physics in Medicine & Biology 2011 56 18 5949 5967 10.1088/0031-9155/56/18/011 21860076
30. Yu W Wang C Huang M Edge-preserving reconstruction from sparse projections of limited-angle computed tomography using ℓ 0-regularized gradient prior Review of Scientific Instruments 2017 88 4 043703 10.1063/1.4981132 28456252
31. Xu L Yan Q Xia Y Jia J Structure extraction from texture via relative total variation ACM Transactions on Graphics 2012 31 6 1 10
32. Gong C Zeng L Adaptive iterative reconstruction based on relative total variation for low-intensity computed tomography Signal Processing 2019 165 149 162 10.1016/j.sigpro.2019.06.031
33. C. Tomasi, “Bilateral filtering for gray and color images”, Proc.ieee Inter.conf.computer Vision, 1998.
34. Kundu R A Novel Technique for Texture and Edge Preservation Using Bilateral Filter National Academy Science Letters 2020 44 2 117 124 10.1007/s40009-020-00921-3
35. Kak AC Slaney M Wang G Principles of computerized tomographic imaging Medical Physics 2002 29 1 107 10.1118/1.1455742
| 36443529 | PMC9707190 | NO-CC CODE | 2022-12-01 23:20:21 | no | J Digit Imaging. 2022 Nov 28;:1-10 | utf-8 | J Digit Imaging | 2,022 | 10.1007/s10278-022-00720-w | oa_other |
==== Front
J Digit Imaging
J Digit Imaging
Journal of Digital Imaging
0897-1889
1618-727X
Springer International Publishing Cham
36443529
720
10.1007/s10278-022-00720-w
Article
Bilateral Weighted Relative Total Variation for Low-Dose CT Reconstruction
He Yuanwei 12Yuanwei He
received the B.S. degree in Information and Computing Science from Chongqing University, Chongqing, China, in 2017. She is currently a doctoral student with the College of Mathematics and Statistics, Chongqing University, China. Her interest is in image processing.
http://orcid.org/0000-0001-6109-9583
Zeng Li [email protected]
12Li Zeng
received his B.S. degree in 1986 from Sichuan Normal University, M.S. degree in 1989 from Dalian University of Technology and Ph.D. degree (equivalently) in 1997 from Chongqing University. He was a postdoctor at the University of Basel, Switzerland, from 2000 to 2001. Now, he is a professor and PhD supervisor in Chongqing University. His research interests include industrial CT and digital image processing.
Chen Wei 3Chen Wei
received the Ph.D. degree from Third Military Medical University, Chongqing, China, in 2001. He is currently a Professor in Southwest Hospital, Army Medical University, Chongqing. His research interests include medical image processing and radiomics.
Gong Changcheng 4Changcheng Gong
received his B.S. degree in Information and Computing Science from Chongqing University in 2014, Ph.D. degree in the College of Optoelectronic Engineering from Chongqing University in 2019. He is currently an associate professor in Chongqing Technology and Business University, China. His interest is in image processing.
Shen Zhaoqiang 12Zhaoqiang Shen
received the B.S. degree in applied mathematics from Chongqing University, Chongqing, China, in 2016, where he is currently pursuing the Ph.D. degree with the College of Mathematics and Statistics. His research interest includes image reconstruction.
1 grid.190737.b 0000 0001 0154 0904 College of Mathematics and Statistics, Chongqing University, Chongqing, 401331 China
2 grid.190737.b 0000 0001 0154 0904 Engineering Research Center of Industrial Computed Tomography, Nondestructive Testing of the Education Ministry of China, Chongqing University, Chongqing, 400044 China
3 grid.416208.9 0000 0004 1757 2259 Department of Radiology, Southwest Hospital of AMU, Chongqing, Chongqing, 400038 China
4 grid.411578.e 0000 0000 9802 6540 College of Mathematics and Statistics, Chongqing Technology and Business University, Chongqing, 400067 China
28 11 2022
110
30 1 2022
14 10 2022
17 10 2022
© The Author(s) under exclusive licence to Society for Imaging Informatics in Medicine 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Low-dose computed tomography (LDCT) has been widely used for various clinic applications to reduce the X-ray dose absorbed by patients. However, LDCT is usually degraded by severe noise over the image space. The image quality of LDCT has attracted aroused attentions of scholars. In this study, we propose the bilateral weighted relative total variation (BRTV) used for image restoration to simultaneously maintain edges and further reduce noise, then propose the BRTV-regularized projections onto convex sets (POCS-BRTV) model for LDCT reconstruction. Referring to the spacial closeness and the similarity of gray value between two pixels in a local rectangle, POCS-BRTV can adaptively extract sharp edges and minor details during the iterative reconstruction process. Evaluation indexes and visual effects are used to measure the performances among different algorithms. Experimental results indicate that the proposed POCS-BRTV model can achieve superior image quality than the compared algorithms in terms of the structure and texture preservation.
Keywords
Low-dose computed tomography (LDCT)
Image reconstruction
Relative total variation
Structure preservation
http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 61771003 Zeng Li Graduate Research and Innovation Foundation of Chongqing, ChinaCYB21044 He Yuanwei General project of Chongqing Natural Science Foundationcstc2021jcyj-msxmX0679 Gong Changcheng Science and Technology Research Program of Chongqing Education Commission of ChinaKJQN202000808 Gong Changcheng
==== Body
pmcIntroduction
Computed tomography (CT) is a technique based on X-ray attenuation to reconstruct tomographic images of an object [1]. As one of the greatest scientific and technological achievements at the end of twentieth century, CT have had a revolutionary impact on clinical screening, diagnosis, image-guided surgery, image-guided radiotherapy, and other aspects [2, 3]. However, the widespread use of CT has aroused people’s concern about causing cancer or genetic abnormalities [4, 5]. Therefore, strategies were proposed to lower X-ray dose in CT scanning [6], such as reducing tube current or peak voltage, decreasing the number of projections, and improving the hardware of CT system. All of the above methods can reduce the X-ray dose absorbed by patients, while this paper mainly studies low-dose CT acquired by reducing the radiation tube current or tube voltage. CT imaging artifacts brought by these methods are not too serious. However, due to the low signal-to-noise ratio of the obtained projection data, images reconstructed by traditional methods are usually severely degraded by noise, which might affect radiologists’ diagnosis [7]. In order to reconstruct high-quality images from low-dose computed tomography (LDCT) projection data, methods including projection domain preprocessing, iterative reconstruction, and post-processing have been proposed.
Operations on sinogram, called projection preprocessing method, aim to reduce noise in raw data. Projection data could be considered a noisy image. Denoising filters are usually designed according to the noise characteristics or structure information of the sinogram. Compared with other methods, projection preprocessing takes the statistical properties of noise as prior information. In recent years, models like penalized likelihood [8] and sparsity-based sinogram denoising [9] have been proposed to suppress noise and avoid edge blurring in sinogram. Liu et al. developed a well-designed composite dictionary with discriminative features in sinogram domain [10]. Xie et al. made full use of the prior knowledge and statistical properties of LDCT sinogram to formulate the preprocessing as a standard maximum a posteriori estimation [11]. Karimi et al. proposed simultaneous sparse representation (SSR) method [12], which is similar to sparse representation-based dictionary learning [13, 14] and non-local means algorithm [15, 16]. Hu et al. developed an improved Wasserstein generative adversarial network (WGAN) framework for sinogram to enhance the reconstructed image quality [17].
Post-processing method is not limited by CT supplier. Some outstanding filters and networks have been presented to tackle LDCT restoration, such as BM3D [18, 19], conveying path-based convolutional encoder-decoder network (CPCE) [20], and joint bilateral filter net (JBFnet) [21]. Post-processing methods can be directly applied to the reconstructed CT images and easily integrated into the current CT workflow. Based on the statistical property of CT images, Choi et al. proposed a deep learning approach to statistical image restoration for LDCT [22]. Shiri et al. applied a residual convolutional neural network to cope with ultralow-dose CT images of COVID-19 patients [23]. Gholizadeh-Ansari et al. used dilated convolutions with different dilation rates to capture more contextual information in fewer layers for LDCT [24].
Iterative reconstruction methods can also be used to enhance image quality of LDCT by taking advantages of the measured data information and some prior knowledge. Gao et al. proposed a prior-based machine learning model for Bayesian reconstruction of ultralow-dose CT images [25]. He et al. developed a noise suppression-guided image filtering reconstruction (NSGIFR) algorithm [26] for LDCT reconstruction. They introduced guided image filtering (GIF) [27] and block-matching and 3D filtering (BM3D) [18, 19] to achieve an outstanding performance on noise suppression and texture preservation. As a commonly used regularization method, total variation (TV) [28, 29] uses L1-norm of the image gradient to decrease noise in LDCT image reconstruction. However, TV-based reconstruction results are weak in retaining image structure, at the same time introducing blocking artifacts, which affects the clinical diagnosis. Therefore, Yu et al. considered L0-norm-based LDCT image reconstruction model, only calculating the number of image gradients whose amplitude is non-zero while ignoring penalty for the large ones [30]. Recently, relative total variation (RTV) [31] was proposed to extract main structures from complicated texture patterns. Unlike L0-norm and L1-norm, RTV is defined in terms of the differences between the partial derivatives of textures and structures in an image. Window inherent variation (WIV) and window total variation (WTV) are, respectively, defined as denominator and numerator of RTV. The main structures in images stand out according to the definition of WIV, which could easily distinguish structures and the textures. For WIV, the partial derivatives of textures and noise in a local window tend to cancel each other out. However, both textures and noise contribute to the value of WTV. As a consequence, the textures and noise can be removed by penalizing RTV. In [32], Gong et al. proposed RTV-regularized projections onto convex set (POCS-RTV) reconstruction model. Based on this model, they used L-curve method to set parameters in POCS-RTV, called adaptive POCS-RTV (POCS-ARTV). This model is able to reduce noise by penalizing RTV term.
In this work, we aim to take advantages of prior knowledge in gradient domain to enhance the quality of LDCT. Even though experimental results in [32] show a satisfactory performance on low-intensity CT reconstruction, we observed that POCS-RTV tends to cause fuzzy edges and details in reconstructed results especially for extremely LDCT. We consider this phenomenon is related to the weighting function of RTV, because the weight is only based on distance between pixels, ignoring the importance of image gray-scale values. We draw lessons from the thought of bilateral filtering [33, 34] to propose the bilateral weighted relative total variation (BRTV). Spatial proximity and pixel value similarity are utilized as the weighting function of WIV. Two close and similar pixels make more contributions to the partial derivative in WIV so that a smaller BRTV could be acquired, which could suppress noise while preserve sharp edges and fine details at the same time. Structures, textures, and noise contribute to the value of WTV. Therefore, weights of WTV are set as the original ones in RTV.
Inspired by [33, 34], we first propose BRTV to simultaneously maintain edges and further reduce noise for image restoration, then propose the BRTV-regularized projections onto convex set (POCS-BRTV) model for LDCT reconstruction. Being different from POCS-RTV, the weights of WIV in POCS-BRTV rely on both the closeness to vicinity in the domain and similarity to vicinity in the range, which could better characterize the distance and difference between two pixels. Especially for LDCT, textures and edges are severely degraded by extremely high level noise. It is easy to confuse textures and noise when reconstruction models based on image sparsity in the gradient domain are used to tackle the LDCT images. POCS-BRTV shows outstanding performance on getting sharper edges as well as distinguishing fine details and noise for LDCT reconstruction. The main contributions of this work are threefold. Firstly, we develop an image restoration model, BRTV, based on proximity and similarity between two pixels in a local rectangular. Secondly, BRTV is utilized as the regularization term for LDCT reconstruction model. Penalizing BRTV term can suppress noise and preserve main structures. Finally, we develop an iterative algorithm to solve POCS-BRTV model.
This paper is structurally organized as follows. In the next section, we briefly introduce LDCT imaging model and RTV. In the third section, the proposed algorithm is presented, after which the experimental results of the simulated data are described. The conclusion of this study is given in the final section.
Methods
Low-Dose CT Imaging Model
In LDCT, relationship between pixels and projection data can be expressed as.1 g=Af+e.
Column vector g denotes the N-dimension projection data degraded by noise e; column vector f is M-dimension object image; A represents the projection coefficient matrix.
The task of CT imaging is to reconstruct the object image f from projection data g. In ideal circumstances, CT projection data is noise-free so that the reconstructed images can be obtained according to the inverse of projection matrix if there is enough computer memory available. As a matter of fact, LDCT cannot be directly reconstructed due to the limitation of the computer memory for problem dimension and the degradation by noise. To approximately find f from g, the least squares method of optimal objective function is usually adopted to solve the problem in Eq. (1), which is represented in the following form.2 f∗=argminf⩾0Af-g22+λRf.
The first term in Eq. (2) is a data fidelity term to constrain the reconstructed image and projection data. The second term is a penalty term, usually used for noise reduction and edge preservation. λ is a penalty parameter to balance the two terms just mentioned.
Relative Total Variation
Relative total variation (RTV) [31] can be treated as a special weighting total variation, which enforces the structure part to be sparse in gradient domain. The formulation of RTV can be expressed as.3 RTVp=DxpLxp+ε+DypLyp+ε,
where ε is a positive number to avoid denominators equaling to zero. Dxp and Dyp are windowed total variations (WTV) along x and y directions at the pixel p of image f, using for reflecting the absolute differences between neighboring pixels within a rectangular region Rp centered at pixel p. Structures, textures, and noise in region Rp contribute to WTV, which has no dependence on the sign of each partial derivative. Lxp and Lyp denote the windowed inherent variations (WIV) along x and y directions at pixel p, and the WIV helps to distinguish structures from image f. WTV and WIV are defined as.4 Dxp=∑q∈Rpkp,q·∂xfq,Dyp=∑q∈Rpkp,q·∂yfq,
5 Lxp=∑q∈Rpkp,q·∂xfq,Lyp=∑q∈Rpkp,q·∂yfq,
where ∂x and ∂y are the partial derivatives in two directions. kp,q is a weighting function defined based on the spatial affinity referring to a rectangular region centered at pixel p, setting as6 kp,q∝exp-xp-xq2+yp-yq22σ2,
where xp and yp are the horizontal and vertical coordinates at pixel p, and σ relates to the scale of window Rp.
The Proposed Reconstruction Algorithm
LDCT naturally contains severe noise, and this noise will corrupt the image quality of the complicated texture within the LDCT. RTV tends to blur the distinction between textures and noise so that it is less than satisfactory for LDCT image denoising. In [32], Gong et al. proposed relative total variation-regularized projections onto convex set (POCS-RTV) model. Experiments show a satisfactory performance on low-intensity CT reconstruction, while we observed that this model tends to cause fuzzy edges and textures in reconstructed results especially for extremely LDCT. We consider that this phenomenon is related to the weighting function of RTV, because kp,q is only based on distance between pixels p and q, ignoring the importance of image gray-scale values. We draw lessons from the thought of bilateral filtering [33, 34] to utilize spatial proximity and pixel value similarity as the weighting function of WIV. Therefore, bilateral weighted relative total variation (BRTV) is proposed to simultaneously maintain edges and further reduce noise. Based on the characteristics of BRTV and LDCT, we utilize BRTV as the regularization term of LDCT reconstruction model and propose the POCS-BRTV model.
Bilateral Weighted Relative Total Variation
BRTV considers both closeness to vicinity in the domain and similarity to vicinity in the range of WIV. Considering that WTV would extract edges, textures, and noise of an image at the same time, BRTV takes no account of the modifications on WTV. BRTV is set as7 BRTVp=DxpL~xp+ε+DypL~yp+ε,
where Dxp and Dyp are same as the ones in RTV. L~xp and L~yp utilize a new weighting function hp,q. They are defined as.8 Dxp=∑q∈Rpkp,q·∂xfq,Dyp=∑q∈Rpkp,q·∂yfq,
9 L~xp=∑q∈Rphp,q·∂xfq,L~yp=∑q∈Rphp,q·∂yfq,
where kp,q is the original weighting function in RTV. It was defined based on the closeness between pixel p and pixel q shown in Eq. (6). hp,q, as a composite similarity measure, is proportional to the product of gray similarity and position proximity between pixel p and pixel q:10 hp,q∝exp-xp-xq2+yp-yq22σ2·exp-fp-fq22σ2,
where xp and yp are the horizontal and vertical coordinates at pixel p. fp is the gray-scale value at pixel p. A large hp,q represents a greater closeness between pixel p and pixel q. As a result of that, the partial derivatives make more contributions for WIV to reach a smaller BRTV.
Lxp, L~xp, Lyp, and L~yp of a noisy image are displayed in Fig. 1 to verify the performance of hp,q. It can be observed that L~xp and L~yp show clearer structures with less blurry edges than Lxp and Lyp.Fig. 1 The noisy phantom image and corresponding Lxp, L~xp, Lyp, and L~yp images. The display window is [0, 0. 5]
POCS-BRTV Reconstruction Model
The proposed POCS-BRTV reconstruction model is set as11 minf⩾0Af-g22+λ·∑p∈fBRTVp.
This model is to reconstruct image f from projection data g. A represents the projection coefficient matrix. The first term is data fidelity term, which maintains the constancy between the reconstructed image and the measurements. The second term, BRTV, is applied to remove noise and preserve clearer edges during LDCT reconstruction process. λ is a weight to balance data fidelity and regularization term.
As an iterative algorithm is adopted to solve this problem, we linearize Af-g22 at point fk. Problem (11) could be expressed as12 minf⩾0Afk-g22+ATAfk-g,f-fk+γ/2·f-fk22+λ·∑p∈fBRTVp,
where γ is a positive parameter, and fk is the intermediate image at the kth iteration during reconstruction process. · represents inner product of the two vectors. Optimization problem (12) can be reformulated as the following concise form:13 minf⩾0f-fk-1/γ·ATAfk-g22+2λ/γ·∑p∈fBRTVp.
The term fk-1/γ·ATAfk-g∗ can be approximately implemented by simultaneous algebraic reconstruction technique (SART). The combination of SART and nonnegative constraints, called POCS, is to acquire the intermediate image fk+1/2. Let λ1 be equal to 2λ/γ. Therefore, problem (11) can be solved by a two-step iterative reconstruction algorithm, including POCS step [35] and BRTV step:14 fk+1/2=POCSfk,g,
15 fk+1=argminf⩾0f-fk+1/222+λ1·∑p∈fBRTVp.
To solve Eq. (15), we first discuss the x-direction measure of penalty term. It can be written as16 ∑pDx(p)L~x(p)+ε=∑p∑q∈Rpkp,q·∂xfq∑q∈Rphp,q·∂xfq+ε=∑q∑p∈Rqkp,q∑q∈Rphp,q·∂xfq+ε∂xfq≈∑q∑p∈Rqkp,qL~xp+ε1∂xfq+εg∂xfq2=∑qu~xqwxq∂xfq2
where εg is introduced for numerical stability. u~xq and wxq are set as.17 u~xq=∑p∈Rqkp,qL~xp+ε,wxq=1∂xfq+εg.
A similar process can be used to the y-direction measure in BRTV. Then, Eq. (15) can be written in a matrix form:18 vf-vfk+1/2Tvf-vfk+1/2+λ1vfTCxTU~xWxCxvf+vfTCyTU~yWyCyvf
where vf and vfk+1/2 represent the vector of f and fk+1/2, respectively. Cx and Cy are the matrices of forward difference operator. U~x, U~y, Wx, and Wy are the diagonal matrices consisting of u~xi, u~yi, wxi, and wyi, respectively. The optimization problem of Eq. (18) can be solved using the following equation:19 I+λ1CxTU~xtWxtCx+CyTU~ytWytCy·vft+1=vfk+1/2,
where I is an identity matrix. t represents the number of iterations of BRTV.
The complete workflow of the proposed POCS-BRTV model is summarized as Table 1.Table 1 LDCT reconstruction by the proposed POCS-BRTV model
Experiments and Analysis
In this section, experiments on digital photons are used for demonstrating performance of the proposed model. All experiments ran on a 3.20 GHz Intel(R) Core(TM) i7-8700 CPU with windows 10 64-bit system environments, and all the reconstruction algorithms were implemented by Matlab 2016b combining with Visual Studio 2015. All reconstructed images use root mean square error (RMSE), peak signal-to-noise ratio (PSNR), and structure similarity (SSIM) for quantitative analyses. The size of each phantom is 256×256 for all of the experiments in this paper. Projections from these phantoms are acquired via fan-beam CT, and the detector units are arranged in line. Poisson noise is added to projection data in order to simulate LDCT. The level of Poisson noise is negatively correlated with photon number I0. Specifically, a large I0 corresponds to a higher SNR. We set I0=1.0×104 and 1.0×105 for all digital photon experiments. The geometry scanning parameters in this work are set as Table 2.Table 2 Geometry scanning parameters of the simulated experiments
System parameter Parameter value
The distance from X-ray source to center of rotation 500.0 mm
The distance from center of rotation to detector 0 mm
Interval angle between two views 1
The number of detector units 372
The length of the detector 372 mm
Image size 256 × 256
Pixel size 1.0 × 1.0 mm2
In all of the experiments, the qualities of reconstructed images are highly affected by parameters, which are selected according to the evaluation indexes and visual effect. In this work, we fixed the number of max iterative reconstruction NiterMax=1000. A stopping criteria referring to errors between two successive steps is set for an earlier termination. The relaxation parameters αSART are empirically fixed as 0.15 and 0.25 corresponding to photon number I0 at 1.0 × 104 and 1.0 × 105, respectively. The value of the regularization parameter λ1 can control the smoothness of results. A larger one represents a greater smoothness. Making BRTV iteration number NBRTV larger could preserve the edge sharpness well. Because BRTV is implemented in each iteration during the reconstruction process, NBRTV set as 2 or 3 is enough. σ is a spatial parameter to control the window size for computing the windowed variations. A large σ refers to a bigger window size, as a result of which the images would be strongly smoothed. A bit larger ε helps preserve smoothly varying structures. In other words, a smaller one helps to preserve sharper edges and more fine textures. Besides, we found that the value of ε depends on image complexities, simple images corresponding to a small value of ε. It is worth noting that all of the parameters of POCS-RTV are set the same as those of POCS-BRTV for a fair comparison.
Shepp-Logan Phantom Experiment
The first simulation experiment uses a Shepp-Logan phantom to simulate LDCT case. In this experiment, the photon number is set as I0=1.0×104 and 1.0×105. Parameters for POCS-BRTV model in different corresponding cases are set as follows: (1) λ1=0.0007, NBRTV=2, σ=0.6, ε=0.000001; (2) λ1=0.00045, NBRTV=2, σ=0.5, ε=0.000001.
In Fig. 2, CT images are reconstructed using LDCT projection data via SART, POCS-TV, POCS-RTV, and POCS-BRTV. On the far left one is the noise-free phantom, in which a ROI is marked by a red rectangle and zoomed-in at the corner. It can be observed that images reconstructed by SART are severely degraded by noise. POCS-TV reconstruction can effectively reduce noise, while the results from POCS-TV are blurred by staircase effect (indicated by red arrows). Adjusting parameters could mitigate this problem, while at the same time details might be blurred during the reconstruction process. From the reconstructed images of POCS-RTV, edges indicated by red arrows are blurry, while these artifacts disappear in the results of POCS-BRTV. Clear edges can be seen from the results of POCS-BRTV. Our model represents better noise suppression and edge protection than other three algorithms.Fig. 2 The tomographic results of Shepp-Logan phantom and a zoomed-in region reconstructed by different algorithms with different noise levels. The first column is the reference image, the rest of columns are results reconstructed via SART, POCS-TV, POCS-RTV, and POCS-BRTV. Images from top to bottom are the results reconstructed with different photon numbers: 1.0 × 104 and 1.0 × 10.5. The display window is [0, 0. 5]
Table 3, listed RMSE, SSIM, and PSNR corresponding to Fig. 2, is to quantitatively analyze different reconstruction algorithms under different noise levels. The bold ones are the optimal indexes. POCS-BRTV outdistances the other three algorithms according to the indexes.Table 3 Global RMSEs, SSIMs, and PSNRs of reconstructed Shepp-Logan phantom with different photon numbers
Photon number Method RMSE SSIM PSNR
1 × 104 SART 0.0522 0.6791 25.6493
POCS-TV 0.0117 0.8248 38.6347
POCS-RTV 0.0066 0.9745 43.5711
POCS-BRTV 0.0050 0.9891 46.1031
1 × 105 SART 0.0193 0.8469 34.2898
POCS-TV 0.0041 0.9937 47.8366
POCS-RTV 0.0020 0.9962 53.8810
POCS-BRTV 0.0011 0.9985 59.0495
FORBILD Head Phantom Experiment
The second experiment is performed on the projection data of the FORBILD head phantom. Parameters for our model corresponding to I0=1.0×104 and 1.0×105 are set as follows: (1) λ1=0.01, NBRTV=3, σ=1, ε=0.0002; (2) λ1=0.004, NBRTV=3, σ=0.9, ε=0.00001.
The restored images from LDCT data are presented in Fig. 3, and the reference image is on the far left. Two ROIs are marked with red rectangles. The rest of columns are images processed by SART, POCS-TV, POCS-RTV, and POCS-BRTV, respectively. The zoomed-in ROIs corresponding to Fig. 3 are displayed in Fig. 4.Fig. 3 The tomographic results of FORBILD head phantom reconstructed by different algorithms with different noise levels. The first column is the reference image, the rest of columns are results reconstructed via SART, POCS-TV, POCS-RTV, and POCS-BRTV. Images from top to bottom are the results reconstructed with different photon numbers: 1.0 × 104 and 1.0 × 10.5. The display window is [0.5, 1.5]
Fig. 4 ROIs of the FORBILD head phantom and the reconstructed results in Fig. 3. The images in (a) are extracted from the reference FORBILD head phantom image. Images in (b) and (c) are extracted from results reconstructed with different photon numbers: 1.0 × 104 and 1.0 × 105, respectively. From top to bottom, images are ROI 1 and ROI 2. From left to right in (b) and (c), images are, respectively, reconstructed by SART, POCS-TV, POCS-RTV, and POCS-BRTV. The display window is [0.5, 1.5]
From the zoomed-in ROIs, it can be observed that all of the results from SART are severely disrupted by noise so that three minor dots in images cannot be identified even to the naked eye. Images reconstructed via POCS-TV tend to cause staircase effects, which are not smooth enough. POCS-RTV leads to over-smoothing results. Almost all of the three minor dots are smoothed by POCS-RTV. These problems are improved according to the results of POCS-BRTV. However, there is something that needs to be improved. When the noise in projection data intensifies, even though the POCS-BRTV model can significantly suppress noise, two extremely minor dots cannot be preserved in ROI1. Table 4 presents the quantitative evaluation indexes corresponding to the results in Fig. 3. The bold ones are optimal indexes. The results show that POCS-BRTV model presents the best performance at the two presented cases.Table 4 Global RMSEs, SSIMs, and PSNRs of reconstructed FORBILD head phantom with different photon numbers
Photon number Method RMSE SSIM PSNR
1 × 104 SART 0.1927 0.4384 14.2608
POCS-TV 0.0647 0.7113 23.7786
POCS-RTV 0.0294 0.8334 30.6457
POCS-BRTV 0.0274 0.8499 31.2318
1 × 105 SART 0.0725 0.6243 22.7593
POCS-TV 0.0240 0.9344 33.3779
POCS-RTV 0.0097 0.9671 40.2613
POCS-BRTV 0.0058 0.9867 44.7963
In Fig. 5, the horizontal profiles (177th row) of the reconstructed images from the incident X-ray intensity at I0=1.0×104 are provided to verify the advantages of POCS-BRTV, that it generates the nearest image, especially at edges, to the reference one. From the RIOs in Fig. 5, we can see that POCS-BRTV is able to preserve the sharp edges, while POCS-RTV tends to cause jitter on edges.Fig. 5 Horizontal profiles (177th row) of the reconstructed images shown in Fig. 3 by different algorithms with the incident X-ray intensity of 1 × 104
Discussion and Conclusion
In this paper, we proposed a LDCT reconstruction model based on the prior knowledge in gradient domain. This work extended the thinking of Gong et al. that they utilized the differences between partial derivatives generated by noise and structures, and developed POCS-RTV for low-intensity CT reconstruction [32]. POCS-RTV is defined based on WIV and WTV in the gradient domain. Structures and noise can be adaptively distinguished by WIV. However, the results from POCS-RTV tend to cause blur on edges especially under high-level noise. It might result from the weights of WIV that only depend on the distance between two pixels in a local rectangular. In fact, both nearby spatial location and similar gray values make contributions to characterize the closeness between two pixels. Therefore, we further integrated the similarity measurement with the weighting function of WIV, proposing POCS-BRTV to reconstruct LDCT. POCS-BRTV considers both position closeness and gray similarity to vicinity in the range of WIV, which helps preserve more fine details and sharp edges compared with POCS-RTV.
In summary, we propose a bilateral weighted relative total variation (POCS-BRTV) model for LDCT reconstruction in this work, integrating the thinking of bilateral filtering with POCS-RTV. Experimental results obtained at different noise levels show that the proposed POCS-BRTV model leads to significant improvements on LDCT images and especially achieves better results on preservation of edges and details during the reconstruction procedure.
Acknowledgements
This work was supported by the National Natural Science Foundation of China under Grant 61771003; Graduate Research and Innovation Foundation of Chongqing, China, under Grant CYB21044; General project of Chongqing Natural Science Foundation under Grant cstc2021jcyj-msxmX0679; and Science and Technology Research Program of Chongqing Education Commission of China under Grant KJQN202000808.
Declarations
Conflict of Interest
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Goldman, and W. L., “Principles of CT and CT technology”, J Nucl Med Technol, vol. 35, no. 3, pp. 115–128, 2007.
2. J. H. Siewerdsen, “Cone-Beam CT with a Flat-Panel Detector: From Image Science to Image-Guided Surgery”, Nucl Instrum Methods Phys Res A, vol. 648, no. supp-S1, pp. S241-S250, 2011.
3. Schubert T Jacob AL Takes M Menter T Gutzeit A Kos S CT-Guided Percutaneous Biopsy of a Mass Lesion in the Upper Presacral Space: A Sacral Transneuroforaminal Approach Cardiovascular & Interventional Radiology 2012 35 5 1255 1257 10.1007/s00270-011-0232-9 21789696
4. Frush DP Donnelly LF Rosen NS Computed Tomography and Radiation Risks: What Pediatric Health Care Providers Should Know Pediatrics 2003 112 4 951 957 10.1542/peds.112.4.951 14523191
5. Brenner DJ Hall E Computed tomography—an increasing source of radiation exposure New England Journal of Medicine 2007 357 22 2277 2284 10.1056/NEJMra072149 18046031
6. Naidich DP Marshall CH Gribbin C Arams RS Mccauley DI Low-dose CT of the lungs: Preliminary observations Radiology 1990 175 3 729 731 10.1148/radiology.175.3.2343122 2343122
7. Dhawan AP Rangayyan RM Gordon R Image restoration by Wiener deconvolution in limited-view computed tomography Applied Optics 1985 24 23 4013 10.1364/AO.24.004013 18224156
8. Li T Li X Wang J Wen J Lu H Hsieh J Liang Z Nonlinear sinogram smoothing for low-dose X-ray CT IEEE Transactions on Nuclear Science 2004 51 5 2505 2513 10.1109/TNS.2004.834824
9. J. Shtok, M. Elad, and M. Zibulevsky, “Sparsity-based sinogram denoising for low-dose computed tomography”, in IEEE ICASSP, 2011, pp. 569–572.
10. Liu J Ma J Zhang Y Chen Y Yang J Shu H Luo L Coatrieux G Yang W Feng Q Discriminative Feature Representation to Improve Projection Data Inconsistency for Low Dose CT Imaging IEEE Transactions on Medical Imaging 2018 36 12 2499 2509 10.1109/TMI.2017.2739841
11. Q. Xie, D. Zeng, Q. Zhao, D. Meng, Z. Xu, Z. Liang, and J. Ma, “Robust Low-Dose CT Sinogram Preprocessing via Exploiting Noise-Generating Mechanism”, IEEE Trans Med Imaging, vol. PP, no. 12, pp. 2487–2498, 2017.
12. Karimi D Ward RK Sinogram denoising via simultaneous sparse representation in learned dictionaries Physics in Medicine & Biology 2016 61 9 3536 10.1088/0031-9155/61/9/3536 27055224
13. Lu Y Zhao J Wang G Few-view image reconstruction with dual dictionaries Physics in Medicine & Biology 2011 57 1 173 189 10.1088/0031-9155/57/1/173
14. Q. Xu, H. Y. Yu, X. Q. Mou, L. Zhang, J. Hsieh, and G. Wang, “Low-Dose X-ray CT Reconstruction via Dictionary Learning”, IEEE Transactions on Medical Imaging, 2012.
15. Chun SY Dewaraja YK Fessler JA Alternating Direction Method of Multiplier for Tomography With Nonlocal Regularizers IEEE Transactions on Medical Imaging 2014 33 10 1960 1968 10.1109/TMI.2014.2328660 25291351
16. Buades A Coll B Morel JM Nonlocal Image and Movie Denoising International Journal of Computer Vision 2008 76 2 123 139 10.1007/s11263-007-0052-1
17. Hu Z Xue H Zhang Q Gao J Zhang N Zou S Teng Y Liu X Yang Y Liang D Zhu X Zheng H DPIR-Net: Direct PET Image Reconstruction Based on the Wasserstein Generative Adversarial Network IEEE Transactions on Radiation and Plasma Medical Sciences 2021 5 1 35 43 10.1109/TRPMS.2020.2995717
18. K. Dabov, A. Foi, V. Katkovnik, and K. Egiazarian, “Image denoising with block-matching and 3D filtering”, Proc. SPIE-IS&T Electronic Imaging, pp. 354–365, 2006.
19. Dabov K Foi A Katkovnik V Egiazarian K Image denoising by sparse 3-D transform-domain collaborative filtering IEEE Transactions on Medical Imaging 2007 16 8 2080 2095 10.1109/TIP.2007.901238
20. Kalra LS Cong W Wang G 3D Convolutional Encoder-Decoder Network for Low-Dose CT via Transfer Learning from a 2D Trained Network IEEE Transactions on Medical Imaging 2018 3 7
21. M. Patwari, R. Gutjahr, R. Raupach, and A. Maier, “JBFnet -- Low Dose CT Denoising by Trainable Joint Bilateral Filtering”, 2020.
22. K. Choi, J. S. Lim, and S. K. Kim, “StatNet: Statistical Image Restoration for Low-Dose CT using Deep Learning”, no. 99, pp. 1–1, 2020.
23. Shiri I Akhavanallaf A Sanaat A Salimi Y Askari D Mansouri Z Shayesteh SP Hasanian M Rezaei-Kalantari K Salahshour A Sandoughdaran S Abdollahi H Arabi H Zaidi H Ultra-low-dose chest CT imaging of COVID-19 patients using a deep residual neural network Eur Radiol 2021 31 3 1420 1431 10.1007/s00330-020-07225-6 32879987
24. Gholizadeh-Ansari M Alirezaie J Babyn P Deep Learning for Low-Dose CT Denoising Using Perceptual Loss and Edge Detection Layer Journal of Digital Imaging 2020 33 2 504 515 10.1007/s10278-019-00274-4 31515756
25. Gao Y Tan J Shi Y Lu S Gupta A Li H Liang Z Constructing a tissue-specific texture prior by machine learning from previous full-dose scan for Bayesian reconstruction of current ultralow-dose CT images J Med Imaging (Bellingham) 2020 7 3 032502 32118093
26. He Y Zeng L Yu W Gong C Noise suppression-guided image filtering for low-SNR CT reconstruction Medical & biological engineering & computing 2020 58 11 2621 2629 10.1007/s11517-020-02246-1 32839918
27. K. He, J. Sun, X. Tang, and m. intelligence, “Guided image filtering”, IEEE transactions on pattern analysis, vol. 35, no. 6, pp. 1397–1409, 2012.
28. E. Y. Sidky, Y. Duchin, X. Pan, and C. Ullberg, “A constrained, total-variation minimization algorithm for low-intensity x-ray CT”, Medical Physics, vol. 38 Suppl 1, no. S1, 2011.
29. Tian Z Jia X Yuan K Pan T Jiang SB Low-dose CT reconstruction via edge-preserving total variation regularization Physics in Medicine & Biology 2011 56 18 5949 5967 10.1088/0031-9155/56/18/011 21860076
30. Yu W Wang C Huang M Edge-preserving reconstruction from sparse projections of limited-angle computed tomography using ℓ 0-regularized gradient prior Review of Scientific Instruments 2017 88 4 043703 10.1063/1.4981132 28456252
31. Xu L Yan Q Xia Y Jia J Structure extraction from texture via relative total variation ACM Transactions on Graphics 2012 31 6 1 10
32. Gong C Zeng L Adaptive iterative reconstruction based on relative total variation for low-intensity computed tomography Signal Processing 2019 165 149 162 10.1016/j.sigpro.2019.06.031
33. C. Tomasi, “Bilateral filtering for gray and color images”, Proc.ieee Inter.conf.computer Vision, 1998.
34. Kundu R A Novel Technique for Texture and Edge Preservation Using Bilateral Filter National Academy Science Letters 2020 44 2 117 124 10.1007/s40009-020-00921-3
35. Kak AC Slaney M Wang G Principles of computerized tomographic imaging Medical Physics 2002 29 1 107 10.1118/1.1455742
| 0 | PMC9707191 | NO-CC CODE | 2022-12-01 23:20:21 | no | manag kinderopvang. 2022 Nov 28; 28(6):54-58 | latin-1 | null | null | null | oa_other |
==== Front
J Gen Intern Med
J Gen Intern Med
Journal of General Internal Medicine
0884-8734
1525-1497
Springer International Publishing Cham
36443630
7959
10.1007/s11606-022-07959-7
Perspective
Moving Away from Chaos: Intentional and Adaptive Management of the Non-visit Care River
http://orcid.org/0000-0001-9319-5457
Matulis John C. DO, MPH [email protected]
1
McCoy Rozalina MD, MS 1
Liu Stephen K. MD, MPH 2
1 grid.66875.3a 0000 0004 0459 167X Division of Community Internal Medicine, Geriatrics and Palliative Care, Mayo Clinic, 200 1st ST SW, Rochester, MN 55905 USA
2 grid.254880.3 0000 0001 2179 2404 Department of Medicine, Geisel School of Medicine at Dartmouth, Hanover, NH USA
28 11 2022
15
19 6 2022
15 11 2022
© The Author(s), under exclusive licence to Society of General Internal Medicine 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
In modern primary care practice, clinicians face increasing volumes of asynchronous, electronic, non-visit care (NVC). Systems for completing this work, however, remain under-developed and often lack definition around patient and practice expectations for work completion and team member contributions. The resulting reactive, unstructured, and unscheduled NVC workflows cause and exacerbate physicians’ cognitive overload, distraction, and dissatisfaction. Herein, we propose that primary care practices take an intentional, holistic approach to managing systems of NVC and offer a conceptual model for managing NVC work, analogizing the flow of these tasks to the flow of water through a river system: (1) by carefully controlling the inputs into the NVC system (the tributaries entering the river system); (2) by carefully defining the workflows, roles and responsibilities for completion of common tasks (the direction of river flow); (3) by improving the interface of the electronic health record (obstacles encountered in the river); and (4) by optimizing effectiveness of primary care teams (the contours of the river determining rate of flow). This framework for managing NVC, viewed from a broader system perspective, has the potential to improve productivity, quality of care, and clinician work experience.
KEY WORDS
primary health care
health care quality
professional burnout
workflow
patient portals
==== Body
pmc The pieces weren’t fitting together. My 2:00 p.m. patient had just described a confusing constellation of symptoms and a vague temporal profile; lacking an obvious diagnostic pivot point, I ordered basic labs and promised to follow-up soon. I hoped that taking 5 minutes to organize my thoughts while putting together the clinical note could offer some structure to my evolving diagnostic framework. As I logged on to my computer and opened the electronic health record (EHR), there it was… again. In bright red font, an upward arrow indicating a “high priority” patient message next to a bold number 4, informing me of the number of new messages I had been assigned to address, but hadn’t yet reviewed. The attention switch was unavoidable. What if someone was dying? What if I really screwed something up? After clicking into the message, I learned a patient was asking for an urgent refill of their antibiotics, yet the single line of clinical information was insufficient to judge the best course of action. 15 minutes later, following a clarifying phone call to the patient, a phone note rendered, and a prescription sent, I was able to hit the done button on the message. A quick view of my schedule informed me that my 2:30 p.m. patient has been waiting for me for 15 minutes and my 3:00 patient is well into the process of being roomed. My motivation to think deeply about my puzzling patient lost, I move on and focus on the next patient. I finish the day, go home, have dinner with the family. I help put the kids to bed and log on to the computer at 8:30 pm. I stare at the in basket. The vague unease and resentment I felt last night quickly seeps back in. Tomorrow, I promise, I will leave with all my work done.
William Osler extolled us to “cultivate the power of concentration, which grows with its exercise”.1 Piecing together complex and disjointed information, creating a coherent patient narrative while integrating the unique psychosocial identities of the patient into clinical decision-making demands focus, attention, and clarity of thought. Internal medicine, at its best, enables accessible, high-quality, patient-centered care with timely and accurate diagnoses, cost-effective evaluation and management, and improved health outcomes for patients entrusting their health care to us.2 Reliable, clear, and manageable non-visit care (NVC) workflows are not only important for effectively executing the cognitively demanding work necessary for patient care, but they are also necessary for the physician’s physical, social, and emotional well-being.3,4 The unpredictable flow of non-patient facing work and low levels of control over clinical work are well-recognized contributors to burnout among primary care physicians.5,6
In-basket message volumes have increased in many health care systems since the start of the COVID-19 pandemic,7,8 with time spent on in-basket work higher for female physicians, and those caring for older and sicker patients.9–11 Asynchronous, electronic patient care remains undervalued in modern primary care practice. The time allocated for answering these messages is inadequate, nor purposefully supported by the care team in a way that optimizes the clinician’s role in the process.12–15
The development of asynchronous, directly pushed, electronic NVC has created an unrelenting conveyer belt of unscheduled and reactive messages seen in many primary care practices. Electronically routed NVC tasks do not require delivery from a trusted team member, and can instead be sent on by employees who may have never met the physician or the patient being cared for.13 These unscheduled tasks are relegated to time that does not exist on the physician’s calendar, with other team members unaware of how, when, or where the physician would complete these tasks. The physician, nurse, or administrative staff member may be drowning in messages while their fellow care team members remain busy and unaware, separately sequestered in their own electronic world.
While professional ethics dictate that physicians sacrifice themselves to meet a patient’s needs,16 there is always a breaking point. How many messages with inadequate information, unclear patient expectations, or assigned tasks below their skill level can be sent to the frantically busy primary care physician before that breaking point arrives? That breaking point may result in cutting back clinical FTE, seeking administrative or non-primary care clinical responsibilities, or leaving primary care to reinvent one’s career. We can and must do better. And better is possible.
What Does the Literature Say?
There is a growing body of literature discussing the challenges posed by electronic in-baskets and potential steps for improvement. A thematically organized summary of selected recent literature is described in Table 1. A recent call to action from subject matter experts on primary care transformation translates this literature base into steps which health care systems, payors, and regulators can take to begin addressing NVC overload.17 While these prescriptions are well reasoned, physician leaders and their administrative partners looking for actionable and feasible solutions may find the recommendations daunting and wonder where to begin. Herein, we propose a conceptual framework for NVC that primary care physician leaders and their administrative partners can leverage to develop a more intentional, balanced, and adaptive system of NVC (Table 2). Table 1 Summary of selected NVC literature and recommendations for improvement
EHR in-basket challenge Specific recommendations Citing articles
Low-value message content sent to clinician Automate low-value messages out of in-basket Murphy et al.28, Shah et al.29, Sinsky et al.17, Tai-Seale et al.12
Insufficient time provided Delegation of messages to optimized care team Sinsky et al.17, Murphy et al.28
Sub-optimal team member support Team-based training, expanded roles and responsibilities, protocols and standing orders Murphy et al.28, Dymek et al.30, Winner31, Smith et al.24
Message processing and interface too complex Rapid access to contextual information (appointments, tests, medications, notes) Murphy et al.28, Dymek et al.30
Clinic flows not streamlined “EHR Sprints,” improved documentation tools Sieja et al.26, Shah et al.29
Existing tools in EHR not optimized Create EHR templates and individual work protocols Winner31
Table 2 Intentional and adaptive vs. chaotic NVC systems
NVC system component Chaotic NVC management approach Intentional, adaptive NVC management approach River analogy
Undifferentiated patient requests Variable and unclear criteria regarding which requests proceed through NVC workflows Carefully defined and clearly communicated criteria identifying requests which should and should not proceed through NVC workflows Defined tributaries form a river
Workflow for selected NVC tasks Unclear roles and responsibilities; individuals unsure of their requested action in completing NVC work Clearly defined roles and responsibilities; measured volumes of workload and staffing adjustment as appropriate A river follows a clear and well-defined path to its terminus
EHR interface Static EHR interface rendering a high-cognitive load and limited functionality Iterative improvements to EHR interfaces reduce cognitive load and improve user experience A river adapts and changes course in response to obstacles
Primary care team effectiveness Failure not seen as opportunity to improve, ineffective communication, inadequate care team structures Failures result in learning, effective communication, and dynamic multidisciplinary team structures River contours and gradient enable reliable and consistent flow
A Better Path Forward: the Adaptive NVC System
Our conceptual framework begins with analogizing the flow of NVC tasks to 4 sequential factors influencing the flow of water through a river: (1) the inputs of tasks entering the NVC system, corresponding to the volume of water entering the main river from its tributaries; (2) the specific direction, sequence, and assigned team members that the NVC tasks flow to, corresponding to the course the riverbed follows to reach its destination; (3) the EHR interface of tasks and communications, corresponding to the rocks, islands, and other obstacles found within the riverbed; and (4) the effectiveness of the primary care team in moving NVC work forward, corresponding to the contours of the riverbed, which determine the rate of flow. As NVC tasks move through a system of work, task completion is impacted by the volume and complexity of tasks entering the system, the specific pathway assigned NVC tasks follow to completion, the structural features and obstacles which disrupt planned workflows, and the effectiveness with which primary care teams complete their assigned work. The river flow analogy allows us to visualize the multitude of factors impacting not only NVC completion, but the burden (water turbulence) and collateral damage (erosion) that poorly executed NVC tasks have on the primary care system broadly, and the primary care physician in particular.
If the NVC system has too many tasks entering, if the workflows or EHR processes impede efficient completion of the work, or if the front-line team functions ineffectively, the NVC work will overwhelm the underlying system and the people within it. Just as a river system adapts to changes in precipitation or habitat around its watershed, the primary care practice must adapt its systems of work to changing patient needs and variable primary care resources. To move towards a system with a more balanced and steady flow of NVC work, we propose that the management of NVC work should be targeted towards the following four domains: controlling NVC inputs, defining NVC workflows, improving the EHR interface, and optimizing primary care team effectiveness.
Managing the NVC Tributaries or Controlling NVC Inputs
The selection of tasks which flow into the primary care team’s NVC work system, or the process by which patient requests are transferred from an intake point to the primary care team’s established workflows, is a critical step in calibrating the volume and content of NVC work to the primary care team’s capacity and skillsets. The entry of NVC tasks not only determines the volume of NVC work entering the system, but also defines the content of the daily work of the primary care team. Indistinct boundaries surrounding the work of the primary care team, including lack of clarity on which team member should be assigned a particular task and how that assigned work should be completed, are important contributors to primary care physician dissatisfaction.18 The creation of structured and transparent NVC processes presents an opportunity for primary care physician leaders and their administrative partners to seek input and provide clarification on which tasks should be done by the front-line primary care team and which tasks can be delegated to non-primary care team support staff, as well as whether that work should be completed through visit (traditional or telemedicine) or NVC workflows. Common requests which should and should not enter the primary care team NVC system must be clearly established by the practice, and the best way for patients to make common requests needs to be consistently communicated to both team members and patients. Table 3 provides hypothetical examples of common patient requests that could be considered for entry into the primary care team’s NVC system. Table 3 Hypothetical dispositions for common patient requests
Patient request Disposition of that request Next steps NVC system component addressed Rationale
Patient reports updated blood pressure and blood glucose readings Enter the primary care team NVC system Organize and format necessary data; seek clarifying medication information from the patient when appropriate ➔ communicate results to the physician with clear demarcation of normal vs. abnormal Defining workflows and optimizing primary care team effectiveness Effective chronic disease management requires frequent NVC touches
Patient requesting lab work for new symptoms Do not enter the primary care team NVC system Automatically schedule telemedicine or in-person visit Controlling inputs Patient history and differential diagnosis are required before test ordering to promote evidence-based, cost-effective care
Patient requesting letter or form completion after recent visit Enter the primary care team NVC system Obtain specific information from the patient related to required documentation elements ➔ route to administrative team member to draft letter for physician signature Defining workflows and optimizing primary care team effectiveness External communication is a core NVC service
COVID-19 vaccination or testing questions Do not enter the primary care team NVC system Non-primary care team member responds to request using protocols established by the practice Controlling inputs Protocolized processes improve reliability and efficiency of communicated information
Patient asks question related to a prescribed medication Enter the primary care team NVC system Develop an interface which obtains necessary information related to patient’s question and automatically route the question to the appropriate pharmacist or prescriber Defining workflows and improving EHR interface Limited, well-defined concerns can be efficiently and appropriately addressed by NVC
Patient requesting routine health maintenance task completion Do not enter the primary care team NVC system Non-primary care team members respond to the request using protocols established by the practice or automated health maintenance outreach systems Controlling inputs Where possible, standardized outreach process and protocols should replace physician chart review and order entry
Defining NVC Workflows
The workflow an NVC task follows before reaching its endpoint is where the selected NVC task is clarified, necessary information is added to the request as appropriate, and the task is routed to the most appropriate team member who has a clear understanding of their role in completing this task. Robust workflows in this part of the system are the backbone of a well-managed NVC system. HealthPartners Medical Group in St. Paul, MN,19 after comprehensive restructuring of in-basket folders and redefining team member roles, specified and reengineered in-basket workflows, achieving significant improvements in physician efficiency and satisfaction. Clearly defining workflows for common NVC tasks was felt to be foundational for the improvements observed. The lesson from this work is that once well-designed and clear workflows are established, common tasks can be completed with a reasonable level of variation and improved efficiency when outcomes such as message turn-around time, communication quality, and patient experience are evaluated.
Improving the EHR Interface
Primary care physician leaders and their administrative partners should also address EHR features which create turbulence in NVC workflows. Redesigning the EHR to improve user experience can yield significant improvements in NVC efficiency and physician wellness.20,21 At the University of Colorado, real-time redesign of the EHR interface resulted in improved end-user experience. During brief and focused periods of time, team members, informatics leaders, and process engineers conducted “EHR sprints” consisting of EHR training, user customization, and real-time EHR interface redesign, collectively resulting in significant improvements in end-user satisfaction.21 For practices that are just getting started, simple EHR optimizations such as reducing the number of in-basket categories and removing automated notifications and unnecessary tasks can significantly improve the cognitive burden 22 faced by front-line team members. EHR training, protected time for EHR personalization, and “at the elbow” support to improve user efficiency may also be important adjuncts to reducing this turbulence created in the EHR interface.21
Optimizing the Effectiveness of the Primary Care Team
Successful and sustainable NVC systems require high-functioning primary care teams. Trust among team members, predicated on team members trusting their colleagues to accurately and reliably identify the patient’s need and the urgency for action, serves as an important elixir to the angst provoked by messages arriving in the in-basket with uncertain expected actions and prioritization. The principles that make this team trust possible, and on which high-functioning primary care teams rely, have been previously outlined 23,24 and include promoting shared decision-making among team members, fostering effective interprofessional communication, learning from failure, and creating organizational structures that support multidisciplinary teamwork. For example, a primary care practice at Bellin Health in Wisconsin redesigned care team roles and processes alongside systematic efforts at cultural change and team building resulting in dramatically expanded primary care team capacity.25,26
Finding a Place to Start
Primary care physician leaders and their administrative partners, recognizing the challenge of developing better systems for managing NVC workflows, need not “conquer the in-basket,” but should instead seek to make iterative and continuous improvements to existing NVC systems. Individual leaders will need to focus improvement efforts differently based on their patient population’s needs, clinic resources, and electronic and regulatory environments. Primary care practices which have successfully improved NVC systems have done so by considering the design of the entire NVC system and by clearly identifying the optimal roles and workflows for each care team member.19,25,27 A logical starting point is to ask team members which NVC processes most contribute to team frustration. Involving ancillary and non-clinician staff such as medical assistants, nursing team members, scheduling staff and medical administrative assistants in the redesign will be critical for these staff to maximally utilize their professional skills and optimally contribute to the work of the high-functioning primary care team.
Conclusion
In the modern primary care practice, the evolution of unstructured systems of NVC poses significant challenges to the work experience of primary care physicians and team members. Developing trusted systems for managing NVC, where tasks are intentionally and clearly selected for NVC completion, where selected tasks default to mutually agreed upon, reliable workflows, and where those workflows are executed by a high-functioning primary care team, will significantly improve the experience of the NVC work that makes up a substantial part of the work-life of contemporary primary care teams. Success in these systems is dependent on front-line team member engagement, role clarity, task clarity, and building trusting relationships among team members. While redesigning these processes may appear onerous and costly when compared with continuing the status quo of reactive and ad hoc systems of electronic work, the rewards of diminished frustration, enhanced team member relationships and connection, and better patient care are not only worth the effort, but necessary for the primary care clinic to remain a place where physicians want to continue to work and thrive for the duration of their careers.
Acknowledgements
We would like to thank Dr. Kelly Kieffer, MD, Dartmouth-Hitchcock Medical Center, and Dr. Rajeev Chaudhry, MBBS, Mayo Clinic School of Medicine, for their helpful review and feedback on this manuscript.
Author Contribution
John C. Matulis: Dr. Matulis is an Assistant Professor of Medicine at Mayo Clinic in Rochester, MN. He was involved in the planning, conceptualization, and design of the article and drafted, revised, and finalized the current manuscript. The initial vignette is representative of his experience in balancing non-visit care with other expected clinical work as an early career internist.
Rozalina McCoy: Dr. McCoy is an Associate Professor of Medicine at Mayo Clinic in Rochester, MN. She assisted with the planning, organization, writing, and finalization of the manuscript. She reviewed and revised all sections of the manuscript, and approved the final manuscript as submitted.
Stephen K. Liu: Dr. Liu is an Associate Professor of Medicine at the Geisel School of Medicine at Dartmouth in Hanover, NH. He conceptualized and planned the project with Dr. Matulis. He assisted with the planning, organization, writing, and finalization of the manuscript. He reviewed and revised all sections of the manuscript, and approved the final manuscript as submitted.
Declarations
Conflict of Interest
In the last 36 months, Dr. McCoy has received unrelated grant support from NIDDK, PCORI, and AARP®. She also serves as a consultant to Emmi® (Wolters Kluwer) on developing patient education materials related to prediabetes and diabetes.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Osler W The quotable Osler 2008 Philadelpia, PA. ACP Press
2. What is a Doctor of Internal Medicine, or Internist? The American College of Physicians, 2022. (Accessed May 15th, 2022, at https://www.acponline.org/acp-newsroom/what-is-a-doctor-of-internal-medicine-or-internist-0.)
3. Linzer M Poplau S Babbott S Worklife and wellness in academic general internal medicine: Results from a national survey J Gen Intern Med 2016 31 1004 10 10.1007/s11606-016-3720-4 27138425
4. Beasley JW Wetterneck TB Temte J Information chaos in primary care: Implications for physician performance and patient safety J Am Board Fam Med 2011 24 745 51 10.3122/jabfm.2011.06.100255 22086819
5. Rassolian M Peterson LE Fang B Workplace factors associated with burnout of family physicians JAMA Intern Med 2017 177 1036 8 10.1001/jamainternmed.2017.1391 28492847
6. Olson K Sinsky C Rinne ST Cross-sectional survey of workplace stressors associated with physician burnout measured by the Mini-Z and the Maslach Burnout Inventory Stress Health 2019 35 157 75 10.1002/smi.2849 30467949
7. Holmgren AJ Downing NL Tang M Sharp C Longhurst C Huckman RS Assessing the impact of the COVID-19 pandemic on clinician ambulatory electronic health record use J Am Med Inform Assoc 2021 29 453 60 10.1093/jamia/ocab268
8. Nath B Williams B Jeffery MM Trends in electronic health record inbox messaging during the COVID-19 pandemic in an ambulatory practice network in New England JAMA Netw Open 2021 4 e2131490-e 10.1001/jamanetworkopen.2021.31490 34636917
9. Akbar F Mark G Warton EM Physicians’ electronic inbox work patterns and factors associated with high inbox work duration J Am Med Inform Assoc 2021 28 923 30 10.1093/jamia/ocaa229 33063087
10. Margolius D Siff J Teng K Einstadter D Gunzler D Bolen S Primary care physician factors associated with inbox message volume J Am Board Fam Med 2020 33 460 2 10.3122/jabfm.2020.03.190360 32430380
11. Rittenberg E, Liebman JB, Rexrode KM. Primary care physician gender and electronic health record workload. J Gen Intern Med 2022.
12. Tai-Seale M Dillon EC Yang Y Physicians’ well-being linked to in-basket messages generated by algorithms in electronic health records Health Aff 2019 38 1073 8 10.1377/hlthaff.2018.05509
13. McMahon LF Rize K Irby-Johnson N Chopra V Designed to fail? The future of primary care J Gen Intern Med 2021 36 515 7 10.1007/s11606-020-06077-6 32728962
14. Murphy DR Reis B Sittig DF Singh H Notifications received by primary care practitioners in electronic health records: A taxonomy and time analysis Am J Med 2012 125 209.e1-7 10.1016/j.amjmed.2011.07.029 22269625
15. Murphy DR Meyer AND Russo E Sittig DF Wei L Singh H The burden of inbox notifications in commercial electronic health records JAMA Intern Med 2016 176 559 60 10.1001/jamainternmed.2016.0209 26974737
16. Ofri D. The business of health care depends on exploiting doctors and nurses. The New York Times 2019 June 8, 2019.
17. Sinsky CA, Shanafelt TD, Ripp JA. The electronic health record inbox: Recommendations for relief. J Gen Intern Med 2022:1-2.
18. Agarwal SD Pabo E Rozenblum R Sherritt KM Professional dissonance and burnout in primary care: A qualitative study JAMA Intern Med 2020 180 395 401 10.1001/jamainternmed.2019.6326 31904796
19. AMA Stepsforward. Re-structuring EHR in-baskets in Minneapolis-St. Paul, MN: A case study. Published 2018. Accessed September 18, 2022. https://edhub.ama-assn.org/steps-forward/module/2702709.
20. Melnick ER Harry E Sinsky CA Perceived electronic health record usability as a predictor of task load and burnout among US physicians: Mediation analysis J Med Internet Res 2020 22 e23382 10.2196/23382 33289493
21. Sieja A, Markley K, Pell J, et al. Optimization sprints: Improving clinician satisfaction and teamwork by rapidly reducing electronic health record burden. Mayo Clin Proc; 2019: Elsevier. p. 793-802.
22. Ashton M Getting rid of stupid stuff N Engl J Med 2018 379 1789 91 10.1056/NEJMp1809698 30403948
23. Sullivan EE Ibrahim Z Ellner AL Giesen LJ Management lessons for high-functioning primary care teams J Healthc Manag 2016 61 449 65 28319964
24. Smith CD, Balatbat C, Corbridge S, et al. Implementing optimal team-based care to reduce clinician burnout. NAM Perspectives. 2018 Sep 17.
25. Jerzak J Radical redesign: The power of team-based care Ann Fam Med 2017 15 281 10.1370/afm.2072 28483901
26. Jerzak J Using empowered CMAs and nursing staff to improve team-based care Fam Pract Manag 2019 26 17 22 30645084
27. Jerzak J, Sinsky C. EHR in-basket restructuring for improved efficiency. American Medical Association Steps Forward 2017.
28. Murphy DR, Giardina TD, Satterly T, Sittig DF, Singh H. An exploration of barriers, facilitators, and suggestions for improving electronic health record inbox-related usability: A qualitative analysis. JAMA Network Open 2019;2:e1912638-e.
29. Shah T, Kitts AB, Gold JA, et al. Electronic health record optimization and clinician well-being: A potential roadmap toward action. NAM perspectives 2020;2020.
30. Dymek C Kim B Melton GB Payne TH Singh H Hsiao C-J Building the evidence-base to reduce electronic health record–related clinician burden Journal of the American Medical Informatics Association 2020 28 1057 61 10.1093/jamia/ocaa238
31. Winner J Practical Ways to Manage Your EHR Inbox Fam Pract Manag 2021 28 27 30 34254765
| 36443630 | PMC9707193 | NO-CC CODE | 2022-12-01 23:20:21 | no | J Gen Intern Med. 2022 Nov 28;:1-5 | utf-8 | J Gen Intern Med | 2,022 | 10.1007/s11606-022-07959-7 | oa_other |
==== Front
Anaesthesiologie
Anaesthesiologie
Die Anaesthesiologie
2731-6858
2731-6866
Springer Medizin Heidelberg
36441224
1229
10.1007/s00101-022-01229-1
Einführung zum Thema
Weaning beginnt bereits mit der Beatmung
Zentraler Bestandteil des intensivmedizinischen Alltags
Weaning already begins with ventilationCore component of routine intensive medical care
Moerer Onnen [email protected]
grid.7450.6 0000 0001 2364 4210 Klinik für Anästhesiologie, Universitätsmedizin Göttingen, Georg-August-Universität Göttingen, Göttingen, Deutschland
28 11 2022
13
7 11 2022
© The Author(s), under exclusive licence to Springer Medizin Verlag GmbH, ein Teil von Springer Nature 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
==== Body
pmcDie invasive maschinelle Beatmung ist aus der modernen Intensivmedizin nicht wegzudenken und schafft neben der direkten Behandlung der respiratorischen Insuffizienz v. a. die Zeit für die Behandlung der zugrunde liegenden Ursache [1]. Jedes Jahr werden weltweit mehr als 1 Mio. Patienten wegen eines akuten respiratorischen Versagens beatmet. Die Beatmung stellt eine potenziell lebensrettende Maßnahme dar, und wir haben heute Technologien und -verfahren an der Hand, die eine differenzierte und individualisierte Beatmung ermöglichen. Eine vorzeitige Beendigung der Beatmung oder ein fehlgeschlagener Extubationsversuch mit Reintubation birgt das Risiko einer erhöhten Rate von beatmungsassoziierten Pneumonien und hat Einfluss auf das Outcome der Patienten. Auf der anderen Seite kann eine prolongierte Beatmung zu lebensbedrohlichen Komplikationen führen, ist mit einer unnötig langen Sedierungsdauer, längeren Verweildauer auf der Intensivstation und einem hohem intensivmedizinischen Ressourcenverbrauch, Morbidität und Mortalität verbunden [2, 3]. Das gilt insbesondere für das prolongierte Weaning, ein häufiges und klinisch relevantes Problem [3].
Jährlich werden weltweit mehr als 1 Mio. Patienten wegen akuten respiratorischen Versagens beatmet
Aufgrund des demografischen Wandels und der wachsenden Zahl älterer Menschen in Deutschland sowie der Multimorbidität beatmeter Patienten wird die Zahl der Patienten mit prolongiertem Weaning zukünftig weiter steigen.
Ziel ist es, die Beatmung zum frühestmöglichen Zeitpunkt zu beenden.„Das Weaning beginnt mit der Beatmung“
Damit ist der erste Schritt der, das Team für dieses Ziel zu sensibilisieren und die Rahmenbedingungen zu hinterfragen. Ist die Grunderkrankung ausreichend therapiert? – Muss mein Patient oder meine Patientin so tief sediert sein? Muss sie oder er überhaupt sediert sein? Ist mein Zielwert auf der Richmond Agitation Sedation Scale (RASS) erreicht? – Muss er oder sie kontrolliert beatmet werden? Toleriert sie oder er eine assistierende Beatmung? Und wenn ja, ist die Höhe der Druckunterstützung wirklich erforderlich? … Wie viel Atemarbeit muss mein Patient leisten? Wie viel kann er leisten? … Stimmt die Flüssigkeitsbilanz? Ist die Lunge noch zu feucht? Ist mein Patient mangel- oder überernährt? Ist eine Mobilisation begonnen worden und möglich? …
Die hier aufgeführten Fragen sind exemplarisch zu sehen. Es sind die üblichen Fragen, die sich wie an einer Perlschnur aufreihen lassen und am Ende jedes Organ einbeziehen. Und wir merken nicht nur, dass alle Fragen direkt oder indirekt die Lunge bzw. das respiratorische System ins Visier nehmen, sondern auch, dass wir am Ende Zeit und Personal brauchen, um die Antworten auf unsere Fragen zu erarbeiten und die therapeutischen Konsequenzen zu ziehen.
Beim Großteil der Patienten gestaltet sich der Prozess problemlos, stellt den Intensivmediziner bei Patienten im prolongierten Weaning allerdings vor teilweise große Herausforderungen und verursacht einen erheblichen Teil der Arbeitsbelastung des Personals auf einer Intensivstation [4]. Nicht selten liegt eine hochkomplexe Kombination folgender Faktoren vor:inkomplette pulmonale Resolution des aktuellen Ereignisses (z. B. der Pneumonie),
Folgen der prolongierten Beatmung (eingeschränkte Lungenfunktion infolge beatmungsinduzierter Lungenschäden, ventilatorinduzierte Funktionsstörung des Zwerchfells),
direkte Folgen der kritischen Erkrankung (z. B. Critical-Illness-Polyneuropathie und -Myopathie, septische Kardiomyopathie) und
vorbestehende Komorbiditäten mit direkter Auswirkung (z. B. Herz- und Niereninsuffizienz).
Aufgrund der Bedeutung der Entwöhnung von der mechanischen Beatmung steht das Weaning im kontinuierlichen wissenschaftlichen Fokus, mit dem Ziel der Ermittlung der effektivsten und effizientesten Methoden. Darüber hinaus ist es auch Gegenstand von Initiativen zur Qualitätsverbesserung, um Evidenz und bewährte Verfahren umzusetzen. Hierzu gehört die interdisziplinäre und interprofessionelle Erstellung von Leitlinien [5–7], aber auch die Zertifizierung der Entwöhnung von der Beatmung in der Intensivmedizin, die sich die Deutsche Gesellschaft für Anästhesiologie und Intensivmedizin (DGAI) und andere zur Aufgabe gemacht haben [8].
Der komplexe Prozess des Weanings muss in einem ganzheitlichen Ansatz adressiert werden
In dieser Ausgabe der Zeitschrift Die Anaesthesiologie fassen Prof. Johannes Bickenbach und Dr. Sebastian Fritsch die Herausforderungen zusammen, die die Entwöhnung von der Beatmung an uns stellt [9]. Hierbei ist es den Autoren in ihrem sehr lesenswerten Artikel gelungen, einen großen allumfassenden Bogen zum Thema Weaning zu schlagen. Beginnend mit aktuellen Definitionen und Klassifikationen sowie der Pathophysiologie informiert der Beitrag über allgemeine therapeutische Strategien, Prädiktoren des Weaning-Erfolgs, den Stellenwert automatisierter Weaning-Algorithmen und die Besonderheiten bei COVID-19 bedingtem Lungenversagen. Last but not least werden die außerklinische Beatmung sowie das langfristige Outcome nach einfachem und prolongiertem Weaning behandelt. Die Autoren greifen für den Leser zentrale Aspekte und Empfehlungen zum Weaning auf, wie sie die umfangreichen nationalen Leitlinien [4–6], insbesondere die kürzlich revidierte S2k-Leitlinie zum prolongierten Weaning [5], zum Gegenstand haben und ergänzen diese um aktuelle Studien auf diesem Gebiet. Bickenbach und Fritsch betonen zu Recht, dass der komplexe Prozess des Weanings in einem ganzheitlichen Ansatz adressiert werden muss.
Beatmungskomplikationen vermeiden heißt auch, prolongierte Beatmung und Schwierigkeiten beim Weaning vermeiden. Bickenbach und Fritsch widmen sich deshalb den Komplikationen einer nichtprotektiven Beatmungseinstellung und den Problemen einer kontrollierten Beatmung. Der Vermeidung von „ventilator-associated pneumonia“ (VAP), „ventilator-induced lung injury“ (VILI) und „ventilator-induced diaphragmatic dysfunction“ (VIDD) muss im klinischen Alltag, wenn möglich, eine hohe Priorität eingeräumt werden, um die Inzidenz prolongierter Weaning-Verläufe positiv zu beeinflussen.
Die wichtigsten Schritte im Entwöhnungsprozess, zur Vermeidung einer unnötigen Verlängerung der mechanischen Beatmung, sind das rechtzeitige Erkennen („diagnostic triggering“) der Bereitschaft zur Entwöhnung und der Bereitschaft zur Extubation. Im Rahmen des Screenings (adäquate Oxygenierung? Pulmonale und mentale Funktion? Klinische Stabilität?) und der Anwendung von Screeningtests wie dem weithin etablierten Rapid Shallow Breathing Index (RSBI, Beatmungsfrequenz/Tidalvolumen [f/VT]) wird überprüft, ob die Bedingungen für einen Spontanatmungsversuch erfüllt sind. Der prädiktive Wert des RSBI wird immer wieder hinterfragt, und die Autoren stellen neuere Ansätze wie den Integrative Weaning Index (IWI) vor. Allerdings muss grundsätzlich angemerkt werden, dass der RSBI nicht die „Extubierbarkeit“ testet und uns damit auch nicht von der Aufgabe befreit nachzuweisen, dass der Patient dazu in der Lage ist, eine suffiziente Spontanatmung ohne Unterstützung durch das Beatmungsgerät aufrechtzuerhalten [10].
Fortwährende Kontrolle aller Organfunktionen und Angehen behebbarer Einflüsse sichern den Erfolg
Dem multifaktoriellen Puzzle der Ursachen, die eine erfolgreiche Befreiung vom Beatmungsgerät verhindern, widmen die Autoren einen beträchtlichen Teil ihres Leitthemenartikels. Dieser Schwerpunkt ist meines Erachtens wohlbegründet, denn der immer wiederkehrende Rundumblick auf alle Organfunktionen und das Angehen behebbarer Einflüsse machen am Ende den entscheidenden Unterschied. Die Autoren betonen weiterhin, dass „für die Behandlung dieser hochkomplexen Patienten eine systematische Gesamtstrategie unter Beteiligung zahlreicher Fachdisziplinen und Berufsgruppen erforderlich ist“. Die Erfüllung von Strukturmerkmalen, wie die Durchführung gemeinsamer Visiten und interprofessionellen Fallbesprechungen, kann dazu beitragen.
Der Beitrag kann dem Leser nur empfohlen werden, nicht nur, um in Kürze einen Überblick über den gesamten Prozess der Entwöhnung von der Beatmung und die allgemeinen Behandlungsstrategien zu erhalten, sondern auch, um dafür begeistert zu werden, sich darüber hinaus noch tiefer in einen der Dreh- und Angelpunkte moderner Intensivmedizin zu einzuarbeiten.
Interessenkonflikt
O. Moerer gibt an, dass kein Interessenkonflikt besteht.
QR-Code scannen & Beitrag online lesen
==== Refs
Literatur
1. Zilberberg MD de Wit M Pirone JR Shorr AF Growth in adult prolonged acute mechanical ventilation: implications for healthcare delivery Crit Care Med 2008 36 5 1451 1455 10.1097/CCM.0b013e3181691a49 18434911
2. Adamides AA Cooper DJ Rosenfeldt FL Bailey MJ Pratt N Tippett N Focal cerebral oxygenation and neurological outcome with or without brain tissue oxygen-guided therapy in patients with traumatic brain injury Acta Neurochir 2009 151 1399 1409 10.1007/s00701-009-0398-y 19727549
3. Boles JM Bion J Connors A Herridge M Marsh B Melot C Weaning from mechanical ventilation Eur Respir J 2007 29 5 1033 1056 10.1183/09031936.00010206 17470624
4. Tobin MJ Advances in mechanical ventilation N Engl J Med 2001 344 26 1986 1996 10.1056/NEJM200106283442606 11430329
5. Schönhofer B, Geiseler J, Braune S, Dellweg D, Fuchs H, Hirschfeld-Araujo J et al (2019) Prolongiertes Weaning. S2k-Leitlinie herausgegeben von der Deutschen Gesellschaft für Pneumologie und Beatmungsmedizin e. V. AWMD Register Nr. 020/015. Verabschiedet von den Vorständen der beteiligten Fachgesellschaften am 09.07.2019. https://www.awmf.org/uploads/tx_szleitlinien/020-015l_S2k_Prolongiertes_Weaning_2019_09_1.pdf. Zugegriffen: 26.11.2022
6. Fichtner F, Moerer O, Laudi S, Weber-Carstens S, Nothacker M, Kaisers U (2017) S3-Leitlinie Invasive Beatmung und Einsatz extrakorporaler Verfahren bei akuter respiratorischer Insuffizienz herausgegeben von der Deutschen Gesellschaft für Anästhesiologie & Intensivmedizin, 2017. AWMF Leitlinien-Register Nr. 001/021. https://www.awmf.org/uploads/tx_szleitlinien/001-021l_S3_Invasive_Beatmung_2017-12.pdf. Zugegriffen: 26.11.2022
7. Rollnik JD, Adolphsen J, Bauer J, Bertram M, Brocke J, Dohmen C et al (2016) S2k Leitlinie Besonderheiten des prolongierten Weanings bei Patienten in der neurologisch-neurochirurgischen Frührehabilitation herausgegeben von der Deutschen Gesellschaft für Neurorehabilitation, 21.10.2016. AWMF Leitlinien-Register Nr. 080 – 002. https://www.awmf.org/uploads/tx_szleitlinien/080-002l_S2k_Prolongiertes_Weaning_neurol_neuroch_Fruehreha_2017-04-abgelaufen.pdf. Zugegriffen: 26.11.2022
8. Bingold T Bickenbach J Coburn M David M Dembinski R Kuhnle G Moerer O Weber-Carstens S Marx G DGAI Zertifizierung anästhesiologische Intensivmedizin: Modulares Zertifikat Intensivmedizin der DGAI Anasth Intensivmed 2014 55 316 329
9. Bickenbach J Fritsch S Weaning von invasiver Beatmung – Herausforderungen im klinischen Alltag Anaesthesiologie 2022 10.1007/s00101-022-01219-3 36418440
10. Tobin MJ Meta-analysis of frequency-to-tidal volume ratio: Conflating extubatability with weanability Chest 2022 161 6 e393 10.1016/j.chest.2022.01.069 35680324
| 36441224 | PMC9707194 | NO-CC CODE | 2022-12-08 23:16:07 | no | Anaesthesiologie. 2022 Nov 28; 71(12):907-909 | utf-8 | Anaesthesiologie | 2,022 | 10.1007/s00101-022-01229-1 | oa_other |
==== Front
Netw Model Anal Health Inform Bioinform
Netw Model Anal Health Inform Bioinform
Network Modeling and Analysis in Health Informatics and Bioinformatics
2192-6662
2192-6670
Springer Vienna Vienna
395
10.1007/s13721-022-00395-x
Original Article
Immunoinformatics design of multi-epitope peptide-based vaccine against Haemophilus influenzae strain using cell division protein
AlChalabi Rawaa 1
Al-Rahim Aya 1
Omer Dania 1
http://orcid.org/0000-0001-7427-4483
Suleiman Ahmed AbdulJabbar [email protected]
2
1 grid.411310.6 0000 0004 0636 1464 College of Biotechnology, Department of Molecular and Medical Biotechnology, Al-Nahrain University, Baghdad, Iraq
2 grid.440827.d 0000 0004 1771 7374 Biotechnology Department, College of Science, University of Anbar, Ramadi, Anbar Iraq
28 11 2022
2023
12 1 114 7 2022
20 9 2022
7 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Haemophilus influenzae is a pathogen that causes invasive bacterial infections in humans. The highest prevalence lies in both young children and adults. Generally, there are no vaccines available that target all the strains of Haemophilus influenzae. Hence, the purpose of this research is to employ bioinformatics and immunoinformatics approaches to design a Multi-Epitope Vaccine candidate employing the pathogenic cell division protein FtsN that specifically combat all the Haemophilus influenzae strains. The current research focuses on developing subunit vaccine in contrast to vaccines generated from the entire pathogen. This will be accomplished by combining multiple bioinformatics and immunoinformatics approaches. As a result, prospective T cells (helper T lymphocyte and cytotoxic T lymphocytes) and B cells epitopes were investigated. The human leukocyte antigen allele having strong associations with the antigenic and overlapping epitopes were chosen, with 70% of the total coverage of the world population. To construct a linked vaccine design, multiple linkers were used. To increase the immunogenic profile, an adjuvant was linked using EAAAK linker. The final vaccine construct with 149 amino acids was obtained after adjuvants and linkers were added. The developed Multi-Epitope Vaccine has a high antigenicity as well as viable physiochemical features. The 3D conformation was modeled and undergoes refinement and validation using bioinformatics methods. Furthermore, protein–protein molecular docking analysis was performed to predict the effective binding poses of Multi-Epitope Vaccine with the Toll-like receptor 4 protein. Besides, vaccine underwent the codon translational optimization and computational cloning to verify the reliability and proper Multi-Epitope Vaccine expression. In addition, it is necessary to conduct experiments and research in the laboratory to demonstrate that the vaccine that has been developed is immunogenic and protective.
Keywords
Haemophilus influenzae
Epitope
Cell division protein
Bioinformatics
Linkers
issue-copyright-statement© Springer-Verlag GmbH Austria, part of Springer Nature 2023
==== Body
pmcIntroduction
Haemophilus influenzae (Hi) is a Gram-negative coccobacillus which belongs to the Pasteurellaceae family (Soeters et al. 2018). It is tiny (0.3–1 µm), anaerobic, pleomorphic pathogenic bacteria. The infection caused by Hi is termed as “Haemophilus influenzae disease” (Suga et al. 2018). It can cause pneumonia, otitis media, epiglottitis, and meningitis in the upper respiratory tract (Slack et al. 2020). It usually gets into the lungs via aspiration or hematogenous dissemination. It is the source of an extensive range of bacterial illnesses (Butler and Myers 2018). They are divided into two types: encapsulated and non-encapsulated (Wilson et al. 2020). Non-typeable Hi (NTHi) strains do not express a polysaccharide capsule, but Hi serotypes (i.e., Hia, Hib, Hic, Hid, Hie, and Hif) do, with Type B being the most prevalent cause of pneumonia, infecting predominantly children and immunocompromised people (Wilson et al. 2020). However, NTHi is linked to both invasive and non-invasive illnesses. The Hi genome (Rd KW20) was sequenced first among the members of Pasteurellaceae family (Pinto et al. 2019). The genome size is 1,830,137 base pair with 38% GC content (López-López et al. 2021).
Hi adheres to host cells and exhibits pathogenicity via various modes of transmission (Chen et al. 2018). One of the crucial steps in infection and pathogenesis is the initial adhesion of Hi to the host cells (López-López et al. 2021). Hi adheres to host cells and exhibits pathogenicity such as the capsule, the adhesion proteins, the IgA1 protease, and, last but not least, the lipooligosaccharide (Huska et al. 2022; Vitovski et al. 2002; Kilian et al. 1983). The bacterial capsule confers antiphagocytic properties, and the absence of an anti-capsular antibody promotes bacterial propagation (Potts et al. 2019). As a result, they evade and change the host immune responses, as well as aid in the development of a biofilm-like colony, through comparable interactions (Mirzaei et al. 2020).
The Hi takes as little as a few days to elicit clinical signs but the exact incubation period is still unknown (Saadati et al. 2021). Infected persons spread the disease by coughing or sneezing, which causes minute respiratory droplets containing the bacteria (Soeters et al. 2018). If other people breathe in those droplets, they may become ill. The sickness caused by Hi affects predominantly children under the age of 5 and adults aged 65 and up (Butler and Myers 2018). American Indians, Alaska Natives, or people with specific medical issues are at risk (Soeters et al. 2018; Butler and Myers 2018). The infected organ, and the history and physical examination are the clinical manifestations of Hi. The most prominent symptoms of pneumonia are high-grade fever, chills as well as cough, fatigue, breathlessness, chest pain, and body aches (Wen et al. 2020).
Antibiotic-resistant bacteria has emerged due to the frequent use of antibiotics (Serwecińska 2020). Major challenges to developing a vaccine against Hi are: (i) lack of polysaccharide capsule in non-typeable Hi, necessitating the development of alternative vaccine antigens; (ii) non-typeable strains have a lot of genetic and antigenic diversity, whereas Type B encapsulated bacteria are usually a clonal population, and (iii) the pathogenesis of non-typeable. The contiguous spread of Hi infections from mucosal surfaces implies that a successful vaccination will demand the creation of an alternate vaccine generated by hematogenous dispersion (Vitovski et al. 2002). The use of vaccinology may be a viable strategy for preventing and controlling Hi infections (Kim et al. 2004). With the evolution of bioinformatics, computational tools have simplified computational evaluations for the vaccine candidates, greatly reducing the time required for preclinical and clinical operations (Ko et al. 2012). Advancements in bioinformatics and immunoinformatics have resulted in the rapid epitope evaluation and validation of proteins that could be used as vaccine targets (Colovos and Yeates 1993). To create a viable subunit vaccine, several antigenic markers and adjuvants must be chosen and added to create a vaccine construct. This vaccine construct can then be utilized to evaluate the simulated immunogenic response of the host using immunoinformatics techniques (“Designing a chimeric subunit vaccine for influenza virus, based on HA2, M2e and CTxB 2022).
Except vaccines for Hib, there are no vaccines available that target general Hi strains; therefore, in this research, bioinformatics and immunoinformatics tools are employed to computationally design a subunit Multi-Epitope Vaccine (MEV) against Hi (Fig. 1). This research puts emphasis on the development of subunit vaccinations as opposed to the vaccines generated from entire pathogens. Therefore, a highly conserved and antigenic cell division protein FtsN of Hi was targeted. Therefore, FtsN protein was selected and utilized for major histocompatibility class (MHC)-I and MHC class-II binding T cell and B cell epitopes prediction. These specific epitopes are responsible for inducing interferon-gamma (IFN-γ). The highly antigenic epitopes are therefore utilized for MEV construction using structure prediction to computationally assess the vaccine construct. The immunological and physicochemical attributes of the vaccine construct such as its stability, flexibility, and solubility, as well as its binding efficacy to the Toll-like receptor 4 (TLR4), were evaluated. The MEV was also simulated to evaluate the generation of immune response by the host. Lastly, codon translation enhancement and vaccine cloning (computational simulation) are performed to assure the successful construction and expression of the MEV construct.Fig. 1 Research methodology employed to develop a subunit MEV vaccine against Hi
Methodology
Sequence retrieval and structural analysis
The FASTA sequence for cell division protein FtsN was first obtained from the National Center of Biotechnology Information (NCBI) as a FASTA file (NCBI Resource Coordinators 2013). ProtParam was then used to predict the physical and chemical properties of FtsN (Gasteiger et al. 2005). VaxiJen v.2.0 tool determined the antigenicity of proteins (Doytchinova and Flower 2007). It uses an alignment-free method to predict the antigenicity from physicochemical data. With the help of SOPMA, prediction of the secondary structure was carried out (Geourjon and Deléage 1995). Researchers have not yet figured out the 3D conformation of the FtsN protein, which is crucial in the final stages of cell division. Therefore, the Robetta server estimates the 3D conformation of the FtsN protein (Kim et al. 2004). It employs two different approaches: comparative modeling and de novo 3D conformation prediction. We used GalaxyWEB to fine-tune the predicted 3D protein models (Ko et al. 2012). The SAVES server assessed the predicted models for a quality check (Colovos and Yeates 1993). Moreover, the Ramachandran plot was used to validate our predicted models on the RAMPAGE server (Lovell et al. 2003).
MHC class-I T cell epitopes identification and analysis
The IEDB-AR v.2.22 consensus technique server was employed to identify the 9-mer MHC class-I epitopes (Zhang et al. 2008; Calis et al. 2013). The FtsN protein sequence was input for prediction of T cell epitopes, and the respective alleles were preferred. To narrow down the field of epitopes for further study, those with consensus scores of fewer than two are deemed good binders. The 15-mer MHC class-II T cell epitopes were predicted using NetMHCIIpan 4.0 server (Reynisson et al. 2020). The epitopes were divided into strong, weak, or non-binding at the default threshold based on the percentage value. We chose 2%, 10%, and greater than 10% as the percentiles for determining whether or not a substance is strongly bound. Cytokines activate immune system including other cytotoxic T cells. Vaccine development involves the critical component of cytokines that induce helper T lymphocyte (HTL) and cytotoxic T lymphocytes (CTL). Using the IFN epitope server, IFN- γ epitopes were characterized. It uses the support vector machine (SVM) approach, hybrid motif method, and IFN- γ vs non-IFN-γ model (Dhanda et al. 2013a). In contrast, IL-4Pred and IL-10Pred predicted IL-4 and IL-10 enhancing qualities (Dhanda et al. 2013b; Nagpal et al. 2017). Both servers predicted interleukin-inducing peptides in-silico. According to the IEDB-AR MHC-I immunogenicity tool and the VaxiJen server, antigenicity and immunogenicity were evaluated (Doytchinova and Flower 2007). AlgPred 2.0 was used for allergenic profiling (Saha and Raghava 2006a). ToxinPred was employed to predict toxicity in the study (Gupta et al. 2013). In this way, each epitope was screened and tested.
Prediction and evaluation of B cell epitopes
The promising candidates for vaccine development are B cell epitopes because they induce adaptive immune response (Zhang et al. 2014). Conformational and linear are two subtypes of B cell epitopes. The ABCPred server was used for the prediction of linear B cell type 16-mer epitopes and those 0.5 prediction score were shortlisted (Saha and Raghava 2006b). For the prediction of discontinuous B cell epitopes, the Discotope server was utilized (Kringelum et al. 2012). VaxiJen 2.0, ToxinPred 2.0, and AlgPred 2.0 were used to analyze the antigenic, toxic, and allergic characteristics of the newly discovered B cell epitopes (Doytchinova and Flower 2007; Saha and Raghava 2006a; Gupta et al. 2013).
Shortlisting and conservancy check of identified epitopes
A conservancy check was run through the IEDB-AR v.2.22 for shortlisting of conserved T cell and B cell epitopes (Bui et al. 2006). We focused on epitopes that showed a 100% conservation. Cytokine-prompting capabilities were recognized as a crucial criterion in sorting out effective epitopes. IFN is known for its intrinsically safe reactions and its ability to prevent viral duplication. In addition to this, they can prepare both CTL and HTL, both of which can induce a range of immunological responses. The IFN stimulating potential of expected epitopes was analyzed using IFN epitope server which employs SVM algorithm (Dhanda et al. 2013a).
HLA population coverage analysis
Levels of dispersion and expression vary among HLA alleles based on ethnicities and regions around the world, which influences the development of a successful vaccination (Bui et al. 2006; Immunoinformatics Approach for Epitope-Based Peptide Vaccine Design and Active Site Prediction against Polyprotein of Emerging Oropouche Virus 2022; Nain et al. 2020; Hoover et al. 2003). IEDB-AR v.2.22 Population Coverage tool was employed for the population coverage analysis on the shortlisted epitopes (Bui et al. 2006). Population coverage was evaluated for each epitope for variety of geographic locations based on HLA alleles’ classification. The areas with great significance were identified with the Hi pathogen under the current research.
Designing of MEV
The MEV contains the highly antigenic epitopes that are non-allergenic, 100 percent conserved and overlapping, have extensive population coverage, and have a strong association with a common human allele. As a result, only those epitopes that met the aforementioned requirements were included in the MEV. Following inter-interaction compatibility validation and to enhance the immunological response, the first CTL epitope was linked to the adjuvant using EAAAK linker. This linkage decreases association with other biomolecules through efficient dissociation and enhances stability (Hoover et al. 2003; Mahram and Herbordt 2010; Bjellqvist et al. 1993). With an adjuvant, the immunogenicity of the vaccination may rise. AAY and GPGPG linkers were used for other epitopes to maintain their distinct immunogenicity. The primary goal in the development of MEV is to prevent the induction of junctional epitopes, and both AAY and GPGPG are able to achieve this purpose yet, they enhance immunization and epitope performance (Lovell, et al. 2003; McGuffin et al. 2000). The beta-defensin adjuvant having length of 45 amino acids was used in this investigation that serves as both an antibacterial agent and an immunomodulatory (Hoover et al. 2003; Wiederstein and Sippl 2007).
Primary and secondary structural analyses
The human proteome was examined with the BLASTp using its default settings to make certain that the generated MEV sequence was not similar to any other sequence (threshold: 10; word size: 6; matrix BLOSUM62) (Kuriata et al. 2018). A protein is considered non-homologous if it shares less than 37% of its amino acid sequence. ProtParam evaluated the physicochemical features of the vaccine construct (Gasteiger et al. 2005). Using the Vaxijen v.2.0 server, antigenicity predictions were made possible for vaccine. PSIPRED was employed to determine the 2D structure analysis of the MEV (McGuffin et al. 2000). It examined the extensive chain, coils, alpha-helices, and beta-strands (McGuffin et al. 2000; Kurcinski et al. 2019).
MEV construct 3D conformation prediction, refinement, and evaluation
For 3D conformation prediction of the MEV, Robetta with de novo modeling approach was employed to model because the vaccine is a collection of different epitopes. Secondly, there was not a suitable template with appropriate identity percentage to be used (Kim et al. 2004; Prediction of residues in discontinuous B-cell epitopes using protein 3D structures-Haste Andersen 2006). The projected 3D shape of our vaccine was subjected to a GalaxyWEB server to improve the structure quality (Ko et al. 2012). The Ramachandran plot evaluation of the refined 3D conformation of MEV was performed with the assistance of the RAMPAGE server (Lovell, et al. 2003). The Z-score for MEV was computed via the ProSA-web server (Wiederstein and Sippl 2007). With the help of the SAVES server, we were able to choose and test the best-predicted model (Colovos and Yeates 1993). Additionally, to assess the MEV structural flexibility, CABS-Flex v.2.0 server was used (Kuriata, et al. 2018). The flexibility of the constructed vaccine is critical to its function, and it provides complete picture stability of a protein by simulating its residues (Kurcinski et al. 2019).
Screening for B cell epitopes
The Discotope server is utilized to forecast discontinuous structural B cell epitopes with a threshold of -3.7 which is used for the construction of final MEV (Kringelum et al. 2012). For the prediction of linear B cell epitopes, ABCPred server was employed (Saha and Raghava 2006b).
Protein–protein docking between TLR4 receptor and MEV
If a vaccine reaches the target immune cells in the right way, the host will create an effective immune response. As a result, the MEV’s affinity for human immunological receptors was investigated using molecular docking. The antiviral immune responses are mostly attributed to TLR4 receptor. The TLR4 receptor protein (PDB ID: 4G8A) was docked against the vaccine construct using the ClusPro server, a popular technique for protein–protein docking (The ClusPro web server for protein–protein docking 2022; Mugunthan and Harish 2021). The docked adducts were visualized with PyMOLv.1 (Pymol: an open-source molecular graphics tool—ScienceOpen 2022). PDBsum was used to achieve the schematic diagrams demonstrating the interactions among the docked protein complex (Laskowski 2009).
Immune system simulation for vaccine efficacy
To observe the generated immunological responses by the designed MEV, computational immune simulation server C-ImmSim 10.1 was used (Pandey 2016). The server simulates the lymph node, thymus, and bone barrow. The number of simulation steps (100), injections (1 each), and random seed number (12,345) were set, whereas HLA-binding alleles such as, B0702, B0702, A0101, A0101, DRB1 0101, and DRB1 0101 and volume (10) were chosen.
In-silico E. coli cloning and codon optimization
Each species uses a different set of codons for the translation of the mRNAs into proteins. Therefore, specie to specie usage of codon varies. If codons are not optimal for the host species, gene expression may be lowered. The Java Codon Adaption Tool (JCAT) server was employed to optimize codons for E. Coli strain K12 (Pandey 2016; Grote et al. 2005). The cleavage sites of restriction enzymes and rho-independent transcription termination were chosen to avoid bacterial ribosome binding sites. The codon adaption index score and GC percentage content of the optimized DNA sequence were evaluated (Livingston et al. 2002).
Results
Sequence retrieval and structural analysis
The sequence for the cell division protein FtsN was obtained from NCBI with accession number WP_168769343.1 in FASTA format having 256 amino acids. VaxiJen version 2.0 was then used for antigenicity testing. The results indicated that protein was considerably allergenic with a score of 0.8390. The physicochemical parameters, such as half-life, stability profiling, theoretical pI of 9.92, molecular weight, and aliphatic index, were evaluated with the help of ProtParam (Table 1). The secondary structure analysis using SOPMA revealed more helixes in the protein secondary structure as compared to strands (Table 2, Fig. 2).Table 1 FtsN cell division antigenic protein physiochemical properties
Molecular weight Isoelectric point Instability index Stability profiling Aliphatic index Grand average of hydropathy Half-life
29,139.38 9.92 39.84 Stable 79.65 − 0.761 30 h (mammalian reticulocytes, in vitro)
> 20 h (yeast, in vivo)
> 10 h (Escherichia coli, in vivo)
Table 2 SOPMA secondary structure prediction results of FtsN cell division protein
Amino acids Alpha-helix Beta-turn Coils
256 52.73% 1.56% 35.94%
Fig. 2 SOPMA secondary structure prediction of FtsN protein. Yellow color bars indicate strands, pink bars are representing helices, and grey color lines are coils
The Robetta server and GalaxyWEB application predicted and refined the 3D model, respectively (Fig. 3A). Ramachandran plot assessed the estimated structure (Fig. 3B). The FtsN structure contained the 99.587% allowed amino acids. In SAVES, the optimized model score was 100. As a result, it was clear that the finalized model was of the highest possible quality.Fig. 3 Prediction of FtsN 3D structure and its evaluation: A FtsN refined 3D structure, the cyan color represents the helices in the protein structure and red color represents the beta sheets whereas the pink color represents loops; B identification of favorable amino acid residues in FtsN protein
Identification and analysis of B cell and T cell epitopes
IEDB-AR v.2.22 predicted a total of 1,792 epitopes, whereas 72 HTL epitopes were identified using NetMHCIIpan 4.0 server. ABCPred server predicted a total of 25 linear B cell epitopes, while the Discotope server identified 224 discontinuous B cell epitopes. The IFN triggering capability of 16 predicted epitopes was predicted using the IFN epitope server. It employed SVM hybrid algorithms excluding motif to predict. Several interesting epitopes based on the criteria; 100 percent conserved among proteins have a significant association, do not overlap with human proteins, and be highly antigenic/immunogenic were selected. These CTL epitopes were analyzed for their cytokine-inducing capabilities (Interleukin-10, Interleukin-4, and IFN- γ), and a total of 55 CTL epitopes, 5 HTL epitopes, 5 linear B cell epitopes, and 12 discontinuous B cell epitopes were selected. After the conservancy analysis, eventually, 7 CTL epitopes and 2 discontinuous B cell epitopes were shortlisted as shown in Table 3. Whereas no HTL epitopes met the criteria for non-allergenicity, antigenicity, or toxicity, and none could elicit an IFN immunological response.Table 3 Selected CTL epitopes of from FtsN protein
CTL epitopes Conservancy Toxicity Allergenicity Antigenicity IL4pred IL10pred IFN epitope server Alleles
IKALETRTV 100.00% (21/21) Non-toxin Non-allergen Antigenic Inducer Positive Positive HLA-B*08:01
LETRTVPVD 100.00% (21/21) Non-toxin Non-allergen Antigenic Inducer Negative Negative HLA-B*40:01
MAHRDFAAR 100.00% (21/21) Non-toxin Non-allergen Antigenic Inducer Positive Positive HLA-A*02:01
RPEEVWSYI 100.00% (21/21) Non-toxin Non-allergen Antigenic Inducer Negative Negative HLA-A*11:01
SVEQNMRLT 100.00% (21/21) Non-toxin Non-allergen Antigenic Inducer Negative Negative HLA-A*11:01
SYIKALETR 100.00% (21/21) Non-toxin Non-allergen Antigenic Inducer Negative Negative HLA-A*24:02
YIKALETRT 100.00% (21/21) Non-toxin Non-allergen Antigenic Inducer Negative Negative HLA-B*08:01
HLA population coverage analysis
A total of seven selected CTL epitopes with their corresponding HLA alleles were estimated for the population coverage collectively. HLA alleles vary according to ethnicity and geographical places. As a result, it has an impact on the development of an epitope-based vaccination. For selected epitopes, the analysis found cumulative coverage of 72% of the world population (Fig. 4). For the population of the Philippines, the highest coverage was reported. However, the least population coverage was found within the population of Scotland. HLA alleles are highly variable among ethnic groups and retain T cell epitope responses. T cell epitopes can bind to HLA alleles for better population coverage. For the estimation of the global distribution of HLA alleles, we used CTL and HTL epitopes. According to the results, in several geographical regions around the locations of the world, epitopes with the best coverage of their alleles were picked. According to our findings, population coverage was higher in regions in which Hi cases had been reported in the past. Furthermore, our findings revealed that specific epitopes had the potential to serve as candidates for MEV.Fig. 4 HLA alleles population coverage analysis of FtsN epitopes according to regions
MEV design and evaluation
All nine identified epitopes were subsequently utilized to create MEV construct. Beta-defensin was linked to the EAAAK linker to the MEV N-terminal. Beta-defensin is a 45-amino acid long adjuvant. EAAK linker boost the immune responses. AAY and GPGPG linkers were used to link other epitopes to preserve their different immunogenic activity. The final design of the vaccine contained 149 amino acids (Fig. 5).Fig. 5 Sequence overview of the MEV construction. It has 149 amino acids, including an adjuvant (shown in purple) connected to the N-terminus of MEV via the EAAAK linker (shown in green). The AAY linker (shown in blue) was utilized to connect the CTL epitopes, whereas the GPGPG linker (in pink) connects the HTL epitopes
Physicochemical and immunogenic characterization
Immunogenic and physiochemical properties were evaluated after the vaccine design was created. The findings of evaluating the sequence homology of the created MEV were compared to the human proteome and found to have no resemblance with any part of the human proteome. The immunological analyses were then performed. It was found that the vaccine was non-allergic, non-toxic, and had a high antigenicity (0.7831 at the 0.4 percent threshold). ProtParam was used to analyze the physicochemical properties of the vaccine. The theoretical PI was 9.74 kDa while the molecular weight of the vaccination was 16,588.10 kDa. The designed vaccine has a mean half-life of 30 h in vitro, > 20 h in vivo (Yeast), and > 10 h in vivo (E. coli in-vivo). The GRAVY has been calculated to be -0.516; a negative sign indicates that the vaccine is hydrophilic. Hence, analysis revealed that FtsN protein is a potential vaccine candidate.
Structural analyses
SOPMA server was used to analyze the secondary structure of the vaccine construct (Fig. 6). The analysis revealed the overall secondary structure of the vaccine, which comprises an alpha-helix with 84 residues (56.38%), beta-strands with 8 residues (5.37%), and coils with 45 amino acids (30.20%).Fig. 6 SOPMA secondary structure prediction analysis of the vaccine construct. Yellow colors are indicating strands, helices are shown in pink color, and grey color lines are representing coils
Tertiary structure prediction and refinement
The ROBETTA server was utilized for tertiary structure prediction of the vaccine construct. The deep learning method called RoseTTAFold was employed for the 3D modeling of MEV (Fig. 7A). The GalaxyWEB server optimized the predicted structure. The Ramachandran plot analysis for refined structure revealed that 97.674% are in highly preferred observations, 2.326% are in preferred observations, and 0% are in the outlier region (Fig. 7B). The Z-score obtained is − 4.87 (Fig. 7C). The optimized model received a perfect score during the ERRAT quality check using SAVES. Thus, the refined model possesses good quality.Fig. 7 The 3D structure of the MEV structure prediction and validation: A epitopes are shown in black color in the MEV construct. The adjuvant sequence is shown in green color, EAAAK linker sequence is shown in purple color, the AAY linkers are shown in red color, and the GPGPG linkers are shown in blue colors; B identification of favorable amino acid residues using Ramachandran plot analysis; C MEV refined conformation (alpha-helix are in cyan color, and loops are shown in pink color)
Additionally, the CABS-flex 2.0 server ran 50 simulation cycles, at temperature of 1.4 °C to test the structural flexibility of MEV. Regions adjacent to the N-terminal showed the least variance among the ten final 3D conformations (Fig. 8A). Optimum residue interaction map for the last ten models was collected (Fig. 8B). Lastly, the root mean square variation (RMSF) plot revealed the variations of all amino acids (Fig. 8C). The existence of changes in the MEV structure is indicative of its high adaptability and supports its viability as a vaccine candidate.Fig. 8 CABS-flex 2.0 results: A cartoon representation of the top ten models; B residue contact map; C RMSF graph showing simulation-induced MEV residue variations
B cell epitopes screening
Besides making antibodies, B lymphocytes secrete cytokines, which are important for humoral immunity, and MEV should have B cell epitopes specific to its domain. A total of 26 discontinuous and 15 linear epitopes were predicted from the final MEV construct using ABCPred 2.0 and Discotope (Table 4). Discontinuous B cell epitopes of the MEV were visualized using PyMOL v1.3. The following illustration is shown in Fig. 9.Table 4 Final list of the linear B cell epitopes in the final MEV construct
No Peptide sequence Start position Score
1 LETRTGPGPGKNRAGP 127 0.96
2 DAAYMAHRDFAARAAY 71 0.91
3 PGPGKNRAGPGPGNGE 133 0.89
4 EEQIGKCSTRGRKCCR 27 0.86
5 YSYIKALETRAAYYIK 110 0.83
6 YCRVRGGRCAVLSCLP 10 0.83
7 PEEVWSYIAAYSVEQN 88 0.82
8 SVEQNMRLTAAYSYIK 99 0.81
9 AVLSCLPKEEQIGKCS 19 0.81
10 TRTVPVDAAYMAHRDF 65 0.8
11 CSTRGRKCCRRKKEAA 33 0.78
12 KIKALETRTVAAYLET 50 0.75
12 DFAARAAYRPEEVWSY 79 0.72
13 TRTVAAYLETRTVPVD 56 0.67
14 CRRKKEAAAKIKALET 41 0.59
Fig. 9 Surface representation of the MEV. A Discontinuous B cell epitopes in MEV design are highlighted in red color; B linear B cell epitopes are colored red in MEV
Protein–protein docking between TLR4 receptor and MEV
An immune response is activated as a result of suitable interaction between antigen molecules and immune receptor molecules. Consequently, the server ClusPro 2.0 is utilized for MEV protein–protein docking against the human immunological receptor TLR4. TLR4 possess the ability of eliciting an immune response in response to bacterial identification. The docking analysis revealed a significant interaction between MEV and TLR4. TLR4–MEV has a binding score of -1035.2 kcal/mol (Fig. 10).Fig. 10 Molecular docking between the MEV and TLR4: cartoon representation of the docked TLR4–MEV complex. TLR4 is shown in orange–brown color and MEV is shown in cyan color
Interaction analysis between vaccine and TLR4
To activate immunological responsiveness, immune receptor molecules must form a suitable connection with the antigen molecule. Thus, the ClusPro v.2.0 server was utilized to execute protein–protein molecular docking of the MEV with human immune receptors TLR4. TLR4 is capable of inducing an immune response following the viral invasion. The docking analysis revealed that the MEV had robust interactions with both TLR4 chains A and B. TLR4–MEV binding score was − 1035.2 kcal cal/mol (Fig. 11A). MEV was shown to have 11 hydrogen bond interactions with TLR4 chain A and 9 hydrogen bond contacts with TLR4 chain B (Fig. 11B). Green sticks illustrate MEV residues hydrogen bonding to TLR4, whereas TLR4 residues interacting with MEV are displayed in hot-pink color stick representation.Fig. 11 A Interacting residues of MEV that interact with TLR4 are shown as a hot-pink color, whereas TLR4 residues that interact with MEV are shown as a green color stick representation. The hydrogen bonds are given as yellow dashed lines; B blue lines represent hydrogen bonds, red lines represent salt bridges, and orange lines represent other interactions. Different colored residues indicate amino acid properties (positive: blue color, negative: red color, aliphatic: grey color, aromatic: pink color, proline and glycine: orange color, and cystine residue: yellow color)
Immune system simulation of vaccine construction for vaccine efficacy
The immunogenic responses induced by the vaccine construct were predicted using C-ImmSim 10.1 server. The genuine immune system within the host consists of both secondary and primary immunological responses that contribute significantly to the pathogen invasion. The C-ImmSim server is utilized to perform the simulation of the host immune response to the antigen. It was observed that antigen reduces over time when passing through the secondary and primary phases. In secondary and primary responses, IgG + IgG and IgM concentrations were higher than IgM, IgG1 + IgG2, and IgG1 Fig. 12B. On the other hand, a strong cytokine reaction to the vaccine was observed which indicates that MEV may generate a successful immune response and clearance of the antigen in the host as it did in the simulation Fig. 12A.Fig. 12 Simulated immunological responses to MEV: A cytokine, interleukin production; B immunoglobulin synthesis and B cell isotypes after exposure to varied conditions
In-silico E. coli cloning and codon optimization
The JCAT server implemented the optimization of MEV codons. Codon optimization was executed to efficiently express and produce the MEV in the host E. coli system. According to JCAT results, MEV codons adapt due to codon usage of the E. coli K12 strain. One of the best findings was a CAI score of 1.0, with a GC content of 50.799% projected for the modified DNA sequence.
Discussion
Hi causes pneumonia, bacterial meningitis, and other serious infections, particularly among infants. These infections range from mild to serious, like bloodstream infections and ear infections (Haemophilus influenzae Infections in the H. influenzae Type b Conjugate Vaccine Era 2022). According to the World Health Organization, Hi causes 3 million cases of serious disease and 400,000 deaths worldwide each year (Haemophilus influenzae Infections in the H. influenzae Type b Conjugate Vaccine Era 2022). The majority of incidents occur in unimmunized children and in economically underdeveloped countries. This global health issue could be alleviated with a very effective vaccine. However, there is currently no vaccination available for Hi. One of the most basic and safest methods of preventing infectious diseases worldwide is vaccination (Vitovski et al. 2002; Potts et al. 2019; Haemophilus influenzae Infections in the H. influenzae Type b Conjugate Vaccine Era 2022). The current research focuses on subunit vaccinations in contrast to vaccines generated from the entire pathogen. Different immunogenic parts of pathogens help to build subunit vaccines; therefore, they can induce a strong immune response. The successful development of potent MEVs remains a considerable challenge due to difficulties in selecting epitopes, and adequate antigens (Potts et al. 2019; Fleischmann et al. 1995). As a result, forecasting relevant epitopes in an antigen is a critical step in constructing a MEV (Cripps et al. 2002).
Furthermore, due to the increase in antibiotic-resistant pathogenic bacteria and the limitations of conventional vaccines, it is important to focus on new strategies (Gilsdorf 1987). As a new approach, Multi-Epitope Vaccines (MEV) have been made using immunoinformatics methods, such as predicting epitopes with reliable servers, attaching epitopes with the appropriate linkers and adjuvants, and evaluating immunological, physicochemical, and structural properties with bioinformatics servers and tools (Qamar et al. 2021). These are the basic steps in making MEVs. MEVs stand out from other types of vaccines because they are more cost effective, safe, less time consuming to design, use natural adjuvants, and have good preclinical evaluations. So, MEVs are promising vaccines against invasive Hi infections (Designing a chimeric subunit vaccine for influenza virus, based on HA2, M2e and CTxB: a bioinformatics study 2022; Jalalvand and Riesbeck 2018, p.).
In this study, we aimed at utilizing the potential immunogenic epitopes obtained from the FTsN protein of Hi. Several computational tools were employed to create MEV that provokes both primary and secondary immune responses. IEDB-AR v.2.22, NetMHCIIpan 4.0, and ABCPred servers were used to predict the CTL, HTL, and LBL epitopes, respectively, to find potential vaccine candidates based on the parameters of toxicity, conservancy, antigenicity, allergenicity. B lymphocytes make antibodies, HTL causes both primary and secondary immune responses, and CTL stops the spread of viruses and bacteria by making certain cytokines and killing virally infected cells. Filters such as the epitopes that are immunogenic, antigenic, and non-toxic, have overlapping CTL and HTL epitopes, and most critically, have epitope conservancy. BLAST searches revealed that all of the epitopes were non-self, as they displayed no resemblance to the human proteome. A final MEV includes both the B cell and T cell epitopes for encouraging a complete system of immune reaction. This may help researchers to use the anticipated epitopes in the current study to computationally build therapeutic antibodies against Hi. Rather than depending on comprehensive functional screening from large libraries, they will be able to rationally create functional therapeutic antibodies.
The epitopic sequences linked with adjuvant as well as appropriate linkers make up the vaccine design. The adjuvant (beta-defensin) acts as an antibacterial and immunomodulator. EAAAK linkers which form helix were used to connect the adjuvant to the first CTL epitope of the FtsN protein. The EAAAK linker boosts bifunctional catalytic activity while decreasing protein toxicity. HTL epitopes were linked with GPGPG while remaining CTL epitopes were linked with AAY linkers to enhance immunogenicity. The use of linkers brings the pH of the construct closer to that of the physiological range. The final MEV design contains 149 amino acids and is non-allergenic, antigenic, and non-toxic (16,588.10 KDa). The final design was extremely basic (PI 9.74), suggesting it may enable prolonged normal pH contact. Furthermore, the expected instability index (25.92) represents protein stability even after expression, which increases its use. In addition, the calculated aliphatic index (A1) indicates its stability. Hydrophilicity was revealed through a negative GRAVY value (− 0.516), indicating water molecules form strong interactions with MEV. The 3D MEV conformation was also developed utilizing the Robetta server. Using the SAVES server, structure validation tools prove the accuracy of the predicted model with a 100-quality score. A higher score of − 155.51 predicted using AggreScan3D suggested a high solubility level of MEV. Whereas, the Ramachandran plot analysis revealed that 97.674% are in highly preferred observations. Protein–protein docking was performed between TLR4 immune receptor and MEV to explore the efficacy of this vaccine. If a vaccination interacts effectively with the target immune cells, the host creates an efficient immunological response. As a result, the interaction between the MEV and the immunological receptor TLR4 was investigated with a molecular docking approach. Thus, molecular docking analysis revealed that vaccine and receptor have a strong and stable binding with many hydrogen bond interactions.
In the current research, the immunoinformatic strategy for vaccine designing was utilized to generate MEV capable of eliciting immune responses against Hi. We believe that this vaccination will elicit both primary and secondary immune responses. The interactions and binding patterns between the TL4 and the MEV were highly stable. We may conclude that our vaccine design is capable of eliciting an immunological response in the real life. However, the real immune stimulation potential is needed to be validated to combat Hi, and further wet laboratory investigations are crucial to check their efficacy and safety.
Conclusion
Infection caused by Hi has remained mysterious in terms of treatment and underlying immunological concepts. Hi is now a global disease and issue. There are no proper treatments available for Hi that can treat the infection effectively; however, there are treatments available for management purposes. As such, there is no proper cure available for Hi. Multiple antibacterial medications have been investigated against Hi, but no feasible results have been found. These limitations led to the concept of developing a subunit MEV humoral vaccine in this study. The development is centered around bioinformatics and immunoinformatics analysis that can identify key vaccine candidates to generate a cellular immune response. MEV model was proposed using a combination of computational biology and immunology data to help design a possible vaccine against Hi infection. The analysis elucidates that the proposed vaccine model would be effective against the infection and in Hi treatment and research. This research employs a comprehensive computational approach for developing a subunit vaccine model. Therefore, additional experimental and laboratory research is required to test and analyze the specificity, efficacy, and safety of a vaccine.
Data availability
All data for this work are included in manuscript body.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Bjellqvist B The focusing positions of polypeptides in immobilized pH gradients can be predicted from their amino acid sequences Electrophoresis 1993 14 1 1023 1031 10.1002/elps.11501401163 8125050
Bui H-H Sidney J Dinh K Southwood S Newman MJ Sette A Predicting population coverage of T cell epitope-based diagnostics and vaccines BMC Bioinform 2006 7 153 10.1186/1471-2105-7-153
Butler DF Myers AL Changing epidemiology of Haemophilus influenzae in children Infect Dis Clin 2018 32 1 119 128 10.1016/j.idc.2017.10.005
Calis JJA Properties of MHC class I presented peptides that enhance immunogenicity PLoS Comput Biol 2013 9 10 10 10.1371/journal.pcbi.1003266
Chen D Microbial virulence, molecular epidemiology and pathogenic factors of fluoroquinolone-resistant Haemophilus influenzae infections in Guangzhou, China Ann Clin Microbiol Antimicrob 2018 17 1 41 10.1186/s12941-018-0290-9 30470228
Colovos C Yeates TO Verification of protein structures: patterns of nonbonded atomic interactions Protein Sci Publ Protein Soc. 1993 2 9 9 10.1002/pro.5560020916
Cripps AW Foxwell R Kyd J Challenges for the development of vaccines against Haemophilus influenzae and Neisseria meningitidis Curr Opin Immunol 2002 14 5 553 557 10.1016/S0952-7915(02)00373-4 12183152
Designing a chimeric subunit vaccine for influenza virus, based on HA2, M2e and CTxB: a bioinformatics study|BMC Molecular and Cell Biology|Full Text. https://bmcmolcellbiol.biomedcentral.com/articles/10.1186/s12860-020-00334-6 Accessed 20 Sep 2022
Dhanda SK Vir P Raghava GPS Designing of interferon-gamma inducing MHC class-II binders Biol Direct 2013 8 30 10.1186/1745-6150-8-30 24304645
Dhanda SK Gupta S Vir P Raghava GPS Prediction of IL4 inducing peptides Clin Dev Immunol 2013 2013 263952 10.1155/2013/263952 24489573
Doytchinova IA Flower DR VaxiJen: a server for prediction of protective antigens, tumour antigens and subunit vaccines BMC Bioinform 2007 8 1 1 10.1186/1471-2105-8-4
Fleischmann RD Whole-genome random sequencing and assembly of Haemophilus influenzae Rd Science 1995 269 5223 496 512 10.1126/science.7542800 7542800
Gasteiger E Walker JM Protein identification and analysis tools on the ExPASy server The proteomics protocols handbook 2005 Totowa Humana Press 571 607
Geourjon C Deléage G SOPMA: significant improvements in protein secondary structure prediction by consensus prediction from multiple alignments Comput Appl Biosci CABIOS 1995 11 6 6 10.1093/bioinformatics/11.6.681
Gilsdorf JR Haemophilus influenzae non-type b infections in children Am J Dis Child 1987 141 10 1063 1065 10.1001/archpedi.141.10.1063 3498361
Grote A JCat: a novel tool to adapt codon usage of a target gene to its potential expression host Nucleic Acids Res 2005 10.1093/nar/gki376
Gupta S In silico approach for predicting toxicity of peptides and proteins PLoS ONE 2013 8 9 9 10.1371/journal.pone.0073957
Haemophilus influenzae infections in the H. influenzae type b conjugate vaccine era | J Clin Microbiol. https://journals.asm.org/doi/10.1128/JCM.05476-11 accessed 20 Sep 2022
Hoover DM Wu Z Tucker K Lu W Lubkowski J Antimicrobial characterization of human beta-defensin 3 derivatives Antimicrob Agents Chemother 2003 47 9 9 10.1128/AAC.47.9.2804-2809.2003
Huska B Kubinec C Sadarangani M Ulanova M Seroprevalence of IgG and IgM antibodies to Haemophilus influenzae type a in Canadian children Vaccine 2022 40 8 1128 1134 10.1016/j.vaccine.2022.01.029 35078664
Immunoinformatics Approach for Epitope-Based Peptide Vaccine Design and Active Site Prediction against Polyprotein of Emerging Oropouche Virus. https://www.hindawi.com/journals/jir/2018/6718083/ accessed 20 Sep 2022
Jalalvand F Riesbeck K Update on non-typeable Haemophilus influenzae-mediated disease and vaccine development Expert Rev Vaccines 2018 17 6 503 512 10.1080/14760584.2018.1484286 29863956
Kilian M Thomsen B Petersen TE Bleeg H Molecular biology of Haemophilus influenzae IgA1 proteases Mol Immunol 1983 20 9 1051 1058 10.1016/0161-5890(83)90046-9 6358864
Kim DE Chivian D Baker D Protein structure prediction and analysis using the Robetta server Nucleic Acids Res 2004 10.1093/nar/gkh468
Ko J Park H Heo L Seok C GalaxyWEB server for protein structure prediction and refinement Nucleic Acids Res 2012 10.1093/nar/gks493
Kringelum JV Lundegaard C Lund O Nielsen M Reliable B cell epitope predictions: impacts of method development and improved benchmarking PLoS Comput Biol 2012 8 12 12 10.1371/journal.pcbi.1002829
Kurcinski M Oleniecki T Ciemny MP Kuriata A Kolinski A Kmiecik S CABS-flex standalone: a simulation environment for fast modeling of protein flexibility Bioinformatics 2019 35 4 694 695 10.1093/bioinformatics/bty685 30101282
Kuriata A CABS-flex 2.0: a web server for fast simulations of flexibility of protein structures Nucleic Acids Res 2018 46 W1 W1 10.1093/nar/gky356 29982807
Laskowski RA PDBsum new things Nucleic Acids Res 2009 10.1093/nar/gkn860
Livingston B A rational strategy to design multiepitope immunogens based on multiple Th lymphocyte epitopes J Immunol 2002 168 11 5499 5506 10.4049/jimmunol.168.11.5499 12023344
López-López N Gil-Campillo C Díez-Martínez R Garmendia J Learning from –omics strategies applied to uncover Haemophilus influenzae host-pathogen interactions: current status and perspectives Comput Struct Biotechnol J 2021 19 3042 3050 10.1016/j.csbj.2021.05.026 34136102
Lovell SC Structure validation by Calpha geometry: phi, psi and Cbeta deviation Proteins 2003 50 3 3 10.1002/prot.10286
Mahram A, Herbordt MC (2010) Fast and accurate NCBI BLASTP: acceleration with multiphase FPGA-based prefiltering. In: Proceedings of the 24th ACM International Conference on Supercomputing, New York, NY, USA, 2010, pp. 73–82. 10.1145/1810085.1810099
McGuffin LJ Bryson K Jones DT The PSIPRED protein structure prediction server Bioinforma Oxf Engl. 2000 16 4 4 10.1093/bioinformatics/16.4.404
Mirzaei R The importance of intracellular bacterial biofilm in infectious diseases Microb Pathog 2020 147 104393 10.1016/j.micpath.2020.104393 32711113
Mugunthan SP Harish MC Multi-epitope-Based vaccine designed by targeting cytoadherence proteins of Mycoplasma gallisepticum ACS Omega 2021 6 21 13742 13755 10.1021/acsomega.1c01032 34095666
Nagpal G Computer-aided designing of immunosuppressive peptides based on IL-10 inducing potential Sci Rep 2017 7 42851 10.1038/srep42851 28211521
Nain Z Karim MM Sen MK Adhikari UK Structural basis and designing of peptide vaccine using PE-PGRS family protein of Mycobacterium ulcerans—an integrated vaccinomics approach Mol Immunol 2020 120 146 163 10.1016/j.molimm.2020.02.009 32126449
NCBI Resource Coordinators Database resources of the National Center for Biotechnology Information Nucleic Acids Res 2013 41 D1 D8 D20 10.1093/nar/gks1189 23193264
Pandey R Exploring dual inhibitory role of febrifugine analogues against Plasmodium utilizing structure based virtual screening and molecular dynamic simulation J Biomol Struct Dyn 2016 10.1080/07391102.2016.1161560
Pinto M Insights into the population structure and pan-genome of Haemophilus influenzae Infect Genet Evol 2019 67 126 135 10.1016/j.meegid.2018.10.025 30391557
Potts CC Genomic characterization of Haemophilus influenzae: a focus on the capsule locus BMC Genomics 2019 20 1 733 10.1186/s12864-019-6145-8 31606037
Prediction of residues in discontinuous B cell epitopes using protein 3D structures—Haste Andersen—2006—Protein Science—Wiley Online Library.” https://onlinelibrary.wiley.com/doi/full/10.1110/ps.062405906 accessed 20 Sep 2022
Pymol: an open-source molecular graphics tool—ScienceOpen. https://www.scienceopen.com/document?vid=4362f9a2-0b29-433f-aa65-51db01f4962f accessed 20 Sep 2022
Qamar M Designing multi-epitope vaccine against Staphylococcus aureus by employing subtractive proteomics, reverse vaccinology and immuno-informatics approaches Comput Biol Med 2021 10.1016/j.compbiomed.2021.104389
Reynisson B Barra C Kaabinejadian S Hildebrand WH Peters B Nielsen M Improved prediction of MHC II antigen presentation through integration and motif deconvolution of mass spectrometry MHC eluted ligand data J Proteome Res 2020 19 6 6 10.1021/acs.jproteome.9b00874
Saadati A Kholafazad-Kordasht H Ehsani M Hasanzadeh M Seidi F Shadjou N An innovative flexible and portable DNA based biodevice towards sensitive identification of Haemophilus influenzae bacterial genome: a new platform for the rapid and low cost recognition of pathogenic bacteria using point of care (POC) analysis Microchem J 2021 169 106610 10.1016/j.microc.2021.106610
Saha S Raghava GPS AlgPred: prediction of allergenic proteins and mapping of IgE epitopes Nucleic Acids Res 2006 10.1093/nar/gkl343
Saha S Raghava GPS Prediction of continuous B cell epitopes in an antigen using recurrent neural network Proteins 2006 65 1 1 10.1002/prot.21078 16856179
Serwecińska L Antimicrobials and antibiotic-resistant bacteria: a risk to the environment and to public health Water 2020 12 12 12 10.3390/w12123313
Slack M Haemophilus influenzae type b disease in the era of conjugate vaccines: critical factors for successful eradication Expert Rev Vaccines 2020 19 10 903 917 10.1080/14760584.2020.1825948 32962476
Soeters HM Current epidemiology and trends in invasive Haemophilus influenzae disease—United States, 2009–2015 Clin Infect Dis 2018 67 6 881 889 10.1093/cid/ciy187 29509834
Suga S A nationwide population-based surveillance of invasive Haemophilus influenzae diseases in children after the introduction of the Haemophilus influenzae type b vaccine in Japan Vaccine 2018 36 38 5678 5684 10.1016/j.vaccine.2018.08.029 30122645
The ClusPro web server for protein–protein docking|Nature Protocols. https://www.nature.com/articles/nprot.2016.169 accessed 20 Sep 2022
Vitovski S Dunkin KT Howard AJ Sayers JR Nontypeable Haemophilus influenzae in carriage and disease a difference in IgA1 protease activity levels JAMA 2002 287 13 1699 1705 10.1001/jama.287.13.1699 11926894
Wen S Feng D Chen D Yang L Xu Z Molecular epidemiology and evolution of Haemophilus influenzae Infect Genet Evol 2020 80 104205 10.1016/j.meegid.2020.104205 31981610
Wiederstein M Sippl MJ ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins Nucleic Acids Res 2007 10.1093/nar/gkm290
Wilson C Mitchelmore P Brown A Culture-independent multilocus sequence typing (MLST) screening for Haemophilus influenzae cross-infection in non-cystic fibrosis bronchiectasis (NCFB) Access Microbiol. 2020 2 2 179 10.1099/acmi.fis2019.po0173
Zhang Q Immune epitope database analysis resource (IEDB-AR) Nucleic Acids Res 2008 10.1093/nar/gkn254
Zhang J Zhao X Sun P Gao B Ma Z Conformational B cell epitopes prediction from sequences using cost-sensitive ensemble classifiers and spatial clustering BioMed Res Int 2014 2014 689219 10.1155/2014/689219 25045691
| 36465492 | PMC9707196 | NO-CC CODE | 2022-12-01 23:20:21 | no | Netw Model Anal Health Inform Bioinform. 2023 Nov 28; 12(1):1 | utf-8 | Netw Model Anal Health Inform Bioinform | 2,022 | 10.1007/s13721-022-00395-x | oa_other |
==== Front
J Gastrointest Surg
J Gastrointest Surg
Journal of Gastrointestinal Surgery
1091-255X
1873-4626
Springer US New York
36443553
5535
10.1007/s11605-022-05535-2
GI Image
Transvaginal Evisceration
http://orcid.org/0000-0002-7515-3074
Hermoza Arámbulo Rodrigo [email protected]
Madrazo González Zoilo [email protected]
grid.411129.e 0000 0000 8836 0780 General and Digestive Surgery, Bellvitge University Hospital, Barcelona, Spain
28 11 2022
12
27 10 2022
29 10 2022
© The Society for Surgery of the Alimentary Tract 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
==== Body
pmcA 78-year-old female with past medical history of invasive bladder cancer was treated with neoadyuvant chemotherapy and surgery. In February 2022, it was performed a complete robotic bladder resection, urinary diversion with a Bricker technique, and total abdominal histerectomy with bilateral oophorectomy and lymphadenectomy. The postoperative period was uneventful, and the patient was discharged from the hospital 4 days after the surgery. During the ambulatory follow-up, the patient was diagnosed with vaginal and rectal prolapse.
Eight months later, the patient presented to the Emergency Department with intense pelvic and lower abdominal pain after an episode of nausea and vomiting, associated with a pelvic mass. The physical examination revealed a transvaginal evisceration of the small bowel (Fig. 1). After applying temporary containment of the intestinal loops and irrigating them with topical hypertonic glucose solution, the patient was immediately taken to the operating room. She underwent a midline laparotomy, with reduction of the bowel into the abdominal cavity. It was evidenced a 5-cm defect in the vaginal vault that was repaired with simple stitches (absorbable suture) (Fig. 2). No bowel resection was necessary (bowel without ischemia or perforation), and the Bricker ileal conduit was unscathed. After a mild COVID-19 postoperative infection, the patient made full recovery and was sent home 7 days after the intervention.Fig. 1 Small bowel transvaginal evisceration
Fig. 2 Defect in the vaginal vault
Transvaginal bowel evisceration is a rare surgical condition. The first case was reported in 1864 by Hypernaux et al., and a limited number of cases have been published since then.1 The risk groups for this surgical emergency include postmenopausal women, women who had past medical history of vaginal or uterine surgery, and patients with concomitant pelvic organ prolapse. The increase incidence in this group has been associated with decreased vascularization and vaginal wall atrophy.2 Premenopausal women can also be affected, but it is infrequent and more related to vaginal trauma or sexual activity .1 Many other factors can favor the onset of this surgical complication: elderly, obesity, radiotherapy, hypothyroidism, multiparous women, but most cases reported always described a situation increasing the intra-abdominal pressure in the context of pelvic floor weakness as a trigger moment for the evisceration (trauma, surgery, coughing, constipation, vomits).3 Patients usually present with abdominal pain, vaginal bleeding, and vaginal mass.1 Our patient was a postmenopausal woman who underwent recent bladder and uterine surgery, diagnosed with postoperative vaginal and rectal prolapse, and who experienced a transvaginal evisceration after having nausea and vomits.
Transvaginal small bowel evisceration carries high rates of mortality and morbidity (6–8% and 15–20%, respectively).3 Within its management, emergency surgery is a priority, and any delay in surgical intervention could eventually lead to gut injury, ischemia, bowel resection, sepsis, and death. Prompt reduction of the small bowel is necessary, and it must be carefully examined to evaluate its vitality.2 Transvaginal and transabdominal approaches have been used, and both have equal rates of recurrence and complications; the choice must be done depending on the clinical presentation of the patient. In 20% of cases, it is necessary to perform a bowel resection and anastomosis,1 followed by repair of the vaginal defect.
In conclusion, transvaginal evisceration is a life-threatening and infrequent pathology. It requires urgent diagnosis and emergency surgical treatment to prevent small bowel ischemia and its complications. If the eviscerated bowel is nonviable, resection and anastomosis should be done.
Data Availability
Data available on request from the authors.
Declarations
Ethics Approval
Written informed consent was obtained from the patient for publication of this case report and all images included. The confidentiality of the data is guaranteed by the authors.
Conflict of Interest
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Rana AM Rana AA Salama Y Small bowel evisceration through the vaginal vault: A rare surgical emergency Cureus 2019 11 10 e5947 31799088
2. Negrete JR Molina GA Sanchez AC Transvaginal evisceration of the small bowel a rare and potentially lethal event, a case report Ann Med Surg (Lond). 2021 65 102352 10.1016/j.amsu.2021.102352 34026097
3. Arabadzhieva E Bulanov D Shavalov Z Yonkov A Bonev S Spontaneous transvaginal intestinal evisceration in case of long-standing uterine prolapse BMC Surg. 2022 22 1 157 10.1186/s12893-022-01615-x 35509095
| 36443553 | PMC9707197 | NO-CC CODE | 2022-12-01 23:20:21 | no | J Gastrointest Surg. 2022 Nov 28;:1-2 | utf-8 | J Gastrointest Surg | 2,022 | 10.1007/s11605-022-05535-2 | oa_other |
==== Front
Med Biol Eng Comput
Med Biol Eng Comput
Medical & Biological Engineering & Computing
0140-0118
1741-0444
Springer Berlin Heidelberg Berlin/Heidelberg
2712
10.1007/s11517-022-02712-y
Original Article
Home-care nursing controlled mobile robot with vital signal monitoring
Mireles Caridad [email protected]
1Caridad Mireles
received the B.E. in biomedical engineering from the UPIBI - IPN in 2020. Currently, she is a student of the Ms. C. in Computing Technology from the CIDETEC - IPN.
Sanchez Misael [email protected]
1Misael Sanchez
received the B.E. in biomedical engineering from the UPIBI - IPN in 2019 and the Ms. C. in Computing Technology from the CIDETEC - IPN in 2021.
http://orcid.org/0000-0003-2066-7470
Cruz-Ortiz David [email protected]
2David Cruz-Ortiz
received the Ph.D. degree in control sciences from the Department of Automatic Control, CINVESTAV-IPN in February 2020. His research interests include robust control and robotic systems.
Salgado Iván [email protected]
1Iván Salgado
received the Ph.D. degree in computer sciences from CIC - IPN. His current research interests include the design, control and optimization of robotic systems and adaptive control.
Chairez Isaac [email protected]
23Isaac Chairez
received the Ph.D. degree in control sciences from the Department of Automatic Control, CINVESTAV - IPN in 2007. He is currently with UPIBI - IPN.
1 grid.418275.d 0000 0001 2165 8782 Centro de Innovación y Desarrollo Tecnológico en Cómputo, Instituto Politécnico Nacional, Av. Juan de Dios Bátiz s/n, 07700 Mexico City, Mexico
2 grid.418275.d 0000 0001 2165 8782 Medical Robotics and Biosignals Lab, Unidad Profesional Interdisciplinaria de Biotecnología, Instituto Politécnico Nacional, Av. Acueducto, 07340 Mexico City, Mexico
3 grid.419886.a 0000 0001 2203 4701 School of Engineering and Sciences, Tecnologico de Monterrey, Campus Guadalajara, Zapopan, Guadalajara, Mexico
29 11 2022
122
11 10 2021
27 10 2022
© International Federation for Medical and Biological Engineering 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
This study describes the development (design, construction, instrumentation, and control) of a nursing mobile robotic device to monitor vital signals in home-cared patients. The proposed device measures electrocardiography potentials, oxygen saturation, skin temperature, and non-invasive arterial pressure of the patient. Additionally, the nursing robot can supply assistance in the gait cycle for people who require it. The robotic device’s structural and mechanical components were built using 3D-printed techniques. The instrumentation includes electronic embedded devices and sensors to know the robot’s relative position with respect to the patient. With this information together with the available physiological measurements, the robot can work in three different scenarios: (a) in the first one, a robust control strategy regulates the mobile robot operation, including the tracking of the patient under uncertain working scenarios leading to the selection of an appropriate sequence of movements; (b) the second one helps the patients, if they need it, to perform a controlled gait-cycle during outdoors and indoors excursions; and (c) the third one verifies the state of health of the users measuring their vital signs. A graphical user interface (GUI) collects, processes, and displays the information acquired by the bioelectrical amplifiers and signal processing systems. Moreover, it allows easy interaction between the nursing robot, the patients, and the physician. The proposed design has been tested with five volunteers showing efficient assistance for primary health care.
Graphical Abstract Main stages of the home-care nursing controlled mobile robot
Keywords
Elderly home-care
Nursing robotics
Biosignal processing
https://doi.org/10.13039/501100007161 Secretaría de Investigación y Posgrado, Instituto Politécnico Nacional SIP-20221503 Cruz-Ortiz David https://doi.org/10.13039/501100007161 Secretaría de Investigación y Posgrado, Instituto Politécnico Nacional SIP-20221150 Cruz-Ortiz David https://doi.org/10.13039/501100007161 Secretaría de Investigación y Posgrado, Instituto Politécnico Nacional SIP-20221746 Salgado Iván
==== Body
pmcIntroduction
The average aging of the world population has been increased due to health, social and technological factors, including mainly the fertility rate detriment and the increment of the world population life span [1]. Nowadays, some studies estimate 727 million people who are 65 years old or over in the global population in 2020 [34]. Furthermore, projections are estimating that the aged population in 2050 will increase near to 1.5 billion people, which means that the percentage of 9.3% in 2020 will increase to 16.0% in 2050 [34]. Also, Eastern and South-Eastern Asia are the regions housing 260 million elderly population [33]. Then, Europe and Northern America are the next on the list having a population of 200 million in 2019 [33].
Human aging may originate from the decrease in functional human body productivity, the increment in the risk of cardiovascular episodes, chronic degenerative diseases such as melanomas or skin lesions and muscle-skeletal diseases, among others [26–29]. In consequence, geriatric medical cares are indispensable for reducing physical accidents or clinic problems that put at risk the life of patients [10]. Indeed, some countries have created strategies for taking care of elderly population [1]. They have improved the specialization of health services and the development of biomedical technologies that enhance health conditions for the patients. For example, digital biomedical technologies such as telemedicine, telecare, and e-health can supply long-distance attention. These techniques allow bidirectional, safe, and professional communications between patients and healthcare professionals [3, 11].
Robotic technologies provide interplay channels between patients and healthcare professionals. Indeed, robots are fundamental tools of modern life. These devices provide support in a diversity of industrial and domestic activities [6]. Particularly speaking, in medicine, some applications of robotic devices can perform robotic surgeries, and they can develop intelligent systems to visualize the physiological variables of patients, as well as to give them support in their daily routines [4, 16]. All the assisting robotic technologies have the main objective of improving the quality of life for people who have physical disabilities developing their daily routine [8]. For instance, some hospital facilities have been instrumented with robotic systems that deliver food or medicines under a programmed schedule. In addition, some of these robots support patients in their movements and record clinical information for long-distance medical attention [20, 31]. Such assisting procedures can reduce the complications of healthcare conditions. The interplay between the robotic configuration and the set of biomedical monitoring devices leads to three major medical assistance categories: mobile service, physical assistance, and supporting people in their activities [23]. Usually, all these robotic systems are indistinctly known as nursing robots. Nursing devices have diverse functions such as supervising people, recording information, mechanical support to assist people, measuring temperature, sanitation, and sterilization, among many other functionalities.
Regarding humanoid robots, several examples provide healthcare assistance [5]. Such is the case of ARMAR III and PR2, which have been designed to interact with humans and manipulate objects in different environments [2, 9]. Also, the Care-O-Bot 3, a robot assistant designed to help humans in everyday environments, can manipulate objects [24]. Nevertheless, the main purpose of these devices is not to provide healthcare services. Therefore, even though the mechanical structures satisfy some requirements needed in assistance robotics, their level of specialization should be improved to develop specific tasks associated to an adequate healthcare assistance.
One of the first successful robots with nursing functions is the RIBA robot, which was designed by Honda Robotics. The main purpose of this device is to carry patients over 80 kg weight [23]. Under the SARS-COV-2 pandemic situation, some countries implemented nursing robots to attend sick patients with a teleoperated scheme. Another relevant example is the Cody system, which is a robot assistant designed to help caregivers with the hygiene of the patient [21]. The AIMBOT, which can sanitize and get the temperature of patients while moving around hospital environments. Despite the relevant technical advances, current monitoring and long-distance treatment functions associated with nursing robots are limited to supporting medical staff [25]. Thus, even tough their significant advances, nursing robots are insufficient supporting nursing activities.
Home-care of senior citizens should employ a complete measurement of physiological variables like heart rate, blood pressure, temperature, and oxygen saturation. In addition, assisting robotic devices should help patients to ambulate in controlled spaces. Taking this paradigms into account, this work proposes the development of a home care robot prototype that can help geriatric patients supporting their gait cycle and monitoring their mentioned vital signs. Finally, a GUI visualizes the collected information from the patient, including the acquired physical variables to be transmitted to the physician in charge to obtain an adequate diagnosis.
The main contributions of the proposed nursing robotic device are the following: The development of a prototype of a nursing robotic device that can measure electrocardiography signals, non-invasive arterial pressure, heart rate, oxygen saturation, and skin temperature.
The design of an adequate graphical user interface to allow the interaction between the robotic device and the patient as well as the physicians.
A functional traction system controlled by a robust control algorithm with the objective of localizing and tracking the patient helping him/her to exert an assisted gait cycle.
Pre-clinical validation of the nursing robotic device considering the evaluations of all sections of the nursing robotic device.
This manuscript is organized as follows: Section 2 describes the specific requirements for the nursing robotic device, which serves to propose its main operative characteristics. Section 3 details the mechanical structure and the technical abilities included in the nursing robotic device, such as the medical technologies for monitoring vital signs. Section 4 describes the electronic elements associated with the biomedical technology that measure the physiological variables continuously. Section 5 establishes the automatic controller, which aims to regulate its mobilization towards the patient when needed. Section 6 highlights the results of the proposed controller obtained using a virtualized representation of the assisting nursing robotic device (ANRD). Section 7 contains the details of the experimental evaluation considering the mechanical structure, its medical instrumentation, and the suggested automatic controller. Finally, Section 8 closes the study with some final remarks and future trends.
Design requirements and medical assisting abilities
Medical devices such as assistance robots provide clinical services to patients in different scenarios: home-care or hospital environments. According to [12], the main characteristics to be solved for a medical assistance robot should include and adequate physical design, a safety instrumentation, and a manageable operating system for the patient and the physician. Among others, the physical requirements include robustness, an ergonomic design easily adapted for the patient, static stability to support the patient, among others. Therefore, the mechanical design must be solid and stable for the safety of patients.
Another characteristic must be the adaptation of the nursing robot to different terrains and environmental situations because the system must provide assistance to the user to execute different tasks such as sitting or walking [12]. In addition, the instrumentation and the system operation should be easy to manipulate by users such as nurses, doctors, and patients. According to [7], the instrumentation must be designed considering clinical purposes such as vital signal monitoring. The operating system requisites the implementation of a human-machine interface for analyze the data. The interface needs to be intuitive and safe.
Regarding the physical design, the mechanical structure of the ANRD must allow an appropriate interaction with the patient according to the medical and home care assisting requirements. Notice that taking care of senior citizens implies the fulfillment of several tasks. Therefore, this study considers the walking aiding service under controlled indoor-outdoor conditions, a simplified vital signals monitoring, and autonomous mobilization to search and track the patient when the robot is not with him. The assisted-aided walking function must help patients who suffer from mobility diseases by accidents or the aging deterioration that affect their independent gait cycle. During the aging period, physiological changes are reflected in modifications of either nervous or peripheral mechanisms controlling equilibrium and locomotion systems. Senior citizens have a stance phase reduction from 0.64 to 0.39 s, and the necessity for a double support increases between 0.24 and 0.28 s in the gait cycle [14]. Therefore, the mechanical design for the proposed device has been defined taking into account the anthropomorphic parameters for the elderly population, which are summarized in Table 1. Table 1 Average anthropomorphic dimensions considered for the robot design
Parameters Men Women
Elbow height 104.1 cm 85.7 cm
Height 163.2 cm 150.6 cm
Maximum length of arm 51.3 cm 49.8 cm
Step width 69 cm 61cm
As a consequence of the mobility limitations for senior citizens, the suggested device must develop controlled displacements from one point to another in a non-structured working space. Hence, a class of robust controller can help the nursing robot to track the patient.
Physiological measurements are necessary to monitor the health of the senior population. These clinical indicators provide relevant information about the current status of cardiovascular, respiratory, nervous, and endocrine systems [13]. Furthermore, vital signs define a universal mechanism to know the health condition of the patient, including a preliminary grading of eventual fatal events [32]. Elderly patients suffer geriatrics syndromes that affect human health. These syndromes could be a consequence of functional disorders (involving different systems) such as diabetes, hypertension, heart deceases, cognitive symptoms, dementia, depression, dislocations, and fractures [17].
The proposed nursing robotic device measures heart rate, arterial blood pressure, oxygen saturation, and body temperature as the most significant vital signs. A particular mechanical and instrumentation design yields a simplified method to get the physiological information without the intervention of physicians or nurses. The acquired signals are displayed in a graphical user interface that provides information to the physician in charge of the patient. Also, the proposed interface allows the registration and save the acquired information for each patient in a proposed database that is integrated into the digital section of the instrumentation device of the nursing robot.
In order to evidence the novelty of the proposed ANRD, a comparison between commercial devices and the ANRD was made to evidence the novelty of the proposed device. Table 2 shows the features considered in the design of the ANRD. Here, it can be observed that the ANRD considers novelty characteristics to assist the senior citizen population. Table 2 Feature comparison of the ANRD with commercial systems
Robot Dimensions Sensors Vital signs Functions
ANRD 40 kg GPS Temperature Support
62 x 53 x 133 cm ECG SpO2 Monitor signs
Temperature Blood pressure User interface
Pressure
AIMBOT 50 kg Camera web Temperature User interface
193 x 55 x 55 cm Pressure UV disinfection
LiDAR
RIBA 180 kg Camera – Support
75 x 84 x 140 cm Microphone Voice recognition
Care O-Bot 180 kg Tactile sensor – Manipulation
75 x 55 x 145 cm Camera User interface
Touchscreen
Mechanical design of the nursing device
According to the requirements, a mechanical structure was designed in a computer assisting design (CAD) software considering all the mechanical elements of the nursing robotic device. The dimensions of the robot are in agreement with the anthropomorphic characteristics of averaged elderly patients (see Table 1).
The internal structure of the robotic device consists of aluminum profile model Strut profile 30x30 M8/- material number 3842990721 from Bosch, which according to their technical data, is ideal for supporting the senior citizen weight. The dimensions of the ANRD are 62 cm length, 53 cm width and a 133 cm height. This proposal gives support and stiffness for adequate security of patients when they interact with the nursing robot. A traction mechanism supports the internal structure for mobilizing the system in structured environments with stable motion. This mechanism has been adapted considering anthropomorphic measurements to offer the patient a friendly companion device.
For a comfortable interaction during the patient’s gait, the front part has an aperture at the chest height and supports the hands over the nursing robot considering the height average of senior citizens. The traction mechanism provides motion capacities to the nursing assistance robot. This mechanism incorporates aluminum plates and has two side trolley bands (Fig. 1). Two lateral DC motors actuates the sidebands, one on each side. A polymeric (acrylic) cover protects the external structure offering a friendly image to the patient. In addition, this cover protects all the internal elements. Fig. 1 Components of the ANRD. The robot has two sections; aluminum profiles constitute the mechanical structure covered with self-designed 3D-printed elements. The upper section has the corresponding physiological sensors, while the lower section contains the traction mechanisms
The robot structure consists of two principal sections: the upper part measuring the vital signs and the lower part in charge of the mobilization mechanism, consisting of mobilization actuators, energy sources, and instrumentation boards.
The upper part is integrated by all the elements that measure the physiological signals and the embedded systems that process them. In this particular case, a touch screen LCD with dimensions of 25.6 cm × 17.5 cm shows the information to the patients and gives them the option to send the information to the physician in charge (see Fig. 1).
The hand-shaped molds were designed with specific sections that hold the sensors for each different signal (see Fig. 2). This strategy minimizes the wiring and facilitates the use of the robot by the patients. In addition, the internal structure has two molds for supporting the patient hands. The location of these molds is 90 cm over the robot basis, inside its upper section. Fig. 2 Design of hand-shape molds for supporting sensors and measurements systems of electrophysiological signals
Location of acquisition electrodes
The sensor used to estimate the heart rate consists of an electrocardiography amplifier. This device is in charge of recording the electrical heart signals and requires three electrodes (2 differential and one reference) [18]. Two electrodes are located at each mold (left-hand and right-hand molds) which are in contact with the user’s palms (See Fig. 2). The third one is a clip electrode attached to the inferior part of the nursing robot. Notice that this element is a dry electrode, which only requires the patient to move their leg and touch the electrode to obtain the measure of the biosignal. Although these electrodes were selected due to their reusability and easy collocation, the user should connect the electrode to their left ankle. Thus, when the patient holds the nursing robot, the palms hold the molds so that the electrodes contact the palms yielding in the construction of the Einthoven’s triangle.
Location of bracelet for non-invasive pressure acquisition
The measurement of the arterial blood pressure was taken in the index finger of the left hand. This methodology follows the same strategy used for the traditional measurement over the arm but an instrumentation with higher sensibility characteristics is needed. The acquisition system used a polymer mold corresponding to a bracelet with a rubber chamber, which covers the finger where the non-invasive arterial pressure will be measured.
Position of oxygen saturation variable
The oxygen saturation (SpO2) measurements were acquired using a standard oximetry device attached to a polymer piece placed inside the hand-shaped mold. The oxygen saturation was obtained from the right index finger. The designed mechanism can be adapted to different finger sizes.
Position of temperature monitoring sensor
An infrared sensor is placed over the skin to obtain the body temperature. To measure the temperature, the patient must approach his/her head or face (around cheekbones) to the measuring device. Thus, the proposed device corresponds to the traditional long-distance measuring of temperature, which simplifies the interaction between the patient and the nursing robot device.
Instrumentation of the nursing device
The primary vital signs are periodically acquired to know the patient’s health condition. To this end, the biomedical instrumentation employs a single processing board TIVA TMC1294 (EK-TM4C1294XL, 120MHz 32-bit ARM Cortex-M4 CPU, 1MB Flash, 256KB SRAM, 6KB EEPROM, Integrated 10/100 Ethernet MAC+PHY, Dual 12-bit 2MSPS ADCs, motion control PWMs) based on a microcontroller of Texas Instruments (TM4C1294NCPDT). The digital acquisition of biophysical signals uses a parallel analogical-digital-converter. The data transfer uses protocols like I2C and RS-232 (115200 bauds). The microcontroller sends the information using the serial communication protocol. The data processing is exerted in Matlab®, including frequency analysis, visualization, and storing. Figure 3 shows the general diagram of the suggested bio-instrumentation. Fig. 3 Biomedical instrumentation diagram
Heart rate
The heartbeat rate was monitored using the sensor AD8232. This device is a fully integrated single-lead electrocardiogram (ECG). The main features of this device are a low supply current of 170 μ A common-mode rejection ratio of 80 dB and high signal gain G = 100 with dc blocking capabilities via a high-pass filter). The sensor AD8232 is connected to the processing board’s analog-digital converter (ADC). The Pan-Tompkins algorithm was used to analyze the ECG to obtain the heartbeat (bpm) as HR= 60/T [15]. The processing algorithm obtains the inter-period between QRS complexes.
Body temperature
The selected sensor is the MXL9014, which is an infrared temperature sensor with a range of 7–280 ∘C that employs a serial protocol UART to send the data. Considering that the body has some regions where the temperature is homogeneous, there are zones where the measurements are reliable such as cheekbones or the forehead [30]. This sensor was attached to the upper part of the ANRD. The sensor required a characterization bases on a sequence of tests made on different volunteers.
Oxymeter
The SpO2 measurement was obtained with a MAX 30101 sensor, an oximeter that emits two wavelengths: the oxygen hemoglobin (O2Hb) and no-oxygen hemoglobin (RHb). These signals have different absorption patterns in the infrared and red wavelength spectrum. The RHb absorbs mainly in the visible spectrum at 660 nm wavelength, while O2Hb absorbs infrared light at 940 nm wavelength. The lectures of absorption at infrared and red wavelengths yield in the saturation percentage using an algebraic relationship between the local maximum of the acquired signals [19]. Because of the sensitiveness of the photodiode, this device was enough to complete the acquisition procedure with only one device.
Arterial blood pressure
The arterial blood pressure was measured at the index finger of the left hand, and its measurement requires an automatic procedure that records the Korotkoff sounds. A general diagram of the stages involved in this process can be observed in Fig. 4. In the first step, a microcontroller TM4C1294 activates the pump air to inflate the bracelet until an internal pressure of 195 mmHg is reached. To ensure the mentioned pressure, a microvalve regulates the air inflow towards the bracelet, which is placed around the index finger (see the stage “Bracelet” in Fig. 4). Then, the microvalve opens the air pathway in a controlled way. The output valve is completely opened when the pressure comes to 40 mmHg, releasing the bracelet’s air. The pressure sensor obtains two signals: the raw pressure signal and the filtered pressure signal (high-pass of 1 Hz, low-pass of 4 Hz, and a Notch filter of 60 Hz). The signal is adequate for obtaining the Korotkoff sounds, and in the final stage, the microcontroller measures the pressure signal via an analogical RS-232 transfer method. Fig. 4 General diagram for arterial pressure measure
Consequently, the data are processed considering the application of digital filtering and threshold the Korotkoff sounds corresponding to 80% and 55% of the amplitude signal for systolic and diastolic pressures [19]. This procedure is automatically controlled by the microcomputer interface fixed up into the assisting robot. The algorithm for getting the non-invasive pressure implements a band-pass filter from 1 to 5 Hertz and an amplitude characterization of the oscillatory pressure response at the detected signal. The analysis of such signals leads to the estimation of the non-invasive arterial pressure.
Instrumentation used for the mobilization system
The ANRD is carried out by a four-wheel differential-drive mobile robot, which is a class of autonomous unmanned vehicle (AUV). A general overview of the electronic instrumentation in the AUV is depicted in Fig. 5. From this figure, it can be noticed that the proposed instrumentation considers different electronic elements such as wireless communication devices, power converters (H-bridges or DC-DC), and batteries. Fig. 5 Electronic instrumentation of the AUV
In Fig. 5, the first stage employs a microcontroller TM4C123, which is in charge of generating the PWM signals to regulate the movement of the AUV. Next, the PWM signals serve as an input of the high-power stage integrated by a couple of H-Bridges (DC-DC power converter) after passing the isolation stage, which separates the low-power stage from the high-power stage. Once the AUV is moving due to the PWM signals, a GPS module provides the position coordinates of the nurse robot. Then, a set of wireless transmitters (Xbee devices) sends the AUV position information to a personal computer, where the control algorithm is computed. Finally, the obtained control signal (duty-cycle of the PWM) is sent to the microcontroller in the AUV to close the controller loop.
The traction mechanism instrumentation is located in the lower part of the robot. It consists of mobilization motors, energy sources, and instrumentation boards. The instrumentation of the traction mechanism consists of the following phases: It is necessary to know the relative position between the robot and the patient with respect to a non-inertial frame. This study considers the application of a high-resolution GPS module (SparkFun GPS-RTK2 Board - ZED-F9P). These modules have 10 mm precision on all three dimensions (latitude, longitude, and altitude). With this information, it is possible to define the tracking error used in the controller design. For validation purposes, a camera device was placed over the working space to confirm the effectiveness of the GPS positioning device.
Wireless communication strategy. A proposed peer-to-peer radio-frequency communication system solved the data transmission procedure. A Zigbee module (Silicon Labs EM357 SoC Transceiver chipset, indoor communication range of 60 m, and transmit power of 3.1 mW (+ 5 dBm) / 6.3 mW (+ 8 dBm) boost mode) from the Xbee company implements the communication protocol. The Zigbee devices use the universal asynchronous receive and transmit (UART) protocol communication at 9600 bauds.
Generation of PWM (Pulse Width Modulation). The communication stage transmits the value obtained from the automatic control algorithm. This value corresponds to the average power that must be delivered to each DC motor. The data received with the Xbee receptor (in ASCII) attached to the processing board is converted into a pulse width modulation (Operative frequency of 200 Hz) according to the required differential velocity for solving the tracking problem. Finally, the modulate pulses are transmitted to DC-AC power converters (H-bridges) that provide motors the needed electrical power, direction, and orientation according to the commands provided by the automatic controller.
Isolation. The generated pulses requires an isolation circuit that is implemented with the integrated circuit 4N25 optoisolators.
Tracking control algorithm design
Dynamic description of the mobile nursing robot
Following the study presented in Kozłowski and Pazderski [22], the AUV position is given considering the location of the center of mass with respect to a selected inertial framework O1 located at its relative origin. The vector η:=xyα⊤ corresponds to the AUV position described by the displacement in a two-dimensional plane x and y measured each one in meters and the rotation with respect to the α axis measured in radians. The AUV dynamic model is defined as follows 1 η¨=M−1−A(η˙)+B(η)τ−ξ(η,τ),
where M:ℝ3→ℝ3×3 corresponds to the inertia matrix satisfying M=diagm,m,I with m and I being the AUV mass and inertia term respectively, diag{z} is the diagonal matrix formed with the elements in z. Here, it is considered a homogeneous mass distribution in the ANRD and also it is assumed that M > 0.
The vector A(η˙)∈ℝ3 is given by A(η˙)=Ax(η˙)Ay(η˙)Ar(η˙)⊤ corresponding to the lateral skidding forces in the wheels with respect to the x and y axis and the corresponding moment in the center of mass. The matrix B∈ℝ3×2 establishes the relation of input vector τ with the AUV dynamics given by B=1rcosαcosαsinαsinα−c−c
where r is the radius of each wheel, and c is the half of the AUV width.
The vector τ∈ℝ2 is the torque exerted by the wheels at the left side τL and right side τR respectively, that is, τ=τLτR⊤. The vector ξ∈ℝ3 represents the external forces and perturbations acting over the AUV dynamic satisfying ξ(η,τ)≤ξ+,∀t≥0
Taking into consideration that AUV dynamics given in Eq. 1 does not consider the non-holonomic restrictions, to produce the AUV motion without skidding, the following constraints must be satisfied y˙α˙<β,y˙+r0α˙=0,
where r0 is the radius of rotation for the AUV center of mass in the x-axis and β is a known positive constant. Here, it is necessary to satisfy 0 < r0 < β to avoid skidding the robot. In terms of generalized coordinates, the expression in Eq. 1 is rewritten as C(η)η˙=0,C(η)=−sinαcosαr0,
The dynamic expression considering the non-holonomic restrictions admits the following representation 2 Mη¨=−A(η˙)+B(η)τ−ξ(t,η)+C⊤λ.
with λ being the vector of Lagrange multipliers corresponding to constants proposed in Eq. 1. The relation between the inertial velocity η˙ and the pseudo velocity v with respect to the body framework O2 is expressed as 3 η˙=Q(η)v,Q(η)=cosα−sinαsinαcosα0−1r0.
Considering the AUV dynamics in Eq. 2 together with the expression given in Eq. 3 for the pseudo velocity v and taking λ = 0, the following states are obtained 4 η˙=Q(η)v,v˙=(Q⊤MQ)−1Q⊤−MddtQv−A(η˙)−f(t,η)+B(η)τ.
Let propose a dynamic compensation based on the control action τ such as τ=−(Q⊤B)−1Q⊤MQu+Q⊤MddtQv+Q⊤A(η˙),
where u is an internal control vector given by u=u1u2⊤. Then, the dynamics in the expression (4) becomes η˙=Q(η)v,v˙=u+ξ~,
with ξ~=(Q⊤MQ)−1Qξ1ξ2⊤.
Output feedback based linearization
Let introduce the following artificial output that, according to [22], describes the instantaneous center of rotation of the nursing robot 5 y=x+r0cosαy+r0sinα.
Taking into consideration that the proposed coordinate transformation (by an output feedback transformation) allows to get a linearized version of the dynamics of Eq. 4 (with the inclusion of a nonlinear output feedback form). Then, a feasible state feedback using a dynamical extension in the control input u1 can be expressed as 6 u1=ξ~1+ζ˙,ζ=υ1+ddtξ~1,u2=υ2,
that can be rewritten as 7 u1ζu2=Buυ+ξ~1+ζ˙ddtξ~10,
where the control action is given by the vector υ∈ℝ2 described as υ=υ1υ2⊤ and Bu=02x1I2x2⊤. Here, I2×2 describe the identity matrix.
The corresponding dynamics of the output y = y1 described in Eq. 5 is 8 y˙1=y2,y˙2=y3,y˙3=h1(η,v)υ+h2(η,v)+Γ(ξ~1,ξ˙1,ξ˙2),
where v=v1v2 are the set of admissible velocities defined in Eq. 3 and h1(η,v)=cos(α)1r0v1sin(α)sin(α)−1r0v1cos(α),h2(η,v)=2r0ζv2sin(α)−1r02v1v22cos(α)−2r0ζv2cos(α)−1r02v1v22sin(α).
Finally, function Γ(ξ~1,ξ˙1),ξ˙2) joints the action of the signal perturbations produced by the non-modeled dynamics and uncertainties in ξ in Eq. 2.
Control structure
The nursing robot movement was enforced by a class of state feedback following a class of extended proportional-derivative controller (PD) that satisfies the following dynamics: 9 υ=h1−1−h2(η,v)+Kpe+Kdė+Kddë,e=q∗−q,
where e is the vector of tracking errors, Kp is the matrix of proportional gains, Kd is the matrix of derivative gains, and Kdd is the feedback of the double derivative of the tracking error. The proposed controller can be realized without restriction, considering that the inverse of h1 is well defined all the time, and it depends only on information that can be measured online. With the aim of developing the controller, the Euler derivative obtains the vector of derivative errors ė as well as its second derivative, which is needed in the proposed design. Notice that the proposed controller can be simplified using a state estimator for the time derivative of e. Even more, the inclusion of sliding mode based differentiators can enhance the controller performance tracking the reference trajectories [35].
Numerical evaluation of the mobilization
Once the control design has been proposed, the tuning of its gains used a numerical simulation to run the control action for securing the robot motion under realistic scenarios, while the reference trajectories are tracked. The simulation was based on the Simscape Toolbox of Matlab® software.
The simulation results exports a CAD version Matlab® to create a virtual space where the control action can be evaluated. Furthermore, the simulation included two principal robotic devices: the nursing robot and a biped robot representing the patient. Thus, the biped robot’s trajectories for axis X, axis Y, and orientation were proposed. The movement of the nursing robot must solve three principal motion problems: (1) To approach to the biped position (y¯1,i(t), x¯1,i(t)), (2) to follow a trajectory to reach a second set of coordinates (y¯2,i(t), x¯2,i(t)) and (3) to develop a circular trajectory (y¯3,i(t), x¯3,i(t)). The equations that describe the trajectories corresponding to these three problems are the following: 10 xi∗(t)=x¯1,i(t)ift≤t1,x¯2,i(t)ift1<t≤t2,x¯3,i(t)ift>t2,yi∗(t)=y¯1,i(t)ift≤t1,y¯2,i(t)ift1<t≤t2,y¯3,i(t)ift>t2,
where i = {B,NR}, B and NR denote the biped robot and the nursing robot respectively.
The transitions time happens in t1 = 5 and t2 = 10. Then, the initial points where the mobile robot is located during the first stage correspond to x1,NR(0) = 0 and y1,NR(0) = 1.5. Then, the second set points are x2,NR(0) = − 2.3 and y2,NR(0) = − 2.5. After that, the points where the nursing robot and biped must come to start the third stage are x3,NR(0) = − 4.5 and y3,NR(0) = − 1. Hence, the trajectory during the first stage is originated by two sigmoidal equations between (x1,NR(0),y1,NR(0)) to (x2,NR(0),y2,NR(0)). The second trajectory between (x2,NR(0),y2,NR(0)) to (x3,NR(0),y3,NR(0)) was also proposed using a set of sigmoidals, whereas the last stage is given by x3,NR and y3,NR. All the mentioned functions that describe the reference trajectories satisfy the subsequent equations. 11 x¯1,i(t)=x0+x1−x01+e−2.1(t−0.5),y¯1,i(t)=y0+y1−y01+e−2.5(t−0.5),x¯2,i(t)=x1+x2−x1+0.151+e−2.5(t−7.8),y¯2,i(t)=y1+y2−y1+0.151+e−2(t−8.8),x¯3,i(t)=1.7sinπ2t,y¯3,i(t)=1.7sint+π2+0.45π2,
Table 3 contains the parameters used for testing the PD control in simulation. Table 3 Set of gains used in the controller evaluation
Kp Kd Kdd
Axis X 6000 200 20
Axis Y 6000 300 20
𝜃 15000 200 20
Figures 6 and 7 show the comparison of reference trajectories, the biped coordinates, and the controlled trajectories of the nursing device for both axes X and Y with respect to the time. The red and cyan lines represent the reference trajectories corresponding to the equations presented in Eq. 11. The black and blue lines represent the position of the nursing and the biped robots, respectively. Fig. 6 Trajectory of x-coordinate of biped and nurse robots
Fig. 7 Trajectory of y-coordinate of biped and nurse robots
In this case, the PD controller allows tracking the different parts of the reference trajectory. Figure 8 shows the reference orientation trajectory given by 𝜃=arctanyx. Figure 9 depicts the position for both coordinates x and y, where the first and second trajectory given by the sigmoidal equations are depicted. Notice that the proposed controller solves the tracking of the reference coordinates during the first second of numerical simulation. Fig. 8 Trajectory of 𝜃-coordinate of biped and nurse robots
Fig. 9 Phase plane trajectory of biped and nurse robots
Figure 10 depicts eight frames (a-h) obtained from 15 s of simulation where different positions of the nursing and biped robots are reached. In each frame, two figures can be observed, the right side of each subfigure depicts the isometric view of the robots, while the left side depicts the upper view. Subsections (a-d) correspond to the sequence of times occurring during the first trajectory. Then, the subsections (e–f) correspond to the second part of the trajectories. Finally, the images labeled with letters (g–h) correspond to times included in the third part of the reference trajectory. Fig. 10 Simulation of biped and nurse robot
Experimental evaluation
Biomedical instrumentation system
The integration phase for the experimental system consists of gathering each part of the nursing robot: the mechanical section, the biosignal instrumentation, and the motion system. These sections were interconnected in a microcomputer that is placed over the upper part of the nursing robot. The microcomputer can make the decisions connecting all the operating functions for the patient based on a GUI. Figure 11 shows the general diagram of the assisting robot integration. Fig. 11 Diagram showing the integration elements of the nursing robot
In this particular case, the instrumentation section gets the measurements of each biosignal. Then, the results obtained in the instrumentation phase are the signals corresponding to the ECG, corporal temperature, oxygen saturation, and arterial pressure blood. Each biosignal acquired in the microcontroller is transmitted to the graphical interface on the microcomputer.
In order to evaluate the biomedical instrumentation stage, a pre-clinical evaluation with young people was proposed. Here, it should be noticed that even when the objective of the nursing robot is to assist the elderly population, it is crucial to validate the performance of the nursing robot in a less vulnerable population. Then, young people were selected for pre-clinical evaluation.
In this case, 10 healthy volunteers (four males and six females, in a range age from 23 to 38 years old) participated in this study. The participants were selected according to the inclusion protocol SIP-20211220 approved by the research committee and regulated by the research and postgraduate secretariat (Secretaría de Investigación y Posgrado del Instituto Politécnico Nacional). The inclusion criteria are healthy participants without any neuromusculoskeletal problem, cardiovascular, pulmonary or neurological disease. In addition, the participants can be subjects of both sexes, males or females, between 18 and 40 years old, from any race, religion, and ethnic orientation. For this pre-clinical evaluation, the data from 10 participants within the ages stipulated in the protocol was considered. In Table 4, a summary of the participants in this study is depicted. Table 4 General information of the participants in the pre-clinical evaluation
Participant Age Gender Occupation Physical activity
1 24 F Student Low
2 24 M Student Low
3 37 F Student Low-middle
4 24 M Professor Low
5 31 F Professor Low
6 32 M Professor Low
7 38 F Student Low
8 23 M Student Low
9 24 F Student Low
10 24 F Student Middle
Here, it should be hightailed that all the participants previously signed an informed consent as part of this research project. Then, all participants agreed to take part in the experiments. Also, they have given their permission for the use of their photographs in this manuscript.
The stages that describe the procedure to get the heart rate measurement are summarized in Fig. 12. First, the designed module acquired the ECG signal (solid blue line). Then, the ECG signal’s time derivative is computed to detect the characteristic peaks in the ECG signal. As a result, the signal processed (solid red line) is obtained, corresponding to the moment when the QRS complex appeared in the original ECG. After that, a peaks detection algorithm was implemented in the final stage (solid green line). Once the peaks detection procedure is completed over a given period T, the equation FC = 60/T estimates the heart rate in beats per minute. Fig. 12 Stages of the ECG signal processing
The arterial pressure measurement obtains the two signals appearing in Fig. 13. The red signal represents the pressure information coming from the sensor measured by the selected microcontroller. The blue signal represents the filtered information by three active filters: high-pass of 1 Hz, low-pass of 5 Hz, and a reject band twin filter centered at 60 Hz with a bandwidth of 5 Hz. Then, the signal is acquired by the chosen microcontroller. The blue signal in Fig. 13 shows the Korotkoff components that are processed for obtaining the systolic and diastolic pressure in sections (A) and (B) of the figure. Fig. 13 Signal pressure
The measurements obtained from the 10 participants in the pre-clinical validation were compared with commercial pulse oximeter to validate the heart rate module. In this case, a pulse oximeter manufactured by an conventional commercial device (EleGate HOG15), which provides a heart rate measure and oxygen saturation, was used to make the comparison. The obtained results are evidenced in Table 5. Therefore, as the percentage error evidence, the mean error between the measures obtained with the proposed module and a commercial device is 2.9%, which is admissible in the suggested application. Table 5 Heart rate comparison values
Participants ANRD Commercial device % Error
(heart rate) (heart rate)
1 63.3 64 1.09
2 73.1 76 3.85
3 73.1 76 9.0
3 63.65 70 4.26
4 64.4 67 3.88
5 63.05 64.5 2.25
6 61.7 62 0.48
7 68.35 68 0.51
5 61.9 63 1.75
6 74.8 73 2.45
7 77.9 78.5 0.128
Regarding arterial pressure, the module was also evaluated in the 10 volunteers selected according to the inclusion protocol. Then, the designed module produced an error of 10%, demonstrating that the arterial pressure values measured with the finger instrumentation are similar to those obtained with commercial sphygmomanometer (Model Homecare) measurements. Notice that the device used in this particular case is a manual instrument. Therefore, the measures’ reliability depends on the ability of the person in charge to execute the task. On the other hand, the measures of the oxygen saturation as well as the corporal temperature are shown in Tables 6 and 7, respectively. Table 6 Oxygen saturation comparison values
Participant ANRD (SpO2) Commercial device (SpO2) % Error
1 96 96 0.0
2 100 92 8.69
3 97 92 5.43
4 84 93 9.68
5 96 96 0.0
6 97 96 1.04
7 96 97 1.04
8 99 93 6.45
9 95 93 2.15
10 97 93 4.30
Table 7 Body temperature comparison values
Participant ANRD (∘C) Commercial device (∘C) % Error
1 35.96 35.91 0.28
2 36.2 36.4 1.09
3 36.09 35.8 1.62
4 35.62 35.5 0.68
5 36.7 36.3 2.02
6 36.63 36.7 0.38
7 36.6 35.8 4.47
8 37.05 36.4 3.57
9 35 36.3 7.16
10 37.22 37.5 1.49
From Tables 6 and 7, it can be observed that the mean error for the oxygen saturation is 3.8% whereas the mean error for the corporal temperature is 2.2%. Then, the comparison evidenced that the obtained values are reliable.
Human-machine interaction
The main purpose of the nursing robot is to track the patient position that is inside of specific working space. Here, the patient’s position has been simulated by a GPS sensor. It is crucial to notice that the proposed control algorithm has (in its mathematical structure) functions to consider the position of the robot and patient as input information.
The experimental evaluation also considered a novel method to adjust the control gains that removes any high-frequency oscillation during the transient period while the robot is approaching the patient. This particular fact opens the possibility of developing a class of adaptive control gains to reduce all undesired motions by the nursing robot.
Figure 14 displays some photographs where the relative location between the nursing robot and the patient is reduced by using the GPS information as input information of the state feedback controller. This information was supplied to the microcontroller with a sampling period of 100 ms, that means that the position information provided by the GPS module (SparkFun GPS-RTK2 Board-ZED-F9P) to the nursing robot is updated with a sample frequency of 10 Hz. In Fig. 15, the yellow arrows represent the patient position. Fig. 14 Example of the controller execution over the nursing robot and its relative position with respect to the patient
Fig. 15 Experimental evaluation of the proposed control action
Graphical user interface
The device includes a GUI to show the patient’s clinical status information. The interface was designed using the Graphical User Interface Development Environment (GUIDE) in the Matlab® environment. Each interface element was designed to be considered a class of object-oriented strategies that simplify the software design. The proposed interface consists of four parts shown in Fig. 16: the first section corresponds to the welcome screen (Fig. 16a), the second is the entry of personal information (Fig. 16b), the third corresponds to the monitoring process of vital signals (Fig. 16c), and the fourth is supplying the support to the displacement control (Fig. 16d). All these parts consider the minimum information required by the user and the medical physician in charge of the patient. The following items detail the elements included in the graphical user interface. The first section consists of a welcome screen for the patient.
The second section displays three options. Two of them allow choosing the robot mode, whereas the third option is setting the patient’s personal information.
The third section allows the introduction of the basic personal information of the patient. User data include name, weight, age, and general clinical observations. All the information is saved in a mat file that can be sent to the medical professional.
The monitoring section consists of five buttons. Two buttons allow the displacement between all the sections in the interface. The other three buttons indicate the measurement of biosignals. The “Start” button visualizes the following signals: ECG, oxygen saturation, and corporal temperature. For measuring the arterial pressure, it is necessary to push the cyan button. The same section shows the name and the personal information of the patient.
Fig. 16 GUI of the nursing robot
Conclusions
The developed assistance nursing robot design presents an adequate structure for wandering in structured zones where a patient could ambulate freely, such as a hospital or home-like environments. The nursing robot is designed, instrumented and controlled to satisfy all the requested tasks for a nursing home-care robotic device. The motion control allows an efficient interaction with the patient. Indeed, the movement produced by the proposed state feedback control provides the nursing robot with an adequate response when the patient is ambulating in structured environments. The acquisition of the physiological information is instrumented following the expected characteristics for the selected signals. All the collected information give sufficient knowledge about the essential health conditions of the patient. The assisting nursing robot represents an alternative for home-care of senior citizens monitored at home with a support for their walking cycle. The developed study has potential future trends, including new instrumentation devices with a resilient working strategy and a modern control designs including robust approaches with adaptive gains that help the energetic consumption, which may extend the period of operation of the nursing robot.
Funding
This work received financial support from the Instituto Politénico Nacional provided through the research grants with reference SIP-20221746, SIP-20221503 and SIP-20221150.
Declarations
Ethics approval
This paper or a similar version is not currently under review by a journal or conference. This paper is void of plagiarism or self-plagiarism as defined by the Committee on Publication Ethics and Springer Guidelines.
Conflict of Interests
The authors declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. AgeWatch (2015) Global AgeWatch Index 2015: Insight report. Tech. rep., Global AgeWatch
2. Asfour T, Regenstein K, Azad P et al (2006) Armar-iii: An integrated humanoid platform for sensory-motor control. In: 2006 6th IEEE-RAS international conference on humanoid robots. IEEE, pp 169–175
3. Assesment EPT (2019) Technologies in care for older people: EPTA report 2019. Sveriges riksdag
4. Azar A, Eljamel S (2014) Robotics: Concepts, Methodologies, Tools, and Applications, vol 2, The University of Dundee, UK, chap 9 Medical Robotics, p 1116. 10.4018/978-1-4666-4607-0.ch05410.4018/978-1-4666-4607-0.ch054
5. Azeta J, Bolu C, Abioye AA et al (2017) A review on humanoid robotics in healthcare
6. Ben-Ari M Mondada F Robots and their applications 2018 Cham Springer International Publishing 1 20
7. CalebSolly PH, Chris D et al (2021) Standards and regulations for physically assistive robots. In: 2021 IEEE international conference on intelligence and safety for robotics (ISR
8. Christoforou E, Panayides A, Avgousti S et al (2019) An overview of assistive robotics and technologies for elderly care, p 971. 10.1007/978-3-030-31635-8_118
9. Cousins S Ros on the pr2 [ros topics] IEEE Robotics & Automation Magazine 2010 17 3 23 25 10.1109/MRA.2010.938502
10. Cruz-Jentoft AJ Baeyens JP Bauer JM Sarcopenia: European consensus on definition and diagnosis: Report of the european working group on sarcopenia in older people Age Ageing 2010 39 4 412 423 10.1093/ageing/afq034 20392703
11. Danielsson K, Lindgren H, Mulvenna M et al (2017) Digital technology in healthcare and elderly care. The European Conference:188–190. 10.1145/3121283.3121425
12. Eftychios G, Christoforou S, Avgousti R et al (2020) The upcoming role for nursing and assistive robotics: Opportunities and Challenges Ahead. Front Digit Health
13. Elliot M Coventry A Critical care: The eight vital signs of patient monitoring British Journal of Nursing (Mark Allen Publishing) 2012 21 621 5 10.12968/bjon.2012.21.10.621 22875303
14. Martins FG, Santos D et al (2016) Falls risk detection based on spatiotemporal parameters of three-dimesional gait analysis in healthy adult women from 50 to 70 years old 82:220–226
15. Fariha Z Ikeura R Hayakawa S Analysis of Pan-Tompkins algorithm performance with noisy ECG signals J Phys: Conf Ser 2020 1532 012 022
16. Guntur SR, Gorrepati RR, Dirisala VR et al (2019) Chapter 12 - Robotics in Healthcare: An internet of medical robotic things (IoMRT) Perspective. In: Dey N, Borra S, Ashour AS (eds) Machine learning in bio-signal analysis and diagnostic imaging. Academic Press, pp 293–318
17. Gupta D, Kaur G, Jain A (2016) Geriatric syndromes. Jaypee brothers, chap 335 p 1753
18. Webster GJ Medical instrumentation: Application and design 2010 New York Wiley
19. Jaafar R, Desa HM, Mahmoodin Z et al (2011) Noninvasive blood pressure (NIBP) measurement by oscillometric principle. In: 2011 2nd International Conference on Instrumentation, Communications, Information Technology, and Biomedical Engineering. 10.1109/ICICI-BME.2011.6108622, pp 265–269
20. KEENONRobotics (2020) PEANUT Comercial Delivery Robot
21. King CH, Chen TL, Jain A et al (2010) Towards an assistive robot that autonomously performs bed baths for patient hygiene. In: 2010 IEEE/RSJ international conference on intelligent robots and systems. IEEE, pp 319–324
22. Kozłowski K, Pazderski D (2004) Modeling and control of a 4-wheel skid-steering mobile robot. International Journal of Applied Mathematics and Computer Science 14
23. Mukai T, Hirano S, Nakashima H et al (2010) Development of a nursing-care assistant robot RIBA that can lift a human in its arms. In: 2010 IEEE/RSJ international conference on intelligent robots and systems, pp 5996–6001. 10.1109/IROS.2010.5651735
24. Reiser U, Connette C, Fischer J et al (2009) Care-o-bot®; 3-creating a product vision for service robot applications by integrating design and technology. In: 2009 IEEE/RSJ international conference on intelligent robots and systems. IEEE, pp 1992–1998
25. RobotLAB (2020) AIMBOT Indoor Virus Prevention. UBTECH
26. Singh L, Janghel RR, Sahu SP (2020) Trcsvm: A novel approach for the classification of melanoma skin cancer using transfer learning. Data Technologies and Applications
27. Singh L, Janghel RR, Sahu SP (2021a) A hybrid feature fusion strategy for early fusion and majority voting for late fusion towards melanocytic skin lesion detection. International Journal of Imaging Systems and Technology
28. Singh L Janghel RR Sahu SP Slicaco: An automated novel hybrid approach for dermatoscopic melanocytic skin lesion segmentation Int J Imaging Syst Technol 2021 31 4 1817 1833 10.1002/ima.22591
29. Steves C Spector TD Jackson SHD Ageing, genes, environment and epigenetics: what twin studies tell us now, and in the future Age Ageing 2012 41 5 581 586 10.1093/ageing/afs097 22826292
30. Sund Levander M Grodzinsky E Assessment of body temperature measurement options British journal of nursing (Mark Allen Publishing) 2013 22 942 50 10.12968/bjon.2013.22.16.942 24037397
31. Sakai TALYX (2013) Image-Processing Technologies for Service Robot “HOSPI-Rimo”. Panasonic Techinical pp 624–625. 10.1299/jsmemag.116.1138_624
32. Teixeira C Boaventura R Souza A Vital signs measurement: An indicator of safe care delivered to elderly patients Texto & ontexto - Enfermagem 2015 24 1071 1078 10.1590/0104-0707201500003970014
33. United Nations DoE Social Affairs PD (2019) World Population Ageing 2019 : Highlights. (ST/ESA/SERA/430)
34. United Nations DoE Social Affairs PD (2020) World Population Ageing 2020 Highlights: Living arrangements of elder persons. (ST/ESA/SERA/451)
35. Yu X, Feng Y, Man Z (2020) Terminal sliding mode control - an overview. IEEE Open Journal of the Industrial Electronics Society PP:1–1. 10.1109/OJIES.2020.3040412
| 36445530 | PMC9707198 | NO-CC CODE | 2022-12-01 23:20:21 | no | Med Biol Eng Comput. 2022 Nov 29;:1-22 | utf-8 | Med Biol Eng Comput | 2,022 | 10.1007/s11517-022-02712-y | oa_other |
==== Front
Ir J Med Sci
Ir J Med Sci
Irish Journal of Medical Science
0021-1265
1863-4362
Springer International Publishing Cham
3209
10.1007/s11845-022-03209-1
Abstracts
Irish Thoracic Society Annual Scientific Meeting 2022 Abstract Book 1 – 3 December 2022, Naas, Ireland
29 11 2022
157
© The Author(s), under exclusive licence to Royal Academy of Medicine in Ireland 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
==== Body
pmcDisclosure Statement
Irish Thoracic Society Annual Scientific Meeting 2022
1st2nd and 3rd December – Killashee Hotel Naas
Thursday 1st December 2022
13.30 – 16.30 Specialist Registrar Training
17.30 – 20.30 ITS Case Study Forum
17.30 – 19.00 Case Study Poster Review
19.00 – 20.30 Case Study Oral Presentations followed by prize giving and supper
Friday 2nd December 2022
08.00 – 10.15 Poster Session I
08.00 – 09.00 Poster Review
09.00 – 10.15 Parallel Poster Discussions
1. Asthma
2. Interstitial Lung Disease
3. Lung Cancer
4. Patient Care
5. TB, Pleural and Sleep Disorders
10.15 - 10.45 Tea and Coffee
10.45 – 11.45 Poster Session II
10.45 – 11.45 Poster Review
11.45 – 13.00 Parallel Poster Discussions
6. Cystic Fibrosis
7. COPD
8. COVID-19
9. General Respiratory
10.Telehealth
13.00 – 14.30 Lunch and ITS AGM
14.30 – 16.30 Oral Presentations
1. Use of digital measurement of medication adherence and lung function to guide the management of uncontrolled asthma: The INCA Sun randomized clinical trial
Elaine Mac Hale1, presenting author on behalf of the INCA Sun randomized clinical trial work group.
1. RCSI, Dublin, Ireland.
2. A systematic assessment clinic for COPD reduces hospitalisation rates
Finbarr Harnedy1, Jack Allen1, Michele Cuddihy1, Eimear Ward2, Ciara Feeney2, Maeve Sorohan2, Imran Sulaiman1, Breda Cushen1
1Department of Respiratory Medicine, Beaumont Hospital, Dublin 9
2Respiratory Integrated Care, Beaumont Hub, Dublin North City and County CHO9
3. The Benefit of a Liraglutide-Based Weight Loss Regimen Alone or in Addition to Standard CPAP Therapy on Metabolic Function, Atherosclerosis and Inflammation in Patients with Obstructive Sleep Apnoea - an Explorative, Proof of Concept Study
Cliona O'Donnell1,2, Shane Crilly3, Anne O'Mahony1, Jonny Dodd2,3, Donal O'Shea4, David Murphy2,3, Silke Ryan1,2
1Pulmonary and Sleep Disorders Unit, St. Vincent's University Hospital, Dublin, Ireland
2University College Dublin, School of Medicine, Dublin, Ireland
3Radiology Department, St. Vincent's University Hospital, Dublin, Ireland
4Endocrinology Department, St. Vincent's University Hospital, Dublin, Ireland
4. Evidence of mTORC1 pathway in Diffuse Idiopathic Pulmonary Neuroendocrine Cell Hyperplasia (DIPNECH)
Maylis Alquier1,2, Janet McCormack2, Marissa O’Callaghan4,6, Evelyn Lynn4,6, Orla O’Carroll4, David Murphy5,7, Rachel Crowley7, Dermot O’Toole7, Cormac McCarthy4,6, Aurelie Fabre 2,3,6,7
1. Faculty of Medicine, University of Limoges, France
2. Research Pathology Core, Conway Institute, University College Dublin,
3. Histopathology department, St. Vincent’s University Hospital, Dublin, Ireland
4. Department of Respiratory Medicine, St. Vincent’s University Hospital, Dublin, Ireland
5. Department of Radiology, St. Vincent’s University Hospital, Dublin, Ireland
6. School of Medicine and Medical Sciences, University College Dublin, Dublin, Republic of Ireland
7. ENETS Centre for Excellence, St. Vincent’s University Hospital, Dublin, Ireland
5. Investigation of the Utility of Exhaled Breath Condensate (EBC) as a Liquid Biopsy in the Detection of Spatial Genomic Heterogeneity in Patients with Early-Stage Non-Small Cell Lung Cancer (ESLC)
Robert Smyth1, 2, Simon Furney3, Siobhan Nicholson4, Katherine Sheehan5, Ronan Ryan6, Daniel Ryan2, 7 Liam Grogan8, Oscar Breathnach8, Patrick Morris8, Bryan Hennessy2, 8, Ross Morgan7*, Sinead Toomey2*
1Pulmonary Critical Care, Boston Medical Center, Boston, MA, United States, 2Department of Molecular Medicine, Royal College of Surgeons in Ireland, Dublin, Ireland, 3Genomic Oncology Research Group, Royal College of Surgeons in Ireland, Dublin, Ireland, 4Department of Pathology, St. James's Hospital, Dublin, Ireland, 5Department of Histopathology, Royal College of Surgeons in Ireland, Dublin, Ireland, 6Thoracic Surgery, St. James's Hospital, Dublin, Ireland, 7Department of Respiratory Medicine, Beaumont Hospital, Dublin, Ireland, 8Department of Medical Oncology, Beaumont Hospital, Dublin, Ireland.
6. Circulating soluble P2X7 receptor as a novel indicator of resolution of inflammation in patients with cystic fibrosis.
Azeez Yusuf1, Debananda Gogoi1, Claudie Gabillard-Lefort1, Cedric Gunaratnam1,2, Noel G. McElvaney1,2, Michelle Casey1,2 & Emer P. Reeves1
1The Irish Centre of Genetic Lung Disease, Dept of Medicine, Royal College of Surgeons in Ireland, Dublin, Ireland 2Cystic Fibrosis Unit, Dept of Respiratory Medicine, Beaumont Hospital, Dublin, Ireland
7. Results of the Irish Thoracic Society ( ITS) Bronchoscopy questionnaire in preparation for the National Bronchoscopy Quality Improvement project
Dr Daniel Ryan1, Prof Marcus Kennedy2
1. Department of Respiratory Medicine, Beaumont Hospital, Dublin
2. Department of Respiratory Medicine, Cork University Hospital
8. Incidence and outcomes of pulmonary arterial hypertension in the Republic of Ireland
Sarah Cullivan1, Brian McCullagh1, Sean Gaine1
1. National Pulmonary Hypertension Unit, Mater Misericordiae University Hospital, Dublin, Ireland
16.30 – 17.00 Tea and Coffee
17.00 – 17.45 Guest Lecture I
Developments in sarcoidosis pathogenesis and treatment
Dr Elliott Crouser, Ohio State University
19.00 – 19.45 Prize Giving and Drinks Reception
20.00 – late Gala Dinner
Saturday 3rd December 2022
09.30 – 10.00 Pro/con debate
10.00 – 10.45 Guest Lecture II
Role of the pathologist in the diagnosis of interstitial lung diseases
Professor Aurelie Fabre, St Vincent’s University Hospital Dublin
10.45 – 11.15 Tea and Coffee
11.15 – 12.00 Guest Lecture III
What physiology is essential to the respirologist in 2022?
Professor Franco Laghi, University of Loyola Medical School
12.00 – 12.30 Guest Lecture IV
Update on the National Clinical Programme Respiratory
Dr Stanley Miller, National Clinical Lead Respiratory, Consultant Respiratory
Physician Mater Misericordiae University Hospital Dublin
12.30 Meeting close
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
1. Asthma
1.1 Don’t act on the ACT
Vincent Brennan, Patrick Kerr, Garrett Greene, Lorna Lombard, Helen Doherty, Elaine Mac Hale, Richard Costello
RCSI
The Asthma Control Test (ACT) is a validated tool used in the assessment of asthma control, and, frequently, in clinical trials. We tested the hypothesis that, due to the non-specific nature of respiratory symptoms, the ACT may actually be influenced by comorbid conditions.
Using data from the INCA-SUN trial, gathered during 6 visits over a 32-week period, we compared ACT scores with objective markers of asthma control, including spirometry and measures of T2 inflammation, and evaluated the influence of comorbidities, such as obesity, gastro-oesophageal reflux (GORD) and depression/anxiety. We discovered that the ACT showed no significant correlative relationship to objective measures of asthma control. Multiple logistic regression analyses accounting for age, sex, FEV1, T2 biomarkers and steroid exposure showed significantly lower ACT scores if GORD or depression/anxiety were present. Similar findings were noted using ordered logistic regression analyses for BMI classes. The number of comorbidities was found to correlate negatively with ACT scores at each visit, and patients with a co-morbidity were less likely to improve subjectively throughout the study. (See Fig. 1.) Given the ACT scores’ lack of specificity, treatment decisions should be supported by objective evidence of poorly controlled asthma. Fig. 1 (1.1) A comparison of ACT scores based on the presence of common comorbidities.
Conflict of Interest: None to declare
1.2 Inflammatory Pathways in severe asthma
Vincent Brennan1, Patrick Kerr1, Garrett Greene1, Lorna Lombard1, Helen Doherty1, Elaine Mac Hale1, Heike Hawerkamp2, Padraic Fallon2, Richard Costello1
1. RCSI, 2. Trinity Biomedical Science Institute, Dublin
Asthma is characterised by inflammation and airflow limitation. Eosinophilic inflammation is linked to asthma and the ‘T2 high’ endotype has been well. The corollary of apparent ‘T2 low’ asthma however remains poorly described even though it is thought to account for nearly 50% of patients. We aimed to describe the different inflammatory pathways active in both ‘T2 high’ and apparent ‘T2 low’ patients and to compare these with healthy controls.Longitudinal evaluation of 200 patients from the INCA-SUN trial identified four groups: ‘Consistently T2 low’ (20% n = 40), ‘Usually T2 low’ (24% n = 48), ‘Usually T2 high’ (32% n = 64) and ‘Consistently T2 high’ (24% n = 48). Multiplex assays of inflammatory pathways associated with asthma including Th-1, T2 and Th-17 were performed in these patients, as well as 56 healthy controls. We found that the ‘Consistently T2 high’ cohort had significantly higher values of periostin (p < 0.0001) and IgE (p < 0.0001) as expected but they also demonstrated significantly higher levels of TNF-α (p < 0.0001) and IL-6 (p = 0.0005) when compared to the other cohorts. Patients in the ‘Consistently T2 low’ cohort showed no evidence of an alternate inflammatory pathway, with their cytokine results most closely resembling the cohort of healthy controls.These findings support the hypothesis that the apparent ‘T2 low’ endotype may reflect suppressed inflammation instead of an alternate inflammatory target.
Conflict of Interest: None to declare.
1.3. Screening for Hyperglycaemia in Accordance with ‘Joint British Diabetes Societies for inpatient care’ Guidelines in Patients with Acute Asthma Exacerbations Treated with Systemic Corticosteroids in a Single Centre
Alexandra Brickley1, Dominic Doyle2, Philip Healy3, Ciaran O’Donovan4, Desmond Murphy5
Cork University Hospital
Systemic corticosteroids are employed for both acute and maintenance treatment of asthma, a chronic inflammatory disease. Rates of corticosteroid related hyperglycaemia and Diabetes Mellitus(DM) are estimated at 32.3% and 18.6% respectively1. The ‘Joint British Diabetes Societies for inpatient care’(JDBS) 2022 guidelines ‘Management of Hyperglycaemia and Steroid Therapy2’ provide a framework for managing steroid induced DM and hyperglycaemia. This is a retrospective cohort study assessing patients with acute asthma exacerbations, managed with corticosteroids in accordance with JDBS guidelines in the following three areas: 1) Pre-treatment assessment with HbA1C/fasting/random glucose 1 year prior to treatment and screening for risk factors for hyperglycaemia. 2) Inpatient capillary blood glucose(CBG) monitoring. 3) Screening 3 months post-treatment with HbA1C/fasting/random glucose. 40% had any of the 3 markers checked pre-treatment with 3 abnormal HbA1Cs indicating undiagnosed prediabetes. 30% had any of the 3 markers checked 3 months post-treatment.50% had daily capillary blood sugar checks during inpatient admission. 46.15% of high risk patients had CBG checked daily.
DM is associated with significant morbidity and mortality, with steroids contributing to both de novo and uncontrolled pre-existing hyperglycaemia. This study demonstrated suboptimal screening pre, during and post treatment with systemic corticosteroids which highlights the need to tackle compliance of inpatient management and outpatient follow-up.
Liu XX, Zhu XM, Miao Q, Ye HY, Zhang ZY, Li YM. “Hyperglycemia induced by glucocorticoids in nondiabetic patients: a meta-analysis”.(2014) Ann Nutr Metab. 65(4):324–332.
“Management of Hyperglycaemia and Steroid Therapy”. (2022) Joint British Diabetes Societies for inpatient care’, Available at: https://www.diabetes.org.uk/professionals/resources/shared-practice/inpatient-and-hospital-care/joint-british-diabetes-society-for-inpatient-care
Conflict of Interest: None to declare.
1.4 Herpes Zoster in Mepolizumab Recipients. To Vax or not to Vax?
Tara Byrne, Emma Burke, Hilary Mc Loughlin
Portiuncula University Hospital
Background: Herpes zoster is a significant health burden that affects individuals of any age. It is more common among individuals aged ≥ 50 years, those with immunocompromised status, and those on immunosuppressant drugs. This is likely secondary to the immunosenescence associated with advancing age 2. An increased incidence has been reported in asthmatics 1.Methods:A retrospective review was conducted of eosinophilic asthma patients who have commenced on Mepolizumab 100 mg subcutaneously in Portiuncula University Hospital in 2021-2022.Results:22.29% (n = 2) of patients who received Mepolizumab developed Herpes Zoster after the 1st or 2nd dose of Mepolizumab. Individuals’ ages ranged from 55-71 years and cases were deemed to be mild or moderate.Conclusion:GSK reports a 0%-2% incidence of Herpes Zoster across 10 phase 3 clinical trials. The rationale for the increased risk of Herpes Zoster virus in individuals receiving Mepolizumab is unknown. A population-based case-control study yielded results noting an increased incidence of Herpes Zoster in individuals with asthma 1. The study recommended that consideration should be given to immunizing adults with asthma aged more than 50 years as a target group.Future Recommendations:Patients aged > 50 years with a known history of Asthma receiving mepolizumab are at an increased risk of Herpes Zoster. Vaccination should be considered.
References
1 Kwon, H.J., Bang, D.W., Kim, E.N., Wi, C.-I., Yawn, B.P., Wollan, P.C., Lahr, B.D., Ryu, E., Juhn, Y.J., 2016. Asthma as a risk factor for zoster in adults: A population-based case-control study. J. Allergy Clin. Immunol. 137, 1406–1412. https://doi.org/10.1016/j.jaci.2015.10.032
2 Sheth, P., 2018. Epidemiology, treatment, and prevention of herpes zoster: A comprehensive review [WWW Document]. Indian J. Dermatol. Venereol. Leprol. URL https://ijdvl.com/epidemiology-treatment-and-prevention-of-herpes-zoster-a-comprehensive-review/ (accessed 8.2.22).
Conflict of Interest: None to declare
1.5 A Patient Experience Survey of asthma biologic patients self-administering at home under the care of the asthma service at Cork University Hospital
Casey D1, Vairamani P1, Arnott F 1,Walsh L 1,Murphy J1,Leahy P1,Gomez F1,Plant BJ 1, 2, Murphy DM1,2
1Cork University Hospital, Wilton, Cork,2 University College Cork, Cork
The asthma biologic outpatient service has been in operation in CUH since 2018. A survey to determine patient satisfaction on patients self-administering for a period of 6 months or longer was carried out in July 2021. A questionnaire was distributed to all (90) patients who received more than three infusions within the outpatient setting and had been self-administering at home for 6 months or more.
A 7 item Likert questionnaire was developed. Questions 1-6 were based on a scoring scale of 1-5 to rate the service, and question 7 was open ended to allow for feedback. These questionnaires were discussed with patients with help from administration staff whom the patients had never met previously to exclude any bias. Patients who responded to a text message agreed to being contacted, to complete the questionnaire. A total of 82 patients completed the questionnaire.
All patients felt they would be able to contact the respiratory team in an event of acute deterioration in their asthma symptoms. 100% of patients felt they were well educated on self-administering prior to discharge. Patients stated they were happy that the asthma nurse checked in monthly or 2 monthly to review asthma symptoms.Our results suggest that patients feel empowered to self-manage their asthma when given the opportunity to administer biologics in their own home.
Conflict of Interest: None to declare
1.6 GLYCAEMIC CONTROL IN OUTPATIENTS WITH SEVERE ASTHMA
Dominic Doyle, Alexandra Brickley, Deborah Casey, Hammad Danish, Desmond Murphy
Cork University Hospital
Patients in our severe and difficult to control asthma cohort may be dependent on high and long-term doses of glucocorticoids which deranges their glycaemic control and predisposes to steroid-induced hyperglycaemia, which is not always reversible. We hypothesise that our outpatients with asthma who are on a steroid dose equivalent to prednisolone 2.5 mg daily average and above are at risk of hyperglycaemia and we internally audited this against the guidelines from the Joint British Diabetes Societies 2021. We identified 45 of these high-risk patients from a list of those requiring biologic drugs. Patient characteristics, history of diabetes and type, diabetes medication, HbA1c/glucose measurement and timing, and the average daily steroid dose preceding the HbA1c/glucose measurement by 3 months were recorded. 14 outpatients were on a dose of prednisolone at or above the threshold of 2.5 mg daily average. Among this group, 11 had HbA1c checked within a 3-month window following their course of steroids (see Fig.). Only 1 patient in the steroid group had HbA1c greater than 48 mmol/mol and this was a patient who had diabetes. The authors hope this audit reminds physicians to assess glycaemic control in all of our high-risk patients (1.6).
Conflict of Interest: None to declare.
1.7 Development of a Novel Pre-clinical Model of House Dust Mite Induced Allergic Airway Inflammation Using Transgenic Mice Expressing Human MIF
Hazel Dunbar1, Ian Hawthorne1, Myron Rebello, Brion Kennedy, Seamas Donnelly2, Karen English1
1Maynooth University, Co. Kildare 2 Trinity College, Dublin and Tallaght Hospital, Dublin
The pro-inflammatory cytokine macrophage migration inhibitory factor (MIF) plays an important role in several inflammatory conditions including allergic asthma. A functional promoter polymorphism located at position -794 in the gene promoter for human MIF contains a tetranucleotide sequence, CATT, that is repeated between 5 and 8 times. MIF high (7-CATT) and low (5-CATT) expression allele is a prime example of major differences in the inflammatory microenvironment present in different asthma patient populations with a high correlation between 7-CATT MIF allele and severe asthma and 5-CATT allele with mild asthma.We have developed novel transgenic mice with high or low expression of human MIF. Challenge with house dust mite (HDM) leads to exacerbated allergic airway inflammation in 7-CATT mice compared to 5-CATT or wildtype controls. 7-CATT mice exhibit enhanced immune cell airway infiltration, significantly increased goblet cell hyperplasia, sub-epithelial fibrosis, airway hyper-responsiveness and bronchoconstriction. Eosinophils, total cell counts, IL-4, IL-5 and IL-13 were increased in the bronchoalveolar lavage fluid of 7-CATT mice. Administration of MIF inhibitor ISO-1 or SCD-19 ameliorated the exacerbation effect of human MIF expression in 7-CATT mice. This model provides a novel means to develop a scale of allergic airway inflammation following HDM-challenge using MIF expression.
Conflict of Interest: None to declare
1.8 A randomised trial of a T2-composite-biomarker strategy adjusting corticosteroid-treatment in severe asthma, a post- hoc analysis by sex
Matthew Chad Eastwood1, John Busby, David Jackson, Ian Pavord, Catherine Hanratty CE, Ratko Djukanovic, Arron Woodcock A, Samantha Walker, Timothy Hardman, Joseph Arron, David Choy, Peter Bradding, Chris Brightling, Rekha Chaudhuri, Douglas Cowan D, Adel Mansur, Stephen Fowler, Peter Howarth, James Lordan, Andrew Menzies-Gow, Timothy Harrison, Douglas Robinson, Cecile Holweg, John Matthews, Liam Heaney2 on behalf of the RASP-UK Investigators
University Belfast, Belfast, UK.
5–10% of patients with asthma have severe disease with a consistent female preponderance. Asthma guidelines recommend stepwise, symptom-driven disease control with no differential treatment considerations for either sex. This post-hoc analysis of a 48-week, multicentre, randomised controlled clinical trial, compared a T2 biomarker-defined treatment algorithm with standard-care, stratifying patient outcomes by sex. The primary outcome was the proportion of patients with a reduction in corticosteroid treatment (log-regression models). Secondary outcomes included exacerbation rates, hospital admissions and lung function (linear/Poisson/logistic/cox regression models).301 patients were randomised: 194 (64.5%) females and 107 (35.5%) males. The biomarker algorithm led to more females being on a lower corticosteroid dose vs standard-care, compared to males (effects estimate females: 3.57, 95% CI: 1.14, 11.18 vs. males 0.54, 95% CI: 0.16, 1.80). In T2-biomarker low females, reducing the corticosteroid dose was not associated with increased exacerbations. Females scored higher in all ACQ-7 domains, but with no difference when adjusted for BMI/depression/anxiety. Dissociation between symptoms and T2-biomarkers were noted in both sexes, with more females being symptom high/T2-biomarker low (22.8% vs. 15.5%; p = 0.0002), whereas males were symptom low/T2-biomarker high (11.5% vs. 22.3%; p < 0.0001).We identified females achieved a greater benefit from biomarker-directed corticosteroid optimisation versus symptom-directed treatment.
Disclosure of Conflicts of Interest
Dr Eastwood has nothing to disclose. Dr Busby reports personal fees from NuvoAir outside the submitted work. Dr Jackson reports personal fees from AstraZeneca, BI, Chiesi, Glaxo SmithKline, Sanofi Regeneron, grants from AstraZeneca outside the submitted work. Dr Pavord reports speaker fees, payments for organization of educational events, consultant fees, international scientific meeting sponsorship from Sanofi. Dr Pavord reports speaker fees, payments for organization of educational events, consultant fees, international scientific meeting sponsorship from Regeneron Pharmaceuticals Inc. Dr Pavord reports non-financial support from Excerpta Medica, during the conduct of the study. Dr Pavord reports speaker fees from Aerocrine AB, speaker fees and consultant fees from Almirall. Dr Pavord reports speaker fees, payments for organization of educational events, consultant fees, international scientific meeting sponsorship from AstraZeneca. Dr Pavord reports Speaker fees, consultant fees, international scientific meeting sponsorship from Boehringer Ingelheim. Dr Pavord reports grants, speaker fees, consultant fees and international scientific meeting sponsorship from Chiesi. Dr Pavord reports Speaker fees, payments for organization of educational events, consultant fees, international scientific meeting sponsorship from GSK. Dr Pavord reports Speaker fees, consultant fees from Novartis. Dr Pavord reports Speaker fees, payments for organization of educational events, consultant fees, international scientific meeting sponsorship from Regeneron Pharmaceuticals Inc. Dr Pavord reports Speaker fees, payments for organization of educational events, consultant fees, international scientific meeting sponsorship from Sanofi. Dr Pavord reports Speaker fees; payments for organization of educational events, consultant fees, international scientific meeting sponsorship Teva. Dr Paovord reports consultant fees from Circassia, Dey Pharma, Genentech, Knopp Biosciences, Merck, MSD, RespiVert, and Schering-Plough. Dr Paovord reports consultant fees international scientific meeting sponsorship from Napp Pharmaceuticals from outside the submitted work. Dr Hanratty has nothing to disclose. Dr Djukanovic reports receiving fees for lectures at symposia organised by Novartis, AstraZeneca and TEVA, consultation for TEVA and Novartis as member of advisory boards, and participation in a scientific discussion about asthma organised by GlaxoSmithKline. He is a co-founder and current consultant, and has shares in Synairgen, a University of Southampton spin out company. Dr Woodcock reports personal fees from GlaxoSmithKline, personal fees from Novartis, personal fees from Chiesi, Chairman (Food Allergy Diagnostics) at Reacta Biotech, Chairman (Cough drug in Phase 1) Axalbion, Chairman (Clinical Trials Unit) at Medicines Evaluation Unit outside the submitted work. Dr Walker has nothing to disclose. Dr Hardman has nothing to disclose. Dr Arron is a former employee of Genentech and holds stock in the Roche Group. Dr Arron is a current employee of and holds stocks/options in 23andMe. Dr Choy reports and is an employee of Genentech Inc and owns stocks/options in the Roche group. Dr Bradding reports grants provided from Medical Research Council during the conduct of this study. Dr Bradding reports a relationship with Genentech that includes consulting or advisory. Dr Bradding reports a relationship with Chiesi that includes travel reimbursement. Dr Bradding reports a relationship with Sanofi Genzyme that includes travel reimbursement. Dr Bradding reports a relationship with Teva that includes travel reimbursement. Dr Brightling reports grants and personal fees from Glaxo SmithKline, grants and personal fees from AstraZeneca, grants and personal fees from Sanofi, grants and personal fees from Novartis, grants and personal fees from Chiesi, grants and personal fees from Genentech, grants and personal fees from Gossamer, grants and personal fees from Mologic, grants and personal fees from 4DPharma outside the submitted work. Dr Chaudhuri reports grants and personal fees from AstraZeneca. Dr Chaudhuri reports personal fees from Glaxo SmithKline and Novartis. Dr Chaudhuri reports personal fees and Fees for lectures and support for conference attendance from Sanofi. Dr Chaudhuri reports personal fees and Fees for lectures from Teva. Dr Chaudhuri reports personal fees and fees for lectures and support for conference attendance from Chiesi outside the submitted work. Dr Cowan has nothing to disclose. Professor Mansur reports grants from Medical Research Council during the conduct of the study. Grants, personal fees, institutional payments for lectures, advisory board payments, sponsorships to attend conferences, educational grants and research grants from Teva, Glaxo SmithKline, Novartis, AZ, PI, Sanofi, Chiesi, outside the submitted work. Dr Fowler reported grants and personal fees from Boehringer Ingelheim. Dr Fowler reported personal fees from AstraZeneca Chiesi, Glaxo SmithKline, Novartis and Teva outside the submitted work. Dr Howarth reports employment by Glaxo SmithKline. Dr Lordan has nothing to disclose. Dr Menzies-Gow reports grants and personal fees from Astra Zeneca, personal fees from Glaxo SmithKline, Novartis, Sanofi. Dr Menzies-Gow reported personal fees and non-financial support from Teva outside the submitted work. Dr Harrison reports grants, personal fees and non-financial support from AstraZeneca, grants and personal fees from Glaxo SmithKline, Vectura, Synairgen and Chiesi, outside the submitted work. Dr Robinson has nothing to disclose. Dr Holweg reports that she is an employee from Genentech Inc/Roche. Dr Matthews reports that he is a former employee of Genentech. Dr Matthews is a current employee of 23andMe and holds stocks/options in 23andMe. Professor Heaney reports having received sponsorship for attending international scientific meetings AstraZeneca, Boehringer Ingelheim, Chiesi, Glaxo SmithKline and Napp Pharmaceutical. Professor Heaney has reported personal fees from Novartis, Hoffman la Roche/Genentech Inc, Sanofi, GlaxoSmithKline, Astra Zeneca, Teva, Theravance, Circassia. Professor Heaney reports grants from Medimmune, Novartis UK, Roche/Genentech Inc, and GlaxoSmithKline, Amgen, Genentech / Hoffman la Roche, Astra Zeneca, Medimmune, GlaxoSmithKline, Aerocrine and Vitalograph outside the submitted work. Professor Heaney reports having received sponsorship for attending international scientific meetings AstraZeneca, Boehringer Ingelheim, Chiesi, Glaxo SmithKline and Napp Pharmaceutical. Professor Heaney has reported personal fees from Novartis, Hoffman la Roche/Genentech Inc, Sanofi, GlaxoSmithKline, Astra Zeneca, Teva, Theravance, Circassia. Professor Heaney reports grants from Medimmune, Novartis UK, Roche/Genentech Inc, and GlaxoSmithKline, Amgen, Genentech / Hoffman la Roche, Astra Zeneca, Medimmune, GlaxoSmithKline, Aerocrine and Vitalograph outside the submitted work.
1.9 Asthma patients discharged from the emergency department in Ireland: an unmet need?
Cameron Forward1, Rory O’Loghlin1, Helen Mulryan1, Donna Langan1, Mike Harrison1, Robert Rutherford1, Ruth P Cusack1
Galway University Hospital1
Introduction: Appropriate ED management of asthma including discharge planning is critical. The BTS/ITS have an asthma discharge bundle to improve outcomes. The aim of this audit was to assess patients discharge plans from ED and determine if patients receive suboptimal follow up.
Methods: A retrospective audit of 145 patient visits who attended ED with a diagnosis of an asthma exacerbation from 1/10/2020 to 30/9/2021 in a tertiary referral centre in the west of Ireland.
Results: Data was collected on 99 patients from 145 visits, with 27 patients having > 1 visit with an exacerbation of asthma. A total of 106 patients were discharged from ED. Four patients received inhaler technique education, one received peak flow diary, five patients received an asthma action plan, no patients received trigger avoidance advice. Fifty-five patients were discharged with oral corticosteroids, thirty-six with inhaled corticosteroids and sixty-four with inhaled SABA. Seven patients were referred to the respiratory service, fifty-three discharged to GP, with forty-five who received no follow up.
Conclusion: Adherence with the asthma discharge bundle is poor. We have identified a need for asthma patients that could receive specialist review after ED discharge. We are re-designing the asthma ED pathway to reflect this need.
References
1. Department of Health- Management of an Acute Asthma attack in Adults. National Clinical Guideline No. 14. November 2015. ISSN 2009-6259
2. Respiratory Health of the Nation 2018. Irish Thoracic Society
3. Emergency Asthma Guideline: Management of the Acute adult asthma patient. HSE National Asthma Programme. 2012
4. Improving Outcomes in Asthma: Asthma Care Bundle. Asthma UK. British Thoracic Society. 2016
Conflict of Interest: None to declare
(1.10 Not for publication—Pulmonary rehabilitation versus usual care for adults with asthma – a Cochrane systematic review)
1.11 Establishment of a Specialist Respiratory Pharmacist Outpatient Clinic
Cairine Gormley
Altnagelvin Hospital, Western Health and Social Care Trust (WHSCT), Derry
The National Review of Asthma Deaths (2014), reported asthma deaths in the United Kingdom remain among the highest in Europe. Recent Northern Ireland data from 2012–16 revealed 182 deaths were registered due to asthma.A local audit of asthma management based on the National Asthma and Chronic Obstructive Pulmonary Disease Audit Programme (NACAP) adult asthma clinical audit 2019/20 found good practice in the initial treatment of patients admitted with asthma. However after care post discharged was sub-optimal- only 31% having the asthma care bundle completed.A Specialist Respiratory Pharmacist Outpatient clinic was established. Patients were telephoned one week after discharge with a face to face appointment scheduled one month later. A database was developed to collate data. 74 patients were reviewed between June 2021 and March 2022, with 305 episodes of care. All interventions made were recorded and graded for clinical significance using the Eadon Criteria — a scale from 1 to 6 where a score of 4 or greater represents an improvement in quality of patient care. There were 356 grade 4 interventions. University of Sheffield School of Health and Related Research (ScHARR) model was used to estimate potential healthcare savings associated with the interventions in the region of £20,000–50,000.
Conflict of Interest: None to declare.
1.12 MIF Enhances Human Mesenchymal Stromal Cell Longevity In Vivo In Allergic Airway Inflammation
Ian Hawthorne1, Hazel Dunbar1, Myron Rebello, Brion Kennedy, Seamas Donnelly2, Karen English1
1 Maynooth University 2Trinity College Dublin and Tallaght Hospital
Bone marrow derived mesenchymal stromal cells (MSCs) are currently under investigation in clinical trials as immunomodulatory therapeutics for acute respiratory distress syndrome and other inflammatory conditions. Previously we have identified the capacity for pro-inflammatory signals to enhance MSC therapeutic efficacy. In some cases, increased efficacy is associated with enhanced human MSC survival in vivo. The pro-inflammatory cytokine macrophage migration inhibitory factor (MIF) is present at high levels in severe asthma. This study sought to elucidate the influence that MIF licensing has on human MSC survival and function; cyto-protection and immune modulation.
Moreover, the effect of MIF on MSC therapeutic efficacy in vivo was investigated using a clinically relevant acute house dust mite (HDM) (Dermatophagoides pteronyssinus) model of allergic airway inflammation in humanised MIF mice. MIF enhanced human MSC proliferation in vitro and drove MSC migration in transwell assays in a CXCR4 dependent manner. MIF pre-stimulation enhanced MSC promotion of wound healing in airway epithelial cells in a VEGF dependent manner. MSCs protected against allergic airway inflammation in mice expressing higher levels of human MIF. Importantly, higher level expression of human MIF in our transgenic mouse model of allergic airway inflammation significantly enhanced longevity of human MSCs in vivo.
Conflict of Interest: None to declare
1.13 Effect of Mandatory Face Coverings on Asthma Control For a Severe Asthma Cohort
MR Kelly, C Hayes, M Malouf, E Mulligan, E Moloney, S Lane, M Kooblall
Tallaght University Hospital, Dublin
Prior studies have demonstrated a decrease in asthma exacerbations during the Covid-19 pandemic.[1] One of the primary reasons is thought to be decreased exposure to respiratory viral pathogens due to face coverings and social distancing.[1] Mandatory face covering legislation ceased in Ireland on February 28th 2022. We undertook to assess the impact of mandatory face coverings on asthma control for our severe asthma cohort in Tallaght Hospital.We reviewed the medical records of 60 patients with severe asthma on biologic therapy. 27 patients (45%) were on Omalizumab and 33 patients (55%) on anti IL-5 therapy. We compared patients Asthma Control Questionnaire (ACQ) score, peak expiratory flow rate (PEFR) and number of exacerbations from April 2021 to June 2021 inclusive with the same data for April 2022 to June 2022.55% of patients were female. The mean age was 55 years. We found our cohort had improved asthma control during our chosen time period in 2021 compared to 2022 with lower mean ACQ score and exacerbations and higher mean PEFR.Our cohort demonstrated improved asthma control during the period with mandatory face covering legislation. Several confounding factors exist including difference in lockdown/social distancing rules and Covid-19 incidence during these time periods.
References
SA Shah, JK Quint, A Shiekh. Impact of COVID-19 pandemic on asthma exacerbations: etrospective cohort study of over 500,000 patients in a national English primary care database. Lancet Regional Health- Europe. 2022; 19:100,428.
Conflict of Interest: None to declare.
1.14 Pathway utilization in the Management of Asthma
Elizabeth Kavanagh, Diane Moran, Elaine Hayes
Our Lady of Lourdes Hospital, Drogheda, Co Louth Asthma is the most common chronic respiratory disease in Ireland, resulting in a significant number of presentations to the Emergency Department annually with episodic exacerbations of asthma that could potentially be life threatening. Asthma also imposes a significant burden on health care systems, society and the family. A Respiratory Care Pathway ‘The Management of Acute Asthma’ is currently employed within the Emergency Department in order to enhance the quality of care across the continuum in line with international evidence based best practice.A retrospective chart audit was conducted to measure concordance with the care pathway.
Analysis of the results indicated that the asthma pathway was oftentimes incomplete or even in some cases not utilized.
In response to these findings a review and redesign of the care pathway was undertaken and comprehensive education provided to all stakeholders involved in the implementation and utilization of the care pathway.
Conflict of Interest: None to declare.
1.15 Analysis of the relationship between FeNO and blood eosinophil count in severe asthma patients
James O’Hanlon1, Amar Mainra2, Deirdre Long1, Lorna Lombard3, Elaine MacHale3, Dorothy Ryan1, Richard Costello1,3, Breda Cushen1
1Beaumont Hospital, Dublin,2RCSI Medical School, 3Beaumont Hospital, Dublin
Both FeNO and blood eosinophil count (BEC) are biomarkers of Th2 airway inflammation. FeNO can be easily measured in a clinic setting. We examined whether FeNO alone was sufficient to determine control of airway inflammation.
Sixty-five contemporaneous FeNO and BEC measurements from 18 patients were analysed. Concurrent ACT scores were available for 18 measurements. Non-parametric data was log-transformed prior to pairwise correlation analysis to determine the strength of the relationship, if any, between FeNO, BEC and ACT score. Data is reported as mean(SD) or median(IQR).
All patients had severe asthma prescribed high-dose inhaled corticosteroid plus additional controller therapy. No patients were prescribed asthma biologic therapy. The mean age was 56.7 years, 56% female. Mean FEV1 was 2.01 L. The median FeNO was 14(24) ppb, median BEC 0.28(0.41) cells/mm3 and median ACT score 18.5(7). FeNO and BEC were positively correlated but the relationship was weak (r = 0.406, p = 0.0008). Of FeNO recordings < 25 ppb (66%), 42% had a BEC > 300 cells/mm3. There was moderate negative correlation between FeNO and ACT (r = -0.559, p = 0.02) but no relationship between BEC and ACT (r = -0.243, p = 0.33). FeNO and BEC are independent biomarkers with variable relationship to ACT score. Both biomarkers should be considered when evaluating severe asthma patients.
Conflict of Interest: None to declare.
1.16 Patients’ satisfaction with care provided by a Respiratory Clinical Nurse Specialist in an out-patient setting
Martha Reilly1, Elaine Hayes1, Tidi Hassan1
1Our Lady of Lourdes Hospital, Drogheda
Benefits for nurse-led service provision by a Clinical Nurse Specialist (CNS) are well documented in international literature. The quality of the information and education given in a clinic setting by a CNS directly impact the Asthma patient, asthma control and their quality of life.A patient satisfaction survey was conducted in order to ascertain patient satisfaction with the CNS led service and to highlight areas which required improvement thus allowing any potential deficits in the service to be identified for future development. 56 Questionnaires were posted to patients who attended appointments with the CNS during 2021. 24 completed questionnaires were returned, yielding a response rate of 42%.Analysis of the responses demonstrated overall high levels of satisfaction (95.8%) with the educational value and care provided. Areas of improvement were highlighted with 25% of respondents stating they would like to have discussed other issues and 41.7% stating they did not receive written literature on their condition. This information will be utilised to improve and standardise service delivery. Having under treated asthma can have a negative impact on asthmatic patient health and quality of life. Feedback from the patient plays a fundamental role in improving the quality of service provision.
Conflict of Interest: None to declare
1.17 Challenges in determining suitability for home self-administration of asthma biologic therapies
Paula Hallahan and Claire Sheridan
Mater Misericordiae University Hospital, Dublin
Home administration of targeted asthma biologics has recently become an option for patients with severe Il5 mediated asthma, allowing patients greater independence in the management of their disease and more convenience around work, study and travel. Organisational gains include time and cost savings. In 2022 a directive was issued by the HSE stating that from the third dose patients should be self-administering at home. Fundamentally as experienced Clinical Nurse Specialist’s (CNS) we recognised that despite this directive not all patients would be ready within the time frame indicated, or at all.In addition to the assessment for, management, co-ordination and administration of asthma biologic therapies the respiratory CNS service in the Mater Misericordiae University Hospital is responsible for determining suitability for home administration, a process requiring sound clinical judgement. Challenges identified by our service in determining suitability included poor health literacy, low executive function, lack of commitment to treatment and overall suboptimal adherence to baseline treatment plans.Initially interest in the concept of home administration was high. However, not all patients met the criteria due to one or multiple challenges identified. Conversely a number of patients were initially reluctant and lacked confidence in their ability to self-administer therapy. However, the majority overcame this fear and successfully transitioned to home self-administration with good CNS-led support, education and training.
Conflict of Interest: None to declare
1.18 Safety of Bronchoscopy in Asthma
Noreen Tangney1, Fiona Arnott1, Desmond Murphy1
1Cork University Hospital
The British Thoracic Society reports that up to 10% of patients with asthma can develop respiratory symptoms following bronchoscopy. It is imperative to monitor and record the safety outcomes of such procedures in order to maintain a standard of care in the investigation and management of patients with asthma. Our study retrospectively reviews the occurrence of adverse events in asthmatic patients who have undergone bronchoscopy in our institution in the last 1-2 years. We recruited patients and determined the occurrence of adverse events peri-procedurally by two methods. The first being via review of patient medical notes and the second by telephoning patients individually and enquiring of their experiences within 72 h of the procedure. Specific adverse events that were reviewed included fever, extra ventolin or nebuliser use, antibiotics, steroids, GP visits, or hospital admission. The stage at which these medications were required (pre-emptively, post-procedure, or post-discharge) was examined. Pre-procedural nebuliser use was also recorded. Our study illustrates the safety profile of bronchoscopy in asthma and highlights potential adverse events albeit in the minority of patients undergoing the procedure.
1.19 Real-World Clinical Outcomes of Asthma patients switched from Reslizumab to either Mepolizumab or Benralizumab
Laura J Walsh1, Deborah Casey1, Punitha Vairamani1, Fiona Arnott1, Barry J Plant1 and Desmond M Murphy1,2
1 Cork University Hospital, Cork, 2 University College Cork, Cork
Treatment of severe refractory asthma often involves the use of targeted biological therapy. Randomised controlled trials have shown improvements in clinical parameters with these treatments but real-world data is lacking.The clinical parameters, frequency of exacerbations, number of hospital admissions, asthma control questionnaire score (ACQ), forced expired volume in one second (FEV1) and maintenance corticosteroid dose of twenty asthma patients switched from reslizumab to benralizumab or mepolizumab at 1 year prior and 6 months after switching were compared.The mean frequency of exacerbations (0.7 v 0.3) improved post switch, albeit non-significantly. The mean ACQ was essentially unchanged (1.6 v 1.5). The number of hospital admissions remained at one pre and post switch. 25% were on maintenance steroid before and after switching but one patient required an increased dose post switch resulting in an increase in the mean maintenance oral corticosteroid dose (1.6 mg to 2.4 mg). The mean FEV1 was unchanged (80% v 77.9%) six months post switching. Regarding asthma control (n = 19), 47.4% were controlled pre and post switch (ACQ < 1.5), 36.8% remained uncontrolled despite switching but n = 4 reduced exacerbation frequency, 10.5% improved control while 5.3% disimproved. We present real-world clinical outcomes of asthma patients switched from reslizumab to either benralizumab or mepolizumab with improvements in ACQ and exacerbation frequency observed without a loss of clinical effectiveness in the majority.
Conflict of Interest: None to declare
1.20 Implementation of an Asthma QI project in Galway University Hospital
Donna Langan 1, Helen Mulryan 1, Niki Byrne 2, Cameron Forward 1, Michael Harrison1, Melissa Mc Donnell 1, Ruth Cusack 1
Galway University Hospital (GUH) 1, Respiratory Integrated Care, Galway 2
Following a review of all patients with an asthma exacerbation who visited the emergency department (ED) and Acute Medical Unit (AMU) at GUH, we identified that 50% of the patients received no follow up care from their GP also, they were not referred to the respiratory team. A review of GUH’s asthma discharge care programme was required. A Quality Improvement (QI) project was carried out. It identified a number of key areas for improvement, including personalised asthma action plans, smoking cessation advice, minimising aeroallergen exposure, inhaler therapy and technique education. An Asthma Proforma was also devised to assist the review of asthma patients. Identification and referral of patients was a key component of the QI project.
All patients discharged from the ED/AMU who met the inclusion/exclusion criteria received an asthma education pack on discharge. A virtual consultation one-week post discharge with an ANP/CNSp, with an asthma education and self-management care plan was implemented. All additional investigations including PFTs/FeNo and bloods were performed prior to their face to face review. On review in the Respiratory clinic their asthma self-management care plan was reviewed and if patients were stable they were discharged from clinic.
Awaiting Conflict of Interest
1.21 Has Anyone Checked Your Inhaler Technique? A Safe Asthma Discharge Care Pathway (SADCP) Clinic Quality Improvement Project
Paul McGowan1, 2, Kathryn Ferris1, 2, Katie McMullan2, Catherine Russell2, Gillian Gallagher2, Patrick McCrossan1, 2, Dara O’Donoghue1, 2
1. Queen’s University Belfast, Belfast, 2. Royal Belfast Hospital for Sick Children, Belfast
The National Review of Asthma Deaths (2104) reported that, in children and adults who died with asthma, few had documentation of their inhaler technique.1 As part of a quality improvement project to implement a Safe Asthma Discharge Care Pathway (SADCP), we assessed documentation of newly referred patients to the nurse-led asthma clinic from the Emergency Department to determine the frequency of documentation of inhaler education.A SADCP was developed as a standard record of asthma care following stakeholder consultation and expert consensus. Data was gathered retrospectively from Emergency Department referrals pre-implementation. New referrals to the clinic were then assessed for completed documentation post implementation.Our results include a total of 38 newly referred patients. In 24 patient records (63%) there was no documentation of inhaler technique. In 14 patient records (36%) had documentation included, of which 12 (31%) had complete and 2 (5%) had incomplete documentation. Post referral there was a significant improvement with 100% of new referrals having correct documentation. Going forward, the SADCP will be integral to ensuring documentation of many aspects of asthma care including inhaler technique. Introduction of a SADCP has optimised written documentation of inhaler technique education to help improve paediatric asthma care.
Royal College of Physicians. Why asthma still kills: The National Review of Asthma Deaths (NRAD) 2014
Conflict of Interest: None to declare.
1.22 How inhaler technique is assessed in children and young people: a scoping review
Kathryn Ferris1, Patrick McCrossan1,2, Paul McGowan2, Kathryn Wilson2, Michael Shields1, James Paton3 and Dara O’Donoghue1,2
1Queen’s University Belfast, 2 Royal Belfast Hospital for sick children, Belfast, 3 University of Glasgow
Current guidelines recommend that all asthma review appointments must include an assessment of inhaler technique. However, most guidelines do not provide information on how the healthcare professional should conduct this assessment. The aim of this scoping review was to explore the published literature on methods used to assess inhaler technique.
We searched Medline, Embase, Cinahl and the Cochrane library for studies published from 1st January 1956 to 1st February 2022, on methods of assessing inhaler technique in children and young people with asthma. Two reviewers completed screening and data extraction independently.
Sixty-six papers were identified for full text analysis. Primary themes included assessment through both subjective measures including checklists or global scoring and objective measures including inspiratory flow and drug deposition. Secondary themes included observer variability, validation of checklists, impact on asthma outcomes and remote assessments. A consultation exercise with a parent group provided valuable insights and allowed us to map their experiences to the overarching themes.This scoping review provides a broad overview of currently used methods to assess inhaler technique in children and young people with asthma. The analysis of which will allow us to consider how these methods might be used in clinical practice and research settings.
Conflict of Interest: None to declare.
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
2. Interstitial Lung Disease
2.1 Nintedanib for progressive pulmonary fibrosis: 12-month outcome data from a multicentre observational study
Lavanya Raman1, Iain Stewart2, Shaney Barratt3, Felix Chua1, Nazia Chaudhuri4*, Anjali Crawshaw5, Michael Gibbons6, Charlotte Hogben1, Rachel Hoyles7, Vasilis Kouranos1, Jennifer Martinovic8, Sarah Mulholland3, Kate Myall8, Marium Naqvi8, Elisabetta A Renzoni1, Peter Saunders7, Matthew Steward6, Dharmic Suresh9, Muhunthan Thillai10, Athol U Wells1, Alex West8, Jane A Mitchell2, Peter M George1
Royal Brompton Hospital, London, UK, 2. Imperial College London, London, UK,3.North Bristol NHS Trust, Bristol, UK,4.University of Ulster, Northern Ireland, UK,5.University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK,6.Royal Devon & Exeter NHS Foundation Trust, Exeter, UK,7.Oxford University Hospitals NHS Foundation Trust, Oxford, UK,8.Guys & St Thomas’ NHS Foundation Trust, London, UK,9.Manchester University NHS Foundation Trust, Manchester, UK,10.Royal Papworth Hospital NHS Foundation Trust, Cambridge, UK,11. Presenting Author
Introduction: Nintedanib attenuates non-IPF progressive pulmonary fibrosis (PPF) in clinical trials but real-world safety and efficacy are lacking. We assessed the impact of nintedanib on the clinical course of patients with PPF.
Methods: Eight UK centres collected standardised data retrospectively and prospectively from patients in whom nintedanib was initiated for PPF between 2019-2020 through an early access programme. Primary analysis included ung function change in the 12 months pre- and post-nintedanib initiation. Secondary analyses included symptoms, drug safety and tolerability.
Results: 126 patients were included of which 67(53%) females with mean age 60(± 13) years. At initiation of nintedanib, mean FVC was 58% and DLCO was 33%. 63% of patients were prescribed oxygen, 68% were prescribed prednisolone (median dose 10 mg) and 69% were prescribed steroid sparing agents. In the twelve months after nintedanib initiation, lung function decline was significantly lower than in the preceding twelve months; FVC -113 mL vs -235 ml, (p = 0.013) and absolute DLCO -2.87% vs -5.79%; (p = 0.02). Response to nintedanib was not influenced by ILD diagnosis, age, CT pattern, or MRC dyspnoea grade. 71% of patients reported side effects. 80% of patients were still taking nintedanib at 12 months. There were no serious adverse events.
Conclusion: In PPF, real-world efficacy of nintedanib mirrors that of clinical trials reducing lung function decline by approximately 50%. Nintedanib was safe and tolerable.
Conflict of Interest: None to declare
2.2 – Abstract withdrawn
2.3 – Abstract withdrawn
2.4 – Abstract withdrawn
2.5 Sarcoidosis Prevalence Western Health & Social Care Trust
Liam Coyle, Caoimhe McGarrigle, Jan Kara, Juan Pastrana, Jason Wieboldt
Terence McManus
South West Acute Hospital, Enniskillen
Sarcoidosis is a multi-systemic condition of uncertain aetiology, characterised by granulomatous inflammation. It predominantly affects the lungs but can affect any organ.
On a global scale it is recognised that there are differences in prevalence of sarcoidosis across geographic regions. The British Lung Foundation study [Respiratory health of the nation] demonstrated a consistently higher prevalence of sarcoidosis in Northern Ireland in comparison with other regions of the UK. We aimed to identify if there was a difference in the prevalence of sarcoidosis in the Western Trust [Southern Sector] in comparison to the rest of Northern Ireland and the UK. We also aimed to identify any clusters of sarcoidosis in local population areas.
We collected data by identifying living patients who had been coded with a diagnosis of Sarcoidosis and seen in secondary care over the previous 10 years. There were 207 cases identified, demonstrating a prevalence of 157 cases per 100,000 population. We also divided patients into postcode areas and estimated the point prevalence in each area. Clusters of higher prevalence were identified at BT75, BT92 and BT94. There is a need for further prevalence studies as this could identify an under resourced health care need within specific trust areas.
Conflict of Interest: None to declare.
(Not for Publication)
2.6 Reviewing the Spectrum of Presentations associated with Myositis-Related Interstitial Lung Disease (ILD) – A Single Centre Series
2.7 Clinical characteristic and cyst burden score in patients with Lymphocytic Interstitial Pneumonia
2.8 Clinical Characteristics in Lymphangioleoimyomatosis)
2.9 An Evaluation of an Interstitial Lung Disease Clinic in a Regional Hospital Using Subspecialist Network with an Academic Centre
Sarah Farrell, Moeez Maqbool, Elaine Curran, Tidi Hassan
Our Lady of Lourdes Hospital, Drogheda
Multidisciplinary team (MDT) diagnosis of interstitial lung disease (ILD) has been proposed as gold standard. We review the activities of a dedicated clinic, supported by a nurse practitioner at Our Lady of Lourdes Hospital, using the MDT platform with Beaumont Hospital.From June 2020 to June 2021, 93 patients were referred with confirmed interstitial abnormalities on CT thorax. The mean waiting list time to appointment was 69 ± 21 days. Fifty-nine cases were discussed at the MDM meeting with 42 (71%) confirmed a usual interstitial pneumonitis (UIP) pattern and 29 (31%) confirmed as progressive pulmonary fibrosis on surveillance as per guideline (2022). Three cases were referred to cardiothoracic surgery for VATS biopsy. Nine patients (33%) commenced on an anti-fibrotic therapy. Twelve patients (13%) were commenced on morphine for palliation.
Here we report the feasibility to see high volume of ILD cases in a dedicated regional clinic with MDM support in an academic centre, in line with Slaintecare to deliver care at the right place.
References
Raghu G, Remy-Jardin M, Richeldi L, Thomson CC, Inoue Y, Johkoh T, et al. Idiopathic Pulmonary Fibrosis (an Update) and Progressive Pulmonary Fibrosis in Adults: An Official ATS/ERS/JRS/ALAT Clinical Practice Guideline. American Journal of Respiratory and Critical Care Medicine. 2022 May 1;205(9):e18–47.
Conflict of Interest: None to declare
2.10 Compliance with American Thoracic Society (ATS)/European Respiratory Society (ERS) guidelines for lung fibrosis reporting in CT-Thorax in University Hospital Kerry, and demographic distribution of pulmonary fibrosis in Kerry Group
Asma Mehmood 1 , Anwar Ali 1 , Aimal Khan Danish 1 , Umar Khan 1 , Usman Haider 1
1University Hospital Kerry
Audit is aimed to assess reporting practices of CT-Thorax with lung fibrosis based on ATS/ ERS guidelines in UHK. Retrospective data were collected from 01/05/2022 to 31/07/2022. A total of 221 CT-Thorax were surveyed. CT-Thorax reporting of pulmonary fibrosis should be done according to ATS/ERS guidelines as (1)Usual Interstitial Pneumonia (UIP), (2)Probable UIP, (3)Indeterminate UIP, (4)Alternative diagnosis. Out of 221, 24 were reported as fibrosis (Table 1) and 15 were reported according to ATS/ERS guidelines. An improvement of 62.5% (15/24) was found compared to previous audit (11/07/2021-11/01/2022), where 23.7% (23/74) of the reports were according to the ATS/ERS guidelines.
Table 1 (2.10) Demographic distribution of pulmonary fibrosis on CT-Thorax in UHK Age group Gender Total
Male Female
≥ 80 3 (12.5%) 4 (16.7%) 7 (29.2%)
60 – 79 9 (37.5%) 7 (29.2%) 16 (66.7%)
40 – 59 1 (4.2%) 0 1 (4.2%)
Total 13 11 24
Ct-Thorax reporting as per ATS/ERS guidelines would help physicians for better investigation/management of pulmonary-fibrosis. 95.9% patients with pulmonary-fibrosis were > 60 years old (29.2% from ≥ 80 years group and 66.7% from 60 – 79 years group; Table 1), notably same percentage as national/international data. ATS/ERS guidelines and results of previous 2 audits were discussed with local radiology group on different levels to improve overall reporting practices. Temporary staffing is a factor where guidelines are not properly followed. Proper reporting can reduce referrals to speciality and can help physicians in better investigation/management of patients.
Conflict of Interest: None to declare
2.11 Computed tomography pulmonary fat attenuation volume (PFAV); a novel idiopathic pulmonary fibrosis biomarker?
Marissa O’Callaghan,1,3 John Duignan,1 Elizabeth J. Tarling,2 Darragh Waters,1 Meghan McStay,1
Orla O’Carroll,1 Marcus W. Butler,1,3 Aurelie Fabre,1,3 Michael P. Keane,1,3 David J. Murphy,1
Cormac McCarthy1,3
1St. Vincent’s University Hospital, Dublin 2University of California, 3University College Dublin
There is increasing interest in the role of lipids in pulmonary fibrosis. Based on the o increased focal subpleural fat observed in IPF biopsies and increased mediastinal fat radiologically, we hypothesised that objective measures of pulmonary lipid may have utility as an IPF biomarker. IPF and control lung biopsies were analysed by immunohistochemistry and unbiased lipidomics. Pulmonary fat attenuation volume (PFAV) was assessed on CT images with SyngoVia (Siemens Healthineers). Aerated lung was automatically segmented and PFAV calculated by Hounsfield-unit thresholding of segmented dataset and expressed as percentage of total lung volume. PFAV was compared to objective markers of disease severity.The total lipid content was significantly increased in IPF lung tissue (23.16 nmol/mg) compared to controls (18.66 mol/mg)(n = 10,p = 0.0317). The median PFAV in IPF was significantly higher than control (1.35% vs 0.79%)(n = 75:28,p < 0.0001) and PFAV correlated with FVC (R2 = 0.407,p < 0.0001) and DLCO (R2 = 0.356, P < 0.0001)(Fig. 1A-D). Moreover, PFAV was higher in IPF even with preserved FVC (p = 0.0024) and DLCO (p < 0.0001). PFAV correlates with radiological features of fibrosis. PFAV is a novel non-invasive measure of pulmonary lipids which shows good correlation with pulmonary function, is automated and easily quantifiable. It could function as a novel biomarker for disease severity assessment in IPF.
(2.11)
Conflict of Interest: None to declare
2.12 Do all patients who have a transbrochial biopsy (TBBx) performed during bronchoscopy need a chest Xray: a single centre study
Waqas Mahmood, Owais Rahman, Junaid Rasul, Orlaith Shinners, Shahram Shahsavari, O'Brien A
University Hospital Limerick
Bronchoscopy is one of the most commonly performed procedures in the respiratory specialty. The risk of pneumothorax after TBBX is one of the major complications that can occur. The aim of our study is to identify risk factors for developing a pneumothorax and clinical findings indicating one has occurred.In this retrospective study, 163 patients underwent flexible fibreoptic bronchoscopy with TBBx (without fluoroscopy) with different pulmonary pathologies at UHL. We assessed patient demographics, clinical sign and symptoms, CT and bronchoscopic findings, and final diagnosis.
Of the 163 patients, 11 (6.7%) developed a pneumothorax. 7 were male; the mean age was 69 ± 2 years; 8 were ex-smokers (72.7%). Out of 163 patients 21 patients had chest pain and 16 complained of shortness of breath. All the patients who had pneumothorax had chest pain and shortness of breath. Only 6 patients required chest drain insertion and the remainder were managed conservatively. The mean duration of hospital stays of those patients requiring chest drain were 4 days (± 1).7% of patients who had a TBBx developed a pnemothorax and the most common presenting complaints were chest pain and shortness of breath. We suggest that CXR not be performed routinely after TBBX, and only in symptomatic patients, and thus reducing the burden on radiology and wasting of valuable resources.
Conflict of Interest: None to declare
2.13 Lung aging in a dish – Induced pluripotent stem cell-derived type 2 alveolar epithelial cells undergo telomere length attrition following prolonged 3D culture
Anja Schweikert1,2, Mari Ozaki1,2, Irene Oglesby1,2, Killian Hurley1,2
1 RCSI, Dublin 2 Tissue Engineering Research Group, RCSI, Dublin
Pulmonary Fibrosis can be linked with mutations in surfactant protein production and telomere length (TL) regulation. Mutations causing shortened telomere length in type II alveolar epithelial cells (AT2) result in severe, early-onset disease. We hypothesise that pro-fibrotic signalling is activated by telomere attrition and that AT2 cells generated from induced pluripotent stem cells (iPSC) (iAT2) can be used as a preclinical model.Genomic DNA samples were acquired on Days 0, 3, 200 and 400 during directed differentiation to iAT2. TL was measured by high-throughput quantitative fluorescent in-situ hybridization, real-time quantitative PCR and Southern blot. The expression of pro-fibrotic, inflammatory and senescence genes at the different timepoints was measured by qPCR to investigate associations with telomere shortening.Analogous to the telomere attrition occurring in-vivo, TL decreased 25-65% from D0 to D200 and 65-80% from D200 to D400 in our model, associated with increased expression of senescence, pro-fibrotic and anti-apoptotic genes and decreased expression of pro-apoptotic genes.This is the first reported model of telomere shortening in human iAT2 cells and explores the relationship of telomere length and pro-fibrotic pathways. Our preliminary data showed a negative association between telomere shortening and pro-inflammatory signalling, indicating an important role in the onset of fibrosis.
Conflict of Interest: None to declare
2.14 The role of acute lower respiratory tract infection and microbial colonisation in interstitial lung disease exacerbations
Ruaidhri J. Keane1, Louise Kelly1, Sinead O’Donnell1 Killian Hurley1
1Beaumont Hospital, Dublin
Interstitial lung disease (ILD) is an umbrella term for a heterogenous group of parenchymal lung disorders that are characterised by a similar clinical phenotype, overlapping radiographic disease patterns, and early mortality. It has been reported that in-hospital mortality for acute exacerbations of idiopathic pulmonary fibrosis (IPF) exceeds 50%1.
The aim of this study was to explore the role of acute lower respiratory tract infection (LRTI) and microbial colonisation in ILD exacerbations by analysing sputum cultures in a cohort of patients with ILD.342 patients were reviewed at the ILD centre over a 5 year period. 49%(n = 169) had a diagnosis of IPF. 10% (n = 17) of these patients had a positive sputum culture. Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis, and Haemophilus influenzae were the most commonly isolated bacteria and accounted for 64.5% of all bacteria cultured.There was a low use of sputum sampling in this study likely secondary to patient and care delivery characteristics. The prevalence of gram negative bacteria was an unexpected result and may be due in part to a selective advantage from the repetitive use of antimicrobials in the community in this population. The relationship between IPF exacerbations, LRTI’s and microbial colonisation is complex and requires further study.
References
Brereton CJ, Jo HE. Acute exacerbations of idiopathic pulmonary fibrosis and the role of corticosteroids. Breathe (Sheff). 2020 Sep;16(3):200,086. 10.1183/20734735.0086-2020. PMID: 33,447,274; PMCID: PMC7792795.
Conflict of Interest: None to declare
2.15 Evaluating Patients Admitted with an Exacerbation of Idiopathic Pulmonary Fibrosis (IPF)
Seamus Lehane, Cara M Gill, Omaima Omar, Harry Hughes, Orla O’Carroll, Lindsay Brown Emmet McGrath, Michael P Keane, Cormac McCarthy
St. Vincent’s University Hospital, Dublin
IPF is a chronic progressive lung disease with high associated mortality and 20% annual incidence of exacerbations1. Patients with IPF are also at increased risk of venous thromboembolism (VTE)2. The aim of this study is to review diagnostic approaches in patients admitted with an exacerbation of IPF.A database of patients with IPF was screened for admissions to a single centre from January 2018 to December 2019. Data was collected from health records including demographics, d-dimer, computed tomography and pulmonary function testing.
Overall, 16% of IPF patients were admitted with a mean length of stay of 12.2 days. Median age at exacerbation was 80.2 years. Only 5% had d-dimers measured. Thirty-one percent of patients were evaluated with CT. Of these, 1/3 underwent CTPA with negative studies. Patients who were admitted with an exacerbation had lower forced vital capacity than those who were not admitted (65.4% vs 76% predicted). Patients admitted with an exacerbation had higher mortality at 2 years than those not admitted (81% vs 30%).
A small cohort of IPF patients required admission with an exacerbation resulting in high mortality. Amongst these, few were evaluated for VTE. Factors influencing this should be further evaluated.
References
1Song JW, Hong SB, Lim CM, Koh Y, Kim DS. Acute exacerbation of idiopathic pulmonary fibrosis: incidence, risk factors and outcome. Eur Respir J. 2011 Feb;37(2):356-63. https://doi.org/10.1183/09031936.00159709. Epub 2010 Jul 1. PMID: 20595144.
2Sprunger DB, Olson AL, Huie TJ, Fernandez-Perez ER, Fischer A, Solomon JJ, Brown KK, Swigris JJ. Pulmonary fibrosis is associated with an elevated risk of thromboembolic disease. Eur Respir J. 2012 Jan;39(1):125-32. https://doi.org/10.1183/09031936.00041411. Epub 2011 Jul 7. PMID: 21737559; PMCID: PMC3757572.
Conflict of Interest: None to declare
2.16 A review of investigative procedures in the multidisciplinary diagnosis of Interstitial Lung Disease
O’Sullivan Grace, Musameh Khaled & Henry Michael
Cork University Hospital
Recent research indicates an increase in diagnostic confidence in diagnosing Interstitial Lung Disease (ILD) with Bronchoscopic Lung Cryobiopsy (BLC) over Surgical Lung Biopsy (SLB)1. In this study, we reviewed the investigations performed prior to the diagnosis of various ILD cases discussed in our Multidisciplinary Team Meeting (MDT).A retrospective review of the MDT cases between July 2019 and May 2022 was performed. This yielded a total of 137 ILD cases, of that, 48 of those were diagnosed with Idiopathic Pulmonary fibrosis. Investigations included CT imaging, Bronchoscopy with bronchoalveolar lavage (BAL), BLC or both; SLB and serological testing.Of the 48 IPF patients, BAL was performed in 46, BLC in 19 and SLB in 3. Both CT and serological testing were performed in all IPF cases. Data from other ILD cases of different aetiologies were also collected, 36 patients underwent BLC, of which 23 had histological features that helped establish a diagnosis, SLB was performed in 6 and the remaining cases did not have any tissue sampling. Thus, BLC supports diagnostic confidence in IPF and other ILD within this regional ILD centre and the reliance on SLB is reduced.
1.Tomassetti S., Well A.U., Costabel U., Cavazza A., Colby TV., Rossi G., Sverzellati, N., Carloni A., Carretta E., Buccioli M., Tantalocco P., Ravaglia C., Gurioli C., Dubini A., Piciucchi S., Ryu JH., & Poletti V. (2016) Bronchoscopic Lung Cryobiopsy Increases Diagnostic Confidence in the Multidisciplinary Diagnosis of Idiopathic Pulmonary Fibrosis. American Journal of Respiratory and Critical Care Medicine 193(7), 745- 752.
Conflict of Interest: None to declare
2.17 Practice changing hypersensitivity pneumonitis guidelines?
Marissa O’Callaghan .,1,2 Omaima Omar.,1 Lindsay Brown,1 Michael Keane.1,2, Emmet McGrath, 1,2 Cormac McCarthy 1,2
1St. Vincent’s University Hospital, Dublin;2University College Dublin
Until 2020, most patients who received a diagnosis of hypersensitivity pneumonitis (HP) required an invasive lung biopsy. The novel HP guidelines recommend that a confident HP diagnosis can be made with a typical or compatible HRCT, identifiable antigen and BAL lymphocytosis > 30%. A biopsy is still indicated with diagnostic uncertainty. The aim of our study was to see if our practice has changed in the wake of these guidelines. This retrospective observational study collected data on patients diagnosed with HP at MDT between January 2020 and December 2021.
42 patients were included; 57% male, mean age 59.7 years. 54.7% had a documented exposure with corresponding antibodies in 5/23. HRCT was categorised by degree of fibrosis in 39 cases. 17 patients had a BAL; 8 of whom had lymphocytosis > 30%. 50% of patients had a tissue biopsy (Table 1), only 1 of whom had a BAL lymphocytosis and identifiable exposure. While exposure identification and HRCT categorisation are now synonymous with our practice, BAL was performed in only 40%. 50% of patients still had a tissue biopsy. Going forward we should have a lower threshold for BAL and careful consideration of whether biopsy is truly needed. Procedure (n = 21) No. (%)
VATS Biopsy 10 (23.8%)
Transbronchial biopsy* 10 (23.8%)
EBUS** 3 (7%)
Table 1 (2.17): Method of tissue biopsy. *One patient later underwent VATS biopsy. **One patient later underwent transbronchial biopsy
Conflict of Interest: None to declare
2.19 Idiopathic pulmonary fibrosis & palliative care: a qualitative exploration of nurses’ experiences
O’Sullivan Grace1, Watson Chanel2, Murray Bridget2 & Henry Michael1
1 Regional ILD centre, Cork University Hospital
2Royal College of Surgeons, Dublin
Symptom burden in Idiopathic Pulmonary Fibrosis (IPF) is evident and often multidimensional. Thus,national and international IPF guidelines recommend early palliative care implementation1,2. Despite this, delayed palliative care implementation and inappropriate referral timing in IPF has been identified as a common trend3.
The research aim was to explore nurses’ experiences in relation to the delivery of palliative care in IPF which may have the potential to improve both the future care of patients and the experience of providing care to patients with IPF by nursing staff.
A qualitative description approach was used. Purposive sampling yielded nine nurses working within an acute respiratory department. Data was collected through semi-structured interviews over 10 weeks and analysed using thematic analysis.
Based on the nurses’ experiences, palliative care implementation timing should be assessed on an individual basis due to the unpredictability of one’s disease trajectory. The respiratory nurses suggested that palliative care discussions should not take place at the time of the diagnosis and information should be provided gradually. Advanced care planning (ACP) was identified as a strategy by the nurses to ensure patient centred care in the event of an acute deterioration. While barriers to appropriate palliative care implementation existed within the nurses’ experiences, they deemed education as an essential to improve or resolve the current barriers between palliative care and IPF.
While ambiguity still surrounds the most appropriate implementation time, palliative care should be implemented at an earlier stage to reduce symptom burden. ACP is an initiative that may positively contribute to IPF care going forward.
References
1. Irish Lung Fibrosis Association, (2015). National Patient Charter for Idiopathic Pulmonary Fibrosis. Available from: https://ilfa.ie/wp-content/uploads/2021/10/ILFA_CharterBooklet_lores-1.pdf [accessed on 15/10/22].
2. 3. Raghu, G., Collard, H. R., Egan, J. J., et al. (2011) An Official ATS/ERS/JRS/ALAT Statement: Idiopathic Pulmonary Fibrosis: Evidence-based Guidelines for Diagnosis and Management. American Journal of Respiratory Critical Care Medicine, 183(6), 788–824.
3. Lindell, K. O., Liang, Z., Hoffman, L. A., Rosenzweig, M. Q., Saul, M. I., Pilewski, J. M., Gibson, K. F. & Kaminski, N. (2015) Palliative care and location of death in decedents with idiopathic pulmonary fibrosis. Chest, 147(2), 423-429.
Conflict of Interest: None to declare
2.20 Frequency of pulmonary complications among patients of pneumoconiosis experienced in tertiary care hospital
J Alam, B Ahmad, M Imran, M Yousaf
Pulmonology Unit Lady Reading Hospital, Peshawar, KPK. Pakistan
Pneumoconiosis is a pathological reaction of lung tissue after exposure to inorganic substances, such as crystalline silica, asbestos and coal dust and this is very common in north of Pakistan because of coal mining, Asbestos in soil and marble factories. A wide spectrum of complications occurs in patients with pneumoconiosis including chronic obstructive pulmonary disease, pneumothorax, pleural diseases, progressive massive fibrosis and malignancy etc. This study was conducted to determine frequency of pulmonary complications among patients of pneumoconiosis. This was a cross sectional study which was conducted from 1st July 2019 to 31st December 2019. In this study total of 120 patients were observed. Patients of pneumoconiosis diagnosed on bases of occupational history, Chest X rays and HRCT. Other investigation like U/S chest, ABGs and biopsy etc. was done in selected patients to look for different complications.Mean age was 42 with SD ± 8, 96% were male. The main complications in our study was pneumothorax (22%) and COPD (12%). Pleural effusion (6%), progressive pulmonary fibrosis (4%) other complications (4%) like malignancy and chronic interstitial pneumonia etc. were also observed. Pneumothorax is common complication of pneumoconiosis with COPD being the second most common.
Conflict of Interest: None to declare
2.21 Clinical characteristics and their correlation to genotype and cyst burden in patients with Birt-Hogg-Dube Syndrome
Fiona Hickey,1 Omaima Omar,1 Evelyn Lynn,1 Marissa O’Callaghan,1 Orla O’Carroll,1 Joseph Morrow,1 David J. Murphy,1 Aurelie Fabre,1,2 Michael P. Keane,1,2 Cormac McCarthy1,2
1St. Vincent’s University Hospital, Dublin, 2University College Dublin
Birt-Hogg-Dube syndrome (BHDS) is a rare autosomal dominant disease characterised by pulmonary cysts, pneumothorax, fibrofolliculomas and an increased risk of renal malignancy. It is caused by germline mutations in the folliculin (FLCN) gene. All patients attending the rare lung diseases clinic with a diagnosis of BHDS were included for analysis. Age, genetic mutation, clinical features, history of pneumothorax, PFTs and cyst score were analysed. Twenty-six patients were included. Median age was 52 (Range 23-88) and 58% female. There were 5 different genetic mutations identified. The most common of these was C.17_21delCTCTC. 3 patients had no identifiable known FLCN mutation. 35% had fibrofolliculomas, 8% had renal lesions. 30% had at least one prior pneumothorax. In those with the C.17_21delCTCTC mutation, 44% had a pneumothorax. Mean FEV1 was 87% (Range:44-119%), mean DLCO 80% (Range:56-112%) and the mean cyst score was 2% of total lung volume (range 0.1-5.4%). Cyst scores did not correlate with pulmonary function. Patients with history of pneumothorax had higher cyst burden scores (mean = 2.37% vs 0.9%, p = 0.08) but had no difference in lung function.
The presentation of BHDS is varied, even within families. Further research is required to identify those most at risk and ensure early identification of affected individuals.
Conflict of Interest: None to declare
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
3. Lung Cancer
3.1 Improving the Patient Journey—A Rapid Access Lung Cancer Satellite Service
Gráinne Cogan, Ian Counihan
Our Lady of Lourdes Hospital Drogheda
The lung cancer diagnostic pathway is a stressful and complex journey for patients and their families. Therefore listening to patients is an integral part of improving the quality of our service delivery. The Rapid Access Lung Cancer satellite clinic was developed in 2018 in order to coordinate and improve access to lung cancer investigations and diagnosis for patients at a local level. In total over 2000 patients have availed of this service resulting in the diagnosis of 392 new lung cancers. A patient satisfaction survey was conducted in order to ascertain if patients were satisfied with the service offered, highlight areas which required further enhancement and identify any potential deficits in the service for future development.
100 questionnaires were posted to patients and family members attending the Rapid Access Clinic between 2018 and 2021. Data was collected by means of an adapted validated self-administered questionnaire. There was a response rate of 72%. Analysis demonstrated overall high levels of satisfaction with the care, information, support and efficiency of the service (90.3%). Areas for improvement were highlighted which included lack of space, privacy and long wait times at the clinic (11.1%) This information was utilised to drive service delivery.
Conflict of Interest: None to declare
3.2 Establishment of a respiratory led ambulatory indwelling pleural catheter service in a large tertiary centre
Finbarr Harnedy1, Ross Morgan1,2, Emmet O’Brien1, Imran Sulaiman1,2, Daniel Ryan1
1 Beaumont Hospital, Dublin, 2 Royal College of Surgeons in Ireland, Dublin
Malignant pleural effusion (MPE) occurs in up to 15% of all cancers with significant symptom burden and high incidence of recurrence. Definitive management of MPE is largely carried out via chemical pleurodesis or insertion of an indwelling pleural catheter (IPC). International best practice models have shifted to a predominantly outpatient and elective service model for MPE management using IPC. Hence, a Respiratory led IPC service was established in Beaumont Hospital in September 2021.
To date, 15 IPCs have been completed by this new service. Most, 12 (80%), were performed while an inpatient. Two patients had a post procedure pneumothorax that required brief attachment to an underwater seal with subsequent resolution. There were no cases of post procedure infection, insertion failure, drain fracture or significant haemorrhage. 2 patients had their IPC removed due to effusion resolution and auto pleurodesis by 4 months.
Furthermore, international standard length of stay for MPE is 7 days, 3 cases were carried out through the ambulatory service with an estimated 21 bed days saved and a cost saving of up to €18,333. No ambulatory patients have reported complications to date. Establishment of an ambulatory IPC service is an effective and cost saving measure.
Conflict of Interest: None to declare
3.3 Malignant Pleural Effusions: An audit of current practice in an Irish tertiary teaching hospital
Finbarr Harnedy1, Shane O’Brien1, Seamus Leahane1, John Patrick Jones1, Imran Sulaiman1, 2, Daniel Ryan1
1 Beaumont Hospital, Dublin, 2 Royal College of Surgeons in Ireland, Dublin
Malignant pleural effusions (MPE) are a frequent complication of advanced malignancy associated with high morbidity, mortality, and healthcare costs. We retrospectively audited adherence to American Thoracic Society (ATS) guidelines for the management of MPE in a large tertiary hospital. To do this, we analysed physical charts, discharge summaries and radiological investigations of all patients with a diagnosis of MPE between 2020 and 2021 initially identified through HIPE data. Within this time period, there were 71 patients with a diagnosis of MPE and 67 (94%) were found to be symptomatic with breathlessness. Of these 67 patients, 60 (90%) had large volume thoracentesis. A definitive pleural intervention was performed on 19 (28%) patients; 11 (16%) patients underwent indwelling pleural catheter (IPC) insertion while 8 (12%) patients had talc pleurodesis performed. The average length of inpatient stay for all patients admitted with a malignant pleural effusion was 16 days with a range of 1 to 122 days. In this tertiary hospital the majority of patients with symptomatic a malignant pleural effusion underwent pleural intervention which is in line with ATS recommendations and highlights the need for pleural specialty care.
Conflict of Interest: None to declare
3.4 Retrospective data analysis on levels of Radon exposure in Lung cancer patients in Level 2, 3 and 4 Hospitals in the Republic of Ireland.
Jehangir Khan1,2, Waheed Shah1,2, Marissa O’Callaghan1,2, David Healy1, Marcus Butler1,2 Aurelie Fabre, Cormac McCarthy1, 2 Michael P Keane1, 2,
1 Saint Vincent’s University Hospital, Dublin,2 University College Dublin
Lung cancer is the leading cause of cancer worldwide. Radon is associated with increased risk of lung cancer after tobacco exposure. EPA has published updated data on Radon exposure in Ireland in 2022. Anecdotally we notice different stage of lung cancer presentation depending on geographical location. We hypothesise that certain radon exposure may be associated with increased staging of lung cancer at presentation.
Methods: Retrospective analysis of lung cancer patients was performed. We tracked Eircodes and evaluated exposure to Radon based on their geographical location.
Results: Analysis of total of 567 patients were performed, n:256 fell in the green zone category of radon exposure, n:126 in the red zone of radon exposure and n:123 in yellow zone of radon exposure. Sub analysis based on staging of lung cancer showed that n:35 was in the red of radon exposure in stage 4 lung cancer, while n:90 was in the green zone of radon exposure within stage 1 lung cancer cohort.
Conclusion: Our data suggests that that Radon may be linked to increase stage of lung cancer. It needs further exploration. Fig. 1(3.4): Comparative analysis of Patients with Radon exposure in Level 2,3 and 4 Hospitals.
Conflict of Interest: None to declare
3.5 A Dedicated Solitary Pulmonary Nodule Service Reduces Inappropriate Surveillance CT
Maqbool Moeez, Sarah Farrell, Grainne Cogan, Aidan Quinn, Tidi Hassan
Our Lady of Lourdes Hospital, Drogheda
Our Lady of Lourdes Hospital has a dedicated solitary pulmonary nodule service since 2020 that includes a multidisciplinary (MDT) meeting with the radiology and respiratory team. We review the activities of this service and compared the number of surveillance CT performed before the service was established. A retrospective study comparing incidental CT reports of solitary nodules in 2018 versus 2021 was performed. Eighty-nine percent (n = 113) of nodules reported in 2021 were discussed at the MDT meeting with 28 percent (n = 32) discharged at first review based on the BTS 2015 guidelines. Seventy-one patients (88%) were followed up to completion for surveillance CT with 9 patients referred to the Rapid Access Clinic. Meanwhile, 106 solitary nodules were reported in 2018. Ninety-one (86%) of patients received surveillance CT although 16 cases did not require surveillance as per BTS guidelines. 42 (46%) surveillance CTs were performed outside the BTS guidelines including 6 weeks and 6 months’ timeframe. Nineteen (20%) cases completed surveillance beyond the recommended time which led to an excess of 1.7 CTs for 2018 versus 2021.
Here we report that a dedicated solitary pulmonary nodule service reduces inappropriate surveillance CT which may overall reduce health cost as well as patient’s anxiety.
References
Callister MEJ, Baldwin DR, Akram AR, Barnard S, Cane P, Draffan J, et al. British Thoracic Society guidelines for the investigation and management of pulmonary nodules: accredited by NICE. Thorax. 2015 Jun 16;70(Suppl 2):ii1–54
Conflict of Interest: None to declare
3.6 A retrospective review of lung cancer referrals and initial workup in Mercy University Hospital Cork pre/ during the Covid-19 pandemic and following initiation of patient redistribution
Emily O’Reilly1, Laura Monaghan1, Seán Landers1, Art Kelleher1, Robert Shannon1, Prof Terry O’Connor1, David Curran1
1. Mercy University Hospital, Cork
This retrospective review aimed to assess referral and initial workup of lung lesions in patients referred or presenting to the Mercy University Hospital Cork 2019-2022. The study period encompassed pre, during, and post Covid-19 lockdown era as well as pre and post introduction of medical admission specialty redistribution at the study site.
The Covid-19 pandemic continues to have a profound impact on healthcare in Ireland. The pandemic prompted some hospitals to change the pathway for acute medical admissions.
This review included all patients investigated for malignant lung lesions January 2019- June 2022 in Mercy University Hospital. Age, sex, mode of referral (emergency department/ other medical specialty/ inpatient/ outpatient/ other hospital), stage at diagnosis, and outcome from multidisciplinary team meeting were assessed. 349 patients met inclusion criteria, 152 female, 197 male. Median age was 70 years. There was a significant difference in mode of referral when comparing all four years, Chi-squared test (p = 0.012). Non respiratory team referral 2019 vs. 2022 52% vs. 31%. There was no significant difference in age or sex at presentation when comparing all four years, Kruskal-Wallis (p = 0.46), Chi-squared (p = 0.20) respectively. The introduction of medical admission specialty redistribution may reduce time to respiratory consultant assessment and therefore diagnostics and management decisions.
Conflict of Interest: None to declare
3.7 A Single Centre Retrospective Analysis of Stage of Presentation of Lung Cancer Pre and Post Covid-19 Pandemic
C Ottewill1, F Mulvaney1, F O’Connell1, P Nadarajan1
1. St. James Hospital, Dublin
Introduction/Aims: This study investigated the effect of the Covid-19 pandemic on lung cancer staging at presentation.
Methods: This retrospective cohort study investigated new primary lung cancers staged at lung cancer MDT in a tertiary referral centre. We compared cancer staging in the 6-month period before the Covid-19 pandemic (Apr 2019- Sept 2019, Group 1) to a similar period after start of the pandemic (Sept 2020- Feb 2021, Group 2). Patient demographics, tumour stage and treatment referrals were analysed.
Results: Patients presented on average two years older at time of diagnosis in Group 2. 39% of patients in Group 2 being current smokers, compared to 29.6% in the pre-pandemic group. Median tumour size at diagnosis was larger in Group 2 (5.6 cm vs 4.1 cm), reflecting overall upgrading from T3 to T4. Those in Group 2 had higher stage nodal disease, with 36.5% of patients presenting with N3 disease, compared to 20.9% in Group 1. There was no observed statistically significant increase in rate of metastatic disease at presentation, with 30.7% in Group 1 presenting with metastatic disease, and 31.7% in Group 2.
Conclusion: This study suggests patients are presenting with more advanced lung cancer following the pandemic, with larger tumours, and higher burden of nodal disease. This is likely multifactorial and further study will be required to explore factors such as lung cancer screening and patient education that may prevent this trend towards later presenting disease.
Conflict of Interest: None to declare
3.8 Difference of postoperative pain between video-assisted and robotic-assisted approach in thoracic surgery
Ghaith Qsous, Amber Downes, Beata Carroll, Sinead Rowe, Santy Manoj, Michael Tolan, David G Healy
St.Vincent’s University Hospital, Dublin
Corresponding author* : Ghaith Qsous
In the last decade there was significantly evolving in thoracic surgery with the spreading use the robotic surgery. We aim to evaluate the pain after at least 6 months from the surgery using the robotic and video-assisted approach.This is a retrospective study that included 92 patients who underwent different thoracic surgeries. Patients were divided into two groups; Video-assisted (VATS) (51 patients), and robotic-assisted (RATS) (41 patients). The EQ-5D-5L questionnaire was used to estimate the utility values of QOL. In the VATS group patients underwent more anatomical lung resection comparing to the RATS group with a p-value of 0.005. In the VATS group, 62.7% of patients had no pain at the time of the questionnaire compared to 51.2% in the RATS group but without a statistically significant. Also, 25.5% vs 39% of patients had mild pain in VATS and RATS respectively but without a statistically significant p-value.In conclusion, on the long-term after surgery, our results didn’t show that RATS has better pain control compared with VATS. Though, robotics is known to have higher hospital costs. The continuance of a comparative study with VATS may be essential. And some efforts need to be taken into consideration to reduce postoperative pain and cost.
Conflict of Interest: None to declare
3.9 Diagnostic yield and complication rates from transbronchial biopsies of abnormal lung lesions in a tertiary referral centre for lung cancer
D Quigley, F O’Connell, P Nadarajan
St James’ Hospital, Dublin
Transbronchial lung biopsy (TBBX) is a procedure performed during flexible bronchoscopy under fluoroscopic guidance. We primarily use this modality to evaluate suspicious peripheral lesions. Diagnostic yield of this procedure varies in the literature but is between 35-60%. Pneumothorax is the most common complication (0.5-5%) with significant pulmonary haemorrhage also an identified risk (0.6-1.5%). A total of 105 procedures were done over 9 month period November 2021-August 2022. Patient data was collected using the electronic patient record system (EPR). 105 patients were included in the analysis. 94 were evaluated for diagnostic efficacy. 11 patients were excluded (Table 1). MDT discussions/clinic letters/radiological investigations/bronchoscopy reports/histological reports and relevant admission notes where analysed in all subjects.
Our diagnostic accuracy from transbronchial biopsies is 71% (67/94). A further 12% of patients were diagnosed with endobronchial ultrasound (EBUS-TBNA) that was done concurrently. Our findings demonstrate that transbronchial biopsies can have a high diagnostic yield in addition to having low rates of complication. It prevents a delay in diagnosis while awaiting CT guided biopsies, which itself also carries a higher risk of pneumothorax. We advocate for TBBX to be standard of care in tertiary lung cancer centres and that performing EBUS during same procedure when possible is safe and effective.
Table 1 (3.9)
Conflict of Interest: None to declare
3.10 Appropriateness of Rapid Access Lung Cancer (RALC) referrals
Shahram Shahsavari, Mohammed A. Alabandi, Musab Eltayeb, Orlaith Shinners, Owais Rahman, Aidan O’Brien
University Hospital Limerick
RALC clinic plays a pivotal role in early diagnosis and treatment of patients with cancer. There is an established guideline from National Cancer Control Programme (NCCP) for patient referrals by GPs and medical practitioners. It is important that patients referred are appropriate to ensure efficient use of resources. In this retrospective study we recruited referrals sent to RALC using consecutive sampling in Feb 2022 and compared it to NCCP guidelines. We assessed appropriateness of referrals as well as rate of proven outcomes.
Out of a total of 48 patients, 30 Patients (62.5%) were referred with lung mass / nodule, 5 patients (10.4%) with non-resolving infiltrates, 5 patients (10.4%) with haemoptysis and 8 patients (16.6%) with other abnormalities.21 patients (43.7%) were referred by hospital clinicians 11 of which were seen as inpatient consults and 27 patients (56.2%) were referred by their GPs. While 17 patients (35.4%) still required further follow up, only 4 patients (8.3%) were actually proven to have malignancy at the time of the initial presentation. 7 patients (14.5%) did not have appropriate imaging before referral and 14 referrals (29%) were deemed as inappropriate. This study concludes that further education regarding NCCP guidelines is required to reduce number of inappropriate referrals to RALC clinics.
Conflict of Interest: None to declare
3.11 THE POWER OF RESPIRATRY CYTOLOGY; A DIAGNOSTIC AND THERAPEUTIC PARADIGM SHIFT
Diarmuid O’Connor 1, David Gibbons 1, 2, Aurelie Fabre 1,2
1.St. Vincent’s University Hospital, Dublin, Republic of Ireland
2. University College Dublin, Dublin, Republic of Ireland
Lung cancer is the leading cause of cancer related death worldwide. 70% of cases are inoperable at diagnosis, therefore respiratory cytology specimens represent crucial sources of tissue. The use of cytology has expanded beyond the diagnostic sphere to also encompass a key role in therapeutic molecular studies. Our aim was to demonstrate the diagnostic and therapeutic value of respiratory cytology for patients with lung cancer.We analysed all respiratory cytology specimens received at out institution between 2020 and 2021. We recorded the number of malignant specimens, their concordance to available biopsies and whether they were used to perform therapeutic molecular studies.1432 respiratory cytology specimens were analysed. Of these, 15.6% (n = 223) were malignant. Of the 223 malignant specimens, 61% (n = 136) were non-small cell lung cancer (NSCLC), incorporating 102 patients. Of these patients, 37.3% (n = 38) has molecular studies successfully performed on cytology specimens with another 14.7% (n = 15) having PD-L1 studies alone performed. Of the 136 NSCLC specimens, 79.7% of cytology specimens with a corresponding biopsy successfully classified NSCLC into adenocarcinoma or squamous cell carcinoma.
Our study highlights the value of respiratory cytology as an effective diagnostic and therapeutic tool, showing value in both classifying NSCLC and in performing essential molecular studies.
3.12 2021 Audit of Endobronchial Ultrasound-guided Transbronchial Needle Aspirates in an Irish Lung cancer Tertiary Referral Centre
Daniele Di Capua, Aurelie Fabre
St Vincent’s University Hospital, Dublin
Endobronchial ultrasound (EBUS) guided transbronchial needle aspiration is a minimally invasive technique used to investigate mediastinal and hilar lymphadenopathy. This retrospective audit of all EBUS guided samples collected between 1st January and 31st December 2021 assessed the diagnostic yield, adequacy, and malignancy subtypes detected by EBUS and use of rapid on-site evaluation (ROSE) in an Irish tertiary referral center.212 lymph nodes were sampled in 168 EBUS procedures, representing 20% increase from 2020 (140 EBUS). Mean patient age was 60.7 years and 41.4% (69/168) female. ROSE was performed in 89.9% (151/168) of cases, with a positive yield in 84.5% (142/151). Overall inadequacy rate was 6.5% (5% in 2020).EBUS malignancy rate was 37.5% (63/168) overall and 56% (61/109) in cases of suspected malignancy or staging, with pulmonary adenocarcinoma being the most common subtype (28.6%). Granulomas were observed in 24.4% (41/168) of EBUS procedures overall, and in 69.4% (25/36) of EBUS procedures performed for investigation of sarcoidosis.EBUS is effective for the diagnosis of both malignant and non-malignant disease with excellent adequacy rates and provides source material for molecular analysis. EBUS and malignancy rates increased, while ROSE rates slightly decreased in 2021, and overall inadequacy rate was slight higher.
Conflict of Interest: None to declare
3.13 Audit – A study of guideline adherence in pulmonary nodule surveillance
Ruaidhri J. Keane1,Claire Bolton1,Michael Emmet O’Brien1
1Beaumont Hospital, Dublin
Pulmonary nodules are a common imaging finding encountered by physicians. Incidental detection of pulmonary nodules has risen significantly with increased Computed Tomography scanning. The British Thoracic Society and Fleischner Society guidelines are used to guide longitudinal pulmonary nodule surveillance (PNS) in Ireland. There is considerable variability in physician adherence to these guidelines, and is typically less than 50% [1]. The aim of this single centre audit was to evaluate the adherence to guidelines in patients being followed up by the PNS service.Over a 24 month period, 162 patients were reviewed by the PNS service. 71 continued surveillance, 68 were discharged with stability, 7 were referred to the rapid access lung cancer clinic, and 17 discontinued surveillance due to death (n = 5) or patient preference (n = 12). 47.5% (n = 77) had multiple pulmonary nodules. 145 patients were appropriate for analysis of guideline adherence. 60% underwent compliant surveillance. Delayed or early imaging, and prolonged period of surveillance were the most common reasons for failure of adherence.The adherence to PNS guidelines is greater in this centre than in other similar centres in published studies [1]. Barriers to guideline adherence include resource availability, patient compliance, SARS-CoV2 pandemic, and physician interpretation of guidelines.
References
McDonald, Jennifer S et al. “Addition of the Fleischner Society Guidelines to Chest CT Examination Interpretive Reports Improves Adherence to Recommended Follow-up Care for Incidental Pulmonary Nodules.” Academic radiology vol. 24,3 (2017): 337–344.
10.1016/j.acra.2016.08.026
Conflict of Interest: None to declare
3.14 Lung Cancer in Women; A Growing Health Concern?
Art Kelleher, Laura Monaghan, Emily O’Reilly, Sean Landers, David R. Curran, Terence M. O’Connor
Mercy University Hospital, Cork
Lung cancer is a serious women’s health issue and kills more women annually than breast, ovarian and uterine cancers combined. The demographic makeup of lung cancer has evolved dramatically. According to literature, diagnosis of Lung Cancer in women is increasing with women more likely to be diagnosed at a younger age and earlier stage of disease. An increased incidence of adenocarcinoma has also been described.
We retrospectively analysed data of 808 primary Lung Cancer diagnoses in the Mercy University Hospital, Cork over a twelve-year period (2010-21). Over that timeframe, there was no increase in either the outright number of diagnoses in women or in the percentage in women as a fraction of the total annual incidence. There was no evidence of presentation in women at a younger age than males with a m. We identified that the percentage of women presenting with stage 4 disease was higher than in males. However, we are beginning to appreciate an increase in adenocarcinoma diagnoses in women.
Broadly, the data we have accumulated suggests a remarkably different trend in demographics compared to described literature. Further data collection and analysis would aid in better comprehending these findings.
Conflict of Interest: None to declare
3.15 Two years on, how do we measure up? A re-audit of Rapid Access Lung Cancer Clinic Deferral Letter Process
O. Shinners, S. Shahsavari, S. Noor, J. Rasul, Z. Ali, R. Murphy, A. O’Brien
University Hospital Limerick
The National Cancer Control Programme recommends a 14 day window for review in Rapid Access Lung Clinic (RALC) from day of referral. University Hospital Limerick is the primary centre for RALC referrals in the Mid-West and thus receives large volumes. This results in a strain to meet this set standard. Many referrals do not need formal RALC review and can be deferred with advice for repeat imaging or for no action required following review of the referral. An initial audit of this deferral letter process was undertaken in 2020 and concluded that 1/3 of patients referred to RALC can be safely deferred back to their initial referrer, however 15% did not have the appropriate follow-up arranged as advised. Following this, our protocol was updated to include a follow-up telephone call. This study aims to evaluate RALC deferral letters from January-June 2021 and to compare outcomes to those reported in 2020. NIMIS Radiology System was utilised to evaluate if appropriate re-imaging was carried out as advised. 280 RALC referrals were received during the period of review, of which 15% (n = 43) met deferral criteria. Of these, 42% required no action and 58% had furthering imaging recommended. Of those requiring further imaging, 80% had this undertaken as appropriate with 15% re-referred due to persistent abnormalities. The percentage of patients who did not have follow-up as advised represented 12% of all deferred patients (n = 5). This study again highlights that a particular subset of patients do not need formal RALC review and can be safely triaged back to the initial referrer. The updated protocol has reduced the number of patients without appropriate follow-up.
Conflict of Interest: None to declare
3.16 Beyond COVID-19: An Evaluation of the Timepoints in Lung Cancer Diagnoses 2019 vs 2022
O. Shinners1, K. Cinnamond1, O. Rahman1, S. Shahsavari1, R. Murphy1, E. Maltseva1, A. O’Brien1
University Hospital Limerick 1
Lung cancer proves a major healthcare challenge. The diagnosis, staging and treatment of lung cancer requires a multi-disciplinary approach and should be undertaken in a timely manner. The National Cancer Control Program (NCCP) advises a 14 day window for patients to be evaluated in the Rapid Access Lung Clinic (RALC) from initial referral. Guidance on time to diagnosis (TTD) is less clear. This varies within the literature from 28-60 days from initial referral. This study aims to evaluate and compare time to first RALC review and TTD of lung cancer patients in 2019 versus 2022. These dates have been specifically chosen to exclude the years impacted by the Covid-19 pandemic and to assess for any residual delays.
In 2019, 64 patients were diagnosed with a primary malignancy via RALC. Mean time to RALC was 15.2 days (median 13), with mean TTD 37.9 days (median 13). In comparison 65 primary malignancy diagnoses were made in 2022, mean time to RALC was 13.1 days (median 10), with mean TTD 50.6 days (median 15). While time to review in RALC improved from 2019 to 2022, TTD disimproved. Those with TTD > 60 days had a more complex path to diagnosis including repeat imaging, PET CT scans, MDM discussions as well as the requirement for CT guided biopsy. Further analysis was undertaken to identify any specific steps in the diagnostic process that led to longer TTD.
Conflict of Interest: None to declare
3.17 An SpR focused pictorial review to improve detection of primary lung malignancy on conventional radiography—lessons learned in a regional lung cancer centre
Grainne N. Allen; Patrick Cooper; Réiltín Hayden; Iain Irvine; Shane Davy; Zaid Tabba; Mark Rogan; Susan Foley; Michael Anthony Farrell
University Hospital Waterford
Lung cancer is the leading cause of cancer related deaths worldwide and earlier diagnosis of the disease improves prognosis. While X-ray will never replace CT in detection of lung cancer, conventional radiography, and the chest x-ray in particular, is often the primary imaging modality that detects or suggests a neoplasm. Here we present a pictorial review of lung masses on plain film, based on our experiences as a regional lung cancer centre.We will discuss the review and check areas on chest x-ray and the use of inversion windowing to highlight areas such as posterior to the heart or diaphragm domes. We will show examples of lung masses picked up on non-dedicated studies, including thoracic spine and shoulder radiographs, and stress the importance of reviewing all areas on an x-ray, regardless of the clinical question. We will place particular emphasis on comparison with prior radiographs and on repeating poor quality x-rays. Our goal is to provide a review to improve detection rates of lung cancer on conventional radiographs, with a focus on teaching for both respiratory and radiology Specialist Registrars.
Conflict of Interest: None to declare
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
4. Patient Care
4.1 A retrospective analysis of referrals and outcome measures of the Respiratory Integrated Care Hub, Chronic Disease Management Programme, Cork City
Murphy J1, Casey D1, O’Riordan U1, Vairamani P1, Tangney N1, Gomez F1,2
1Chronic Disease Hub, Respiratory Integrated Care, SMOH, Cork, 2 Cork University Hospital
Respiratory disease, specifically COPD and Asthma account for the highest numbers of in-patient hospitalisations and bed days compared to other diseases within our Irish health system. Thus leading to the timely introduction of Respiratory hubs within the Integrated Care Programme for prevention and management of chronic disease (ICPCD). Our hub opened its doors to referrals and patients in June 2022. We performed a retrospective analysis of referrals received and triaged by RIC and outcomes including diagnosis and further tests required. Patients who fit the inclusion criteria were referred by their GP. These referrals are triaged by the respiratory team in the hub to the appropriate pathway for diagnostics and review. The patients were reviewed by a CNS and ± the consultant. If patients do not require a direct review by the consultant, their case is discussed at a weekly MDT and a report with diagnosis and management plan is sent to the GP. Follow-up with the RICT is often required. Data was collected through our referral database and patient outcomes.
A total of 40 patients were triaged and seen from mid-June to 31st July. 39 patients were seen in either spirometry only clinics (n = 15) or full assessment (consultant and spirometry) (n = 24). 1 patient was referred for PFTs only (waiting list). Of 15 patients seen for spirometry, n = 4 new diagnosis of COPD, n = 5 new diagnosis of Asthma, n = 2 new diagnosis of ACOS. 1 patient had a change in diagnosis and 3 patients needed further review with PFTs. The results of the full respiratory assessment clinics has not been included in this data as patients are awaiting blood results and further investigations.
Our expert care, diagnostics and knowledge means accurate diagnosis, disease staging and appropriate management, of respiratory patients living with a chronic disease in conjunction with their GPs.
Conflict of Interest: None to declare
4.2 Developing a toolkit for the multidisciplinary team on Inhalers on Carbon Footprint
S.Green1, A. Kirk1, J.O’Callaghan3, C.Wynne3, E.Hurley4, S.Owens5, J.Allen6, K.M.A. O’Reilly1, 2
1. Mater Misericordiae University Hospital, Dublin, 2. University College Dublin, 3. Respiratory Integrated Care in Dublin North City and County, Ireland 4. Trinity College Dublin, 5. Blackrock Medical Centre, Dublin, 6. Suttoncross Surgery, Dublin
Pressurised metred dose inhalers (MDIs) contain a propellant (hydrofluroalkanes) which has a disproportionate global warming potential compared to dry powdered inhalers (DPIs) and soft mist inhalers (SMIs). In an online survey of respiratory multidisciplinary team members, 94% felt it was important to consider carbon footprint when choosing an inhaler. However, over half did not feel confident identifying patients who do require an MDI and those who could suitably be managed with a DPI or SMI. A multidisciplinary group performed a literature review which informed the development of a toolkit for clinicians on inhalers and carbon footprint. The toolkit includes information on the available inhalers carbon footprint and distance travelled equivalent as well as dosage, cost and correct inhaler technique. This toolkit is designed to be easily utilised in the clinical setting and to guide clinicians into choosing the right inhaler for the right patient with the lowest possible carbon footprint. Three documents, including a brief guide, a device choice, technique and carbon footprint table and an evidence based document were generated. The ‘inhalers and carbon footprint’ toolkit addresses a previously unmet need to provide clinicians with the necessary information to reduce the carbon footprint of inhalers while improving patient care.
(4.2)
Conflict of Interest: None to declare
4.3 Patients’ satisfaction with care provided by a Respiratory Clinical Nurse Specialist in an out-patient setting
Martha Reilly1, Elaine Hayes1, Tidi Hassan1
1Our Lady of Lourdes Hospital, Drogheda
Benefits for nurse-led service provision by a Clinical Nurse Specialist (CNS) are well documented in international literature. The quality of the information and education given in a clinic setting by a CNS directly impact the Asthma patient, asthma control and their quality of life.
A patient satisfaction survey was conducted in order to ascertain patient satisfaction with the CNS led service and to highlight areas which required improvement thus allowing any potential deficits in the service to be identified for future development. 56 Questionnaires were posted to patients who attended appointments with the CNS during 2021. 24 completed questionnaires were returned, yielding a response rate of 42%.Analysis of the responses demonstrated overall high levels of satisfaction (95.8%) with the educational value and care provided. Areas of improvement were highlighted with 25% of respondents stating they would like to have discussed other issues and 41.7% stating they did not receive written literature on their condition. This information will be utilised to improve and standardise service delivery.Having under treated asthma can have a negative impact on asthmatic patient health and quality of life. Feedback from the patient plays a fundamental role in improving the quality of service provision.
Conflict of Interest: None to declare
(4.4 Not for publication 4.4 New Guidance on Azithromycin Prophylaxis- A Starting Point)
4.5 Mid-West Respiratory Physiotherapy Optimisation Clinic – A Service Evaluation
Sarah Cunneen1, Jarlath Healy2, Aidan O’Brien3, Máire Curran1, Brian Fitzgibbon1, Lauren Kennedy1, Maura Cleary1
1 Physiotherapy, Integrated Care Programme for Chronic Disease, HSE CHO3, 2 Toomevara Health Centre, 3 University Hospital Limerick
A pilot community-based respiratory clinic has been established in the Mid-West with the aim of optimising respiratory management and enhancing patient care of those living with chronic respiratory diseases. It enables GP practices in North Tipperary to refer patients at the lowest level of complexity directly to respiratory services in line with the HSE Integrated Model of Care for the Prevention and Management of Chronic Disease. The clinic, ran by a Clinical Specialist Physiotherapist, provides one-to-one appointments to patients with chronic respiratory diseases with an opportunity for participation in community based Pulmonary Rehabilitation Classes if required. The face-to-face clinic runs in locations as close to the patients home as possible. Continuous data acquisition will take place with a snapshot clinical audit at 4 months (November 2022). The objectives of the audit are displayed in the table below.
(4.5) Outcome Measures Assess and record the respiratory physiotherapy treatments delivered
Assess and monitor the appropriateness of GP referrals to the service
Patient experience surveys
GP experience surveys
Process Measures Track uptake of the service (percentage of referred patients who attend)
Monitor wait times for the service
Balancing Measures Resources required for the service
Onward referrals to other services
The outcome of the audit will evaluate the pilot clinic and in addition inform future respiratory service planning in the Mid-West.
Conflict of Interest: None to declare
4.6 Evaluation of community based pulmonary rehab programmes in Limerick: A 6-month analysis
Máire Curran1, Brian Fitzgibbon1, Lauren Kennedy1, Sarah Cunneen1, Liam O’Connell1, Louise Crowley1, Maura Cleary1, Patricia O’Rourke1, Aidan O’Brien2
1 Physiotherapy, Integrated Care Programme for Chronic Disease, HSE CHO3
2 University Hospital Limerick, Limerick
Pulmonary rehabilitation (PR) has established itself as a key management strategy in the treatment of people with chronic respiratory disease. Care should be focused in the community as much as clinically appropriate in line with the HSE Integrated Model of Care for the Prevention and Management of Chronic Disease. Historically, PR was only delivered in one acute site, at University Hospital Limerick (UHL). Due to COVID-19 and subsequent cancellation of outpatient services, particularly for group settings, as well as a vulnerable population, this resulted in a two year wait time for PR in Limerick.
PR in the community was established as part of the Respiratory Integrated Care service. In February 2022 there were 234 clients waiting for PR with a longest time of 122 weeks at UHL. There were 189 additional referrals between February and August 2022. Various PR venues have been risk assessed and utilised by the PR service such as GAA community halls and hotels ensuring that the venue was suitable, easily accessible to clients in view of choice of locality and had adequate parking.
(4.6) Initially Currently
Wait time 122 weeks 15 weeks
Number of clients waiting 234 94
Parking €4 per class All venues sourced have free parking and car park is more accessible.
Location UHL only PR programmes have been completed in Limerick City, South Limerick and West Limerick.
Locations closer to clients’ home.
Reduced travel expenses for clients.
Other considerations As there is more physical space in community venues and increased staffing levels this enables more clients to be enrolled in each PR programme.
However, there is a rental cost for each PR venue which should be considered.
(4.6)
In conclusion, this new service has reduced PR waiting times and ensured a more accessible location to those living with a chronic lung condition in the community.
Conflict of Interest: None to declare
4.7 Cork University Hospital review of inhaler education initiative within the inpatient service
Vairamani P1,Casey D1, O’Sullivan G1, Bowen B1, Murphy J1,Noonan C1, O’Grady M1, Arnott F1, Kennedy M1, Plant BJ1, Gomez F1, Henry M1,Murphy DM1
1 Cork University Hospital, Wilton, Cork
2 Health Research Board, Clinical Research Facility, University College Cork, Cork
Asthma and COPD are common respiratory chronic diseases with substantial symptom burden on the healthcare system both nationally and internationally (GINA, 2021 & GOLD, 2021). Despite, advancements in treatments and management guidelines there has been an increasing incidence on the prevalence of these conditions globally in the recent years. Inhaled drug delivery remains the cornerstone of treatment for both COPD and Asthma. However, 90% of these patients do not use their inhalers correctly and thereby experience frequent exacerbation of their symptoms some of which can be life threatening requiring hospitalization (Lavorini et al., 2008). We provided 30 education sessions in 2021 to over 150 healthcare professionals, on wards within CUH 190 assessments were completed on inpatients requiring inhaler education and disease management review. Each of the patient assessed, received face to face teaching with placebos for each appropriate device prescribed. More than 90% of these patients were reviewed for first time education on how to use their prescribed inhalers. It was found that 80% of patients had poor inhaler technique. Lack of education re their disease and inhaler adherence accounted for 60% of these patients reviewed.
If inhaler technique is reviewed on every possible patient contact this will help to improve the importance of inhaler adherence and be associated with reduced symptom burden. This will also help to reduce the burden on hospital admissions. Patients are more likely to use their inhalers effectively, when inhaler technique is demonstrated and educated to an appropriate level.
References:
https://ginasthma.org/wp-content/uploads/2021/05/Whats-new-in-GINA-2021_final_V2.pdf . (Accessed on 25 th August 2022)
https://goldcopd.org/wp-content/uploads/2020/11/GOLD-2021-POCKET-GUIDE-v1.0-16Nov20_WMV.pdf . (Accessed on 25 th August 2022)
Lavorini F., Magnan A., Dubus J.C., Voshaar T., Corbetta L., Broeders M., Dekhuijzen R., Sanchis J., Viejo J.L., Barnes P., Corrigan C., Levy M. & Crompton G.K. (2008). Effect of incorrect use of dry powder inhalers on management of patients with asthma and COPD. Respiratory Medicine. 102(4):593-604. https://doi.org/10.1016/j.rmed.2007.11.003. Epub 2007 Dec 20. PMID: 18083019.
Conflict of Interest: None to declare
4.8 Time from referral to urgent bronchoscopy in Beaumont Hospital: an audit of practice Jan-March 2022
D Barry1, PC Ridge1, Da Ryan1
Beaumont Hospital, Dublin1
The British Thoracic Society recommend that all urgent bronchoscopies be completed within 7 days of referral. We retrospectively examined all bronchoscopy referrals over a three month period to examine compliance with this guideline. A total of 269 bronchoscopies were performed over the study period. While 101 were identified as urgent, only 96 were included in this study, due to incomplete data. Table 1 outlines the indications for urgent bronchoscopy. The median wait for an urgent bronchoscopy was 4 days and 68.75% (66/96) were performed within 7 days of referral. For the indication of malignancy the median time to bronchoscopy was 3 days and 76% were performed within seven days.In this audit we have shown that our wait times for urgent bronchoscopy, are generally within the recommended guidelines of 7 days. A designated referral system for all urgent bronchoscopies, to improve waiting times, is currently in progress.
Table 1 (4.8) Indictations for Urgent Bronchosopy Patient Numbers Malignancy 63/96
Haemoptysis 18/96
Potential TB 7/96
Non resolving consolidation 5/96
Immunosuppressed with lung infiltrates 3/96
Conflict of Interest: None to declare
4.9 Respiratory Integrated Care (RIC): Evaluating the Patients Perspective
Niki Byrne, Ruth Kelly, Philippa Needham
Respiratory Integrated Care, Integrated Care Programme Chronic Disease, Galway City Ambulatory Hub
This study sought to evaluate the patient’s experience of attending a community based, Nurse-led Respiratory Clinic, ensuring that the care provided is of a high standard and is effective in meeting patient’s needs. RIC Nurse-led clinics were established in County Galway in 2016. The service was last evaluated from a patient perspective in 2018. A patient experience questionnaire (PEQ) was posted to patient’s attending the clinic between January 2021 and August 2022. Responses were collated and analysed. Results showed a 61% response rate. 100% of respondents felt that attending the RIC Nurse-led clinic was worthwhile. Patients were asked about their confidence in understanding and self-managing their respiratory condition. Overall, there were improvements seen in these areas (60% & 57% respectively). The majority (73%) of attendees had a preference to attend in a community setting, rather than a hospital environment. The current RIC service is being absorbed into the Integrated Care Programme for Chronic Disease. This study looked at what aspects from patient feedback need to be incorporated when planning patient pathways to the Chronic Disease Hub Respiratory Team. This will ensure the new service is patient focused; leading to better outcomes through higher levels of patient engagement and patient satisfaction.
Conflict of Interest: None to declare
4.10 The HSE Tobacco Free Campus Policy: A survey of awareness and implementation
Stanley DW Miller1, Edward Moloney2, Emmet O’Brien3, Hilary McLoughlin4, Angela Radley5, Lisa Glynn6, Suzanne McCormack7, Luke Clancy8. The Irish Thoracic Society Tobacco Control Advisory Group
1 Mater Misericordiae University Hospital, UCD School of Medicine, and CHO9, Dublin; 2 Tallaght University Hospital, Dublin; 3 Beaumont Hospital, Dublin; 4 Portiuncula University Hospital, Ballinasloe; 5 Tipperary University Hospital, Clonmel; 6 University of Galway; 7 Irish Thoracic Society; 8 TobaccoFree Research Institute Ireland, TUD
The ITS Tobacco Control Advisory Group in collaboration with the HSE Tobacco Free Ireland Programme (TFIP) aims to improve implementation of the HSE Tobacco Free Campus Policy (2012).
An online questionnaire (26 questions) was circulated to the ITS membership (453 members) via an email link. All responses were anonymous. The response rate was 20.3% (92) with all membership categories represented. Institutions included Acute Hospitals, Ambulatory Care Hubs and Primary Care Centres. 92.22% of respondents were aware of the policy. 72.22% stated that the policy had been implemented in their institution. 74.44% did not know who was responsible for policing the policy. Only 39.56% institutions had a clearly highlighted smoking boundary. 91.01% respondents had not been trained to communicate the policy in a non-confrontational way. 68.13% would not feel safe approaching a smoker to communicate the policy. 88.89% regularly observe patients and 62.22% regularly see staff members smoking on site. 63.33% need to walk through smoke/vapour to access an entrance door to their institution. These results provide a better understanding of the awareness of the policy with ‘real world’ feedback from ITS members. We plan to share the findings with the HSE TFIP aiming to make our healthcare campuses truly smoke-free.
References
HSE Tobacco Free Campus Policy. HSE. 2012.
Conflict of Interest: None to declare
4.11 Smoking Cessation Advice Documentation in a Respiratory Outpatient Department
D. Halim 1 , J. Khan 1 , M. Higgins 1 , C. Dempsey 1 , N. White 1 , D. O’Callaghan 1
1-Mater Misericordiae University Hospital
Smoking cessation advice is a crucial component of maintaining respiratory health.. However, this comparatively simple and cost-effective intervention is frequently overlooked. We evaluated the adherence rate to the recommendations made in the National Stop Smoking Clinical Guideline (2022) at our centre.Consecutive charts of patient who attended different Respiratory Clinics were reviewed. The primary outcome was change in percentage documentation of smoking status and smoking cessation advice before and after intervention to improve adherence rates.
One hundred patients were included (mean age 62, range 20-87). 53% were male and COPD was the leading respiratory diagnosis. Smoking status was documented in 67% at most recent clinic appointment (12 smokers, 34 ex-smokers, and 21 never smokers). Documentation was more likely in new appointment but similar between Face-to-Face and Virtual Clinics (Table 1). Of 36 patients that had no smoking status documented, 16 had smoking status previously documented within the past 3 years or since referral. All 12 current smokers were advised on smoking cessation. Intervention was carried out in the form of stamp in the patient’s charts as a guide and reminder. Re-audit showed modest improvement in the documentation while smoking cessation advice was given to 100%.
Smoking cessation advice is consistently provided to current smokers in the clinics. However, more efforts are needed to achieve 100% target for documentation of status. Documentation Yes
(n = 67) No
(n = 33) P-value
Clinic
Face-to-Face
Virtual
50
17
21
12
0.25
Appointment
New
Return
18
49
3
30
0.04
Table 1 (4.11) Comparison between documentation and type of clinics
Conflict of Interest: None to declare
4.12 Expanding Pulmonary Function Testing (PFT) outside the Hospital walls and into the community: Our first month’s findings
1Orla Wynne, 1Jeff Murphy, 1Julius Gramba, 1Craig McDonnell, 1Deirdre Fitzgerald, 1Eddie Moloney
1 Tallaght University Hospital (TUH) & HSE, CHO 7 -Hub3 Russell Building
Tallaght Pulmonary Function Community testing facility opened in the Russell Centre in July 2022, staffed with Senior Respiratory Physiologists rotating from Tallaght University Hospital. In the first month of opening, 79 patients (50 female) attended for testing, of these 79 patients, 29 were current smokers, 26 ex-smokers and 24 never-smokers. Age ranged between 20 and 83 years.
All patients had spirometry, the majority (76 patients) also had Diffusion Capacity (DLCO) testing, 3 patients were unable to perform. In total, 27.8% of patients had an abnormal FEV1, and 46% percent of patients had an abnormal DLCO. 19.7% of patients had both an abnormal FEV1 and DLCO. 26% of patients had normal spirometry but abnormal DLCO. 33 patients had reversibility testing. Of which, 48% met the criteria for reversible airflow obstruction after 400mcg of Salbutamol was administered.
From this data, particularly that 26% of patients had normal Spirometry but abnormal DLCO, we highlight that it is important to provide more comprehensive pulmonary function testing in the community setting, other than just spirometry, in order to make a more accurate diagnosis. Over a quarter of patients would have been given a normal result based on spirometry testing alone.
Conflict of Interest: None to declare
4.13 Long Term Oxygen Therapy prescribing practice and adherence
Emma McArdle1, Margaret Gleeson1, Laura Piggott1, Minesh Kooblall1, Eddie Moloney1, Stephen Lane1
Tallaght University Hospital, Dublin1#
Long term oxygen therapy (LTOT) has been shown to improve survival in patients with chronic hypoxaemia who use it for > 15 h a day. We report an audit carried out to assess the Long Term Oxygen Therapy (LTOT) prescribing practice and adherence in our department. All 48 patients recruited attended the Oxygen Clinic in Tallaght University Hospital. Data including demographics, diagnosis, oxygen prescription and compliance were collected. LTOT was prescribed as per Irish Thoracic Society guidelines. 54% of patients were male. 58% had an underlying diagnosis of COPD while the other 42% had Interstitial Lung Disease (ILD). All patients had Ambulatory oxygen therapy (AOT) prescribed with 52% on LTOT additionally. Of the LTOT group only 56% were compliant with treatment, however this is in keeping with other similar studies. Despite advances in oxygen delivery devices making both domiciliary and ambulatory LTOT more practical and more convenient for patients adherence remains suboptimal. Dedicated pre-assessment, training and continued follow up may help improve compliance.
Conflict of Interest: None to declare
4.14 An Audit on the Compliance of CXR follow up – Beaumont Hospital
Jack Allen1, Sile Toland1, Peter Branagan1
1. Beaumont Hospital
British Thoracic Society (BTS) Guidelines state that all patients, regardless of admission status to hospital, with a community acquired pneumonia (CAP) should have a follow up chest x-ray (CXR) within 6 weeks, particularly those at a higher risk of underlying malignancy [> 50 years old or a smoker]. 1We describe a retrospective audit of Beaumont Hospital’s compliance with BTS guidelines in relation to follow up CXR imaging for patients diagnosed with a radiographically significant community acquired pneumonia. Through HIPE data, we reviewed the records of 50 random patients with a documented diagnosis of “Pneumonia” on their discharge letter from the period of January to March 2022. We confirmed this with their imaging reports. 42% were female, 58% male. Median age of 60.5. 78% were deemed to have a high risk of underlying malignancy. 64% had repeat imaging. Of those who had repeat imaging, 59% had repeat imaging within 6 weeks, 18% in 6-12 weeks and, 23% in greater than 12 weeks. 38% fulfilled the BTS guidelines. With this degree of compliance, detecting other underlying lung pathologies may be delayed. Our plan is to re-audit in 6 months time after completing educational teaching with the NCHD cohort.
References
1. British Thoracic Society Guidance on Respiratory Follow Up of Patients with a Clinico-Radiological Diagnosis of COVID-19 Pneumonia [Internet]. British Thoracic Society; 2020.
Conflict of Interest: None to declare
4.15 An Audit of Oxygen Prescriptions in Medical Patients in a Model 3 Hospital
Patrick Doyle1, Claire Quigley1 & Elaine Hayes1
1 Our Lady of Lourdes Hospital, Drogheda
Medical oxygen is one of the most commonly used drugs in acute care settings. Inadequate prescribing is common, as shown in a UK national audit where only 57.5% of patients had a valid prescription.1 This audit aims to evaluate if oxygen was prescribed as per Irish Thoracic Society (ITS) guidelines in medical inpatients.2 A snapshot audit of oxygen prescription in 196 medical inpatients was undertaken on the 3rd July 2022. Adherence with ITS Oxygen prescription guidelines should include documentation of SpO2, flow rate, target SpO2, delivery system and duration/review date.
Out of 196 patients, 43 (21.94%) were on oxygen therapy however none had an oxygen prescription that was compliant with ITS guidelines. 5 (2.55%) charts included a target saturation. 4 (2.04%) charts had a flow rate prescribed and 4 (2.04%) charts had a delivery system specified. The medication chart used in our centre lacks a review date in the oxygen prescription section, therefore all charts failed to meet this aspect of prescribing standards.
Our results highlight a significant absence of correct oxygen prescribing across all medical inpatients. A multifaceted intervention encompassing prescriber education, nursing checks on medication rounds and medication chart redesign is planned to improve compliance with ITS guidelines.
O’Driscoll, R. British Thoracic Society Emergency Oxygen Audit Report National Audit Period: 15 August – 1 November 2015. United Kingdom. 2016.
Irish Thoracic Society. Irish Guidelines on the Administration of Oxygen Therapy in the Acute Clinical Setting in Adults 2017. Ireland. 2017.
Conflict of Interest: None to declare
4.16 From Zero to Three Hundred—The Power of Interdisciplinary Collaboration
Rosaleen Reilly1 & Agnes Barry2
Meath Integrated Respiratory Service, Navan, Co Meath
The Meath Integrated Respiratory Service (MIRS) established in March 2021 aims to provide the right care in the right place at the right time. Patients previously travelled afar to access this care.
With Slaintecare funding an interdisciplinary team was formed (Respiratory RANP (RANP), Clinical Specialist Respiratory Physiotherapist and clerical officer). Clinical governance is provided by two Respiratory Consultants in Our Lady’s Hospital, Navan (OLHN).Care is based primarily in the home setting, encompassing disease education, self- management, therapy optimisation, exacerbation support and pulmonary rehabilitation. Our RANP clinic (commenced September 2021) receives referrals triaged from OLHN Respiratory OPD waiting list. 22 patients attended for initial assessment, diagnosis, treatment and education. 195 patients enrolled in the Pulmonary Outreach (PO) programme facilitating hospital discharge and exacerbation support. This resulted in 323 home visits and 174 episodes of support. 23 patients enrolled into Pulmonary Rehabilitation (PR), delivered in virtually and Face to Face formats. Pre assessment confirms disease and optimises therapy.
Within 12 months, MIRS has embedded an integrated service in Co Meath. 300 patients have accessed one or more elements of our service. Collaboration between members of MIRS, primary and secondary care teams has resulted in positive patient outcomes.
Conflict of Interest: None to declare
4.17 Has Covid made us “contactless”?
An Appraisal of the “Making Every Contact Count” Model of Intervention in Hospital Inpatients for Smoking Cessation
Susan Crotty, Maria Kurian, Chisanga Bwalya, Fionnbarr McDermott Long, Muhammad Saad Zaheer, Brian Canavan, Kenneth Bolger
St. Luke’s General Hospital Carlow/Kilkenny
The HSE framework “Making Every Contact Count” emphasises the role of behaviour change interventions for chronic disease prevention. Evidence shows that ongoing, episodic advice from healthcare workers is often effective in aiding smoking cessation. Pooled data from 17 trials on brief advice showed significant increase in rates of quitting. An audit was carried out to assess how often advice or information was given to patients regarding smoking cessation while an inpatient in SLGH, in accordance with HSE guidelines.Information gathered will be used to improve smoking cessation bundles in SLGH in conjunction with the health promotion officer.
Conflict of Interest: None to declare
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
5. TB, Pleural and Sleep Disorders
5.1 Audit of the Prevalence and Management of Positional Obstructive Sleep Apnoea in Beaumont Hospital
D.Barry1, C Campbell2, P.Ridge1 T.Kane1 I. Sulaiman1,2
Beaumont Hospital, Dublin1
RCSI, Dublin2
Positional Obstructive Sleep Apnoea (P-OSA), is defined as a 50% reduction in AHI when non-supine, and has been reported in 50-60% of OSA cases. Positional therapy can sometimes be an alternative to positive airways pressure (PAP), in those with mild OSA or intolerant of PAP. We sought to evaluate the prevalence and treatment of P-OSA in the Beaumont Hospital Sleep Service.
We reviewed of all sleep studies completed from 2019- 2022, looking for the mention of ‘positional’ in the reports. In addition we reviewed all the notes to identify the different treatments.
500 studies were carried out, and 23.4% (117) of those had P-OSA. Of the P-OSA cases identified the mean ± SD AHI was 12 ± 9; Supine 21.1%, 35.5%, 43.5% had mild, moderate, severe disease respectively. Non-Supine 21%, 2.5%, 0.8% had mild, moderate severe disease respectively. 56 (47.8%) patients were commenced on PAP, 5 (4.2%) referred for MAD, 27 (23%) were advised on positional therapy, 26(22%) are awaiting follow-up. 6 had repeat studies in position, which was successful in normalizing AHI for 4 (66%) patients.Positional OSA is common. Identification of P-OSA may provide alternative treatment modologies for patients, although a repeat sleep study is important to ensure efficacy.
Conflict of Interest: None to declare
5.2 Cardiovascular Risk Factors in Patients with Obstructive Sleep Apnea
Sarah Carolan1, Mark Sheehy2
Midland Regional Hospital, Mullingar, Co. Westmeath
Obstructive Sleep Apnea (OSA) is a common condition which can result in significant adverse effects with hypertension, myocardial infarction (MI) and stroke all potential complications1. This observational study analysed the relationship between cardiovascular risk factors and OSA and the occurrence of related events within this cohort of patients.A convenience sample of 31 patients with a recent diagnosis of OSA were assessed for risk factor profile and OSA severity. Data was collected using chart review and hospital IT system.29 of the 31 (93.5%) patients included had one or more risk factor for cardiovascular disease including obesity, hypertension, hypercholesterolemia, smoking, family history of cardiac disease and Diabetes Mellitus (Fig. 1). 45% had at least 3 risk factors. None of the patient’s reviewed had a documented history of MI and only one patient had a prior stroke. OSA severity did not appear to have a direct relationship with risk factor prevalence with the mean number of risk factors highest in those with mild OSA (2.52) and mean of 1.66 and 2 in the moderate and severe categories respectively.
Fig. 1 (5.2) Prevalence of Cardiovascular Risk Factors in patients with Obstructive Sleep Apnea
Diagnosing OSA is important due to its associations with cardiac disease and stroke. Although this study was small in size and most patients included had not yet been affected by MI or stroke, it depicted a high prevalence of risk factors which is important for monitoring and optimising to prevent future complications.
References
Slowik JM, Sankari A, Collen JF. Obstructive Sleep Apnea. [Updated 2022 Jun 28]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2022 Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK459252/#
Conflict of Interest: None to declare
5.3 Prediction of Obstructive Sleep Apnea Resolution Post Bariatric Surgery
Caoimhe Coghlan, Liam Doherty, Colm O’Boyle, Derbrenn O’Connor
Bon Secours Hospital, Cork
Obesity is the biggest risk factor for developing Obstructive Sleep Apnea [OSA]. Approximately 80% of bariatric surgery patients have confirmed OSA. When Body Mass Index [BMI] is reduced in these patients, a corresponding reduction in OSA is expected. Despite significant weight loss post-surgery, however, 30% of patients have unresolved OSA. It remains unclear which factors successfully predict resolution of OSA following bariatric surgery.A quantitative, retrospective analysis of 115 patients who underwent bariatric surgery and completed both pre-operative and post-operative sleep studies in the Bon Secours Hospital Cork between 2008 and 2019 was undertaken.
Candidate variables included: age, sex, pre-operative Apnea Hypopnea Index [AHI] pre-operative BMI, change in BMI, pre-operative Supine AHI, change in Supine AHI, pre-operative Epworth Score, pre-operative Oxygen Desaturation Index [ODI], change in ODI, neck circumference at baseline, and type of surgery.The initial median AHI was 42.6 (25.0–62.0) events per hour. After surgery, there was a significant change in median AHI at 11.5 (5.4–19.2) events per hour (p < 0.001). Baseline AHI was the only statistically significant predictive factor of post-operative OSA. The pre-operative parameters of high baseline AHI, and high pre-operative Supine AHI were significant predictors of persistent moderate to severe OSA.
Conflict of Interest: None to declare
5.4 What do the Patients Think?
The Results of a Patient Satisfaction Survey of the Sleep Service in Mallow General Hospital
Catherine Hanlon, Mary Osborne, Doireann O Donovan, Filipe Afonso, John Kiely
Mallow General Hospital, Co. Cork
In November 2021, Catherine Hanlon, Respiratory CNS completed an audit of the MGH Sleep service using a patient satisfaction sleep questionnaire. This was sent to 100 patients (70 male and 30 female)The inclusion criteria for the study being patients currently on CPAP who attended the sleep service within the last 5 years. The questionnaire contained ten questions varying from time on waiting list, information received in clinic to quality of service received from the CPAP supplier.
Seventy-five replies were received over a 3 week cut off period. Each response was analysed. This study has provided valuable feedback as to how the service is performing and how it can be developed in the future.One of the key takeaways from the responses was that patients clearly favoured telemedicine CNS follow up when established on CPAP therapy. This sleep questionnaire will be used as a valuable tool to analyse the service going forward.We are now using this questionnaire in our newly established Weekly compliance clinic when patients are seen at an initial 3 months and then again at 12 months on CPAP therapy.
Conflict of Interest: None to declare
5.5 Audit of Obesity Hypoventilation Syndrome Prevalence and Management in the Beaumont Sleep Service
Linda Keegan PA-IC1,Christinia Campbell1, Imran Sulaiman1
1 Beaumont Hospital. Royal College of Surgeons Ireland
Obesity Hypoventilation syndrome (OHS) is a combination of obesity, causing sleep disordered breathing, and daytime hypercapnia. In 2018 the prevalence of OHS in Ireland was 17.2%, however, if current obesity trends continue this Fig. is expected to rise. OHS is has strong links to higher rates of mortality and chronic diseases therefore, the American Thoracic Society (ATS) devised evidence based guidelines to address gaps in the diagnosis and management of OHS. The aim of this audit is to establish the prevalence of OHS in patients seen in Beaumont Hospital and compare their management to ATS guidelines. The diagnosis of patients seen in the sleep clinic, at Beaumont Hospital, between 2021-2022 was reviewed. Current treatments, ventilation settings, follow up and if applicable, their weight, PaCO2 and serum bicarbonate were documented. Of the 1464 patients reviewed, 426 patients were identified as having a sleep pathology. Three-hundred and ninety-four had OSA, 10 had OHS and 22 had OSA/OHS. Within the OHS and OHS/OSA group 41% were on CPAP therapy and 50% were on Bi-Level Ventilation. The OSA group are managed in sleep clinic and the OHS and OHS/OSA patients are managed in a ventilation clinic. The management of the OHS and OHS/OSA patients at this clinic was compared to the ATS guidelines.
Clinicians should have a high index of suspicion when assessing obese patients in the acute or out-patient setting. Ventilation clinics can aid in ensuring the most up to date guidelines are followed and patients are on the adequate treatment and settings for their condition.
Conflict of Interest: None to declare
5.6 DIAGNOSIS OF OBSTRUCTIVE SLEEP APNOEA USING HOME SLEEP TESTING COMPARED TO POLYSOMNOGRAPHY: A SERVICE EVALUATION
Sarah Waicus1, Edward Moloney123, Paul Byrne2, Jeff Murphy3, Orla Wynne3
1 Trinity College Dublin; 2 Peamount Healthcare, Dublin, 3 Tallaght University Hospital, Dublin
The COVID-19 pandemic has limited access to overnight hospital stay polysomnography (PSG) testing for patients with obstructive sleep apnoea (OSA). Home sleep apnoea testing (HSAT) and virtual-sleep services have been implemented as a response to provide diagnostic testing during the pandemic. The purpose of this study was to compare patient demographics, experiences and treatment adherence among those who underwent PSG or HSAT. A service evaluation of patient data (n = 100) was collected from two sites for those who underwent HSAT at Tallaght University hospital or PSG at Peamount Healthcare. Telephone consultations commenced after each patient underwent either PSG or HSAT to assess patient attitudes, treatment adherence, and sleepiness using likert scales. For demographic data, t-tests were used to compare continuous patient data and chi-squared tests for categorical data. There was no significant difference between age, BMI, sleep duration, sleepiness, apnoea-hypopnoea index scores, and patient preference between those who underwent PSG compared to HSAT (p > 0.05). There was a significant increase in witnessed apnoeas and supine position sleeping in those who underwent PSG testing (p < 0.05). The use of HSAT and virtual services is feasible and may result in greater patient accessibility to sleep apnoea testing for a timely diagnosis and better management.
Conflict of Interest: None to declare
5.7 A Retrospective Analysis on Increased Prevalence of Central Sleep Apnoea in Post Covid—19 Subjects
A.M. O’Connell, A.M. Curran, T. Quadri, Y. Vapra
Naas General Hospital
Aim: Prior to Covid -19 (C-19) when analysis of polysomnography studies is performed the number diagnosed with Central Sleep Apnoea has historically been very low. Since Covid-19 arrived in Ireland – early 2020, it was noted that an increasing number of patients have been diagnosed with Central Sleep Apnoea. The aim of this study was to analyse a set of polysomnography readings from 2019 (pre Covid-19) and a set of studies where patients had reported confirmed post Covid-19 infection, to ascertain any difference in Central Sleep Apnoea prevalence.
Method: A set of readings from 100 sequential patients attending for polysomnography studies during 2019 were compared to an equal sequential set of reported confirmed Covid-19 readings. An Apnoea/Hypopnea Index, (AHI) of > 5 was deemed as positive for Sleep Apnoea with a greater number of Central Apnoea’s deemed as Central Sleep Apnoea.
Results: Table 1 shows a comparison of analysed polysomnography data from 100 non C-19 subjects’ verses 100 post C-19 infected subjects, presented as total percentage change.
Table 1 (5.7) – Polysomnography Analysis Results of C-19 Negative v Post C-19 Subjects Pre/Post Infection (C-19) Negative Obstructive Mixed Central
Pre C-19 (2019) 6 88 3 3
Post C-19 Infection 4 67 3 26
Total % Change ↓ 33% ↓ 23% 0% ↑ 766%
Conclusion: These results show a marked increase in the prevalence of Central Sleep Apnoea in C-19 positive subjects, an increase of 766%.
Conflict of Interest: None to declare
5.8 Covid 18. Obstructive sleep apnoea as a manifestation of Long Covid Syndrome
Michelle Uno, Aisling O’Connor, Sarah Farrell, Tidi Hassan
Our Lady of Lourdes Hospital, Drogheda
With 6500 reported deaths as a result of SARS-CoV-2 infection in Ireland, the exact number of survivors affected with persistent symptoms remain unknown. In Our Lady of Lourdes Hospital Drogheda, we run a post-Covid respiratory clinic to follow patients with persistent respiratory symptoms. Over 350 patients have been reviewed and common symptoms include fatigue. An Epworth Sleepiness Score and STOP-BANG questionnaire were performed as part of the assessment. Twenty-seven patients were referred for a home sleep apnoea testing (HSAT) with a mean body mass index was 32 (± 7) kg/m2 and STOP-BANG of 5 (± 2). All patients had an abnormal HSAT (apnoea-hypopnoea index (AHI) > 5) with a mean AHI 36.8 (± 19). Eleven patients (27%) were categorised as severe range (AHI > 30). Ninety percent (n = 25) of patients were commenced CPAP and 82% (n = 21) felt symptomatically improved. Due to the scarcity in literature on Long Covid, we illustrate that obstructive sleep apnoea is a potential manifestation of Long Covid syndrome and intervention is effective as already established.
Conflict of Interest: None to declare
5.9 A study on awareness of pneumothorax management and chest drain apparatus care among NCHDs
Junaid Rasul Awan, Kaitlyn Cinnamond, Owais Rahman, Orlaith Shinners, Aidan O’Brien
University Hospital Limerick
Pneumothoraces are commonly encountered in clinical practice. Incorrect insertion and care of chest drains can lead to serious complications. The study aimed to assess awareness of management of pneumothorax in emergency and non-emergency settings and chest drain apparatus care among NCHDs.50 NCHDs,including 17 Interns, 16 SHOs and 17 Registrars, routinely partaking in acute medical emergencies, were invited to complete a basic respiratory related questionnaire including optimum thoracocentesis needle placement, safety triangle landmarking, and basic chest drain management. 61.2% of NCHDs had some form of exposure to clinical respiratory medicine; however, no statistically significant correlation was found between previous clinical respiratory exposure and correctly answering basic questions on needle decompression (p = 0.506) and safety triangle identification (p = 0.21). 56% incorrectly illustrated needle decompression landmarks, with registrars being least likely to answer correctly (29.4%). 58% participants were unable to recognise triangle of safety borders. 92% interns had little/no confidence in their chest drain knowledge. 54% NCHDs were unable to identify appropriate management of chest drain water seal lid.
These findings illustrate the importance of dedicated respiratory teaching sessions for NCHDS including practical/simulation workshops to maximise patient care; and the benefit of these can be studied by auditing participants post teaching.
Conflict of Interest: None to declare
5.10 Transitioning from Radiology led to Respiratory-led Pleural Service
Sarah Farrell1, Mohammed Yaseen1, Elaine Curran1, and Elaine Hayes1
1 Our Lady of Lourdes Hospital, Drogheda
Internationally the incidence of pleural disease is increasing. Pleural disease is accepted as a sub-specialty within respiratory medicine, with the recognition that subspecialty development advances standards of care.A retrospective analysis of ultrasound-guided pleural procedures was undertaken in OLOLH between January 2020 and August 2022 to determine the impact of transitioning from a predominantly Radiology led to a Respiratory led pleural service. There were 226 pleural procedures completed during this 30 month period in OLOLH. Only 14% (32) were performed by radiology vs 86% (194) by the respiratory service. The respiratory team diagnosed malignancy in 28% of cases, with a further 4% highly suspicious, thus allowing coordinated follow-up. An ambulatory model of care is an important component of the pleural service with 8 indwelling pleural catheters inserted for the management of chronic pleural effusions and 11% (21) of pleural interventions completed as a day procedure.The change to a predominantly respiratory led pleural service, has been progressed by the appointment of a dedicated pleural advanced nurse practitioner. Standardized pathways, patient information leaflets and consent forms have been developed as part of efforts to improve patient safety and experience. Expansion of the pleural team will allow further improvements in ambulatory provision.
Conflict of Interest: None to declare
5.11 Are Chest X-Rays Being Reviewed At An Appropriate Time Following Chest Drain Removal?- A Single Irish Centre Audit And Literature Review
Farah Kazi1, Wael Shabo1, Mohamad Salama1, Haresh Perthiani1, Sherif El Masry1, Eleanor Carton1
1 Our Lady of Lourdes Hospital
Insertion of chest drains is a commonly performed procedure for pneumothorax, haemothorax, pleural effusion and empyema. Daily chest x-rays (CXR) to confirm position, interval change and decide on removal are advised. Post-removal chest x-rays should be performed and reviewed in a timely manner to ensure early detection of complications or recurrence. Current care bundles cover documentation for insertion and management of drain but do not specifically include removal. Our aims are to assess time interval between a chest x-ray post chest drain removal being filmed and being reviewed at our institution and assess quality of documentation of chest drain care.This is a retrospective study of chest drains, between March 2020 and December 2021. Category Result
Total No. (n) 91
Duration of drain 4.2 days (1–24)
Average time for CXRs to happen 208 min (1–1314)
Average time to review CXRs
Not reviewed > 2 weeks
108 min (1–1077)
4
Table 1 (5.11) Showing results obtained
20 randomly selected charts – documentation reviewed. Chest x-rays are essential in assessing resolution of underlying condition. post- removal x-rays must be reviewed by a clinician in a timely manner to detect complications early. We recommend utilization of the chest drain form which includes post removal management as part of chest drain care bundle.
Conflict of interest: None to declare
5.12 Outcome and clinical characteristics in pleural empyema patients in an inner city hospital: a retrospective study
S Meagher 1 , D Curran 1 , TM O’Connor 1
1 Mercy University Hospital, Cork
A retrospective analysis of patients diagnosed with empyema evaluated patient demographics, co-morbidities, prognostic factors (predisposing diseases, early pleural drainage, intrapleural fibrinolysis, thoracic surgical treatment, nosocomial status) and outcome.Data from the medical records of patients with a diagnosis over a 12 month period were evaluated retrospectively. Empyema diagnosis was established by (a) positive pleural culture or (b) grossly purulent pleural fluid with low pleural pH.
13 patients were identified (n = 7 males, mean age 60, mean length of stay 26.8 days). Median time to drain was 1 day (range 1-5); 5 patients required more than one drain. 10 patients received an intrapleural fibrinolytic and Dornase alpha regimen for 72 h; no intrapleural haemorrhages occurred. Streptococcus species was isolated in fluid cultures of 5 patients. Mycobacterium tuberculosis was isolated in 2 patients. The majority of the patients had clinical and radiological resolution. 4 patients proceeded to transfer for VATS decortication in the local Cardiothoracics centre (2 to rule out malignancy, 2 for trapped lung). 2 patients were admitted to ICU. 2 patients died during their admission.Pleural empyema is associated with considerable morbidity and mortality. These patients have complex clinical courses and prolonged lengths of stay. Early identification and pleural drainage are vital to patient care.
Conflict of Interest: None to declare
5.13 Length of stay of patients with an indwelling chest drain
Marie Talty,Daniel J Ryan, Dorothy Ryan
Beaumont Hospital, Dublin
Insertion of a chest drain is a common procedure for respiratory inpatients, as well as a respiratory consult service. In our institution, lack of comfort and experience with chest drains on outlier wards had anecdotally led to issues with drain management.Through retrospective review of National Quality Assurance & Improvement System data, patients from 2019 & 2020 were grouped both by their ward location and primary team as well as indication for drain insertion. Seventy-three patients were included (mean age 57 and 55 male). The indication for drain insertion in 30 patients was pneumothorax, with remaining 44 pleural effusion. Seventeen patients had a secondary diagnosis of malignancy. 40 patients were managed on a core respiratory ward, with 10/33 patients on outlier wards being under the care of the respiratory team. Overall LOS was 12.1 days, 8.2 days in the respiratory ward group and 18.3 in the outlier group. Pleural effusion patients had a longer mean LOS of 15.1 days.Overall, patients requiring pleural intervention on non-specialist respiratory wards had a significantly longer length of stay in hospital, highlighting the need for cohorting and specialist nursing support for the safe and cost effective management of this complex patient group.
Conflict of Interest: None to declare
5.14 BIPAP Indications, compliance and usage – a retrospective audit in Beaumont Hospital
Jack Allen1, Sile Toland1, Peter Branagan1
Beaumont Hospital, Co. Dublin
The use of non-invasive ventilation (NIV) in an appropriate manner in acute respiratory failure is used to improve outcomes. BTS guidelines recommend that patients who meet criteria for acute NIV should start NIV within 60 min.We describe a retrospective audit of the indications, and adherence of BIPAP initiation in patients managed on the Respiratory Wards of Beaumont Hospital, during a 4-week calendar period from July – August 2022. Hospital guidelines recommend that those commenced on NIV, a prescription be completed along with the most recent ABG results recorded. It is recommended that for any adjustments to settings be recorded and updated in the BIPAP prescription.10 patients were chosen at random from the cohort commenced on BIPAP, in a 70% to 30% ratio of female to males. The median age was 63.5 years. 60% of those commenced had a previous diagnosis of underlying COPD, with 50% of that population group having been established on home NIV. 50% of those commenced on NIV were for infective exacerbations of COPD. Other outcomes analysed showed poor compliance with time to NIV commencement, response to NIV using arterial blood gases, and compliance with local hospital BIPAP documentation. Our plan is to expand our cohort of patients, with further audits in 6 months post education with the NCHD cohort.
Conflict of Interest: None to declare
5.15 Improving Non Invasive Ventilation Out of Hours
Junaidy Mohd Ishak1, Elaine Curran1, and Elaine Hayes1
1Our lady Of Lourdes Hospital, Drogheda
The purpose of this study was to evaluate current NIV prescribing and documentation practice within our hospital’.The establishment of a specialist respiratory unit has enabled the broader use of ventilatory support in the form of non-invasive ventilation (NIV), as a treatment modality for both Type 1 and Type 2 respiratory failure outside of the ICU/HDU setting. Often NIV is started out of hours and outside the respiratory unit. A retrospective chart review of 30 patients who had been commenced on NIV and subsequently transferred to Respiratory unit was undertaken.The study found the rationale for NIV therapy is often inaccurate resulting in patients being placed on NIV inappropriately, with therapy altered on presentation to the Respiratory unit. Findings also revealed that the initial prescription, subsequent changes, goals of therapy, inclusive of the escalation plan are often not clearly recorded. The findings highlight the potential for patients to receive ineffective ventilation leading to therapy failure or treatment escalation. Therefore, in order to provide safe, effective quality care, current NIV prescribing protocols and associated documentation will be renewed in accordance with best available evidence and practice standards.
Conflict of Interest: None to declare
5.16 Fallout from the War- Treating Multidrug Resistant TB among refugees in Ireland
E Mc Nally1, AM McLaughlin1,2, C Ottewill1
1. St James Hospital, Dublin, 2. Trinity College Dublin
This year, we report a sharp increase in cases of multidrug resistant TB (MDRTB). A number of these cases have been reported in Ukrainian refugees.The incidence of TB in Ukraine is currently 67.6 per 100,000. Ukraine has the fourth-highest TB incidence in the WHO European Region and the fifth-highest number of confirmed cases of extensively drug-resistant TB in the world. 29% percent of cases of TB in Ukraine are MDR or XDRTB, 8,000 cases per annum(1).The UN report that 6.3 million people have fled Ukraine since the start of the invasion(2), thus about 4,300 patients who have TB are among those refugees. 26% of Ukrainian patients with TB are HIV co-infected(1).We present 4 cases of MDRTB in Ukrainian refugees who have been treated at the National TB Centre, St James’s Hospital. Specifically, we present the diagnostic and treatment challenges in patients who are refugees and live in congregate settings. We note that working with Ukrainian doctors in the team has facilitated patient management and communication with TB centres in Ukraine. We discuss the challenges in screening of war refugees for TB, so that they may be diagnosed and treated earlier.
References
1. Operational considerations for the prevention and control of infectious diseases—Russia’s aggression towards Ukraine.:18.
2. Situation Ukraine Refugee Situation [Internet]. [cited 2022 Aug 13]. Available from: https://data.unhcr.org/en/situations/ukraine
Conflict of Interest: None to declare
5.17 An Analysis of The Cost of New Treatments for Multi-Drug Resistant Tuberculosis
Niamh O’Flaherty1, Ciara Ottewill1, Emma McNally1, Lorraine Dolan1, Aoife O’Reilly1, Anne-Marie McLaughlin1
1 St James’s Hospital, Dublin
The prevalence of Multi-Drug Resistant (MDR) TB in Ireland is increasing, with seven cases diagnosed to date in 2022. Its treatment has significant financial implications, often requiring inpatient isolation and prolonged treatment courses.This was a retrospective cohort study, analysing all patients who were treated for MDR TB in the National Tuberculosis Centre over a 15 year period. Medications prescribed to this cohort were reviewed with an aim to analyse the relative cost of newer TB regimens. A total of 42 patients were diagnosed with MDR TB in our centre between 2007 and 2022.
During this time there were several major shifts in its treatment from the use of intravenous aminoglycosides to the advent of Bedaquiline and all oral regimens, followed more recently by the six month BPaLM regimen. The costs of each Group A and B drugs with predicted cost for BPaLM vs WHO vs IV regimens are listed below. While the newer medications used are expensive, their price is offset by advantages including significantly shorter treatment duration and removal of costs associated with intravenous medications including hospital bed days, nursing care and OPAT services. This study investigates the cost-analysis of the newer regimens compared to previous regimens. Medication Cost (euro)
Bedaquiline 7,926 per month
Linezolid 692 per month
Pretomanid 278 per month
Moxifloxacin 68 per month
Cycloserine 574 per month
Clofazamine 192 per month
BPaLM Regimen (estimate) 68,090 (6 months total)
WHO Regimen Containing 3 × Group A and 2 × Group B drugs (Estimate) 174,109 (18 months total)
Older regimens including IV aminoglycosides
(Multiple medication options with courses up to 24 months)
Varied from 42,965 up to 97,433
(total cost)
Table 1 (5.17) Drug costs
Conflict of Interest: None to declare
5.18 Empyema thoracis in the Saolta University Health Care Group: has the Covid-19 pandemic led to earlier detection?
Gráinne Keehan1, Jack Whooley1, Alan Soo1
Galway University Hospital
Pleural empyema remains a significant healthcare burden associated with substantial morbidity and mortality. Early recognition and triage down appropriate pathways is central to patient outcomes in this complex disease process. Despite optimal medical therapy however, some patients may fail to improve and require surgical intervention.
The aim of this study was to review whether the distribution and determinants of pleural empyema requiring decortication have changed since the onset of the Covid 19 pandemic.
This single-centre retrospective observational study was conducted between June 2016 and June 2022. Patients that met the inclusion criteria (n = 46) where stratified into those that underwent decortication for empyema between June 2016-June 2019, and those that underwent decortication between July 2019-June 2022.
A reduction in the number of patients presenting with pleural empyema requiring decortication around or after the onset of Covid-19 was noted, compared to the previous 3-year period (n = 16 vs. n = 30). Mean-time from initial diagnosis to referral to a cardiothoracic centre was also reduced (16.31 vs. 21.68 days). In addition to this, a greater proportion of patients in the July 2019-June 2022 were eligible for a VATS decortication (18.75% vs. 3.33%)
This study suggests that empyema is being detected earlier since the onset of Covid-19.
Conflict of Interest: None to declare
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
6 Cystic Fibrosis
6.1 Retrospective review of the prevalence and change in organisms grown sputum culture in the cohort of CF patients on CFTR modulator Kaftrio
D. Barry1, L. Keegan1, M. Casey1, G. McElvaney1, 2
Beaumont Hospital, Dublin1
IRCSI, Dublin2
Lower respiratory tract infections are a challenge in the management of CF patients due to the colonisation of multiple resistant pathogen. Current evidence suggests in already established lung disease that the CFTR modulators are unable to eradicate organisms however may have an effect on the younger less severe population.
Our aim was to establish the prevalence of different organisms in the sputum cultures of CF patients in Beaumont Hospital and also to review the change in organisms after patients have commenced on Kaftrio.We reviewed microbiology results for all CF patients established on Kaftrio in Beaumont hospital before and after commencing the CFTR modulator. Data was collated on an excel spreadsheet.75 patients were prescribed Kaftrio, 58 patients were included in the study,67% were men. 19 organisms were identified,main organisms are outlined in the Table 1 below. In total there was an overall reduction in organisms grown including a reduction of 46.8% in Ps. Aeruginosa, 83% reduction in H.Influenzae, 51% of S.Aureus. 17.2% cultured no organisms post treatment.
We have shown a significant reduction in pathogen burden in Cf patients which may lead to less frequent respiratory tract infections however larger studies are needed to validate this data.
Table 1 (6.1) Organism %Before (n) %After (n)
Pseudomonas aeruginosa 81% (47) 41%(25)
Candida 93% (54) 60.3% (35)
Stenotrophomonas Maltophilia 44.8% (26) 6.8% (4)
H. Influenzae 51% (30) 8.6% (5)
Staph Aureus 84% (49) 41. 3% (24)
MRSA 13.7% (8) 12% (7)
Aspergillus 51% (30) 8.6%(5)
Conflict of Interest: None to declare
6.2 Steps Ahead: Optimising physical activity in adults with cystic fibrosis: A pilot randomised trial using wearable technology, goal setting and text message feedback
Máire Curran1 Audrey C. Tierney1,,7, Louise Collins1, Lauren Kennedy1, Ciara McDonnell1, Andrew J. Jurascheck1, Ali Sheikhi1, Cathal Walsh1, Brenda Button5,6, Brian Casserly1, Roisin Cahalan1
1University of Limerick, 5The Alfred, Melbourne,6Monash University, Melbourne,7La Trobe University, Melbourne
Regular participation in physical activity (PA) is encouraged for people with Cystic Fibrosis (CF). This study aimed to assess the effectiveness of an intervention using wearable technology, goal setting and text message feedback on PA and health outcomes in people with CF. This was a pilot randomised trial conducted at University Hospital Limerick. Participants were randomly assigned to the intervention (INT) or active comparator (AC). The 12-week intervention consisted of wearable technology (Fitbit Charge 2) which was remotely monitored, and participants set step count goals. Participants were sent a one-way text message once a week over 12 weeks to positively reinforce and encourage PA participation. The AC group received the wearable technology alone. Follow up was assessed at 24 weeks. Outcomes assessed were PA, aerobic capacity, lung function, sleep, quality of life and wellbeing.
Table 1 (6.2) Overall week-to-week step count percentage change for the intervention and active comparator groups Intervention Active Comparator p value*
Week to week % change: Weeks 1–12 + 27.8% -1% p = 0.023*
Week to week % change: Weeks 13–24 -2.2% + 6.5% p = 0.559
Week to week % change: Weeks 1–24 + 25.6% + 5.5% p = 0.007*
VO2 peak (ml/kg/min) significantly increased for the INT group only at 12 weeks (24.4 ± 7.65 to 26.13 ± 7.79, p = 0.003) but not at 24 weeks (24.45 ± 7.05, p = 0.776). There was no significant effect on lung function, sleep, well-being, or quality of life for either group.
A personalised PA intervention using wearable technology, goal setting and text message feedback increased PA and aerobic capacity in people with CF.
Conflict of Interest: None to declare
6.3 Increased usage, new indications and the emergence of single use flexible bronchoscopes in Adult Cystic Fibrosis. A review of practice at Cork University Hospital 2012-2022
Kevin F Deasy1,2, Hisham Ibrahim1,2, Mairead McCarthy1, James Dorgan1, Claire Fleming1, Ciara Howlett1, Yvonne McCarthy1, Sarah Twohig1 Paul O’Regan4, Laura Kirwan4, Marcus P Kennedy1,2 Michael T Henry1,2, Desmond M Murphy1,2, Barry J Plant1,2
1. Cork University Hospital, 2. University College Cork,4. Cystic Fibrosis Registry of Ireland, Dublin
Traditionally bronchoscopy has been used in paediatric patients-with-cystic-fibrosis (PWCF) for microbiological assessment, however in adult PWCFs this has not been necessary.
We performed a retrospective review of medical records between 2012-2022 of all adult PWCFs attending the service (n = 223) to identify those attending for bronchoscopy. Demographics, baseline metrics, indication and outcome were recorded.
69 bronchoscopies were performed, representing 56 unique patients, including 9 (16%) post-lung-transplant. At time of procedure, mean FEV1 (% predicted) increased from 60% (2012-2014) to 71% (2021-2022) and age from 30.3 to 35.8 years. There was an increase in bronchoscopies with a mean of 6 per-annum (2012-2019), compared with 13 in 2021, and 7 year-to-date 2022. Fig. 1 highlights the changing indications. Since 2019, bronchoscopy as a primary tool to assess airway culture has emerged, with all bronchoscopes changing to single-use-flexible-bronchoscopes (SUFB) since 2021. 78.9% (n = 15/19) of SUFB patients were culture positive compared with 83.7 (n = 41/49) of reusable bronchoscopes. 1 post-lung-transplant patient required admission due to an exacerbation within 4-weeks of bronchoscopy.Increased usage and changing indications for bronchoscopy is probably explained by a reduction in sputum production as a consequence of increasing CFTR modulation. The emergence of SUFBs allows analysis which avoids cross-contamination and sample contamination.
(6.3)
Fig. 1 (6.3) Indications for bronchoscopy in adult PWCFs 2012-2022
Conflict of Interest:None to declare
6.4 The Use of Combination Therapy Drug KAFTRIO in the Treatment of Cystic Fibrosis, and How Pulmonary Function Test Results Can Significantly Change Post Therapy: The Role of the Physiologist
Joy Earls1,2
Technological University Dublin 1 , St. Vincent’s University Hospital 2
Cystic fibrosis (CF) is a multi-organ progressive genetic disease. Approximately 70-80% of CF mutations are Class II, linked to a specific deletion of three base pairs of DNA at position 508 of the gene, known as F508del. KAFTRIO, a combination drug, is utilized in the treatment of these patients. It increases CFTR protein quantity and function at the cell surface and restores cellular transport of sodium and water, reducing mucosal production.
Pulmonary Function Tests are the cornerstone in the clinical management of CF determining the severity of lung disease, capturing clinical course, evaluating therapeutic efficacy, and monitoring disease progression. Forced expiratory volume in 1 s (FEV1) being the established marker of CF. Respiratory physiologists play a pivotal role in the care and management plan of a patient by ensuring validity, reliability, and repeatability of PFT’s. An emerging pattern of significant improvement was noted in PFT results of returning patients who had been prescribed KAFTRIO. A specific case study is presented in Table 1 showing a marked increase in PFTs.
Table 1 (6.4) – Pulmonary Function Test results, Weight and BMI pre- and post-KAFTRIO Date FVC FEV1 PEF Weight BMI
29/10/2020 (Pre Kaftrio) 53%
2.66L
32%
1.66L
53 55 kg 18.2
26/01/2022 (Post Kaftrio) 74%
3.65L
56%
2.37L
90 77 kg 25.4
Conflict of interest : none to declare
6.5 Significant real-world improvement in clinical, radiological and systemic inflammatory outcomes post CFTR modulation with elexacaftor/tezacaftor/ivacaftor in cystic fibrosis patients homozygous for the Phe508del mutation
Hisham Ibrahim1, Kevin Deasy1, Alexander T O’Mahony,1, Mairead McCarthy1, James Dorgan1, Claire Fleming1, Ciara Howlett1, Yvonne McCarthy1, Sarah Twohig1, Paul O’Regan6, Laura Kirwan6, Michael M Maher1, Owen J O’Connor1, Barry J Plant1
7. University Hospital, Cork, 6. Cystic Fibrosis Registry of Ireland
In clinical trial treatment with elexacaftor/tezacaftor/ivacaftor (ETI) in CF patients homozygous for Phe508del mutation was associated with significant improvement in FEV1, sweat chloride, weight and quality-of-life (CFQR-R).We assessed these outcomes in a real-world setting post ETI therapy in our clinic. In addition to clinical trial data, changes in ultra-low-dose CT imaging, peripheral-blood inflammatory-cytokines and patient-reported outcomes (PROMs) were measured.
The first 49 CF patients homozygous for Phe508del mutation attending our standard clinic commenced on ETI were assessed at baseline (time zero) and prospectively at three and 6 months. Table-1 summarizes the outcomes with significant improvements in FEV1, weight, BMI, and sweat chloride post-ETI treatment(P < 0.0001). Validated patient-reported outcome measures also improved significantly for chronic rhinosinusitis/SNOTT(P = 0.0064), respiratory quality-of-life CFQR-R(P < 0.0001), and fatigue score(P = 0.0049). Ultra-low-dose CT imaging scores demonstrated reductions in peri-bronchial thickening, mucus plugging(P < 0.0001), collapse/consolidation(P = 0.0425), and improvements in total Bhalla score (P < 0.0001). Significant changes in the systemic inflammatory status of our cohort was seen with a reduction in interleukin (IL)-6 and IL-8(P < 0.0001), along with increasing IL-10(P = 0.0004).Based on clinical trial parameters, ETI responders, in addition demonstrate significant improvements in CT imaging, circulating cytokines and PROMs which may be of further use to evaluate treatment response in an era of evolving CFTR modulation.
Table 2 (6.5) Changes in Clinical parameters, patients-reported outcomes measures, radiological and systemic inflammatory parameters at baseline and 6 months post initiation of Elexacaftor/Tezacaftor/Ivacaftor (ETI) therapy Baseline 6 months post ETI therapy Delta (Δ)
Variable N Mean SD N Mean SD p-value
FEV1%pred 49 65.5 18.6 43 75.5 19.9 < 0.0001
FEVC %pred 48 83.9 15.8 43 92.1 15.0 < 0.0001
Weight (Kg) 49 67.4 12.4 43 71.7 12.9 < 0.0001
BMI 48 23.2 2.89 42 24.5 3.54 0.000147
CFQR-R 47 70.7 17 37 89 12.8 < 0.0001
Sweat Cl (mmol/L) 47 79.7 15.7 41 40.4 16.3 < 0.0001
Fatigue 48 41.5 9.21 36 44.9 8.73 0.004925
SNOTT 47 7.79 6.79 37 4.7 5.63 0.006413
IL8 (pg/ml) 49 8.03 6.65 40 4.04 2.60 < 0.0001
IL6 (pg/ml) 49 1.48 1.17 40 0.814 0.421 < 0.0001
IL10 (pg/ml) 49 0.910 1.48 29 4.68 16.1 0.000451
Total Bhala Score 22 16.1 4.06 22 13.8 4.11 < 0.0001
Conflict of Interest: None to declare
6.6 Effect of Elexacaftor-Tezaftor-Ivacaftor on a West of Ireland cohort of patients with Cystic Fibrosis (CF) and comparison of “smart” monitoring of patients compared to traditional lab and clinic based measurements
P. Ridge; S. Bradish; I. Maguire; C. Edwards; R. Cusack; M. O’Mahony
Galway University Hospital, Galway
CF is a rare autosomal recessive multisystem disorder secondary to cystic fibrosis transmembrane protein (CFTR) gene dysfunction. ETI has been shown to improve a range of outcomes in CF patients internationally. We performed a prospective, non-randomized, non-controlled study to assess if ETI had similar effect in our real life West of Ireland cohort over a 9 month period. Data was collected in the traditional method but also remotely using a novel monitoring mobile phone app in conjunction with smart hand-held spirometers and weighing scales. We aimed to assess the accuracy of this form of remote monitoring compared to traditional laboratory monitoring.18 patients commencing ETI therapy were recruited to the study. Characteristics are outlined in Table 1. Pre commencing ETI the mean predicted FEV1 in our cohort was 59.6% (± 18.8). This increased to 73.7% (± 23.2) (p = < 0.0001) with therapy. ETI therapy also resulted in significant improvements in body mass index, sputum volume and in quality of life and symptom scores (Table 1). Total combined days spent in hospital among the cohort fell from 434 days the year prior to starting ETI treatment to 39 days the following year. Our study confirmed the hand-held spirometer correlated strongly with those measured in the pulmonary function lab.
Conflict of Interest: None to declare
Table 1 (6.6) Demographics and clinical characteristics pre-and-post treatment
Characteristics Start of study End of Study
Male sex: no. (%) 14 (78%)
Age
Mean: year (SD) 28.7 (± 8)
Genotype
Homozygous Phe508del- no. (%) 15 (83%)
Heterozygote Phe508del- no (%) 3 (17%)
Clinical characteristics
Height- cm (SD) 170.6 (± 6.4)
Weight- Kg (SD) 67.8 ± 12.3) 72.8 (± 11.8)
Body Mass Index- Kg/m2 (SD) 23.3 (± 3.9) 25.1 (± 3.9)
Percent Predicted FEV1- % (SD) 59.6% (± 18.8 73.7% (± 23.2)
Sweat Chloride- mmol/L (IQR) 84 (66–101) 46 (32–58)
Sputum volume: ml/day (IQR) 28.8 (12–37.5) 1 (0–10)
CFRSD- no. (IQR 34 (23–41) 14 (0–23)
CFQ-R- no. (IQR) 78 (66–83) 94(92.3–100)
6.7 An audit into the optimization of bone health in the paediatric CF population
Róisín Taplin1, Muireann Ní Chróinín2
Cork University Hospital
This audit looks as how the treatment provided in CUH compares to the European guidelines as well as the prevalence of osteoporosis and osteopenia. Cystic fibrosis (CF) is one of the most common life-limiting autosomal recessive disorders, with Ireland having a high incidence of CF. (1)
This study is a cross-sectional prospective chart-review. The European CF bone mineralisation guidelines were used as a benchmark. Participants included 98 children with CF between 6 months—16 years.
The lowest compliance was with optimal BMI (69.4% BMI > 50th centile) and vitamin K supplements (91%). There was full compliance in monitored calcium intake and exercise regime. No patients required bisphosphonates or long-term steroid use. No patients reported a history of fragility fractures. The average vitamin K was 3.5 ng/ml. The average vitamin D was 77 nmol/L. The average DEXA score was 1.3 and the average age at the first DEXA scan was 10 years 3 months. The prevalence of osteopenia was 2% and no incidence of osteoporosis.
This shows that the guideline’s recommendations have been implemented in CUH leading to better bone density outcomes for these patients, and a better understanding of screening practices and allocation of resources.
References
HSE national clinical programme; Cystic fibrosis 2015 Cystic Fibrosis—HSE.ie accessed 12/06/2022
Conflict of Interest: None to declare
6.8 The CFTR modulator combination elexacaftor/tezacaftor/ivacaftor restores CFTR protein expression in circulating neutrophils of patients with cystic fibrosis
Azeez Yusuf1, Michelle Casey1,2, Debananda Gogoi1, Claudie Gabillard-Lefort1, Cedric Gunaratnam2, Noel G. McElvaney1,2 & Emer P. Reeves1
1 Royal College of Surgeons in Ireland, Dublin
2 Beaumont Hospital, Dublin, Ireland
Cystic Fibrosis (CF) is caused by mutations in the CF transmembrane conductance regulator (CFTR) gene and is characterized by sustained inflammation. Studies have demonstrated that neutrophils account for ~ 70% of the total cell count in CF bronchial lavage fluid. Neutrophil dysfunction in people with CF (PWCF) is due to lack of CFTR function and chronic inflammation, causing altered cell chemotaxis, oxidant production, degranulation and apoptosis. CFTR modulator therapy (elexacaftor/tezacaftor/ivacaftor (ETI)), improves FEV1 in PWCF. The aim of this study was to explore the impact of ETI on neutrophil CFTR protein expression. PWCF eligible for ETI-therapy (n = 16, mean FEV = 81%) were recruited, non-paired samples collected pre- (n = 9) and 12 months post-ETI treatment (n = 7). Neutrophil whole cell lysates were analysed for CFTR and NHERF1 expression, as NHERF1 mediates CFTR translocation to plasma membranes. Cytosolic [Ca2+] and calpain activity were assessed fluorometrically. Markers of neutrophil adhesion (CD16b) and degranulation (CD66b) were assessed by flow cytometry.Compared to healthy control neutrophils, CFTR expression in CF neutrophils was significantly decreased (p = 0.04). Significantly increased cytosolic activity levels of the Ca2+-dependent cysteine protease calpain were detected in circulating CF neutrophils (p < 0.0001), which lead to significantly decreased NHERF1 expression (p = 0.005). ETI significantly decreased cytosolic Ca2+ levels (p < 0.0001) and calpain activity (p = 0004), with a corresponding increase in NHERF1 expression (p = 0.001). Of importance, ETI increased neutrophil CFTR protein expression compared to the pre-therapy group (p = 0.0029), and reduced the percentage of CD16b+/CD66b+ cells (p = 0.018).
Our results conclude that ETI significantly increases CFTR expression in circulating blood neutrophils of PWCF and decreases markers of neutrophil activation. This is of particular interest as not all patients are currently eligible for ETI or other CFTR modulator therapy.
Conflict of Interest: None to declare
6.9 Investigating peptidyl-arginine deiminases as novel antimicrobial agents against bacterial pathogens of patients with cystic fibrosis
Rory D. Baird, Debananda Gogoi, Azeez Yusuf, Michelle Casey, Grace Murray, Mark Murphy and Emer P. Reeves
RCSI University of Medicine and Health Sciences, Beaumont Hospital, Dublin
Persistent microbial colonisation of the airways is common in patients with cystic fibrosis, causing chronic inflammation and pronounced lung function decline. Successful eradication of pathogens by antibiotic therapy is rare, owing to the characteristics and resistance of certain microorganisms. Peptidyl-arginine deiminases (PADs) mediate calcium dependent, irreversible conversion of peptidyl arginine residues to citrulline. In the current study, our aim was to explore the potential of PADs to kill the Gram negative bacterium most common in CF, Pseudomonas aeruginosa.
Neutrophils were isolated from healthy control subjects. Following N2 cavitation and subcellular fractionation by sucrose gradient ultra-centrifugation, PAD2 and PAD4 were localised in neutrophil fractions by western blot analysis. For PAD2, PAD4 or bactericidal permeability-increasing protein (BPI; a positive control with potent activity against Gram-negative bacteria) killing assays, P. aeruginosa (PA01) was exposed to increasing concentrations of antimicrobial protein and colony-forming units were enumerated.Ex vivo, PAD2 and PAD4 were localised to neutrophil primary granules. Localised to this cellular compartment supports their participate in microbial killing. In vitro, PAD2 and PAD4 used in combination, dose-dependently killed up to 74 ± 19% of P. aeruginosa after 1 h incubation (p < 0.001). Moreover, post 1 h exposure to 20 nM enzyme, PAD4 killed significantly more bacteria compared to PAD2 (p = 0.024) or BPI (p = 0.035). Results demonstrate PAD2 and PAD4 participate in microbial killing. Overall, this study supports the rational design of novel PAD-based antimicrobial therapeutics.
This study was supported by Pfizer Healthcare Ireland (Educational Grant, 2021).
Conflict of Interest: None to declare
6.10 Decreased anti-neutrophil cytoplasmic autoantibodies (ANCA) against bactericidal/permeability-increasing (BPI) post CFTR modulator therapy
Debananda Gogoi 1, Michelle Casey 1,2, Azeez Yusuf 1, Claudie Gabillard-Lefort 1, Cedric Gunaratnam 2, Noel G. McElvaney 1,2 & Emer P. Reeves1
1 Royal College of Surgeons in Ireland, Dublin, Ireland
2 Beaumont Hospital, Dublin, Ireland
Pseudomonas aeruginosa (P. aeruginosa) is the dominant lung bacteria in patients with cystic fibrosis (PWCF), chronically infecting up to 75% of the adult CF population. Published reports have shown that BPI-ANCA correlate better with lung function impairment and long-time prognosis than anti-P. aeruginosa serology, and has similar ability to identify patients with chronic P. aeruginosa. Therapeutic interventions specifically targeting defective CFTR protein have improved the outlook for PWCF. Of importance however, there is a gap in our knowledge, and whether the titre of BPI-ANCA declines post CFTR modulator therapy is unknown. Accordingly, the aim of this study was to assess the impact of modulator therapy, elexacaftor/tezacaftor/ivacaftor (ETI), on titres of BPI-ANCA.Plasma samples were collected from patients receiving ETI, post 6 (n = 18) or 12 months (n = 12) treatment, and healthy controls (n = 3). Anti BPI-IgG autoantibodies were measured by ELISA. Results demonstrate that plasma levels of anti BPI-IgG autoantibodies post ETI therapy were significantly reduced at 6 months (p = 0.0001) and 12 months (p < 0.0001), as compared to pre- modulator therapy samples. The association between P. aeruginosa colonization and anti-BPI antibodies is complex, and our findings are the first to demonstrate reduced BPI-IgG autoantibodies post CFTR modulator therapy. Further work is underway to determine whether anti-BPI-IgA and anti-pseudomonas serology similarly decrease post CFTR modulator therapy.
Funding: Pfizer Healthcare Ireland, Educational Grant, 2022.
Conflict of Interest: None to declare
6.11 Closed Loop Audit – Cystic Fibrosis Related Diabetes
Ruaidhri Keane1, Ruth Hannon1, Noel G. McElvaney1, Cedric Gunaratnam1
1Beaumont Hospital, Dublin, Ireland
Cystic fibrosis-related diabetes (CFRD) is an extrapulmonary complication of cystic fibrosis (CF). It is associated with increased morbidity and mortality. The American Diabetes Association (ADA) published guidelines for the management of CFRD in 2010.The aim of this single centre closed loop audit was to re-evaluate the adherence to ADA guidelines following an audit in 2016.
120 patients with CF were reviewed over a 5 year period, 2017 to 2022. 25.8% had CFRD, 93% had a confirmed diagnosis, 84% were on insulin therapy, 35% were reviewed quarterly, 42% had a HbA1c < 7%, compared to 32.4%, 49%, 74%, < 5%, and 43% respectively in 2016. This population had a median FEV1 of 71%, BMI of 22.6, and exacerbation rate of 0, compared to 56%, 21.8, and 2.6 respectively. 16% of the non-CFRD population had an annual OGTT compared to 46% previously.
Adherence to ADA guidelines is greater in almost every parameter. The clinical improvement in the CFRD population in terms of FEV1, BMI, and exacerbation rate is likely influenced by guideline adherence and the introduction of triple combination therapy. SARS-CoV2 pandemic and the Health Service Executive ransomware attack are unique to this re-audit and likely account for the reduction in CFRD screening.
Table 1 (6.11) Clinical characteristics CFRD n = 35 [2016] CFRD n = 31 [2022] Non-CFRD n = 73 [2016] Non-CFRD n = 89 [2022]
Age (median, range) 26 (19 – 47) 35 (19 – 55) 27 (17 – 54) 30 (17 – 72)
Male (n, %) 19 (54%) 19 (61%) 43 (58%) 43 (48%)
Pancreatic insufficiency (n,%) 33 (94%) 30 (97%) 57 (78%) 71 (80%)
Osteoporosis (n, %) 10 (28%) 7 (23%) 18 (24%) 12 (13%)
Indwelling Venous Access Device (n, %) 24 (68%) 17 (55%) 51 (69%) 37 (42%)
Pseudomonas Colonisation (n, %) 23 (65%) 22 (71%) 47 (64%) 44 (49%)
FEV1% (median, range) 56.5 (22 – 100) 71 (26 – 113) 61 (15 – 122) 84 (22 – 130)
BMI kg/m2 (median, range) 21.7
(16.7 – 30.5)
22.6
(18 – 30.2)
21.7
(16.3 – 40.1)
24
(17.7 – 47.4)
Exacerbation rate (median, range) 2.6 (0 – 13) 0 (0 – 4) 1.8 (0 – 10) 1 (0 – 10)
Conflict of Interest: None to declare
6.12 A Tool for the Assessment of Adherence to International Guidelines in the Routine Monitoring of Paediatric Patients with Cystic Fibrosis
Gráinne O’Mahoney1 and Muireann Ní Chróinín 1
Cork University Hospital, Cork, Ireland
Cystic Fibrosis (CF) requires extensive routine monitoring, further complicated in paediatric populations by differing age-related requirements. The purpose of this study was to assimilate necessary investigations and clinical assessments into a single score. A scoring tool was devised following review of National Institute for Health and Care Excellence, European Cystic Fibrosis Society and Cystic Fibrosis Foundation guidelines. (Table 1) Comparison between the years 2019 and 2020, in the context of Covid-19 restrictions, was used to assess the ability of the scoring tool to reflect changes in clinical practice. A retrospective chart review was conducted of paediatric patients with CF attending Cork University Hospital from 2019-2020. Analysis was conducted in SPSS v28 using Chi Square and Mann Whitney U Tests. The proportion of patients with a score of 100% was 20.21% in 2019 and 2.27% in 2020, a decrease of 89%. (p < 0.005) The median score was 71.43% (IQR 54.17%-87.5%) in 2019 and 42.86% (IQR 25.0%-56.75%) in 2020, a decrease of 40%. (p < 0.005)
This tool illustrates an ability to concisely summarise overall monitoring and quantify the impact of alterations in service provision. Future use would allow for comparison between centres and assessment of quality improvement initiatives.
Table 1 (6.12) – Scoring Tool
Score = (Total Conducted / Total Applicable) * 100
Conducted + 1 if: Applicable + 1 if:
Sputum Cultures Four or more sputum cultures All patients
CXR CXR within the past 12 months All patients
Growth Parameters Four or more weight measurements and four or more height measurements All patients
Fat-Soluble Vitamins Completed measurement of Vitamin A, D, E and K All patients
Liver Bloods Completed measurement of AST, ALT and GGT All patients
Miscellaneous Bloods Completed measurement of WCC, Aspergillus serology and Serum IgE All patients
Faecal Elastase Faecal elastase measured No diagnosis of pancreatic insufficiency
Spirometry Four or more FEV1 measurements and four or more FVC measurements Aged 6 or above
Diabetes Screening OGTT conducted Aged 11 or above and no diagnosis of diabetes
Conflict of Interest: None to declare
6.13 An Audit of screening of aetiological tests in adults with diagnosis of non-CF bronchiectasis
MK Rana, S Lane
Peamount healthcare Dublin
Bronchiectasis is a disorder characterised by abnormal dilation of bronchi and a clinical syndrome of breathlessness, cough, sputum production and recurrent respiratory tract infections.HRCT thorax is the most appropriate diagnostic test when bronchiectasis is suspected.The critical first step following diagnosis of bronchiectasis is to determine the underlying cause. The European respiratory society guidelines recommend minimum bundle of aetiological tests with a new diagnosis of bronchiectasis that includes, differential blood count, serum immunoglobulins(total IgG,IgA and IgM), test for allergic bronchopulmonary aspergillosis(total IgE,specific IgE and IgG to aspergillus) and test for non-tuberculous mycobacteria (NTM) by sending sputum for AFB.Other tests are guided by history,clinical presentation and physical exam.The aim of our audit to see if the patients who had confirmed diagnosis of bronchiectasis had screening with minimal bundle of aetiological tests according to ERS bronchiectasis guidelines.A total of ten patients with known diagnosis of bronchiectasis were admitted in Peamount respiratory unit for pulmonary rehab in may 2022.All patients had confirmed diagnosis of bronchiectasis on Ct thorax. We retrospectively reviewed in our lab data if these patients had screening with minimal bundle of aetiological tests.The results showed that all 10 patients had test for differential blood count, serum immunoglobulins and sputum for culture and sensitivity. Out of 10 only 6 patients had test for allergic bronchopulmonary aspergillosis and 7 patients had sputum test for NTM.This audit showed that there was a lack in screening test for aspergillosis and non-tuberculous mycobacteria.We recommend that all patients with new diagnosis of bronchiectasis should be screened with minimal bundle of aetiological tests according to ERS guidelines and other tests should be performed based on the clinical presentation. These tests can considerably alter the clinical management of bronchiectasis.
Reference
European Respiratory Journal 2017 50: 1700629; 10.1183/13993003.00629-2017
Conflict of Interest: None to declare
6.14 Nebulized bone marrow derived extracellular vesicles ameliorate bacterial induced pneumonia in vivo
Hector Gonzalez 1 , Claire Masterson 1 , Sean McCarth1, Declan Byrnes 1 ,, Ronan MacLoughlin 3 , John Laffey 2 , Daniel O’Toole 1
1 NUI Galway, Ireland, 3 Aerogen Ltd., Galway
Razionale: Acute respiratory distress syndrome (ARDS) is an inflammatory disease with high mortality and a lack of specific therapy. Mesenchymal stromal cell (MSC) therapy is emerging as a potential therapeutic in a wide range of conditions, including pneumonia. We wished to utilise bone marrow extracellular vesicles (BM-EVs) in a rodent pneumonia model to explore clinically relevant administration using direct lung nebulization.Methods: BM-EVs were delivered intravenously or nebulized into the lungs using a vibrating mesh nebulizer to rats who had undergone bacterial lung installation, with administration delayed by 1 h. 48 h later, animals were assessed for lung physiological, inflammatory and infection parameters.Results: Nebulized BM-EVs delivered 1 h after bacterial installation increased arterial partial pressure of oxygen and reduced lung bacterial load. Proinflammatory cytokines such as IL-1β and IL-6 were ameliorated in bronchoalveolar lavage (BAL).Conclusion: BM-EVs therapy is a promising ARDS therapeutics, and their direct nebulization offers a novel delivery route.Fig. BM-EVs nebulized after bacterial pneumonia induction reduced BAL IL-6 concentrations compared with vehicle and was not statistically different from the IV group. († p < 0.05 wrt Sham; * p < 0.05 wrt Vehicle. NS no significant difference between delivery route. (Sham N=4. Vehicle, BM-EVs NEB, BM-EVs IT: N = 8)
(6.14)
Conflict of Interest: None to declare
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
7. COPD
7.1 A case series investigating a rare phenotype FZ Alpha-1 antitrypsin deficiency in Ireland
Lameese Alhaddah, Daniel Fraughen, Ronan Heeney, Tomás P. Carroll, Mark Murphy, Cedric, Gunaratnam, Gerry McElvaney
1Beaumont Hospital, Dublin 9. 2 University of Pavia, Italy
Alpha-1 antitrypsin deficiency (AATD) presents as a genetic disorder arising from a mutation in the SERPINA1 gene found on the long arm of chromosome 14 at 14q3- 32.1. The most common form of severe AATD is the Z homozygous form. The most common manifestation of severe AATD (ZZ) is emphysema which can occur even in the absence of a smoking history but which is markedly exacerbated by cigarette smoking, cirrhosis, and panniculitis. There are over 200 SERPINA1 mutations in existence. Roughly 5000 people have been diagnosed with moderate AATD (MZ & SZ) and 450 people with severe AATD.Interestingly, there have been 13 patients found with the rare FZ genotype identified through the National Targeted Detection Programme for AATD. The F mutation (Arg223Cys) causes a qualitative deficiency and is not associated with serum AAT deficiency, making it difficult to diagnose. This case series characterises patients with FZ genotype by reporting AAT levels, pulmonary function tests, lung imaging and liver testing. We also discuss clinical progression of these patients and compare them to the more common deficiencies.
Conflict of Interest: None to declare
7.2 A multi-site audit of alpha one deficiency testing in Ireland
Lameese Alhaddah, Ronan Heeney, Tomás P. Carroll, Cedric, Gunaratnam, Gerry McElvaney
1RCSI Beaumont Hospital, Dublin, Ireland, 3Beaumont Hospital, Dublin, Ireland, 4Centre for Diagnosis of Inherited Alpha-1 Antitrypsin Deficiency
Alpha-1 antitrypsin deficiency (AATD) is a well known genetic condition caused by SERPINA1 gene leading to an array of clinical conditions including chronic obstructive pulmonary disease (COPD) and liver disease. In Ireland, there has been roughly 22,000 individuals screened for AATD since 2004 following ATS/ERS guidelines as part of the HSE funded national targeted detection programme. Commonly quantification of alpha-1 antitrypsin levels (AAT) is done by turbidimetry, and AAT phenotyping to confirm AATD is performed with isoelectric focusing. It is recommended by HSE National Laboratory Handbook guidelines to perform AAT phenotyping to investigate AATD when AAT levels are lower than < 1.0 g/L. We performed a retrospective multi-site audit of AAT levels performed in Beaumont Hospital in the last 5 years to evaluate phenotype requesting patterns and diagnosis rates in our respiratory service. Results show that targeted AAT testing in respiratory cohorts allied to rapid phenotyping of low AAT levels are effective in tackling the continuing under-diagnosis of AATD.
Conflict of Interest: None to declare
7.3 Establishing diurnal variation of alpha-1 antitrypsin and acute phase proteins in healthy individuals and alpha-1 antitrypsin patients
Lameese Alhaddah, Ronan Heeney, Tomás P. Carroll, Mark Murphy, Gerry McElvaney, Cedric, Gunaratnam
1RCSI Beaumont Hospital, Dublin, 3Beaumont Hospital, Dublin 4Centre for Diagnosis of Inherited Alpha-1 Antitrypsin Deficiency
Alpha-1 antitrypsin deficiency (AATD) is a genetic disorder arising from a mutation in the SERPINA1 gene. The most common severe AATD is the Z homozygous form. Manifestation of this deficiency can lead to emphysema, cirrhosis and panniculitis. Alpha-1 Antitrypsin (AAT) is a serine proteinase inhibitor and acute phase protein with significant immunomodulatory effects on cytokines and chemokines.The circadian rhythm is a 24 h cycle of daily oscillations in physiology which is governed by the suprachiasmatic nucleus (SCN) in the brain. However, all cells in the body possess a molecular timer. This has a profound impact on all aspects of physiology and disease. We aim to describe the diurnal variation of AAT in healthy individuals and AATD patients to establish baseline Fig.s to further understand its role in disease, as this has not been done previously.We investigated the diurnal variation of acute phase proteins by measuring C-Reactive Protein (CRP), iL-6 and AAT levels in healthy individuals at three different time points over one day. We also performed the same methodology of testing with stable AATD patients to establish a comparative results to further characterise AAT levels in deficient patients.
Conflict of Interest: None to declare
7.4 The development and implementation of a MDT-focussed clinic for COPD
Jack Allen1, Finbarr Harnedy1, Michele Cuddihy1, Eimear Ward2, Ciara Feeney2, Maeve Sorohan2, Imran Sulaiman1, Breda Cushen1
1Beaumont Hospital, Dublin, 2Beaumont Hub, Dublin North City and County CHO9
COPD is a heterogenous disease. We introduced a systematic assessment outpatient clinic(CARA) for COPD patients with high symptom burden and healthcare use. Patients undergo multidisciplinary review with relevant laboratory, radiology and oxygen assessments, Fig. 1. The clinical characteristics and necessary CARA clinic interventions of the first 34 patients are described.The mean age was 69(7.8) years, 56% female, and BMI 28.5 (9.4)kg/m2. All were smokers(38.2% current). The mean FEV1 was 47.5(19.6)%. One-third were prescribed oxygen therapy (LTOT and/or ambulatory) and 27% home NIV. Eosinophil count was ≥ 300cells/m3 in 57.6%. Mean hospitalisation rate pre-CARA was 1.85(1.81)/per year. Co-morbidity was prevalent -asthma(26.5%), cardiovascular disease(53%), bronchiectasis(17.7%), GORD(8.8%), OSA(8.8%), pulmonary hypertension(11.8%), rhinosinusitis(23.5%) and polypharmacy(97%). Maintenance inhaled corticosteroids(ICS) were prescribed for 78% pre-CARA and 81% post. Half of those prescribed high dose ICS(31%) had ICS dose reduction at the clinic; 40% required a change of inhaler device. Four patients commenced oxygen therapy and two home NIV. One third required airway clearance education and 72% pulmonary rehabilitation referral. Additional diagnoses made at CARA included impaired immune response(n = 4), pseudomonas colonisation(n = 2), patent foramen ovale(n = 1), and diabetes mellitus(n = 1).
A multidisciplinary outpatient COPD clinic facilitates a personalised approach to the assessment and treatment of patients with this complex condition.
Fig. 1(7.4) COPD At-Risk Assessment (CARA) Clinic Outline
Conflict of Interest: None to declare
7.5 To evaluate prescribing practices of corticosteroids in patients admitted to hospital with acute exacerbations of chronic obstructive pulmonary disease (COPD)
CM Gill1, O O’Carroll1,, O Collins1, E McGrath1, E Kelly1
1 St Vincent’s University Hospital, Dublin
It has been shown that short courses of corticosteroids (5 days) in the treatment of acute exacerbations of COPD (AECOPD) are associated with improved treatment outcomes and increased time to re-exacerbation rate, when compared with conventional longer tapering regimes1,2. We have anecdotal evidence that patients admitted to hospital with AECOPD frequently receive more than 5 days of steroid therapy.
We retrospectively analysed all patients admitted to our centre over a two month period in 2022. We used HIPE coding system to identify all patients with AECOPD listed as the primary diagnosis. We used electronic and paper records to collect data.
Median age of patients admitted (n = 38) was 63.5 yrs, with a female majority (n = 21). Thirty four percent of patients received steroid therapy for more than 5 days, of which 63% were male. The presence of pulmonary infiltrate, requirement of supplemental oxygen or non-invasive ventilation were not associated with increased likelihood of longer steroid courses. Patients not under the care of a respiratory team were more likely to receive longer regimens of steroid therapy (57% vs 33%).
This audit highlights that patients may be inappropriately receiving longer courses of steroid. Further evaluation of influencing factors is important to avoid adverse effects.
References
Leuppi JD, Schuetz P, Bingisser R, et al. Short-term vs Conventional Glucocorticoid Therapy in Acute Exacerbations of Chronic Obstructive Pulmonary Disease: The REDUCE Randomized Clinical Trial. JAMA.2013;309(21):2223–2231.
Sivapalan P, Lapperre TS, Janner J, Laub RR, Moberg M, Bech CS, Eklöf J, Holm FS, Armbruster K, Sivapalan P, Mosbech C, Ali AKM, Seersholm N, Wilcke JT, Brøndum E, Sonne TP, Rønholt F, Andreassen HF, Ulrik CS, Vestbo J, Jensen JS. Eosinophil-guided corticosteroid therapy in patients admitted to hospital with COPD exacerbation (CORTICO-COP): a multicentre, randomised, controlled, open-label, non-inferiority trial. Lancet Respir Med. 2019 Aug;7(8):699–709. 10.1016/S2213-2600(19)30176-6. Epub 2019 May 20. PMID: 31,122,894.
10.1001/jama.2013.5023
Conflict of Interest: None to declare
7.6 Patient Perspectives on COPD Self-management: COVID-19, Remote Tools and the Need for a Multifaceted Approach towards Patient Care
Neil Jackson1, Lanie Bruce2, Michael Coakley2, Michael Drohan2, Sylvia Hughes2
1 Alpha-1 UK Support Group, 2 COPD Support Ireland
As patients living with chronic obstructive pulmonary disease (COPD) during the COVID-19 pandemic, we experienced fear, isolation and limited or no access to regular healthcare.
On 2 June 2022, we spent 3 h on Zoom discussing our day-to-day experiences of living with COPD and brainstorming areas for change that could improve patient care. We discussed COPD self-management, communication between patients and healthcare professionals, and online support.One thing we all agreed on was that patients need more information about COPD and self-management strategies, particularly at diagnosis. This would help us to adopt the behaviours needed to manage our condition effectively and feel more empowered.Additionally, support groups and online tools have been essential for our management of both the physical and psychological aspects of our condition, allowing us to connect with each other and share our experiences.Overall, we would like to see an improvement in education and online resources that enable patients to self-manage COPD effectively and confidently. Additionally, we hope that public health bodies, pharmaceutical companies and telemedicine providers will work collaboratively to develop optimal knowledge-sharing platforms. Furthermore, we encourage healthcare professionals to conduct annual reviews to ensure the continued suitability of our medications.
Disclosures
Medical writing support for this abstract was funded by Boehringer Ingelheim. The patient authors were not reimbursed for their participation in the focus group and all thoughts expressed are their own. All authors fulfil the ICMJE criteria for authorship.
Conflict of Interest: None to declare
(7.7 Not for Publication Valaciclovir for Epstein-Barr virus suppression in moderate-to-severe COPD (EViSCO): an 8-week, randomised, double-blind, placebo-controlled trial)
7.8 Evaluation of AAT phenotyping in patients with low alpha-1 antitrypsin levels
Maqbool Moeez 1 , Tomás P. Carroll 2 , Ronan Heeney 2 , Elaine Hayes 1
1. Our Lady of Lourdes Hospital, Drogheda
2. Alpha-1 Foundation Ireland, Beaumont Hospital
Alpha-1 antitrypsin deficiency (AATD) is a hereditary condition that leads to low levels of an important antiprotease. The main function of Alpha-1 antitrypsin (AAT) is to protect the lungs, with deficiency leading to risk for lung, liver and skin disease. According to the HSE National Laboratory Handbook (1), phenotyping is recommended if AAT levels are low (< 1.0 g/L) (1). It is also recommended that C-reactive protein (CRP) is performed in conjunction, as AAT is an acute phase reactant.
We performed a retrospective review of samples sent for AAT testing from Our Lady of Lourdes Hospital to determine compliance with testing guidelines.
We reviewed 83 AAT test results with AAT levels < 1.0 g/L, sent from our laboratory. Excluding duplicates, only 18 (25%) out of 72 individuals were phenotyped. 13/18 (72%) had clinically significant AAT deficiency. CRP was performed in only 37% of samples.
Patients who have been identified with low levels of AAT and who have not had phenotyping will be recalled for testing. A new protocol for AAT testing has been proposed with Beaumont Hospital and the RCSI Alpha-1 Foundation Ireland laboratories to ensure phenotyping of patients with AAT levels less than 1 g/L. This will lead to a more rapid diagnosis of suspected AATD cases and earlier intervention.
References
National Laboratory Handbook HSE, Ireland. www.hse.ie/eng/about/who/cspd/ncps/pathology/resources/lab-testing-for-alpha-1-antitrypsn-antibodies.pdf
Conflict of Interest: None to declare
7.9 Not for Publication An Evolving “Admission Avoidance” Service Sees a Shift Towards Increased Patient Case Management Activity and a Potential Decline in Patients Requiring Home Visits
7.10 COPD Discharge Proforma Audit
R. O’Loghlin, C. Y. Song, C. Connolly, E. Nic Dhonncha, R. Cusack, M. McDonnell
University Hospital Galway
Background: Chronic obstructive pulmonary disease (COPD) is characterized by progressive airway obstruction, worsening exercise tolerance, and is associated with significant morbidity and mortality. It is the third most common cause of respiratory death in Ireland(1). A new discharge proforma introduced to University Hospital Galway aims to standardize management and follow-up for patients admitted with exacerbations of COPD. Usual management
(N = 27) Post Performa
(n = 28) P value
Smoking status (%):
Current
Ex-smoker
Non-smoker
Unknown
9 (33.3)
13 (48.1)
4 (14.8)
1 (3.7)
15 (53.6)
13 (46.4)
0 (0)
0 (0)
0.09
Inhaler technique review (%) 6 (22.2) 28 (100.0) < 0.01
Target oxygen level in notes (%) 17 (63.0) 26 (92.9) 0.1
Chest physio review during admission 24 (88.9) 24 (85.7) 1.0
Smoking cessation advice (%) 7 (77.8) 15 (100.0) 0.13
Referral to smoking cessation officer (%) 3 (33.3) 4 (26.7) 0.36
COPD support pack provided (%) 2 (7.4) 28 (100.0) < 0.01
COPD self management plan (%) 2 (7.4) 27 (96.4) < 0.01
Pulmonary rehab referral made (%):
Yes
No
Not suitable
Refused
3 (11.1)
18 (66.7)
6 (22.2)
0 (0)
16 (57.1)
1 (3.6)
7 (25.0)
4 (14.3)
< 0.01
Completed pulmonary rehab (%): 0 (0) 6 (21.4) < 0.01
COPD outreach referral (%) 8 (29.6) 27 (96.4) < 0.01
Follow up (%):
Respiratory
GP
None
19 (70.4)
55 (18.5)
3 (11.1)
27 (96.4)
1 (3.6)
0 (0)
0.03
Table 1 Difference in management before and after the introduction of a COPD discharge proforma
Objectives: To assess the use of the new COPD Discharge Proforma and compliance with the recommendations of the Irish NCEC National Clinical Guidelines on Management of COPD (2020).
Method: A retrospective audit of 55 patients admitted to University Hospital Galway with a diagnosis of “exacerbation of COPD” was performed. 27 patients attended the hospital in 2021 prior to the creation of the COPD discharge proforma, and 28 patients after its initial implementation in 2022. Data was collected from the clinical notes, discharge letters, and investigations available on their electronic patient record.
Results: The adherence to COPD management guidelines in the patient groups with and without the COPD Discharge Proforma is compared in Table 1.
Discussion: Areas for improvement of adherence to COPD management guidelines identified include referrals to smoking cessation officers, COPD support and management plan provision, and pulmonary rehabilitation referrals. Many of these areas needing improvement demonstrated significant improvement following the introduction of the COPD Discharge Proforma.
References
N. Brennan, T.M. O’Connor. INHALE Report. 2004. https://doi.org/10.1183/09031936.04.00025604
10.1183/09031936.04.00025604
Conflicts of Interest
No conflicts of interest to declare
7.11 Applying a Lean Approach to the Physiotherapy Management of Exacerbations of COPD in Medical Wards of a Model Four Hospital 1/2.
Angela Ryan1, Caoimhe Barry Walsh1, Amy Cullinane1, Megan O’Mahony1, Aoife Leahy1, Dr Aidan O’Brien1
1University Hospital Limerick
Lean healthcare involves minimising waste and increasing value in implementing service improvements. A Lean approach was used to improve the quality of physiotherapy input for patients admitted to hospital with exacerbations of Chronic Obstructive Pulmonary Disease (COPD). This involved piloting an Enhanced Recovery Model for COPD over a 6-week period.
Lean tools were used including the seven wastes, five whys and swimlanes. The root cause was identified as variation in assessment. Training and documentation resources were developed for error-proofing and to align assessment and management with national and international clinical guidelines. Outcomes were measured in a 6-week period prior to the service changes and again for the duration of the 6-week pilot of the Enhanced Recovery Model for COPD.Improved quality and efficiency were evident following implementation, with associated cost savings to the organisation. Decreases in the average length of stay (23.17%), frequency of re-admissions within 90 days (10.14%), a slight decrease in the average number of physiotherapy contacts (1.55%) and an increase in the average duration of physiotherapy contact time (11.62%) were observed. Compliance improved with national guidelines. Lean tools proved useful in root cause identification and providing a structured approach to implementing standardisation, resulting in service quality improvements.
Conflict of Interest: None to declare
7.12 Applying a Lean Approach to the Physiotherapy Management of Exacerbations of COPD in Medical Wards of a Model Four Hospital 2/2
Angela Ryan1, Caoimhe Barry Walsh1, Amy Cullinane1, Megan O’Mahony1, Aoife Leahy1, Dr Aidan O’Brien1
University Hospital Limerick 1
To address variation in physiotherapy input for patients admitted to hospital with exacerbations of Chronic Obstructive Pulmonary Disease (COPD), an Enhanced Recovery Model for COPD was piloted over a 6-week period. Resources were developed to support physiotherapists including education and documentation to align with the GOLD Guideline for the Management of COPD (2021).Patients from UHL medical wards referred to Physiotherapy for management of COPD were assessed and managed using the novel resources. Metrics gathered from a six-week period prior to development of the resources (T1) and compared with a six-week period of the Enhanced Recovery Model (T2) included the quality of the physiotherapy assessment (COPD Assessment Tool [CAT] score, modified Medical Research Council Dyspnoea Sub-Scale [mMRC] and the GOLD ABCD assessment tool) and quality of physiotherapy management (including appropriate referrals to Smoking Cessation and Pulmonary Rehabilitation).
Results: 14 patients were included in the pre-implementation phase (T1); 17 in the post-implementation period ( T2). Documentation of all key metrics improved: CAT score from 28.57% (T1) to 92% (T2); mMRC: 0% (T1) to 92% (T2); GOLD ABCD: 0% (T1) to 77% (T2); referral to smoking cessation increased from 0% (T1) to 83% (T2); referral to pulmonary rehabilitation increased from 17% (T1) to 44% (T2).
Conclusion: The quality of physiotherapy assessment and management increased considerably following introduction of relevant education and documentation.
Conflict of Interest: None to declare
7.13 Clinical Characteristics of Patients Admitted to CUH in 2019 with COPD Exacerbation
Noreen Tangney1,2, Punitha Vairamani1,2, Deborah Casey1,2, Una O’Riordan1,2, Jill Murphy1,2
1Cork University Hospital 2Integrated Care Respiratory Hub, Cork
Chronic Obstructive Pulmonary Disease (COPD) is the third leading cause of death worldwide. Ireland has the highest mortality rate from any respiratory cause in Europe (134/100,000), and the third highest in COPD-specific mortality (58.15/100.000). The Irish National Audit of Hospital Mortality 2019 shows that COPD is the only leading cause of death with an increasing mortality rate.We reviewed the clinical characteristics of patients admitted to Cork University Hospital with COPD exacerbations in 2019. Patients transferred to other hospitals, discharged the same day, or incorrectly coded were excluded. 491 patients were coded by the HIPE department as COPD exacerbations and 317 met the inclusion criteria. 31 patients (9.7%) had a spirometry-confirmed COPD diagnosis documented. The mean FEV1 of this group was 61.6%. Of the total 317 patients 35 (11.0%) were on LTOT, 45 (14.2%) required non-invasive ventilation, 132 (41.6%) were admitted under the Respiratory Service, and 135 (42.6%) have died within three years. Among this cohort of patients, the rate of spirometric diagnosis was low (9.7%) and mortality was high. This lack of spirometry could ultimately lead to an overdiagnosis of COPD and contribute to poor health indices of the disease in Ireland. Gender Male Female Total T-test
Total number (%) 151 (47.6) 166 (52.4) 317 -
Mean age, years ± standard deviation 73.2 ± 9.8 71.1 ± 9.8 72.1 ± 9.8 0.05
Mean length of stay, days ± standard deviation 9.0 ± 9.0 9.7 ± 9.7 9.4 ± 9.6 0.59
Spirometry (%) 29 (47.5) 32 (52.5) 61 -
Post-Bronchodilator Spirometry (%) 16 (41.0) 23 (59) 39 -
Obstruction post bronchodilation (%) 13 (41.9) 18 (58.1) 31 -
Mean FEV1, % ± standard deviation 61.4 ± 24.6 61.7 ± 24.7 61.6 ± 24.7 0.54
Long term oxygen therapy (%) 14 (40) 21 (60) 35 -
Non-invasive ventilation (%) 17 (37.8) 28 (62.2) 45 -
3 year mortality (%) 69 (51.1) 66 (48.9) 135 -
Patients under Respiratory service 132
Table 1 (7.13) Characteristics of patients admitted to CUH with COPD exacerbation
Conflict of Interest: None to declare
7.14 Make mine a DECAF
A program to increase awareness and understanding of the role of DECAF Scoring in St. Luke’s General Hospital Carlow/Kilkenny
Muahmmad Saad Zaheer1, Edward Kelly2, Fionnbarr McDermott Long3, Tayyabba Mahmood4, Brian Canavan, Kenneth Bolger
St. Luke General Hospital Carlow/Kilkenny
DECAF score is a validated scoring tool that allows healthcare workers to estimate the mortality risk of patients admitted with an exacerbation of COPD. Road blocks to its use in acute settings are i) under-awareness of its value and specificity and ii) the presence of robust pathways to aid management of patients scoring as low risk, moderate risk and high risk.
Our aim was to evaluate awareness of the DECAF score among the healthcare workers (HCW) in the acute hospital setting. An online questionnaire was generated to assess familiarity and understanding of DECAF, and was circulated to all HCW involved in inpatient care of COPD exacerbations.Preliminary data suggests that comprehension about DECAF is fair, and it is not routinely calculated or implemented as it is poorly understood and does not affect change in patient management. Calculation of the score is reported as difficult with over half of responders reporting this. Irrespective of score outcomes, confidence is lacking in referral to palliative care.
We aim to offer tutorials to all personnel involved in acute COPD care to increase understanding of the utility of DECAF and introduce pathways to better manage patients based on the score.
Conflict of Interest: None to declare
7.15 A systematic assessment clinic for severe COPD is feasible and a positive experience for patients
Michele Cuddihy1, Eimear Ward2, Ciara Feeney2, Maeve Sorohan2, Imran Sulaiman1, Breda Cushen1
1Beaumont Hospital, Dublin,2Respiratory Integrated Care, Beaumont Hub, Dublin North City and County CHO9
The COPD At-Risk Assessment(CARA) clinic was established in 2021 for COPD patients with advanced disease, high healthcare use and symptom burden. All undergo systematic evaluation with Medical, ANP, and Physiotherapy review. Laboratory, radiology and oxygen assessments are performed. The clinic aims to provide patient-centred, individualised assessment and management of COPD. We sought to establish patient satisfaction of this new clinic and identify areas for improvement.An adapted HSE patient experience survey was sent to 22 clinic attendees and returned anonymously. Survey questions examined patient satisfaction (3 questions), and knowledge of their condition (7 questions) with “Yes/No” responses. Overall clinic experience was examined using a 5-point Likert scale from very poor to excellent. A free text comments box was provided.Fifteen(68%) completed surveys were returned. All patients (100%) reported satisfaction with the treatment, time and privacy provided to them. Reassuringly, 80% felt knowledge of their condition, its management and treatment was improved. Overall, the experience of the clinic was excellent(53%) or very good(47%).A multidisciplinary systematic assessment clinic for COPD is feasible to run and is beneficial and acceptable to patients. To improve this service we are focussing on developing strategies to ensure the educational needs of all patients are met.
Conflict of Interest: None to declare
7.16 Identification and characterisation of SERPINA1 Null mutations Causing Severe Alpha-1 Antitrypsin Deficiency
Ronan C. Heeney (1, 4), Geraldine Kelly (1), Daniel Fraughen (4), Orla Cahalane (2), Ilaria Ferrarotti (3), Noel G. McElvaney (4), Cedric Gunaratnam (4) and Tomás P. Carroll (1, 4)
(1) Alpha-1 Foundation Ireland, Beaumont Hospital, Dublin 9. (2) Beaumont Hospital, (3) University of Pavia, Italy. (4) RCSI Education & Research Centre, Beaumont Hospital
Alpha-1 antitrypsin deficiency (AATD) is a genetic disorder which can lead to lung, liver, and skin disease. In Ireland, the most common pathological mutation is Z, found in 1 in 25 Irish people. Rare null or nonsense (SERPINA1) mutations are associated with a complete absence of AAT.
> 23,000 individuals have been screened following ATS & ERS guidelines in a national targeted detection programme. Serum Alpha-1 antitrypsin (AAT) quantification is by turbidimetry. AAT phenotyping is by isoelectric focusing with allele-specific genotyping, when required. Rare and novel mutations are identified by SERPINA1 gene sequencing.
Isoelectric focusing revealed an apparent normal (M) phenotype pattern in 10 individuals with lower than expected AAT levels. Seven individuals had apparent SS or ZZ phenotypes but unusually low serum AAT. One individual had no detectable AAT on phenotyping.
SERPINA1 gene sequencing confirmed the presence of 10 M/Nulls, 2 S/Nulls, 4 Z/Nulls, and 1 Null/Null cases. Six different null mutations have been described among the 17 confirmed cases Q0bolton, Q0dublin, Q0porto, Q0cork, Q0amersfoort, and Q0lisbon
Null mutations are underreported due to unique diagnostic challenges. A multi-faceted approach including phenotyping, allele-specific genotyping, SERPINA1 sequencing, and the ability to interpret dissonant results is required to achieve an accurate AATD diagnosis.
Conflict of Interest: None to declare
7.17 Ten years on: A report on the Chronic Obstruction Pulmonary Disease (COPD) Outreach service in Tallaght University Hospital (TUH)
Judith Maxwell, Ciara Scallan, Elaine Joyce, Rachel Egan, Prof. Eddie Moloney, Louise Cullen, Clare Baily-Scanlan, Sarah Cunneen, Maeve NiChleirigh, Alice O’Dwyer, Sherin Varghese
Tallaght University Hospital
The COPD Outreach (COPDOR) model of care was developed as part of the national COPD Quality in Clinical Care Program 2011 to reduce hospital length of stay (LOS) for patients admitted with an acute exacerbation of COPD. TUH launched its COPDOR service in July 2012 when the average LOS for COPD exacerbations was above the national average at 10.2 days. In the first 6 months 153 patients were reviewed and 47 (31%) patients were accepted to the program. Thirty-eight (80.85%) of these patient’s LOS was less than 3 days, with the average LOS reducing from 10.2 to 5.3 days for all patients accepted to the program, and 9.3 days for all patients admitted with a COPD exacerbation in TUH. Over the last 10 years the TUH COPDOR service has reviewed 6,238 patients admitted with an exacerbation of COPD and 1,529 (24.5%) were accepted to the COPDOR program. In 2021, 921 patients were reviewed for inclusion with 181 patients (20%) accepted to the program. Eighty patients had a LOS less than 3 days (44%). The average LOS for patients accepted to the program was 4.4 days with the average LOS for all COPD exacerbations in TUH being 9.2 days.
Conflict of Interest: None to declare
(7.18 Not for Publication) Enhancing the Care of Chronic Obstructive Pulmonary Disease
7.19 New Non-Invasive Ventilation Protocol for Acute Hypercapnic Respiratory Failure in COPD
Derek Nash1, Ciara Gough1, Ciara Dolan1, Helen Mulryan1, Donna Langan1, Shuaib Goga1, Souliman Eldomi1, Tom McEnery1, Fatima Gargoum1
University Hospital Galway
Acute hypercapnic respiratory failure (AHRF) in patients with chronic obstructive pulmonary disease (COPD) carries a high mortality. Non-invasive ventilation in conjunction with standard medical treatment has be shown to significantly reduce this mortality. One in five patients presenting to hospital with an acute exacerbation of COPD will develop AHRF. A quality improvement project was undertaken to create a protocol for patients receiving non-invasive ventilation for AHRF presenting to the hospital. The protocol was developed in conjunction with the respiratory, emergency and critical care departments using best available evidence. A pilot protocol was initially trialed before the final protocol was implemented. The aim was to create a user-friendly document to improve outcomes as measured by the British Thoracic Society Audit toolkit. Audit data from pre-final protocol (2019 to 2021 n = 30) to post protocol implementation (n = 7) demonstrated a relative improvement in outcomes of target oxygen saturations, early post NIV arterial blood gas measurement and overall success of NIV of 80%, 43% and 20% respectively. These are encouraging early results and the focus will be on continued audit and training of hospital clinical staff in use of the protocol.
Conflict of Interest: None to declare
7.20 Patient Knowledge of COPD Exacerbation Symptoms and Management
Mairéad. O’Sullivan1., S.L. O’Beirne1,2
1.St. Michael’s Hospital, Dun Laoghaire,2.St. Vincent University Hospital
Chronic obstructive pulmonary disease (COPD) exacerbations are associated with morbidity and mortality. Symptoms are non-specific, resulting in a wide variety of patient reported complaints1. Poor awareness of the symptoms of exacerbation can lead to a delayed presentation to health care professionals (HCP), therefore delaying recovery2.
My aim was to assess patient knowledge regarding the symptoms of COPD exacerbation from a cohort of COPD Outreach patients. 20 patients were randomly selected. After obtaining informed consent, patients were surveyed in person, using an open-ended questionnaire.Increased fatigue (29%) and difficulty with daily activities (25%) were reported during exacerbations. Anxiety and agitation (18%) were triggers for shortness of breath. Exhaustion and back pain were frequently reported. Most patients (68%) contacted their GP for assistance, with (32%) of patients attending an emergency department (ED). Only (20%) of patients contacted COPD Outreach. Only (30%) of patients contacted a HCP on the day they noticed a change in their symptoms, with (50%) of patients waiting a few days before making contact.Early recognition of the range of symptoms associated with COPD exacerbation could lead to more prompt exacerbation management, improved utilization of the COPD Outreach service, thus reducing the need for GP referral and ED admission.
Global Initiative for Chronic Obstructive Lung Disease. 2017. Pocket Guide to COPD diagnosis, management, and prevention. A guide for healthcare professionals 2017 report.
S Costi, D Brooks, RS Goldstein. Perspectives that influence action plans for chronic obstructive pulmonary disease. Canadian Respiratory Journal 2006;13 (7): 362–368.
S Costi, D Brooks, RS Goldstein. Perspectives that influence action plans for chronic obstructive pulmonary disease. Canadian Respiratory Journal 2006;13 (7): 362–368.
Conflict of Interest: None to declare
7.21 A Novel Self-referral Pathway for Exacerbating COPD Patients to Prevent Attendance at the Emergency Department (ED) – A Pilot Study with COPD (Chronic Obstructive Pulmonary Disease) Outreach in Tallaght University Hospital (TUH) in Conjunction with the National Ambulance Service (NAS)
Ciara Scallan1, Elaine Joyce1, Judith Maxwell1, Prof Eddie Moloney2, Dr Minesh Kooblall2, Dr Rosie Kelly2, Bridget Clarke3, Ann Mc Dermott3, Mark Foley3, Catriona Kinch3, Sean Brady3, Paul Gallen3, Martin O’ Reilly4, John Guilfoyle4, Brendan Mc Nicholas4, Kristin Mc Kenzie Vaas5
1 COPD Outreach, Tallaght University Hospital, 2 Tallaght University Hospital, 3 National Ambulance Service, 4 Dublin Fire Brigade, 5 Grant Thornton
COPD Outreach in TUH and the NAS started a novel symbiotic self-referral service in June 2022 to help prevent exacerbating COPD patients presenting to ED. If they become symptomatic, patients contact COPD Outreach during office hours, or the NAS out of hours, and if suitable, they are seen and treated at home.To date, 14 patients have contacted the service. Nine (65%) were suitable for COPD Outreach and visited at home. One patient (7%) was seen by the NAS. The other 4 patients (28%) were referred to their General Practitioner (GP) for non-COPD related symptoms. The 9 suitable patients were advised to commence the rescue prescription previously given. This prevented them having to attend their GP or ED.On their last presentation to ED, these 10 patients waited on average of 15.5 h (range from 3.5 to 30 h) to be seen. This new pathway allowed these patients to be treated in the comfort of their own home.This service is only in its infancy but already 10 patients have been treated at home and avoided presenting to ED. This project will continue over the next year with the hope that further ED attendances for COPD patients can be prevented.
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
8. COVID-19
8.1 Intrinsic coagulation in early Covid-19; A central role for the Contact-Kinin System
Melanie Bailey1, Dermot Linden 1,2, Olivia Earley1, Tunde Peto 1,2, Danny McAuley 2, Cliff Taggart 2 and Joe Kidney1
Mater Hospital, Belfast Health and Social Care Trust 1 and Queen’s University Belfast 2
Covid-19 causes venous thromboembolism in 17% of hospitalised cases. It also causes ground glass opacities and increased IL-6.Our hypothesis is that the contact-kinin system is canonical in the initiation of Covid-19 inflammation; Factor XII is activated to FXIIa, simultaneously triggering the intrinsic coagulation pathway and the kallikrein-kinin-system (KKS), whereby pre-kallikrein is converted to kallikrein, cleaving high molecular weight kininogen. The released bradykinin causes increased IL-6 and oedema. Ground glass radiological changes suggestive of oedema are seen in early Covid-19. We wished to determine the coagulation response in Covid-19. Blood results from 98 hospitalised patients were recorded by day of symptoms. The intrinsic pathway activation, measured by mean APTT, peaked at day 7 and returned to normal by day 10. Simultaneously, mean platelets dropped until day 6 before rising.
The increased APTT and decreased platelets demonstrates a consumptive coagulopathy. This suggests activation of the contact-kinin system which occurs in parallel with radiological GGO seen in early Covid-19. The KKS releases bradykinin which causes oedema and increased IL-6. A study is underway to inhibit bradykinin in early hypoxic Covid-19 patients.
(8.1)
Funding: National Institute Health Research/ R&D office & Belfast Health & Social Care.
Trust
Conflict of Interest: None to declare
(8.2 Not for Publication A physiotherapy prospective observational study of post-intensive care COVID-19 patients in Tallaght University Hospital)
8.3 Post-Covid Assessment in a Specialist MDT Clinic: A Retrospective Review of Covid-19 Patients requiring High Dependency Oxygen Requirements or ventilation during Hospital Admission
Erica Bajar1, Orla Price1, Maria Leitermann1, Deirdre Garvin1, Claudia Oliveira1, Jeananne Garavan2 and Cyril Rooney1
1 Mayo University Hospital, 2 Community Healthcare West, Mayo
Severe Covid 19 infection and its associated complications can lead to wide ranging persistent problems after the acute infection has resolved. Post–covid clinics require a multi-disciplinary assessment and intervention to minimise the long term effect of acute severe Covid infection.
We describe our experiences using a multi-disciplinary post-Covid clinic where medical, functional, and psychological sequelae of patients requiring high dependency respiratory support due to COVID-19 from March 2020-June 2021 in MUH were assessed. Eligible patients attended a review clinic consisting of medical, physiotherapy, nursing, occupational therapy, and psychology assessments.Thirty-eight patients consented to review. Most demonstrated full resolution of radiological findings and normal lung function tests. Most common persistent symptoms were breathlessness (n = 19), fatigue (n = 23), pain (n = 15), and cognition/memory deficits (n = 17). High BMI > 30 identified in 23 patients. Screening for anxiety (n = 14), depression (n = 9), and PTSD (n = 9) identified need for a psychology review.
Outcome: Referral to Respiratory clinic (n = 20); 14 patients were discharged to GP. Predominant physiotherapy requirements were for physical conditioning and breathing pattern disorders. Sleep and diet/lifestyle education were the predominant nursing interventions. Fatigue, functional issues, and cognition was the prevalent issue addressed by occupational therapy. Clinical psychology provided psycho-education to all patients attending with 50% referred to community mental health services.
Conflict of Interest: None to declare
8.4 Prevalence of Fatigue and Sleepiness in a Post-Covid Clinic
Christina Campbell[1], Linda Keegan[2], Killian Hurley[1,2], Imran Sulaiman[1,2]
[1] Royal College of Surgeons in Ireland, Dublin. [2 Beaumont Hospital, Dublin
Since the emergence of the COVID19 pandemic, there has been the secondary recognition of a 'Post-COVID’ or 'Long COVID' syndrome. Fatigue is one of the most common symptoms post Covid, but the incidence of sleepiness has not been evaluated. Furthermore, a clear link has emerged between severe COVID19 disease and co-morbid Obstructive Sleep Apnea (OSA). Therefore, it is possible that undiagnosed OSA may contribute to fatigue and sleepiness in a post-COVID population. In a Post-COVID clinic we assessed the incidence of fatigue and sleepiness via the Chalder Fatigue Scale (CFS), Epworth Sleepiness Scale (ESS), Pittsburgh Sleep Quality Index (PSQI) and Berlin Score. Patients with excessive daytime sleepiness were then referred for a Limited Sleep Study via the WatchPAT device.One hundred and nine patients completed at least one questionnaire, with 74 (68%) completing all four. The majority of patients were male (52%) and the mean age ± SD was 57 years ± 14 years. In evaluating the questionnaire data, Median (IQR) score was 18 (13-24) for CFS, 7 (3-12) for ESS and 10 (6-13) for PSQI. Of these patients, 40% had an elevated ESS consistent with excessive daytime sleepiness while 84% had an elevated PSQI consistent with significant sleep disturbance. Additionally, by the Berlin Score 27% of patients were deemed high risk for Obstructive Sleep Apnoea. To date 25 WatchPAT studies have been performed. Of these, 20 (80%) had some evidence of OSA; 8 (40%) mild, 3 (15%) moderate and 9 (45%) had severe. Excessive daytime sleepiness and fatigue are both common symptoms among patients after SARS-CoV-2 infection and may be indicative of undiagnosed Obstructive Sleep Apnoea.
Conflict of Interest: None to declare
8.5 Are invasively ventilated COVID-19 ICU patients more likely to develop a pneumothorax or pneumomediastinum? A retrospective cohort study
Amber J Downes1, Sinead M Campbell1, Jiamin Ke1, Donal Ryan1, David G Healy1
1.St Vincent’s University Hospitals
Severe respiratory syndrome coronavirus 2, and the resultant coronavirus disease 2019, increased the number of patients requiring invasive ventilation in intensive care units (ICU) internationally. In our institution, the requirement for chest drains in ventilated ICU patients appeared to increase during the COVID-19 pandemic. Our study aimed to identify if patients requiring invasive ventilation in ICU were more likely to sustain a pneumothorax and/or pneumomediastinum if they were positive for COVID-19.A retrospective cohort study was completed, reviewing all invasively ventilated patients in ICU between 1st March 2020 and 28th February 2021. Four patient groups were created based on two criteria: presence or absence of COVID-19 and pneumothorax/pneumomediastinum. 344 patients met inclusion criteria for the study. 5 patients were COVID-19 positive with an identified neumothorax/pneumomediastinum. 12 patients were COVID-19 negative with an identified pneumothorax/pneumomediastinum. The risk was 11.4% in patients in the COVID-19 positive group and 5.9% in patients in the COVID-19 negative group, with a significance of 0.035. COVID-19 positive patients have a risk 2.84 times higher of developing a pneumothorax/pneumomediastinum.In our patient cohort, being COVID-19 positive presents a significant increased risk of developing a pneumothorax and/or pneumomediastinum while invasively ventilated, compared to their COVID-19 negative counterparts.
Conflict of Interest: None to declare
8.6 Consultant respiratory pharmacist input to COVID-19 Respiratory Support Unit—An Evaluation
Cairine Gormley1, Martin Kelly1, Brendan Moore1,Rose Sharkey1, Maureen Spargo2, Glenda Fleming2
1. Altnagelvin Hospital, Western Health and Social Care Trust (WHSCT), Derry, 2. Medicines Optimisation Innovation Centre, Antrim
Medicines optimisation in patients admitted to the COVID-19 Respiratory Support Unit during the COVID-19 pandemic presented new problems for healthcare staff.
The objectives were to
To quantify medication-related interventions made by the consultant pharmacist (CP)
To determine potential clinical significance and estimate potential cost-savings
Medicines reconciliation and optimisation occurred on admission [1]. Patients were reviewed daily. Interventions, recorded over 10 weeks, were graded for clinical significance using Eadon Criteria [2] —a score of 4 or greater represents improvement in quality of care. A selection were independently reviewed by consultant respiratory physician (MK). University of Sheffield School of Health and Related Research (ScHARR) model [3] was used to estimate potential cost savings associated with the interventions (Table 1) [1].
Table 1 (8.6) Eadon Criteria [2], Grade and Number of Interventions [1]. Eadon Criteria
(Grade) Number of Interventions ScHARR Lower Limit (£) ScHARR Upper Limit (£)
Significant: improvement in standard of care (Grade 4) 826 53,690 123,900
Very significant: prevents major organ failure or adverse reaction of similar importance (Grade 5) 53 42,930 65,296
Potentially life-saving (Grade 6) 2 2,464 3,520
This service demonstrated an improvement in the standard of care. Applying ScHARR model yielded potential annual health economy savings between £495 k and £960 k.
1. Gormley, C.; Spargo, M.; Fleming, G.; Moore, B.; Scott, M.; Sharkey, R.; Friel, A. Medicines Optimisation for Respiratory Patients: The Establishment of a New Consultant Respiratory Pharmacist Role in Northern Ireland. Pharmacy 2021, 9, 177. https://doi.org/10.3390/pharmacy9040177
2. Eadon H. Assessing the quality of ward pharmacists’ interventions. International Journal of Pharmacy Practice 1992;1:145–47. https://doi.org/10.1111/j.2042-7174.1992.tb00556.x
3. Karnon J. et al. Modelling the expected net benefits of interventions to reduce the burden of medication errors. Journal of Health Services Research & Policy 2008; 13(2):85–91.
Conflict of Interest: None to declare
8.7 Angiotensin-Converting Enzyme (ACE) And Clinical Outcomes In Long COVID Patients
Gabriele Gusciute1, Seamas C Donnelly1,2, Patrick D Mitchell1,2
1 Trinity College Dublin, 2 Tallaght University Hospital, Dublin
While evidence of the association between coronavirus disease 2019 (COVID-19) and the renin-angiotensin system (RAS) is established, the role of serum angiotensin-converting enzyme (s-ACE) remains to be elucidated in long COVID. In this retrospective study, we examined the relationship between s-ACE and clinical outcomes on standardised measures in a sample of long COVID patients (n = 75). The medical records of 75 patients who attended a long COVID clinic (at least 12 weeks post infection) during a ten-month period in 2021 were included. Demographic and clinical variables, as well as performance on measures such as the 6-min walk test (6MWT), pulmonary function tests, the Fatigue Severity Scale (FSS), the Hospital Anxiety and Depression Scale (HADS), St George’s Respiratory Questionnaire for COPD (SGRQ-C), the International Physical Activity Questionnaire-Short Form (IPAQ) and the Sniffin Sticks Test-12 Items (SST-12), were recorded. Elevated s-ACE levels were positively correlated with body mass index (BMI; rs = 0.279) and maximal inspiratory (MIP; rs = 0.255) and expiratory (MEP; rs = 0.275) pressures. A negative association was found between s-ACE and time since COVID-19 diagnosis (rs = -0.258). 21.3% of the sample population had elevated s-ACE (> 65 U/L). Pairwise comparisons between patients with elevated and normal ACE levels revealed higher BMI, and shorter diagnosis to assessment interval, in the high ACE group (p < .05). There was no relationship between s-ACE and any of the standardised scales used. This study is the first to report on s-ACE as a biomarker in long COVID. Further elucidation of this relationship is required and may be additive in informing resource allocation and response to the ever-growing challenge of long COVID.
Conflict of Interest: None to declare
8.8 Follow up of patients discharged from a Respiratory High Dependency Unit after Covid 19 infection
M MC Closkey, M Mc Geady R Laverty, H Khan, C King, MG Kelly, R Sharkey
Altnagelvin Area Hospital Respiratory Unit, Londonderry
The first Respiratory High Dependency Unit(HDU) was started in this District General Hospital in September 2020 in response to the Covid pandemic. All patients requiring high Flow Nasal Oxygen(HFNO) and or CPAP but not immediate invasive ventilation were admitted to this unit. The long term outcome of all patients who survived to discharge was monitored.Patients were followed up post discharge, initially by community respiratory nursing team. Consultant review was at two months and thereafter depending on need. A number of variables were recorded including: number of outpatient attendances, additional investigations performed, duration of oxygen use, requirement for psychology input, need for referral to the Long Covid service and final outcome.A total of 298 patients survived to discharge from the resp HDU, 111 females(F), mean age 51.8 and 186 males (M), mean age 57.8. Average length of stay was 10.5 days F and 12.5 days M. Oxygen was required in 43%(48) F and 50% M(94) on discharge home. The majority of patients were able to come off oxygen after 2 to 4 months. Lung function was done on 25.5%. Three patients have post Covid fibrosis. Eleven patients attended psychology. Fifteen patients (5%) have been referred to Long Covid service.The short to medium term follow up of patients discharged from this Covid Resp HDU showed that the majority of patients recovered well as long as review was timely and patients felt supported on discharge.
Conflict of Interest: None to declare
8.9 Pulmonary Function Tests in COVID-19 Recovered Patients Attending a Pulmonary Rehabilitation Programme
G. Nolan1, MY Tran1, P. Tonge2, S. O’ Beirne1,2
1St Vincent’s University Hospital, Dublin, 2..St Michael’s Hospital, Dun Laoghaire
Pulmonary Rehabilitation (PR) is a well-recognised intervention in chronic respiratory disease. Its role following COVID-19 infection is under investigation. A retrospective review of pre and post pulmonary function tests (PFT’s) performed in forty patients participating in an eight week PR programme, was completed. All PFTs were carried out according to the ATS/ERS 2005 guidelines.
52% of participants were male with a mean age of 52 ± 13.2 years and BMI 33 ± 8.3 kg/m2. Twenty five patients reported resolution of dyspnoea after the PR programme. Mean FVC, FEV1, DLCO and MEP values pre and post PR were normal, with a significant improvement in the FVC post-PR. Mean MIP was mildly reduced pre-PR with a significant improvement post-PR but it did not normalise.
Table 1 (8.9) PFT Results Pulmonary Function Tests Pre-PR Post-PR P value
FVC %predicted 100.5 ± 27.2 105.3 ± 18.7 0.006
FEV1%predicted 97.6 ± 20.8 100.5 ± 17.7 0.084
DLCO %predicted 83.8 ± 17.0 85.5 ± 14.7 0.397
MEP cmH20 100.9 ± 42.6 102.4 ± 39.1 0.527
MIP cmH20 70.4 ± 32.8 75.9 ± 31.2 0.014
Reduced inspiratory muscle strength is associated with dyspnoea given the post-PR improvement in this value it may be useful for patients undergoing PR post-COVID-19 infection.
Conflict of Interest: None to declare
8.10 Outcomes of patients reviewed in a Post-Covid Respiratory Clinic
Aisling O’Connor, Michelle Uno, Sarah Farrell, Tidi Hassan
Our Lady of Lourdes Hospital, Drogheda
Covid-19 infection can result in persisting and debilitating symptoms impacting on a persons’ quality of life. Here we present data from patients attending a dedicated Post-Covid Respiratory clinic.Over 350 patients have attended the Post-Covid Respiratory Clinic, with 9% having required ICU admission, 61.5% required non-ICU hospital admission and 29.5% were referred from community. Of the total patients reviewed, 43.4% were reviewed and discharged through virtual triage clinic with 56.6% requiring face to face clinic review. Of the patients attending face to face review, 67% completed a 6MWT, with 9% of these showing episodes of significant desaturation. In contrast, a high number (> 52%) of patients attending the clinic report persisting or worsening shortness of breath since their acute infection. In addition > 50% of patients attending the clinic reported severe fatigue or brain fog, with 63% reporting they had not returned to their premorbid level of functioning.
Covid-19 symptoms may persist beyond the acute infection stage. There does not appear to be a correlation between initial severity of disease and the development of persisting symptoms. Our data shows the multisystem nature of symptoms and the need for multidisciplinary input in the follow up of patients with persisting symptoms of Covid-19.
Conflict of Interest: None to declare
8.11 Pulmonary embolism rates during different variants of SARS Cov2 an Irish tertiary hospital experience”
Joshua Olaniyi 1,2, Mary Rose Kelly 1, Ahmad Basirat 2, Carol Buckley 1, Padraig Donovan 1, Aoife O’Connor 2, Kashif Rana 1, Robert Trueick 1, Minesh Kooblall, Edward Moloney1, Stephen Lane 1, Seamas Donnelly 1,2, Patrick Mitchell 1,2
1Tallaght University Hospital,2 Trinity College Dublin
Introduction: Variants of SARS CoV2 has been well reported. Reports suggest, compared to non-SARS CoV2 critically ill patients, a positive SARS CoV2 diagnosis confers an increased risk of pulmonary emboli (PE). Little exists on whether variations in SARS CoV2 confers a change in rates of PE. Aim: Perform a retrospective cohort inpatient review at an Irish tertiary hospital with SARS CoV2 and the rates of PEs during different peak variant timelines.
Method: Two data points identified, January 2021- June 2021 (Point A), predominantly Alpha and Beta, and December 2021-January 2022 (Point B), Delta and Omicron in Ireland. Inpatients >18 years with a positive nasopharyngeal swab for SARS CoV2 were included.
Results: 640 patients were identified, 510 in Point A and 130 in Point B with 34 confirmed PEs, (5.3%). 308 were of clinical concern, with a positivity rate of 11%. In Point A, 258 were of clinical concern for PE, and 26 were positive, (9%). In Point B, 50 were of clinical concern for PE and 8 were positive, (16%). Greater rates of positive cases in Point B, Delta and Omicron predominant, were seen, a predominantly more severe or infectious variant respectively.
Conclusion: Little discussion exists on if variations in severity and or infectivity of SARS CoV2 confers an increased risk of PEs. Our data suggest, there may be an increased rate of PEs depending on the variant type of SARS CoV2.
Conflict of Interest: None to declare
8.12 Lung Function and Airway Impedance in Patients Attending Post-COVID Clinic
SF Raza1, A Yunes1, I Delagua1, M Rahaman1, E Judge1, J Faul1, L Cormican1,A McGowan 1, D Ampazis1, A Subramaniam1,3
1. Connolly Hospital Blanchardstown, Dublin, 3. Respiratory Integrated Care, Dublin North City & County
Forced oscillometry is used to evaluate the Respiratory system resistance (Rrs) and reactance (Xrs). It is typically used to assess patients with asthma and more recently in post-COVID patients. Our aim was to evaluate the advantage of using oscillometry in post-COVID patients attending our centre.We conducted a cross-sectional study evaluating lung function test in two group of post-COVID patients; Hospitalised patients (10 received ICU care, severe disease) versus Non-hospitalised (15, mild disease).Twenty five patients were studied, 7% patient in the mild group and 60% in the severe group had impaired lung function; restrictive pattern was more common. Alterations observed in the mild group include 20% patients with partial reversibility, 13% high resistance (Rrs) and 7% high reactance (Xrs) whilst 60% were normal. The severe group had high Xrs in 40% of cases. Oscillometry detected more abnormal cases than spirometry, 33%, mostly in the mild group. There was also a positive association between restriction and high reactance, particularly in the severe group (40%) suggesting tissue stiffness.Thus, oscillometry may be a more sensitive test than spirometry in detecting impaired lung function, particularly fibrosis in the post-COVID patients. However larger studies are necessary to better understand the utility of oscillometry in this cohort.
Conflict of Interest: None to declare
8.13 Compliance with COVID-19 protocols
Kieran Skehan, Emma Dolan, Lisa Murphy, Elaine Hayes
Our Lady Of Lourdes Hospital, Drogheda
Our Lady of Lourdes Hospital created local guidelines for management of COVID-19 patients on admission. Our audit assessed compliance with these protocols.The protocol included risk stratifying patients along with recommended investigations and drug prescriptions. Guidelines also encompassed nutritional strategies developed by OLOLH’s dieticians.
A prospective review was undertaken over a 2 week period of patients presenting with COVID-19 pneumonia. Investigations reviewed included chest x-ray, arterial blood gas (ABG), C-reactive protein (CRP), procalcitonin and d-dimer. Treatments evaluated comprised of oxygen administration, anticoagulation, steroid use and nutritional supplementation.
Our initial Audit found poor compliance with the protocol. Subsequent interventions involved organized teaching sessions, development of a PDF protocol for phones and an acronym for COVID-19 prescribing was developed and displayed in the Emergency Department.Our interventions led to significant improvements in correct prescribing related to COVID-19, with correct nutritional supplement, steroid and anticoagulation prescribing increasing 58%, 29% and 25% respectively. Rates of ABG and d-dimer also increased by 54.4% and 28% respectively.
Conflict of Interest: None to declare
8.14 Six Minute Walk Test vs Sit To Stand Test in assessment of the post COVID-19 Patient
S. Toland, I. DelAgua, D. Ampazis, A. Subramaniam, L. Brien, A. McGowan, L. Cormican
Connolly Hospital Blanchardstown
The 6-min walk test (6MWT) is one of the most commonly used tests to assess exercise capacity in chronic respiratory conditions, however, it is a relatively time-consuming test. Previous studies have shown that Sit To Stand Test (STST) can be used as an alternative for 6MWT in patients with COPD, it is a simpler method and is just as effective to determine functional status, as well as being less time consuming and producing less hemodynamical stress compared to 6MWT.Given the large number of patient that required assessment in the post COVID setting, we wanted to assess whether STST could be used as an alternative to 6MWT in the assessment of post COVID-19 patients. We conducted a retrospective study of 80 patients, 40 of whom underwent a 6MWT and 40 a STST, they were matched according to age, gender and clinical course of their COVID-19 infection. The results in terms of dysponea, BORG score and heart rate post 6MWT and STST were similar between the two cohorts. As like as 6MWT, STST can determine functional status post COVID-19, whilst been less time consuming. STST should be considered as an alternative for 6MWT in assessment of patients post COVID-19.apnoea is a potential manifestation of Long Covid syndrome and intervention is effective as already established.
Conflict of Interest: None to declare
8.15 Dysfunctional Breathing Phenotype in Post COVID Patients
A Yunes1, SF Raza1, I Delagua1, A Valenzona1, M Rahaman1, E Judge1, J Faul1, L Cormican1, D Ampazis1, A Subramaniam1,3
1. Connolly Hospital Blanchardstown, 3. Respiratory Integrated Care, Dublin North City & County
There is increasing awareness of dysfunctional breathing (DB) causing debilitating symptoms with negative impact on functional status and quality of life. Post COVID patients often present with persistent dyspnoea after acute illness and identifying DB in this cohort can help tailor management. We aimed to describe the DB phenotype in patients attending the post COVID clinic in our centre.A retrospective review was conducted on patients attending the service over 5 months. All patients were screened for DB using Nijmegen questionnaire (NQ) and results were correlated with age, gender, BMI and smoking status.124 patients attended the service in this period, of which 25 patients (20%) had a positive NQ score, confirming DB. Mean age was 55 ± 13 years, and 68% were female. Majority had an elevated BMI; 20% were overweight and 64% were obese; mean BMI was 33 ± 10 kg/m2. Fifty-two percent were either current or ex-smokers.In conclusion, DB is not uncommon in post COVID patients. It is more likely to be observed in female, younger patients and those with elevated BMI. Smoking status showed no relation with DB. Larger studies on DB phenotypes in post COVID patients would enable better recognition and timely treatment.
Conflict of Interest: None to declare
(8.16.Not for Publication Analysis of T-cell subsets and their mitochondrial status response to SARS-CoV2 spike peptide activation in COVID-19 patient PBMC samples)
8.17 Follow-Up of Survivors of Critical Illness Related to Covid-19 Infection; A Pilot study
Cara Weldrick 1, SN Mary Flavin 2, Seosaimh Ni Riain 3
University Hospital Limerick
Recent improvements in ICU care has seen a significant increase in the number of patients surviving critical illness. However patients face a wide variety of challenges after discharge. Our study aimed to establish whether patients who had been admitted to an ICU with Covid- 19 pneumonia had persisting deficits 1 year after admission. We conducted a qualitative pilot study of patients that were admitted to the ICU from March 2020 to March 2021 with Covid- 19 infection. A structured interview approach was used to collect outcomes assessing physical, functional and psychological domains. 80% of patients reported that they had not returned to their pre-morbid activity level. 93.3% of patients reported at least one physical impairment. Only 45% of patients had returned to full time employment. A total of 80% of patients reported psychiatric impairment and 40% of patients had an abnormal Hospital Anxiety and Depression scale. The results of our study illustrate the presence of significant physical, psychological and cognitive symptoms experienced by ICU survivors up to 1 year post discharge. The data from our study data is a strong argument on the need for a comprehensive follow-up service post ICU admission to address the complex sequelae of critical illness.
Conflict of Interest: None to declare
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
General Respiratory
9.1 Evaluating Liver Disease in Alpha-1 Antitrypsin Deficiency (AATD)
Fiachra O’Meara, Simon Carty, Daniel Fraughen, Tomás Carroll, Hassaan Yousuf, Cedric Gunaratnam, John Ryan, Noel G. McElvaney
Royal College of Surgeons Ireland
Alpha-1 Antitrypsin Deficiency (AATD) is the commonest genetic cause of COPD, while also increasing susceptibility to liver disease. This study aimed to examine the extent of liver disease in AATD patients in Ireland.This cross-sectional study included 169 patients from the Irish National AATD registry. “PiZZ”, “PiSZ” and “PiMZ” genotypes were assessed using FibroScan; a transient elastography (TE) instrument for non-invasive evaluation of liver steatosis and fibrosis. TE’s Controlled Attenuation Parameter (CAP [dB/m]) and Liver Stiffness Measurement (LSM [kPa]) assesses steatosis and fibrosis, respectively.32% of “PiZZ” patients showed LSM values ≥ 7.1 kPa (stage 2 fibrosis) with an overall mean of 8.41 kPa. In “PiSZ” and ‘PiMZ” patients, 24% LSM values were ≥ 7.1 kPa with a mean of 6.76 kPa and 6.68 kPa, respectively. All cohorts showed significant levels of steatosis with 71% of “PiMZ’ and 60% of “PiSZ” patients yielding a CAP value ≥ 238 dB/m (stage 2 steatosis).
Liver steatosis and fibrosis is under-recognised in AATD. TE is a quick, quantitative, non-invasive imaging modality that is more sensitive than ultrasound in identifying steatosis and fibrosis and could be more readily used to facilitate the work-up of AATD patients and identify those with asymptomatic advancing liver disease.
Table 1 (9.1) Characteristics of “PiZZ”, “PiSZ” and “PiMZ” Phenotypes ZZ SZ MZ All Phenotypes
Demographics
n = 74 25 70 169
Male 34 10 34 78
Female 40 15 36 91
Age (Mean) 52.88 (17.14) 57.90 (11.58) 53.17 (15.19) 53.74 (15.64)
Height (cm)
(mean ± SD)
169.83 (9.05) 163.14 (9.03) 168.43 (9.55) 168.29 (9.43)
Weight (kg)
(mean ± SD)
80.19 (20.7) 74.37 (13.87) 79.10 (16.57 78.87 (18.64)
BMI (kg/m2)
(mean ± SD)
27.76 (6.8) 28.09 (5.46) 27.92 (5.63) 27.86 (6.25)
FibroScan Finding
CAP (dB/m) 253.78 (72.91) 258.84 (53.06) 276.96 (60.7) 264.13 (65.92)
n = S0 (< 238) 33 10 20 63
n = S1 (238–260 6 4 3 13
n = S2 (260–290) 7 3 15 25
n = S3 (> 290) 28 8 32 68
LSM (kPa) 8.41 (8.55) 6.76 (4.83) 6.68 (4.56) 7.45 (6.66)
n = F0-1 (< 7) 50 19 53 122
n = F2 (7–11) 14 4 10 28
n = F3 (11–19) 4 1 5 10
n = F4 (> 19) 6 1 2 9
Conflict of Interest: None to declare
9.2 Calibrated automated thrombography demonstrates a hypocoagulable profile in subjects with pulmonary arterial hypertension
Sarah Cullivan1,2, Brian McCullagh1,Barry Kevane,2,1, Patricia Maguire2, Fionnuala Ni Áinle2,1 and Sean Gaine1
Mater Misericordiae University Hospital, Dublin,2 SPHERE Research Group, University College Dublin
There is considerable interest in the role of platelets and blood coagulation in the pathobiology of pulmonary arterial hypertension (PAH). The aim of this study is to characterise thrombin generation in PAH using calibrated automated thrombography (CAT). Institutional ethical approval was granted. Thrombin generation using CAT was performed using a Fluoroskan Ascent® Plate Reader and Thrombinoscope™ software.In the period between July 2020 and July 2022, 20 individuals with PAH and 20 healthy volunteers were recruited to this study. Platelet counts were significantly lower in individuals with PAH relative to healthy controls (p = 0.0053). Mean platelet volume was significantly higher in PAH subjects at 10.8 fL versus 8.2 fL (p = 0.0001) respectively. CAT parameters also differed between groups, as individuals with PAH had significantly lower endogenous thrombin potential, peak thrombin generation and thrombin generation velocity index in both platelet-rich and platelet-poor plasma, suggesting a hypocoagulable profile (Table 1).This data demonstrated a hypocoagulable profile in subjects with PAH. This may represent an ‘exhaustion effect’ of coagulation factors due to sustained and prolonged activation of the procoagulant pathways in the pulmonary circulation. The clinical implications of these results are immensely relevant and support the shift away from empiric anticoagulation in this population.
Table 1 (9.2) Calibrated automated thrombography in pulmonary arterial hypertension
Table 1 Provides an overview of laboratory parameters and thrombin generation characteristics in platelet-rich plasma and platelet-poor plasma in subjects with pulmonary arterial hypertension (PAH) and healthy controls. Values are expressed as mean ± standard deviation unless otherwise specified. *Median (Interquartile range). Abbreviations: APPT: Activated partial thromboplastin time; VI: Velocity index.
References
Vrigkou, E., et al., Coagulation Profiles of Pulmonary Arterial Hypertension Patients, Assessed by Non-Conventional Hemostatic Tests and Markers of Platelet Activation and Endothelial Dysfunction. Diagnostics (Basel), 2020. 10(10)
Melnichnikova, O., et al., The dynamics of thrombin formation in patients with pulmonary arterial hypertension. Thrombosis Research, 2021. 208: p. 230-232.
Conflict of Interest:None to declare
9.3 Inhibition of C3d signalling downregulates NLRP3 inflammasome activation and IL-1β production in monocytes of individuals with alpha-1 antitrypsin deficiency
Debananda Gogoi1, Michelle Casey1,2, Azeez Yusuf1, Daniel Fraughen1, Malcolm Herron1, Lameese Alhaddah1, Tomás P. Carroll1, Noel G. McElvaney1,2 & Emer P. Reeves1
1 Royal College of Surgeons in Ireland, Dublin,2 Cystic Fibrosis Unit, Beaumont Hospital, Dublin
Alpha-1 antitrypsin (AAT) deficiency (AATD) is characterised by sustained inflammation. Elevated levels of the complement activation product C3d were previously detected in plasma and airway samples, and correlated with airway obstruction. The aim of this study was to investigate C3d as a trigger of monocyte activation that drives inflammation in AATD.Blood was collected from patients with AATD (n = 9, mean FEV1: 45.16% ± 22.88) or healthy control donors (HC) (n = 16). Purified monocytes were challenged with C3d and key markers of monocyte activation, including phosphorylated p85, AKT, as well as caspase-1 and NLRP3 required for IL-1β activation, were measured by qPCR, ELISA or western blot analyses. In AATD, increased monocyte membrane expression levels of C3d (P = 0.006) and cognate receptor CR3 (P = 0.003), triggers IL-1β secretion (P < 0.001). Mechanistically, C3d, but not C5, increased IL-1β expression and secretion through the PI3/AKT/NF-κβ pathway, activating the NLRP3 inflammasome (P < 0.001). In corroboration, AATD monocytes demonstrated increased cytosolic calcium levels (P < 0.001) and caspase-1 activity (P < 0.0001), effecting increased plasma IL-1β levels (P < 0.001). In vitro, exogenous, glycosylated AAT, binds C3d (P = 0.0008) and modulates inflammasome activation (P = 0.001).These results demonstrate that the C3d:CR3-inflammasome axis may represent a key target in regulating the inflammatory response in AATD monocytes, and is significantly downregulated by exogenous AAT. This is of particular interest as intravenous AAT to treat AATD-related lung disease, remains unavailable for people with AATD. This project was funded by the US Alpha-1 Foundation (Grant # 615848) and HRB/HRCI Ireland (Grant # HRB /MRCG-2018-04).
Conflict of Interest: None to declare
9.4 Use of High Flow Nasal Oxygen in Beaumont Hospital and Development of A Local Guideline
C Hayes, D O’Flaherty, C McGeoghegan, B Cushen
Beaumont Hospital, Dublin
High flow nasal oxygen (HFNO) as a non-invasive respiratory support is increasingly used in the management of acute hypoxaemic respiratory failure (AHRF)[1]. International society guidelines with specific recommendations on settings, titration and weaning protocols are lacking. We undertook an audit to assess the use of HFNO in Beaumont Hospital and developed a local guideline, Fig. 1.
We reviewed the medical records of inpatients on general wards using HFNO over the course of one week. Patient demographics, indication, presence of written prescription and level of monitoring with arterial blood gas/vital signs were recorded. A local guideline to improve the use of HFNO in our hospital was developed based on current literature.The medical records of 13 patients, 54% female, were reviewed. The mean age was 68 years. The most common indications for HFNO were AHRF (54%), airway clearance (31%) and type 2 respiratory failure (15%). 31% of patients had a documented ABG prior to commencing HFNO, and 46% ABG post-HFNO initiation. Only 15% of patients had a written HFNO prescription.
We found several areas for improvement in the prescribing, documentation and use of HFNO. We will re-audit the use of HFNO following introduction of our guideline to assess its impact.
References
Oczkowski S, Ergan Büm, Bos L, et al. ERS Clinical Practice Guidelines: high-flow nasal cannula in acute respiratory failure. Eur Respir J. 2022; 59(4).
Conflict of Interest: None to declare
9.5 Physiotherapy-led Lung Ultrasound in a Cardiothoracic Unit: A Service Evaluation
Aisling Hennessy
Blackrock Clinic, Dublin
The use of lung ultrasound (LUS) by respiratory physiotherapists is increasing due to greater diagnostic accuracy of LUS compared to conventional assessment tools of auscultation and chest x-ray interpretation in identifying post-operative pulmonary complications (PCC) such as consolidation, pleural effusion, pneumothorax and interstitial syndrome. The aim of this service evaluation was to explore the use of LUS by physiotherapy in a post cardiothoracic surgery population.All cases between September 2021 and April 2022 where physiotherapy assessment incorporated LUS were retrospectively reviewed. Data collected included trigger for LUS assessment, LUS findings and recommendations based on findings. Data was analysed using descriptive statistics.Forty-two patients were assessed using LUS. Triggers for LUS assessment included failure to wean from mechanical ventilation (n = 3), failure to wean from oxygen therapy (n = 28) and dyspnoea limiting rehabilitation (n = 11). Findings are outlined in Fig. 1. Based on findings further physiotherapy treatment including airway clearance and volume recruitment was recommended in 11 cases. In 31 cases findings were not amenable to physiotherapy and were escalated to the medical team. Timely and appropriate treatment of PPC hinges on accurate diagnosis. Use of LUS aids the identification of PCC amenable to physiotherapy and expedites escalation of cases not amenable to physiotherapy.
Fig. 1(9.5) Lung ultrasound assessment findings
Conflict of Interest: None to declare
9.6 Missed opportunities—need for venous thromboembolism prophylaxis not being identified at time of discharge following hospitalisation during pregnancy
L. Madden Doyle1, D. Lalos2, R. Varley2, C. McCarthy1, A. McErlean1, K. Ewins2,1, DM. Ryan1,2
1. Beaumont Hospital, Dublin, 2. Royal College of Surgeons in Ireland;3.Rotunda Hospital, Dublin
Background: Maternal death enquiries have highlighted venous thromboembolism (VTE) as the fourth leading cause of maternal death1,2
Aim: to audit quality of care in the diagnosis and discharge prophylaxis of VTE in pregnant patients admitted medically
Standard: Royal College of Obstetrics and Gynaecologists (RCOG) Green-top Guidelines 37a and 37b3,4
Method: Pregnancy-related admissions to Beaumont Hospital (BH) from 2019-2021 were identified using Hospital In-Patient Enquiry. Retrospective chart review was performed against RCOG guidelines-based audit tool.
Results: 54 patients had diagnostic code for pregnancy. 20 (37%) were respiratory presentations, of which 14 (70%) were query pulmonary embolus with 1 (7%) subsequently confirmed. 6 of 20 (30%) had delayed time to chest x-ray (CXR), with 2 (14%) never undergoing CXR. Nine (65%) underwent CT pulmonary angiogram, 2 (14%) had perfusion scanning. 28% had no VTE diagnostic imaging despite working diagnosis of VTE, of which x (%) had empiric treatment. There was no documentation of VTE risk assessment on discharge, despite 57% meeting RCOG criteria for VTE prophylaxis during pregnancy.
Conclusion: this audit identified delays and omissions in diagnosis and prophylaxis of VTE in admitted pregnant patients. Addressing thrombo-prophylaxis at time of discharge in this at risk group is a priority audit outcome.
References
NPEC 2019
MBRRACE UK 2021
RCOG green-top guideline 37a
RCOG green-top guideline 37b
Fig. (9.6) Percentage of pregnant patients admitted with working diagnosis of pulmonary embolism who received recommended standard of care in diagnostic imaging and anticoagulation during admission and on discharge
Conflict of Interest: None to declare
9.7 Are We On The Right Trach? A Tracheostomy Service QI in an Acute Teaching Hospital
Eimear McCormack (PT), Sarah Rowland (SLT), Siobhan Connors (CCO), Paul Tierney (Anaesthetics)
Tallaght University Hospital, Dublin
Both the Intensive Care Society and UK National Confidential Enquiry into Patient Outcomes & Death recommend implementation of tracheostomy MDT's. Our tracheostomy MDT consisted of anaesthetics, critical care outreach, clinical specialist PT and SLT. Weaning assessments/goals, tube changes and decannulation plans were discussed with outcomes documented in health care records (HCR). Following ethical approval, 14 HCRs were retrospectively reviewed for baseline data. This was compared to 14 HCRs post MDT. Inclusion: tracheostomy to facilitate mechanical ventilation (MV) wean. Exclusion: tracheostomy for airway obstruction. We studied the effect of weekly MDT meetings on; (i)time to decannulation, (ii)evidence of MDT weaning assessments/plans, (iii)MDT goal setting, (iv)compliance with 30-day timeframe for tube changes and (v)time to wean from MV. Our findings included; decrease in median time to decannulation from 42 to 26 days. Increase in compliance with documented weaning assessment/plans and goal-setting from 0 to 100%. Decrease in median time to wean from MV from 12 to 8 days. Pre MDT compliance with tube changes was 58.3% with 41.7% failing to meet 30-day timeframe with no medical cause documented. Post MDT 57.2% of patients were decannulated before the 30-day tube change timeframe. The remaining 42.8% had tube changes on-time or delayed with medical cause.
References
Cameron, T., McKinstry, A., Burt, S., Howard, M., Bellomo, R., Brown, D., Ross, J., Sweeney, J., & O’Donoghue, F. (2009). Outcomes of patients with spinal cord injury before and after introduction of an interdisciplinary tracheostomy team. Critical Care and Resuscitation, 11(1), 14–19.
Cetto, R., Arora, A., Hettige, R., Nel, M., Benjamin, L., Gomez, C., Oldfield, W. & Narula, A. (2011). Improving tracheostomy care: a prospective study of the multidisciplinary approach. Clinical Otolaryngology, 36(5), 482–488.
de Mestral, C., Iqbal, S., Fong, N., LeBlanc, J., Fata, P., Razek, T., & Khwaja, K. (2011). Impact of a specialized multidisciplinary tracheostomy team on tracheostomy care in critically ill patients. Canadian Journal of Surgery, 54(3),167–172.
Hernández Martinez, G., Fernandez, R., Sánchez Casado, M., Cuena, R., Lopez-Reina, P., Sergio, Z., Luzon, E. (2009). Tracheostomy tube in place at intensive care unit discharge is associated with increased ward mortality. Respiratory Care, 54(12):1644–52.
Pandain, V., Miller, C., Mirski,., Schiavi, A., Morad, A., Vaswani R., Kalmar, L., Feller-Kopman D., Haut, E., Yarmus, L., Bhatti, N. (2012). Multi-disciplinary team approach in the management of tracheostomy patients. Otolaryngology Head Neck Surgery, 147(4): 684–691.
Santos, A., Harper, D., Gandy, S., & Buchanan, B. (2018). The positive impact of multidisciplinary tracheostomy team in the care of post-tracheostomy patients. Critical Care Medicine, 46(1), 591–591.
Conflict of Interest: None to declare
9.8 Do we appropriately investigate our patients with Pulmonary Emboli to stratify severity using local management guidelines?
Eimear Connolly1, Elaine Hayes1, Tidi Hassan1, Ian Counihan1
1 Our Lady of Lourdes Hospital, Drogheda
Pulmonary embolism (PE) is a common presentation to acute hospitals. A local guideline based on the 2019 European Society of Cardiology guidelines was developed to optimise management of patients diagnosed with a PE and guide thrombolysis administration. It uses laboratory and radiologic investigations along with vital signs to stratify patients with PE into non-massive, low risk submassive, high risk submassive and massive, to guide management accordingly, based on severity. This audit was preformed to assess adherence to this PE management guideline. All patients discharged with a diagnosis of PE over a 6-month period (January 2022- June 2022) were identified, and charts reviewed to assess if patients had the required investigations done namely echocardiogram, arterial blood gas (ABG), Troponin, BNP and lactate levels. 43 patients were initially identified. Only 25 charts were available for inclusion (n = 25). Of these patients, only 36% had an ABG (n = 9), 60% (n = 15) had an echocardiogram, 32% (n = 8) had a BNP test, while 72% (n = 18) had a troponin checked. Only 1 patient had all 5 relevant investigations done. This highlights a lack of awareness of how to stratify PE severity and the potential for improving management of PEs with increased education around this local guideline.
Conflict of Interest: None to declare
9.9 Profiling patient demographics and patient journey in an unscheduled care respiratory inpatient rehabilitation service using a prospective database
Elaine Cribbin1, Michelle Fitzgerald1, Lijji Joseph1, Patrick Mitchell1,, Catherine Slattery1, Sindhu Xavier1, Eddie Moloney1, Stephen Lane1, Minesh Kooblall1
1 Peamount Healthcare, Dublin
Peamount Healthcare is a specialist inpatient multidisciplinary rehabilitation unit. Capturing patient profile and service key performance indicators (KPIs) are essential to support service quality and development. The aim was to profile patient demographics and service provision using KPIs in a new database. A new prospective database was developed. All patients (N = 19) admitted for inpatient respiratory rehabilitation from acute hospitals over a one-month pilot-period were included. Data was analysed using MS excel. The most common presenting conditions were COPD (N = 8), post- covid (n = 6) and other (n = 5). Most patients were female (n = 11); average age 68.5 (range 42-82). Physiotherapy was the most common HSCP referral (n = 19), followed by dietician (n = 9) and occupational therapy (n = 8). The average Charlson Comorbidity Index was 4.8 (range 3-7), and Rockwood Frailty Score was 4.2 (range 2-7). Of the patients discharged home during the pilot (n = 10), the average LOS was 11 days (range 3-18). A new database was successfully implemented for a one-month pilot, resulting in the creation of a profile of both patient demographics and service provision in an inpatient respiratory rehabilitation service. Further analysis of data is warranted to discern missing data and staff experience of the database. It is planned to continue with the database and expand the measures included.
Conflict of Interest: None to declare
9.10 Hereditary Haemorrhagic Telangiectasia – an update on national numbers
Sean Landers1, Art Kelleher1, Laura Monaghan1, Emily O’Reilly1, Margaret Murphy2, Adrian P Brady1, Terence M O’Connor1
1 Mercy University Hospital, Cork
Since 2003, the Mercy University Hospital Cork has been the national centre for Hereditary Haemorrhagic Telangectasia (HHT) in Ireland. The centre’s mission is to identify as many affected individuals and families as possible and offer screening for and management of the condition’s many manifestations. Complications include vascular malformations (lung, brain, liver typically), epistaxis and telangectasia. Screening consists of clinical history, genetic testing, bubble study echocardiography, CT-Thorax and MRI Brain. The Mercy University Hospital has assessed 704 patients thus far through tracking index cases and their relatives, with 298 confirmed diagnoses of HHT.
It is estimated that only 10% of affected individuals with HHT have been formally diagnosed globally. We show that a large proportion of patient with HHT develop severe complications (epistaxis (360 patients) and pulmonary AVMs (300 positive bubble study echocardiograms) being the most common complications seen).
Conflict of Interest: None to declare
9.11 Improving Patient Preparation for Bronchoscopy and Endobronchial Ultrasound-guided Transbronchial Biopsy
Caoimhe McGarrigle1, Terence McManus2, Juan Pastrana3, Jason Wieboldt4, Jan Kara5, Frederick Opoko6
South West Acute Hospital, Enniskillen
Bronchoscopy and endobronchial ultrasound-guided transbronchial biopsy (EBUS-TBNA) is a procedure preformed for the diagnosis of chest pathology1. It is a carried out in the day procedure unit and elective patients should have investigations including bloods, electrocardiogram (ECG) and pulmonary function testing and a review of medications and comorbidities completed prior to procedure. It was noted that errors in patient preparation was leading to delays in procedures, late cancellations and medication errors. The aim of this quality improvement project was to reduce the mean number of errors per patient in patient preparation by half in 6 months to provide an efficient and safe service for patients.We retrospectively assessed the preparation of patients undergoing bronchoscopy ± EBUS-TBNA from January 2022 using 12 different parameters which should be addressed in procedure preparation. This baseline data showed a mean error per patient of 0.53. The baseline data was also used to identify patterns and barriers that prevent adequate patient reparation and targeted interventions were used to address these including introducing plans with laboratories for samples, feedback to consultants and the introduction of a check list locally.
Nakajima, T., Yasufuku, K. & Yoshino, I. Current status and perspective of EBUS-TBNA. Gen Thorac Cardiovasc Surg 61, 390–396 (2013).
Conflict of Interest: None to declare
9.12 A Comparison of the ERS 1993 and GLI 2012 Reference Equations for Spirometry in a Waterford Cohort of Patients
Rebecca Murphy1, David Collins1, Mark Rogan1, Susan Foley1
1.University Hospital Waterford
This study compared two different sets of reference equations for spirometry: the European Coal and Steel Community (ECSC) 1993 update and the Global Lung Function Initiative (GLI) 2012 reference equations for spirometry to investigate if there is an impact on disease classification. Retrospective spirometry data was collected from 250 subjects from the University Hospital Waterford patient database over an eight month period, subdivided by gender and grouped by age, with five age categories ranging from 36 to 85 years of age with 25 males and 25 females in each category. Predicted values and LLNs for FEV1, FVC, and FEV1/FVC ratio were then calculated using both sets of reference equations. The values were then analysed for each subject and was noted as correctly classified or misclassified.Under the ECSC reference values, 11.20% of male subjects and 7.20% of female subjects were misclassified using the FEV1/FVC ratio (18.4% of patients were misclassified as having disease). In subjects aged 56 to 85, there was a 24-30% rate of misclassification. We propose that GLI reference equations are universally adopted to standardise Spirometry testing in Ireland.
Conflict of Interest: None to declare
9.13 Investigating the Role of PPAR-δ in Mesenchymal Stromal Cell Therapy for Acute Respiratory Distress Syndrome
Courteney Tunstead, Myron Rebello, Karen English1
1 Maynooth University, Maynooth, Co. Kildare, Ireland
Mesenchymal Stromal Cells (MSCs) have attracted attention as a cell-based therapy due to their immunomodulatory and immunoregulatory properties. MSCs have been investigated in clinical trials for the treatment of a variety of inflammatory conditions, such as Acute Respiratory Distress Syndrome (ARDS). Studies have shown that MSCs require activation, known as ‘licensing’, in order to be considered efficacious. The disease micro-environment is thought to contribute; primarily due to the presence of pro-inflammatory cytokines, such as TNF-ɑ and IFN-γ, which are known to license MSCs. Other factors, such as the Peroxisome Proliferator-Activated Receptor delta (PPAR-δ), have shown to negatively regulate MSCs; reducing MSC efficacy in vivo upon stimulation. Free fatty acids (FFAs) are thought to be natural ligands for PPAR-δ and this implies that patients, or delivery sites, with abundant FFAs will show poor outcomes to MSC therapy. Interestingly, our data also demonstrates that antagonism of PPAR-δ can enhance MSC therapeutic efficacy. This study kick-starts a dialogue about factors that may be present in patients that can negatively affect cell-based therapeutic efficacy. To investigate the mechanisms associated with PPAR-δ modulation of MSC function, we have developed chemical antagonists, agonists and shRNA knockdown approaches and use an Acute Lung Injury (ALI) mouse model.
Conflict of Interest: None to declare
9.14 Why has the decline in smoking among Irish teenagers stopped?
Salome Sunday1 Joan Hanafin1, Luke Clancy1
1TobaccoFree Research Institute Ireland (TFRI), TU Dublin—FOCAS Research Institute
As reported previously1, in Ireland, teen smoking decreased from 41% in 1995 to 13.1% in 2015. But, in 2019, current smoking increased overall to 14.4% from 13.1% in 2015, with the prevalence in 2019 being greater in boys than girls (16.2%) vs (12.8%). This is a serious threat to Ireland’s tobacco endgame.We used data from 1,493 (2015 wave) and 1,949 students (2019 wave) of ESPAD Ireland (European School Survey Project on Alcohol and other Drugs) to analyse the use of tobacco products by teenagers and offer an explanation for the increase. Adjusted incidence risk ratios (IRR) for current smoking were estimated using a Poisson regression analysis.The increase in current cigarette smoking in 16-year-olds in Ireland between 2015 and 2019 was associated with an increasing use of e-cigarettes as well as with having peers who smoke. Lack of parental regulation increased the odds of smoking.
Our findings highlight the negative impact that increased youth e-cigarette use has had on current teenage cigarette smoking. We recommend extending tobacco control legislation regarding minors to include e-cigarettes. This may be desirable to prevent exposure to secondhand aerosol (SHA) but also because of the possible renormalisation of smoking.
Conflict of Interest: None to declare
Funders: RCDHT Grant No. 184; Department of Health ESPAD Grant.
1 Sunday S, Hanafin J, Clancy L. Increased smoking and e-cigarette use among Irish teenagers: A new threat to Tobacco Free Ireland 2025. ERJ Open Res 2021; (https://doi.org/10.1183/23120541.00438-2021).
Table 1 (9.14) Sample Characteristics, and Poisson regression results (Adjusted Incidence Risk Ratios - IRR) of factors associated with current smoking in 16-year-olds, based on data from the Irish 2015 & 2019 ESPAD Surveys Sample Characteristics Poisson Regression Results Current smoking (16-year-olds)
2015 n (%) 2019 n (%) Total Adjusted IRR (95% CI) Male Adjusted IRR (95% CI) Female Adjusted IRR (95% CI)
Year
2015 1493 –- 1 1 1
2019 –- 1949 0.91 (0.85, 0.98)* 0.91 (0.82, 1.00) 0.95 (0.86, 1.05)
Gender
Male 752 (51.1) 946 (48.5) 1
Female 720 (48.9) 1003 (51.4) 1.00 (0.93, 1.07) N/A N/A
Ever-used e-cigarettes
No 1088 (77.0) 1219 (62.7)* 1 1 1
Yes 325 (23.0) 723 (37.2) 1.27 (1.16, 1.40)* 1.33 (1.17, 1.51)* 1.27 (1.11, 1.45)*
Current e-cigarette use
No 1270 (89.9) 918 (84.5)* 1 1 1
Yes 143 (10.1) 351 (18.1) 1.45 (1.31, 1.61)* 1.39 (1.21, 1.60)* 1.49 (1.26, 1.75)*
Household composition
Single parent 262 (17.8) 371 (19.0) * 1 1 1
Two parents 1109 (75.3) 1490 (76.4) 0.90 (0.83, 0.98)* 0.88 (0.78, 0.99)* 0.93 (0.82, 1.05)
Blended family 101 (6.9) 88 (4.5) 0.84 (0.71, 0.99)* 0.87 (0.68, 1.11) 0.81 (0.64, 1.03)
Parental Regulation
Know always 906 (62.7) 1194 (63.2) 1 1 1
Know quite often 337 (23.3) 455 (24.1) 0.97 (0.89, 1.06) 0.97 (0.86, 1.09) 0.97 (0.86, 1.10)
Know sometimes 128 (8.9) 166 (8.8) 1.16 (1.04, 1.30)* 1.15 (0.98, 1.34) 1.16 (0.98, 1.38)
Usually don’t know 73 (5.1) 74 (3.9) 1.25 (1.07, 1.45)* 1.28 (1.06, 1.53)* 1.23 (0.92, 1.63)
Peer smoking
None
A few/some
Most/all
478 (33.4)
802 (56.1)
150 (10.5)
558 (29.8)
1125 (60.1)
188 (10.1)
1
1.00 (0.92, 1.08)
1.20 (1.06, 1.35)*
1
0.99 (0.89, 1.11)
1.08 (0.92, 1.28)
1
1.01 (0.90, 1.13)
1.33 (1.13, 1.57)*
*Statistical significance at p < .05
15 Factors Associated with E-Cigarette Use in Young Adults
Salome Sunday1, Luke Clancy1, Joan Hanafin1
1TobaccoFree Research Institute Ireland (TFRI), TU Dublin – FOCAS Research Institute
The increase in e-cigarette ever and current use among young people represents a growing concern for tobacco control efforts in Ireland. Identifying factors associated with increasing prevalence offers the opportunity to develop appropriate preventative interventions.Using cross-sectional data from 5,190 20-year-olds in the Growing up in Ireland (GUI) survey (Wave 4, 2019), we performed stepwise logistic regression analyses to examine e-cigarette prevalence and factors associated with e-cigarette use in this age group. The prevalence of e-cigarette ever-use was 47.8% and current use was 13.4%. E-cigarette ever and current users were significantly more likely to be male (AOR 1.63 vs 1.71 for females), ever-smokers (AOR:2.99 vs AOR:5.25), current smokers (AOR:1.68 vs AOR:3.22), and ever cannabis users (AOR:2.28 vs AOR:1.97). E-cigarette ever-users were more likely to ever-use alcohol (AOR:3.65) and non-prescribed drugs (AOR:1.44). E-cigarette ever and current use was inversely associated with high leaving certificate points (ever-use: AOR 0.30; current use: 0.49) and being in two-parent families (current use AOR: 0.61).Ever and current e-cigarette use was high among 20-year-olds in this study. Our findings suggest that poly-drug use needs to be addressed simultaneously with e-cigarette prevention strategies to be successful.
Table 1 (9.15) Multivariable (stepwise) logistic regression of e-cigarette ever and current use among young adults N (%)
N = 5190 Ever used e-cigarettes
N = 2460 (47.8%) Current e-cigarettes use
N = 690 (13.4%)
AOR (95% CI) AOR (95% CI)
Gender
Female
Male
2548 (49.10)
2642 (50.90)
Reference
1.63 (1.37, 1.94)
Reference
1.71 (1.30, 2.25)
Religion
None
Catholic
Other Christian
Other
1502 (28.99)
2831 (54.65)
758 (14.63)
89 (1.72)
Reference
-
-
0.85 (0.72, 1.01)
-
Leaving Cert points
Low Achieving
Medium Achieving
Medium High Achieving
High Achieving
1110 (23.32)
1438 (30.21)
1439 (30.23)
773 (16.24)
Reference
0.58 (0.45, 0.76)
0.45 (0.35, 0.58)
0.30 (0.23, 0.40)
Reference
-
0.67 (0.49, 0.92)
0.49 (0.32, 0.75)
Ever smoked
No
Yes
1344 (26.10)
3807 (73.90)
Reference
2.99 (2.36, 3.78)
Reference
5.25 (2.66, 10.34)
Current smoking
No
Yes
3204 (62.20)
1947 (37.80)
Reference
1.68 (1.38, 2.04)
Reference
3.22 (2.40, 4.31)
Ever alcohol use
No
Yes
185 (3.58)
4967 (96.42)
Reference
3.65 (1.54, 8.62)
-
Ever tried cannabis
No
Yes
1992 (39.46)
3055 (60.54)
Reference
2.28 (1.88, 2.76)
Reference
1.97 (1.41, 2.74)
Ever used non-prescribed drugs
No
Yes
3673 (71.55)
1460 (28.45)
Reference
1.44 (1.14, 1.82)
-
Household type
Single
Couple
1139 (23.31)
3748 (76.69)
- Reference
0.61 (0.43, 0.85)
Conflict of Interest: None to declare
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
Telehealth
10.1 A Stimulating Review: Use of Video Consultation in the Management of Narcolepsy
Louise Byrne1, Lisa Slattery1, Deirdre O’Rourke1, Barry Kennedy1, Brian Kent1
1St. James’s Hospital, Dublin
Narcolepsy is an uncommon chronic sleep disorder, characterised by excessive daytime sleepiness, cataplexy, and fragmented nocturnal sleep. The majority of people with narcolepsy require ongoing pharmacotherapy with frequent adjustment of agent and/or dosage. A video assessment service was established to address this and increase access to our national centre for all patients.We retrospectively analysed narcolepsy patient data from a two-year period in a National Referral Centre for Narcolepsy in Ireland. The aim of this study was to establish patient characteristics, and identify trends in medication use.
This study included 52 patients, and 77% had a diagnosis of Type 1 narcolepsy (narcolepsy with cataplexy). 33% had documentation of receiving the Pandemrix H1N1 flu vaccination prior to symptom development. 98% of patients attending were on at least 1 stimulant medication, and 54% of patients were on 2 or more. The most commonly prescribed medication was Concerta XL in 44% of patients, and was concomitantly prescribed with Modafinil in 26%, and Pitolisant or Venlafaxine in 30% of cases respectively. The majority of patients reported significant improvements in quality of life.
This highlights the utility of digital health in the ongoing management of patients with a rare disease in an Irish healthcare setting.
Table 1 Patient Demographics and Medication Use in Narcolepsy Patients Demographics No. (%) of patients
Sex Female 59%, Male 41%
Mean Age 31 ± 12.7
Received Pandremix H1N1 Flu Vaccine 33% (N = 17)
Type 1 Narcolepsy (With Cataplexy) 77% (N = 40)
Type 2 Narcolepsy 23% (N = 12)
Medications No. (%) of patients prescribed medication(s)
Modafinil 21% (11)
Concerta XL 44% (N = 23)
Pitolisant 28.8% (N = 15)
Venlafaxine 25% (N = 13)
Ritalin 25% (N = 13)
Sodium Oxybate 3.9% (N = 2)
% Patients on 1 medication 44% (N = 12)
% Patients on 2 medications 30% (N = 16)
% Patients on 3 medications 23% (N = 12)
Conflict of Interest: None to declare
10.2 Early Experiences of a Pilot Remote-Access Obstructive Sleep Apnoea Video Clinic in a Tertiary Referral Centre
Louise Byrne1, Peter Coss1, Orla Gavin1, Eileen McGinnis2, Una Kearns3, Brian Kent1
1. St. James’s Hospital, Dublin,2ResMed PEI,3MyPatientSpace
Obstructive Sleep Apnoea (OSA) is a highly prevalent sleep disorder characterised by repeated episodes of airway obstruction during sleep. The global burden of sleep disordered breathing is enormous, affecting > 10% of adults worldwide. Access to sleep clinics is limited, with long wait times for assessment.To improve access to care, an online OSA video clinic was launched in St. James’s Hospital in July 2022 for new patients. Prior to video assessment, patients downloaded an app and completed a variety of sleep questionnaires. Home testing was conducted via a single use diagnostic device (WatchPAT-One), and data was reviewed via a cloud-based platform before patients received a video consultation through the app. An initial cohort of patients (N = 8; Table 1) were triaged for suitability, requiring internet access and basic technological proficiency. Average wait time from referral to clinic day was 2.6 months, compared to 7.25 months for in-person OSA clinic. 75% of patients received a diagnosis of OSA on their first visit, and of these, 67% were commenced on CPAP therapy.These early results are promising, and indicate that remote access video clinics can aid in decreasing wait times, increase access to healthcare, and hasten commencement of CPAP therapy in OSA patients.
Table 1 (10.2) Patient demographics Demographics Result
Sex Male 50%, Female 50%
Age 43 ± 7
BMI 33.6 ± 7.3
Apnoea-Hypopnoea Index (AHI) 17 ± 16.8
Commenced CPAP 50%
Wait Time 2.6 months ± 0.70
Conflict of Interest: None to declare
10.3 The ‘SATELITE Project’ – Sleep Apnoea TELephone IniTiativE. A Quality Improvement Study
Peter Corry, Frederick Okpoko, Caoimhe McGarrigle, Jan Kara, Juan Pastrana, Jason Wieboldt, Terence McManus
Western Health & Social Care Trust, Northern Ireland, South West Acute Hospital, Enniskillen
In January 2020, over 400 patients from the Southern Section of the Western Trust were awaiting assessment for Obstructive Sleep Apnoea (OSA). Some had been waiting for almost 4 years. We implemented a quality improvement approach to evaluate existing systems and generate potential solutions: the rate-limiting step in our existing model was insufficient slots available in physical outpatient clinics; we devised a new phone consultation pathway aiming to reduce our waiting list by 100 patients within 6 months.The Project consisted of 4 Plan Do Study Act (PDSA) cycles: 1) Devise a standardised phone consultation template for use by non-consultant doctors to safely assess patients via phone; 2) Transpose the template to an electronic document; 3) Training sessions for non-consultant doctors; 4) six month trial of ad hoc phone assessments.After 6 months, despite regular new referrals, we had reduced our waiting list to 127 patients. Qualitative analysis via random survey of patients revealed high satisfaction with the pathway. Phone consultations are safe, effective and time efficient for the assessment of OSA, enhancing patient care by dramatically reducing waiting times. They necessitate minimal additional financial costs as they utilise existing resources. The model is potentially transferable to other common referrals.
Fig. (10.3) Run Chart showing Number of Patients awaiting assessment for Obstructive Sleep Apnoea (OSA)
Conflict of Interest: None to declare
10.4 An evaluation of patient outcome measures comparing hospital based pulmonary rehabilitation (PR) with virtual pulmonary rehabilitation at Tallaght University Hospital (TUH)
Elaine Joyce1, Ciara Scallan1, Judith Maxwell1, Sara Keane1, Avril Kilkelly1, Alanah Quinsey1, Eddie Moloney1, Minesh Kooblall1
1Tallaght University Hospital
Pulmonary rehabilitation (PR) is the most cost effective intervention for reducing exacerbations and hospital admissions in COPD and other respiratory conditions. Studies which have compared virtual PR to traditional face to face (F2F) hospital based classes have found virtual PR to have similar clinical improvements in the 1 min Sit To Stand (1 min STS) and disease specific outcome measures.Virtual PR was introduced in TUH in June 2020 as a response and solution to Covid 19 restrictions, and F2F PR was reintroduced in Sep 2021 when restrictions were eased. A comparison of patients who completed greater than 12 classes of F2F PR (n = 28) and virtual PR (n = 43) in TUH found that patients who completed F2F PR had a greater improvement in outcome measures when compared against the minimal clinical importance difference (MCID) than those who completed virtual classes. Ninety-six per cent of F2F participants achieved the MCID for 1 min STS test, compared with 79% in virtual PR. Further, 59% of F2F participants achieved the MCID in the COPD Assessment test, compared with 50% in virtual PR. Moreover, 52% of F2F participants achieved the MCID improvement in the Modified Medical Research Council Dyspnoea Score in the F2F class, compared with 29% improvement in virtual PR.
Conflict of Interest: None to declare
10.5 A remote monitoring project for post lung transplant recipients using a patientMpower digital mobile application (pMp app)
1Lourdu Anitha Anthonycruze, 1Sara Winward, 1Patricia Ging, 1Michelle Murray
1. Mater Misericordiae University Hospital, Dublin.
Background: Lung transplant recipients in Ireland receive lifelong follow up at the National Lung Transplant Centre. The focus of this project was the early identification of complications post-transplant, using pMp app. The project commenced in March 2021 with funding from HSE Slaintecare Integration Fund.
Methods: From March 2021 to March 2022, 49 patients were enrolled. There are currently 37 active participants. Bluetooth enabled handheld spirometer, pulse oximeter, weighing scales and blood pressure monitor were provided and linked to the pMp app. An algorithm was developed to standardise review of patient data. A nurse monitored data during office hours. Service evaluation was completed between August 2021 and March 2022.
Results: More than half of the participants (51%; n = 19) remained stable at home. Among the remaining (49%) who had episodes of decline in lung function or other concerning readings, 22% (n = 8) were found to have infection; 3% (n = 1) with rejection; 8% (n = 3) were false alarms due to poor technique. 16% (n = 6) had other issues such as hypo/hypertension and weight gain/loss. Six patients self-withdrew. Three monthly satisfaction questionnaires were issued to all active participants. In March 2022 there were 28 responses (75.6%). 96% of the participants found pMp app easy to use; 82% felt that the app gave them confidence and reassurance; 100% supported continuation of the service.
Conclusion: Remote monitoring was successfully introduced. It demonstrated a positive impact on patient experience and the clinical care with the early identification of decline in lung function allowing timely investigation.
Conflict of Interest: None to declare
10.6 Remote Forced Vital Capacity (FVC) Monitoring in Patients with Connective Tissue Disease related Interstitial Lung Disease (CTD-ILD)
Wan Lin Ng1, Anja Schweikert2, Donough Howard1,2, Laura Durcan1,2, Killian Hurley2
1 Beaumont Hospital, Dublin, 2 RCSI, Dublin
Background: Interstitial lung disease (ILD) is an essential comorbidity to address in patients with connective tissue diseases (CTD). We aim to determine if home spirometry monitoring can detect early progressive fibrosing ILD in CTD-ILD resulting in timely treatment leading to improving mortality and morbidity.
Method: Patients with CTD-ILD, idiopathic pulmonary fibrosis (IPF) and familial pulmonary fibrosis (FPF) were recruited and followed prospectively for 12 months. Participants were given a handheld spirometer and oximeter linked to a real-time electronic health journal via a smartphone app. The highest readings of the day are used for analysis.
Results: 113 patients were recruited; 62 CTD-ILD, 31 IPF and 21 FPF with median age of 66, 71 and 69 years respectively. 47.8% were males. Preliminary data on 93 patients demonstrated median forced vital capacity (FVC) predicted were 79.55% in the CTD-ILD, 77.70% in IPF and 78.88% in FPF cohorts. The median oxygen saturation (SpO2) was 96% the CTD-ILD and FPF, and 95% in the IPF cohort.
Table 1 (10.6) Distribution of FVC and SpO2 readings between males and females in patients with interstitial lung disease CTD-ILD IPF FPF
Male Female Male Female Male Female
No. of Patients 23 28 17 8 7 10
Median Age (years) 71 63 73 65 62 69
No. of FVC Readings 1103 1640 1203 725 259 569
Median FVC (L) 3.00 2.05 2.76 2.32 3.00 1.61
Median FVC Predicted (%) 74.66 84.17 72.67 91.08 72.52 79.29
No. of SpO2 Readings 1124 1612 1135 831 255 617
Median SpO2 (%) 96 96 95 96 96 96
Conclusion: Patients with CTD-ILD were younger and have a higher percentage predicted FVC. There were distinct differences in the FVC between men and women with ILD in all categories. Monitoring FVC remotely is feasible and acceptable to patients with CTD-ILD despite impaired hand function.
Conflict of Interest: None to declare
10.7 Patient Appraisal of New Innovative In Car Respiratory Service
A.M. Curran, A.M. O’Connell, T. Quadri
Naas General Hospital, Kildare
An In Car Respiratory Service was set up in Naas General Hospital to provide patients who were already known to the respiratory services an efficient and safe way to avail of certain pulmonary function tests. Patients attending had numerous conditions including COPD, asthma, goitres and sleep investigations.The service involved driving to a designated area in the car park. Spirometry, FeNO testing and overnight oximetry services were provided. The patient remained in their car during testing.
Methods: The last 50 service users were contacted and asked to appraise their experience. The questions were as follows:
Table 1 (10.7) Questions asked Q1 Easier? 50 100%
Q2 Safer? 50 100%
Q3 Quicker? 50 100%
Q4 Use again in future? 50 100%
Q5 Drive yourself? 35 72%
Q6 If no, easy to find transport? 14 93%
Q7 Prefer in hospital appointment next? 3 6%
Results: 100% of patients thought the service was—a) felt safer b) more convenient patient experience c) quicker than attending laboratory. 100% would use service again.72% of patients drove themselves. 93% found it easier/convenient transport wise. 6% (3 patients) would prefer to attend in hospital for future testing.
Conclusion: This survey concluded that patients found it safer, more convenient and easier to use than in lab services.
Conflict of Interest: None to declare
(10.8 Not for Publication—The effects of a 6-week virtual COVID19 recovery programme on exercise capacity, fatigue scores and quality of life in individuals recovering from COVID-19)
10.9 COVID-19 Remote Monitoring Programme in Our Lady of Lourdes Hospital Drogheda
Michelle Uno, Aisling O’Connor, Sarah Farrell, Tidi Hassan
Our Lady of Lourdes Hospital, Drogheda
The COVID-19 Remote Monitoring (RM) programme was developed as a response to the increased demands on the acute care services, to provide a supported discharge option to the patients with COVID-19 who can be safely managed at home with additional home monitoring. PatientMpower smartphone application for COVID-19 links to a portable pulse oximeter and is used to monitor oxygen saturation, pulse rate, breathlessness, and other symptoms. Information captured in the application is immediately available to the RM team via a secured patient data portal and the team can make an appropriate decision about the ongoing care of patients being monitored at home. Between April 2020 and the end of December 2021, a total of 386 patients were enrolled in the program, and a total of 39 patients were readmitted to the hospital via fast-track access to the Respiratory team. Remote monitoring is an effective service in controlling hospital capacity when there is an increased demand in hospital beds. It improves patient flow by reducing the length of stay and avoiding unnecessary hospital admissions and identifying patient deterioration at an earlier stage reducing the need for ICU admissions.
Conflict of Interest: None to declare
10.10 The use of remote electronic spirometry to assess patient adherence to peak flow monitoring in a tertiary teaching hospital
Finbarr Harnedy1, Deirdre Long1, Dorothy Ryan1
1 Beaumont Hospital, Dublin
The use of peak flow measurements is a core tenant of asthma management but adherence levels to peak flow diaries are often low. Self-monitoring is the cornerstone of chronic disease management with studies suggesting an increase in adherence to self-monitoring may be linked with increased treatment adherence. New technologies have enabled remote monitoring of patient symptoms, adherence to medication, assessment of technique and provide objective measures of disease such as peak flow readings and at home spirometry readings. In this study we analysed patient adherence to remote handheld spirometry monitoring over a 90-day period. Data from 17 patients, 71% female, were examined. Mean ± SD age was 45 ± 18 years (Range 17 – 74). 71% (n = 12) recorded a spirometry reading greater than 66% of the time. Mean number of days where patients recorded one or more readings was 67 ± 27% with a median value of 77%. In the month of July 2022, 97% of spirometry reading were recorded in less than 2 min. Home monitoring of patient spirometry is an effective, well tolerated, and convenient way to monitor patient ability to effectively perform peak flow recordings and monitor response to treatment.
Conflict of Interest: None to declare
10.11 VIRTUAL PULMONARY REHAB VERSUS PHYSICALLY ATTENDING PULMONARY REHAB IN A RURAL AREA
Dr. John Kiely, Mary Osborne, Christina Tobin, Bryan Jacob, Filipe Afonso, Doireann O Donovan and Catherine O Hanlon
Mallow General Hospital, Cork
In 2020, we started up a new Pulmonary Rehab(PR) programme for patients in North Cork, as with any new service there are challenges but COVID restrictions were in addition. Due to COVID 19 and public health restrictions the delivery of the programme had to change!In October 2020, 6 patients participated in Virtual PR using the Webex platform. The target population in North Cork is rural and a significant cohort of patients availing of the service are over 65, with limited IT literacy and access to internet. Unlike the city catchment area for PR, internet connection proved to be problematic but was overcome by the support of family, neighbours, landline telephone support and an exercise You tube video and exercise booklet. Due to the HSE Cyberattack the second cycle of virtual PR had to pivot back to physically attending PR. Overall, the virtual programme displayed similar clinically significant outcome measurements to our physically attending PR programme. However, following a qualitative approach to data collection, the results indicated that the majority of patients preferred physically attending PR, except for a small number of participants. To conclude the patient’s satisfaction survey was very positive for both programmes.
Conflict of Interest: None to declare
10.12 National Rollout of Virtual Pulmonary Rehabilitation
Angela Ryan, Susan Curtis
HSE, National Clinical Programme for Respiratory
Pulmonary Rehabilitation (PR) is a priority service for the National Clinical Programme for Respiratory (NCPR) in the management of Chronic Obstructive Pulmonary Disease (COPD). Prior to March 2020, PR was routinely delivered in in-person in-hospital / health-centre settings. In March 2020, these services ceased to provide in-person PR due to the onset of COVID-19. NCPR aimed to provide services in line with the Models of Care for COPD and Asthma, while adhering to Public Health Guidance. The NCPR developed clinical guidance to support the delivery of online / virtual services, secured the appropriate clinical expertise to approve the guidance (through the Irish Thoracic Society) and worked with national professional bodies to support implementation of the guidance.Health Care Professionals who provided online PR identified factors that facilitated and those that challenged the online service delivery. Facilitators included: management and colleague support, prompt publication of the national clinical guidance document among several internal and external factors. Challenges included a range of technological and personal factors.NCPR are updating the national clinical guidance document, in line with current international clinical guidelines.NCPR are engaging with the Enhanced Community Care and eHealth Programmes regarding current and future provision and procurement.
Conflict of Interest: None to declare
10.13 Evaluating the Effectiveness of Home-based Pulmonary Rehabilitation with Telephone Support for Patients with Chronic Respiratory Disease: A Service Review
Orla Threadgold1, Eve Stanley1, Danielle Moynihan1
Chronic Disease Management Hub, Carlow/Kilkenny1
Home-based pulmonary rehabilitation with telephone support has been suggested as an alternative rehabilitation model for patients with a chronic respiratory disease who, for a variety of reasons, cannot attend centre-based pulmonary rehabilitation and do not have access to appropriate information technology devices to participate in virtual pulmonary rehabilitation [1] 1.In March and April 2022, 67 patients on a pulmonary rehabilitation waiting list were offered either virtual pulmonary rehabilitation or a home-based programme with exercise support.Of the 27 patients that opted for home-based pulmonary rehabilitation, 9 completed an 8-week programme and 2 were lost to follow-up. Patients had a diagnosis of COPD (n = 4) and ACOS (n = 3) and had a mean age of 74.14.
Patients received a home-based exercise programme consisting of progressive aerobic and strengthening exercises and a weekly telephone call from a Physiotherapist to monitor and progress the programme. Patient outcomes were measured with the CAT and mMRC. Following the 8-week programme, the mean CAT score increased by 5.5 points, indicating a worse outcome, however the mean mMRC score reduced from 1.5 to 1.1.The results of this service review indicate that although there is an emerging role for home-based pulmonary rehabilitation as another form of service delivery, further evidence is required to establish the most effective model for patients with chronic respiratory disease.
Conflict of Interest: None to declare
References
Holland, A.E., Cox, N.S., Houchen-Wolloff, L., Rochester, C.L., Garvey, C., ZuWallack, R., Nici, L., Limberg, T., Lareau, S.C., Yawn, B.P. and Galwicki, M., (2021) ‘Defining modern pulmonary rehabilitation. An official American thoracic Society workshop report’, Annals of the American Thoracic Society, 18(5), pp.e12-e29.
10.14 Experiences and Perceptions of Telemedicine in Patients Living with Advanced COPD
Úna O’Connor1, Geraldine Crilly1, Claire Quinn3
St. John’s Hospital, Limerick, 3.University of Galway
Chronic Obstructive Pulmonary Disease (COPD) is a chronic, progressive respiratory condition. It results in increasingly debilitating shortness of breath, which impacts greatly on the daily lives of patients in advanced stages of this life-limiting illness. Telemedicine is increasingly deployed to address significant challenges, including practicalities of attending face-to-face consultations, disease progression and functional decline.There is limited published research to date exploring COPD patients’ experiences of telemedicine as a remote healthcare approach.This qualitative evidence synthesis (QES) explored experiences and perceptions of telemedicine for patients living with advanced COPD. Systematic literature search of twelve electronic databases was undertaken.Independent screening was undertaken by two authors. Quality appraisal was conducted using Critical Appraisal Skills Programme (CASP). Framework synthesis was applied using the COM B model. Qualitative findings were synthesized using Thomas and Harden’s thematic synthesis method. GRADE-CERQual assessed confidence in review findings.
Eight studies were selected for synthesis. Four analytical themes were generated (1) Ownership and Control (2) Virtual Relationships/Being Heard (3) Telemedicine: Benefit versus Burden (4) Trust in Virtual Health Services. The lived experience of COPD patients and their engagement with telemedicine may serve to inform policymakers in the delivery of remote healthcare services. Further research on patient and healthcare professionals’ views on using telemedicine in the palliative stage of COPD is valuable.
Conflict of Interest: None to declare
There is limited published research to date exploring COPD patients’ experiences of telemedicine as a remote healthcare approach.This qualitative evidence synthesis (QES) explored experiences and perceptions of telemedicine for patients living with advanced COPD. Systematic literature search of twelve electronic databases.
10.15 The design process and development of a digital solution which uses respiratory rate as a key value in the monitoring of patients with Chronic Obstructive Pulmonary Disease (COPD) in a Community Virtual Ward led by an Advanced Nurse Practitioner: Providing a bespoke hospital avoidance solution using digital technology to support community based care
Antoinette Doherty 1, Prof Vera Keatings 1, Gintare Valentelyte 2, Myles Murray 3
1 Letterkenny University Hospital, Co Donegal, 1 University of Galway Medical Academy,2 RCSI University of Medicine and Health Sciences, Dublin,3 PMD Solutions, Cork
Background: Individuals with exacerbating COPD display a pattern of repeated illness culminating in hospital admission. A community virtual ward (CVW) with a bespoke platform that incorporated respiratory rate trends was designed, implemented and offered as an alternate to inpatient care in Co Donegal between 16/05/2022 – 16/08/2022 to 15 eligible individuals with COPD.
Aims: To explore the feasibility of a proof of concept CVW which aims to support individuals with COPD become partners in their disease management.
Methods: Patients with COPD were admitted to the CVW for remote monitoring, optimisation of existing care plan and provision of a rescue script for suitable individuals. The objective and subjective patient data was reviewed daily by the advanced nurse practitioner. When a trend from normal became apparent the patient was assessed and necessary support was provided virtually. If applicable their rescue script was activated.
Results: Patient data from10 admitted individuals was used in the final report. Six individuals had one or more exacerbations with a total of 19 courses of antibiotics and 8 courses of steroids being dispensed to participants during the study. The average length of stay in CVW was 74 days with no hospital admissions recorded. An increase individuals understanding of their COPD was demonstrated also.
Conflict of Interest: None to declare
10.16 ‘Wellbeing via dance caught my mind’: A qualitative study exploring health impacts of online dance classes among patients with pulmonary fibrosis
Vikram Niranjan1, Giampiero Tarantino1, Jaspal Kumar2, Nicola Cassidy3, Liam Galvin3, Gemma O'Dowd3, Tracey Barnes4, Finola O'Neill3, Matthew Cullen3, Ray O’Connor5, Andrew O’Regan5
1 UCD,2 National University of Singapore, 3 Irish Lung Fibrosis Association, Dublin,4Dancing for Health CIC, Derbyshire,5.University of Limerick,
Dance is considered to be an excellent physical activity (PA) to improve physical and mental health outcomes among various groups of patients. This qualitative study aimed to investigate the perspectives of adults living with pulmonary fibrosis (PF) who had participated in a dance intervention.
A group of patients with PF, members of the Irish Lung Fibrosis Association, participated in 75 min online dance intervention for eight weeks delivered by an experienced choreographer. An exploratory qualitative study using thematic analysis of semi structured interviews was carried out to understand feasibility as well as health and wellbeing impacts of dancing.Eight participants (6 Female, 2 male; mean age 72.3 years) were recruited. Four key themes emerged: 1) Dance is fun – we’re not exercising 2) Improved sense of wellbeing 3) Positive impact of own online social space 4) Connecting dance impacts to clinical health. Overall, our virtual dance intervention was acceptable, enjoyable, preferable, and feasible among participants; who strongly perceived health benefits especially breathing efficiency and mental health improvements for managing their day-to-day struggles with PF.Emerging themes could influence the development and evaluation of dance as an alternate form of PA for patients with PF, exploring its benefits and sustainability.
Keywords: Dance, wellbeing, quality of life, community-engaged research, chronic disease, pulmonary fibrosis, pulmonary diseases
Authors Contributions: VN conceptualized supervised and obtained funding for the study. VN, NC, LG, GOD, ROC AOR developed the methodology. GT JK and VN performed data collection and analysis. GT and VN wrote the manuscript with input from TB, NC, LG, GOD, FON, MC, ROC AOR and JK. All authors have seen and approved the final version of the manuscript for publication.
Ethics approval and consent to participate: Ethical approval for this study was granted by the institutional Human Research Ethics Committee – Sciences UCD with reference no. LS-21–94-Niranjan and all participants provided written informed consent.
Conflict of interests: No author has any conflict of interest
Irish Thoracic Society Poster Review and Discussion
Friday 2nd December 2022
11. Oral Presentation
11.1 Use of digital measurement of medication adherence and lung function to guide the management of uncontrolled asthma: The INCA Sun randomized clinical trial
Elaine Mac Hale1, presenting author on behalf of the INCA Sun randomized clinical trial work group
1. RCSI, Dublin, Ireland
Disclosures
Grant funding: Health Research Board DI59, Medical Research Council, MR/M016579/1 (RASP), and ILP-HSR-2017-005 2210, INTEREG Europe (CHITIN CHI5436/18) and Investigator Initiated project grant from GSK.
Elaine Mac Hale has no potential conflicts of interests to disclose
Objective: Treatment decisions guided by digitally-acquired data on adherence, inhaler technique and peak flow were compared with current methods.Design: Patients >18 years with severe uncontrolled asthma were enrolled to a 32-week multicentre single-blind randomized clinical trial (INCA Sun) comprising a 1-week run-in period, 3 education visits over 8-weeks, and 3 treatment adjustment visits over 24-weeks in ten asthma clinics.
Intervention: The active group had personalized biofeedback on inhaler adherence, technique, and PEF. Treatment decisions were informed by digital data. The control group had adherence coaching, inhaler training, and an action plan. Treatment was adjusted based on pharmacy refill rates, asthma control, and risk of exacerbations. Both groups used a digitally-enabled inhaler and digital PEF.
Results: Of the 220 patients who consented to participate, 213 were randomized (control: 105; active: 108) and 200 completed the 32-week study. The mean age was 47[SD, 14.9] years, 137[64%] women participated. At week 32, 11(11%) active and 21(21%) control patients required add-on biologic therapy OR 0.42, 95%CI [0.189-0.95], p=0.038). Three of 19(16%) active and 11 of 25(44%) control patients who started on FP 500mcg/day were increased to 1,000mcg/day (OR 0.26; 95%CI [0.07–0.99], p=0.049). Twenty-six of 83(31%) active and 13 of 73(18%) control patients who started on FP 1000mcg/day were reduced to FP 500mcg/day, OR 2.11, 95%CI [1.01–4.74], p=0.047). Despite a lower treatment burden were no differences in asthma control, lung function, T2 inflammation, nor exacerbations between the two groups. No safety differences were observed.
Conclusions and relevance: Evidence-based care informed by digital data safely led to a significantly lower treatment burden.
Conflict of Interest: None to declare
11.2. A systematic assessment clinic for COPD reduces hospitalisation rates
Finbarr Harnedy1, Jack Allen1, Michele Cuddihy1, Eimear Ward2, Ciara Feeney2, Maeve Sorohan2, Imran Sulaiman1, Breda Cushen1
1 Beaumont Hospital, Dublin 9, 2 Beaumont Hub, Dublin North City and County CHO9
Hospitalisations for exacerbations of COPD reduce quality of life and increase mortality. Systematic assessment clinics for severe asthma reduce exacerbations. We examined the impact of a new systematic assessment clinic for severe COPD on hospitalisation rates.
At the COPD At-risk Assessment (CARA) clinic, COPD patients with high symptom burden and high healthcare use undergo multidisciplinary medical, nursing ANP, and physiotherapy assessment. Individualised COPD optimisation strategies are implemented. The number of hospitalisations in the 12 months pre-CARA and up to 6 months post-CARA was compared.
Data from 34 patients, 56% female, was examined. The mean age was 69(7.8) years. One-quarter had GOLD stage 4 COPD, 32% GOLD 3, 35% GOLD 2, 9% GOLD 1. One-third were prescribed ambulatory oxygen, 30% LTOT and 27% home NIV. Three patients died at 6 months. Of the remaining 31 patients, COPD hospitalisations reduced by 73% from 55 in the 12 months pre-CARA to 15 in the 6-months post. The mean hospitalisation rate was 1.77(1.84) pre-CARA and 0.48(0.89) post-CARA, p=0.0002. 64% of those with ≥1 hospitalisation in the 12 months pre-CARA had no further hospitalisations by 6-months post.A systematic assessment approach to outpatient COPD care has substantially reduced hospitalisations in this cohort of severe COPD patients.
Conflict of Interest: None to declare
11.3 The Benefit of a Liraglutide-Based Weight Loss Regimen Alone or in Addition to Standard CPAP Therapy on Metabolic Function, Atherosclerosis and Inflammation in Patients with Obstructive Sleep Apnoea - an Explorative, Proof of Concept Study
Cliona O'Donnell1,2, Shane Crilly3, Anne O'Mahony1, Jonny Dodd2,1, Donal O'Shea4, David Murphy2,1, Silke Ryan1,2
1 St. Vincent's University Hospital, Dublin, Ireland
2 University College Dublin, School of Medicine, Dublin
Obstructive sleep apnoea (OSA) is associated with cardiometabolic disease processes, but the benefit of continuous positive airway pressure (CPAP) therapy is uncertain. Here, we evaluated the effect of CPAP vs. a glucagon-like peptide (GLP)-1 (Liraglutide)-mediated weight loss strategy (LWR) across a range of cardiometabolic outcomes. These are data from a randomized proof-of-concept study (clinicaltrials.gov: NCT04186494). 30 patients with moderate to severe OSA without diabetes were randomised to CPAP, LWR alone or both in combination for 24 weeks. All patients underwent analysis of vascular inflammation by FDG-PET CT at baseline and at study completion, as well as evaluation of metabolic markers, anthropometrics, endothelial function and inflammatory markers.
30 subjects (50±7 years, 75% males, apnoea-hypopnoea index (AHI) 50±19/hr, body mass index (BMI) 34.8 ±3 kg/m2) completed the study. There was a significant decrease in BMI with LWR vs CPAP (-1.85±1.1 vs +0.86±1.5, p<0.01), while the reduction in AHI was greater with CPAP vs LWR (-39±24 vs. -15±23, p = 0.053). Greater improvements in metabolic outcomes occurred with both LBW and combination therapy, but only subjects on CPAP had a significant decrease in vascular inflammation.CPAP therapy improves vascular inflammation in OSA, while combination therapy and LBW have beneficial effects on metabolic function.
Conflict of Interest: None to declare
11.4 Evidence of mTORC1 pathway in Diffuse Idiopathic Pulmonary Neuroendocrine Cell Hyperplasia (DIPNECH)
Maylis Alquier1,2, Janet McCormack2, Marissa O’Callaghan3,2, Evelyn Lynn3,2, Orla O’Carroll4, David Murphy3, Rachel Crowley3, Dermot O’Toole3, Cormac McCarthy2,3, Aurelie Fabre 2,3
1. University of Limoges, France,2.University College Dublin,3.St. Vincent’s University Hospital, Dublin
Diffuse Idiopathic Pulmonary Neuroendocrine Cell Hyperplasia (DIPNECH) is a rare disease affecting essentially women. It is considered pre-neoplastic and is commonly associated with pulmonary carcinoids and less commonly in other lung malignancies. Histologically, there is a proliferation of neuroendocrine cells in the epithelium of the bronchioles (linear, nodular, tumourlets) with occasional peribronchial fibro-inflammatory changes.As the mTOR pathway is described to be involved in the development of pulmonary carcinoids, we explored downstream proteins RPS6 and 4EBP1 on DIPNECH and control lung tissue sections using immunohistochemistry.The cohort of patients included 15 women, aged 26-80. Five were symptomatic (cough, dyspnoea); two had MEN syndrome, 4 previous breast cancer, 11 carcinoids, 2 multiple tumourlets.7 controls lung tissue without DIPNECH from lobectomies sampled away from carcinoids, lung carcinoma and sarcoma metastases were included.4EBP1 is ubiquitously expressed in resident pulmonary epithelial cells and expression was observed in all cases with neuroendocrine cell hyperplasia and tumourlets. RPS6 expression varies within resident cells, and was expressed in the majority of neuroendocrine cells but not as diffusely as 4EBP1.This preliminary data shows that mTORC1 pathway protein are expressed in DIPNECH and might support the role of mTOR inhibitors in the treatment of symptomatic patients.
Conflict of Interest: None to declare
11.5 Investigation of the Utility of Exhaled Breath Condensate (EBC) as a Liquid Biopsy in the Detection of Spatial Genomic Heterogeneity in Patients with Early-Stage Non-Small Cell Lung Cancer (ESLC)
Robert Smyth1, 2, Simon Furney2, Siobhan Nicholson4, Katherine Sheehan2, Ronan Ryan4, Daniel Ryan2, 7 Liam Grogan7, Oscar Breathnach7, Patrick Morris7, Bryan Hennessy2, 8, Ross Morgan7*, Sinead Toomey2*
1 Boston Medical Center 2 Royal College of Surgeons in Ireland, 4 St.James's Hospital, Dublin, 7 Beaumont Hospital, Dublin
Spatial genomic heterogeneity is implicated in treatment resistance and recurrence and may lead to sampling bias in diagnostic biopsies. Liquid biopsy has the potential to overcome the challenges of tumour heterogeneity. This study investigated the potential of using EBC as a liquid biopsy to detect spatial heterogeneity in ESLC.Four quadrants of the resected tumour, a lymph node and normal lung tissue from 8 patients were analysed by whole exome sequencing. Patient specific panels of 5 mutations were designed based on the sequencing results. Cell free DNA (cfDNA) extracted from EBC and plasma, collected before surgery and at 6 and 12 weeks post-operatively, was analysed using digital droplet PCR.The median level of spatial genomic heterogeneity was 30.2% (IQR 11.4-57.8) raising the possibility of sampling bias in biopsy samples. 39/40 mutations (97.5%) were detected in pre-operative EBC samples compared to 40/40 (100%) mutations in plasma. A trend towards reductions in mutation copy numbers was observed in post-operative EBC and plasma. In 3/8 biopsy samples not all mutations were detected by ddPCR. Using this ultra-sensitive technology, almost all selected mutations were detected in EBC. Both liquid biopsies were superior to the original biopsy in the detection of spatial genomic heterogeneity (p=0.0001).
Conflict of Interest: None to declare
11.6 Circulating soluble P2X7 receptor as a novel indicator of resolution of inflammation in patients with cystic fibrosis
Azeez Yusuf1, Debananda Gogoi1, Claudie Gabillard-Lefort1, Cedric Gunaratnam1,2, Noel G. McElvaney1,2, Michelle Casey1,2 & Emer P. Reeves1
1 Royal College of Surgeons in Ireland,2, Beaumont Hospital, Dublin
In patients with cystic fibrosis (PWCF), loss of normal anion transport leads to impaired mucociliary clearance, chronic airway infection and inflammation. The P2X7 receptor (P2X7R) is an important regulator of inflammation and we have recently demonstrated the role of P2X7R in promoting ATP-induced inflammasome activation in CF monocytes. Moreover, published studies have shown increased soluble P2X7R (sP2X7R) protein levels in plasma of patients with sepsis, yet, the presence of sP2X7R in the circulation of PWCF has not been explored. The aim of this study was to assay sP2X7R levels in PWCF pre- and post-elexacaftor/tezacaftor/ivacaftor (ETI) CFTR modulator therapy, to assess the potential use of this protein as a circulating indicator for resolution of inflammation. We evaluated plasma samples prepared from whole blood of 35 PWCF eligible for ETI treatment (pre-ETI n = 17, 12 months post-ETI n = 15, including n = 6 paired samples of pre- and post-ETI treatment). sP2RX7 levels were assessed by sandwich ELISA.A significant reduction in levels of sP2X7R were detected in PWCF 12 months post-ETI therapy compared to sP2X7R levels pre-treatment (p<0.021, paired samples). Similarly, our results showed significantly higher levels of plasma sP2X7R in all patients pre-treatment compared to 12 months post-CFTR modulator treatment (p<0.04). This promising data suggests the use of sP2X7R as an indicator of resolution of inflammation in PWCF, with ongoing studies exploring sP2X7R levels in non-CF bronchiectasis and COPD patient groups.This project is supported by The US Cystic Fibrosis Foundation [Grant Number: REEVES21G0].
Conflict of Interest: None to declare
11.7 Results of the Irish Thoracic Society ( ITS) Bronchoscopy questionnaire in preparation for the National Bronchoscopy Quality Improvement project
Daniel Ryan 1, Marcus Kennedy2
1. Beaumont Hospital, Dublin, 2. Cork University Hospital
The National Bronchoscopy Quality improvement (QI) programme will be the fourth of its kind in addition to established QI projects in GI Endoscopy, Histopathology and Radiology. In preparation for the programme a questionnaire was circulated to public and private hospitals analyzing common trends in Bronchoscopy nationally. 19 hospitals (15 Public, 4 private) with Bronchoscopy units responded. 70% of institutions use the EndoRAAD Bronchoscopy reporting system with the reminder using Unisoft (18.75%) or other systems. Average waiting time for non- urgent essential ronchoscopy is 11.68 days and for suspected lung cancer is 4.10 days. Waiting time for EBUS-TBNA is an average of 7.33 days. Midazolam and Fentanyl are almost universally used for conscious sedation and use of reversal agents documented as an adverse event in every institution. 81.25% of facilities have onsite access to ICU while the remainder have access to immediate ambulance transfer for emergencies. 62.5% of units provide Registrar training with 93.75% supervised by consultants. This joint ITS/RCPI QI programme supported study reveals that waiting times for urgent and non-urgent bronchoscopy are within international recommendations and sedation and safety practices are broadly similar nationally. The questionnaire will help select key performance indicators (KPIs) for the QI programme moving forward.
Conflict of Interest: None to declare
11.8 (Not for Publication) Incidence and outcomes of pulmonary arterial hypertension in the Republic of Ireland List of Sponsors
Astra Zeneca
Aerogen
Air Liquide
ALK Abello
A Menarini
Beechfield Healthcare
BOC Healthcare
Boehringer Ingelheim
Direct Medical
GSK
Home Healthcare
Insmed
Janssen
MSD
Novartis
ResMed PEI
Respicare
Rocket Medical
Sanofi
Spire Medical
Sword Medical
Vertex
Viatris
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
| 36445628 | PMC9707199 | NO-CC CODE | 2022-12-03 23:15:56 | no | Ir J Med Sci. 2022 Nov 29; 191(Suppl 5):129-185 | utf-8 | Ir J Med Sci | 2,022 | 10.1007/s11845-022-03209-1 | oa_other |
==== Front
Cult Med Psychiatry
Cult Med Psychiatry
Culture, Medicine and Psychiatry
0165-005X
1573-076X
Springer US New York
9812
10.1007/s11013-022-09812-2
Clinical Case Study
Violence at Rikers Island: Does the Doctor Make It Worse? A Clinician Ethnographer’s Work Amidst Carceral Structural Violence
http://orcid.org/0000-0002-5594-9198
Sue Kimberly L. [email protected]
grid.47100.32 0000000419368710 Department of Internal Medicine, Yale University School of Medicine, 367 Cedar Street, Room 304A, New Haven, CT USA
29 11 2022
120
17 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
In this article, I describe the dilemmas of working as a physician-ethnographer within the Rikers Island jail healthcare system before and at the beginning of the COVID-19 epidemic in April 2020. The Rikers Island jail system in New York City has been in the national spotlight as a space of violence, trauma, and death amidst calls to decarcerate by community members and abolition advocates. This article is a personal reflection on the labor and subjectivity of healthcare providers and their positionality to multiple axes of structural and interpersonal violence while attempting to provide care in carceral institutions. I observe how COVID-19 functioned as an additional form of structural violence for incarcerated people. Clinical ethnography remains an essential tool for understanding complex social phenomena such as violence. However, physician-ethnographers working in these spaces of structural violence can have unique and conflicting constraints: tasked with providing evidence-based medicine but also simultaneously participating in an unusual form of labor that is an amalgamation of care, social suffering, and punishment. Despite and across at-times conflicting roles and obligations, I propose that these fragmented subjectivities can foment social criticism, propel advocacy toward decarceration, and produce a critically engaged dialogue between fields of anthropology and medicine toward a goal of health justice.
Keywords
Incarceration
Health
Violence
COVID-19
Care
Clinician ethnography
==== Body
pmcBecoming A Jail Physician
Physically arriving to the jail clinic space in the assigned building on Rikers Island requires significant attention and care to enter with minimum hassle and maximum speed. It routinely requires locking up a cellphone and keys (bring your own lock), putting your see-through backpack through the security detector (no liquids, not even water), going through a metal detector yourself, and then walking through and into the trap, which is a space located between two sets of heavy metal doors. In the trap, when one door is open, the other door is closed. Then you travel up and down a set of often broken elevators and bang on the door to the clinic, entering a door that can only be opened with a giant set of metal keys from a correctional officer standing on the inside of the door.
A single cubicle exam room, with a computer that works intermittently, as well as broken or inadequate equipment that is pilfered from the other cubicles, comprises home base for the next eight to sixteen hours, depending on if you’re doing a single or a double shift. Often, the chairs are stolen or switched leaving the last person to arrive in the morning with the worst chair, a chair that’s peeling, or the stuffing on the arms is worn or ripped off. There are sometimes fights between providers over these chairs.
This is the intended site of health and healing for the jail clinician to do their routine primary care, chronic disease management, urgent care visits, and return with patients from emergency runs to housing units. While it has the trappings of a routine clinical space, the scope of what is performed here by the jail clinician is far out of ordinary compared to the work of taking care of patients in a community-based setting.
To understand the work of a jail physician, it is critical to understand the jail system more broadly. Rikers Island is the main New York City jail complex located in the East River that serves as a central location for mostly pre-trial detainees from all five boroughs (Manhattan, Bronx, Brooklyn, Queens, and Staten Island), meaning that most of the people there have been charged with a crime but are awaiting court and have not been convicted. The “Island,” as it is colloquially known, is accessed through a long bridge on the north side of Queens very proximal to La Guardia Airport. The complex itself contains approximately ten separate buildings, with some buildings housing over a thousand people at any one time. Detained people are placed into these buildings based on a variety of factors (e.g., gang affiliation, charge status, conviction, gender identity, health status, among others). There is one building equipped for medically complex infirm patients that require nursing home level of care, such as patients with HIV/AIDS, Type 1 diabetes, or on dialysis. Several of the other boroughs have a jail building that also can house people charged within that borough or to house them temporarily while awaiting local court proceedings.
A prominent campaign known as “Close Rikers” launched in 2016 featuring community coalitions that sought to close the antiquated central island facility, decrease the jail population, and rebuild and rehouse incarcerated individuals into home boroughs. An opposing community group called “No New Jails” positioned themselves as an abolitionist group that demanded both closing the central Rikers facility while also stopping renovation or rebuilding of borough jail complexes, instead arguing that this funding should go toward community alternatives to safety and restorative justice (Eisenberg-Guyot 2021). While these campaigns waged, the New York City system had also been slowly decreasing the standing population of the jail, down to approximately 4,000 people by April 2020 (NYC Press Office 2020). At the end of the campaigns, New York City Council voted 36–13 in favor of Close Rikers, agreeing to shut down the Rikers Island complex by 2026 by renovating borough-based jails, decreasing the population by 75%, and transferring remaining detained people from the island to local borough jails. Within this debate, advocates of improving conditions on Rikers Island for people currently incarcerated argued for the necessity of upkeeping the failing infrastructure on the jail complex and continuing to afford ongoing services, so the NYC budget continued to include both the cost of building four new borough jails to meet Local Law codes for minimum standards, as well as a line item to maintain the aging facility infrastructure until it closed (Rivera 2022).
Many of the calls to Close Rikers and to decarcerate were because Rikers Island had a reputation as a space of violence, self-harm, and injury (particularly increased because of solitary confinement policies), and generally inhumane living conditions. Like many antiquated and dilapidated jail and prison systems around the country, the physical infrastructure of the buildings was poorly maintained. Running to a medical emergency in the building meant navigating out-of-order elevators, exposed plumbing, decayed ceiling and wall infrastructure, broken windows taped over with a sheet of plastic, and flooded floors; reportedly, 79% of weapons were fashioned from decaying building materials of the facility (Rakia 2016). Across these environmental hazards and exposures, patients were transported to and from the clinics from their housing units for medical appointments by at least one or more correctional officers, which made the execution of care plans onerous at best.
In general, incarcerated people have high rates of both acute and chronic health conditions, as well as higher prevalence rates of mental health and substance use disorders than the general population, often requiring frequent and significant interactions with the jail and prison healthcare systems (Binswanger et al. 2009; Massoglia and Remster 2019; Steadman et al. 2009; Fazel et al. 2017). Rikers Island-specific data demonstrated frequent incarceration for people with multiple axes of structural vulnerability, including substance use, homelessness, and mental illness (MacDonald et al. 2015). Given this data, the physician leadership at Rikers Island’s Correctional Health Services were committed to improving the healthcare of incarcerated people despite the significant challenges. Several of the jail medical leadership within the NYC jail system outlined what they call the triple aims of correctional health: patient safety, population health, and human rights (MacDonald et al. 2013). They described the unique epidemiological challenge of encountering patients with increased morbidity and mortality within jail system settings and the urgent need to provide them high-quality prevention and treatment. In fact, incarcerated patients are the only population of individuals in the United States with a legal right to health care as outlined in Estelle v. Gamble (Marshall, Thurgood and Supreme Court of the United States 1976).
However, the provision of this care is made possible only within the context of the security apparatus, including all the jail correctional officers and their associated leadership divisions, which often takes paramount importance over even urgent or emergency health concerns or issues. MacDonald, Parsons, and Venters define “dual loyalty” as a problem unique to the provision of healthcare in correctional settings where the clinician has both a loyalty to the patient but also to the security apparatus, often placing the clinician in a space of ethical conflict (MacDonald et al. 2013). The dilemma of actually enacting clinical medicine within these sites of concentrated structural violence entailed daily direct conflicts and even participation in violence itself.
Many of the principles of providing healthcare in a carceral system were outlined in the medical staff orientation to working in a jail. In many clinical settings, the practice of medicine involves some form of learning by doing, including navigating how to actually get things executed for patients rather than being problems of medical knowledge or clinical competency (Good 1994). Enacting medical care, such as ordering an x-ray of a hand after a patient got into a fight, was easier said than done in the jail infirmary. However, this learning curve, and the medical field’s pedagogical reliance on learning by doing, presented ethical and safety issues within sites of social control and extreme imbalances of power such as carceral systems. I soon learned that while I might see patients for diabetes or high blood pressure or HIV, I was just as often was called to write clearance forms to condone correctional officers spraying detainees with capsaicin or pepper spray, or alternatively, using tasers on them.
Initially, and perhaps profoundly naively, I signed up to work in jail healthcare to ameliorate individual and collective harms to structurally vulnerable populations, yet I wondered how and in what ways I simultaneously became complicit in enacting additional violence or harm. Based on fieldnotes and clinician ethnography in my role as an internal medicine physician at Rikers Island from October 2018–April 2020, before and in the beginning of the NYC COVID-19 pandemic, this paper explores the conundrums of delivering care that providers faced on a daily basis. This exploration builds on seminal work by Carolyn Sufrin, who has also worked as a physician and an anthropologist in a large urban jail (Sufrin 2017). In her paper on this topic, she lays out the “constantly shifting roles and obligations of practicing medicine and anthropology” and explores both tensions and ethical quandaries that come about precisely because “I was a doctor at the jail before I was an ethnographer there,” thus leading her to propose the term “observant participation” for the unique role of physician-anthropologists who are attempting to practice medicine with a critical ethnographic lens (Sufrin 2015).
This clinical work that I intended to do was undoubtedly affected by a deep distrust of carceral systems. As I outlined in previous work, my concern with the expanding range of social and healthcare services offered by prisons and jails coalesced in what I called the “carceral therapeutic state”(Sue 2019), caused by the merging and instantiation of the criminal legal system with therapeutic attention to substance use disorders. I worried that the jail and prisons had become fixed infrastructure in the American society and documented the effects of these carceral institutions on women’s lives as they dealt with opioid use disorder in and out of these spaces. I approached the work at Rikers with trepidation, however, I felt that utilizing the critical inquiry and methodological approaches of ethnography (“observant participation”) could help me to retain a socio-structural lens both within and outside of the direct care that I provided.
Here, I present a case study examining the multi-faceted nature of the violence of care in a jail facility based on self-reflections from routine clinical practice. As a case study, this work falls under university IRB exemption.1 However, to ensure confidentiality and anonymity given the sensitive topic, all individuals have been de-identified and merged into composite cases. Ages, genders, diagnoses, dates/times, and locations have all been changed or removed. Field notes have been edited for clarity and individuals deidentified.
“Contras”: The Disorientation of Everyday Violence
A typical 8-h jail shift routinely involved violence in a myriad of forms. At times during these shifts, there was actual physical violence either perpetuated by incarcerated patients and/or the correctional officers, often involving both parties. I routinely encountered the threat of violence or the aftermath of violence on an average shift.
In addition to this, there was a disorienting everyday violence of simply providing clinical care in a surveilled, restricted carceral environment. Nancy Scheper-Hughes and Philippe Bourgois note that “everyday violence … destroys socially marginalized humans with even greater frequency are usually invisible or misrecognized” (Scheper-Hughes and Bourgois 2004). In this case, violence took numerous forms that was routinized, bland, and knit into the everyday fabric of the jail operations. It might even take a form such as not having an escort to clinic for routine diabetes care in order to adjust insulin dosing or other medications, or having to endure frigid or unbearably hot temperatures for months on end that could exacerbate asthma. Carceral violence operated at numerous levels and was knit into the existence and operation of the institutions, themselves permitted by the existing sociopolitical conditions.
Early on in my time working within the Rikers Island jail system, after a week of orientation to the policies and procedures for new staff physicians, I started shadowing clinical shifts where I would tag along with a seasoned jail clinician. In many cases, this could be a Haitian physician assistant who had worked at the jail for twenty years or more. However, these shifts felt like a blur, and I was taking notes by hand, trying to keep up with a new computer system and figuring out how to actually get things done. After several clinical shifts of shadowing, including one in the intake center of one of the buildings where I was told to ignore a naked man screaming, urinating, and defecating in a tiny plexiglass room within the bustling main intake, I was considered ready to staff my own shifts. However, upon entering my own cubicle at Anna M. Kross Center (AMKC) I realized I knew very little. The clinical medicine I knew was hardly sufficient to fulfill the necessary tasks required of me within the jail infirmary.
One of the first non-medical tasks I was asked to do was evaluate for “contras.” When a security officer strolled into my cubicle, I was immediately on guard. I knew that they would only enter the actual cubicle space if they needed something. It was not technically in my job description to work for them; however, it seemed impossible to deny them their requests. I felt distrustful of them as I saw myself aligned and in spirit with incarcerated patients at the bottom of the power dynamic and social hierarchy. “I need a contra,” he said to me. “A what?” I asked. “A contra,” he repeated. As I wrote in a fieldnote after the incident:Officers come to me for contras and say "we got two contras, we gotta get a spray.” Turns out when it says DOC stun or DOC spray says "NO" it means NO contraindications. Whoops. I told them they can't spray either of them. Whoops? I think I understand now. I will ask one of the physicians when I come in tonight or when they come in tomorrow morning. Yikes. I feel so violent being asked to participate in the use of spray.
The “contra” request, in fact, was the correctional officer asking if the patient had any medical contraindications for the use of the stun shield/taser or use of oleoresin capsicum (“OC” spray). I had learned in my orientation to the jail that OC spray contained a capsaicinoid irritant that would lead to brief, self-limited symptoms that were supposed to resolve in 15–30 min, with reportedly < 1% requiring medical attention. These symptoms ranged across organ systems, including burning and irritation of the eyes and ears, cardiovascular effects (tachycardia, hypertension), respiratory issues (wheezing, drops in oxygen, pulmonary edema), and/or skin redness or blisters. Seeing patients who had just gotten sprayed was a common clinical visit.
When I turned to a physician in a neighboring cubicle to ask how to fill out the forms, I learned that I had, in fact, misunderstood the form. I had checked “no” thinking it meant that the paint could not be sprayed, when in fact checking “no” on the form actually meant “no contraindications to spray.” The contraindications to spray from a medical standpoint included pulmonary illness (such as chronic obstructive pulmonary disease or asthma), medical frailty of any kind, or mental health conditions that could be exacerbated by the use of the spray.
However, the medical policies and procedures clarified that filling out the “contra” form was not “medically clearing” patients for this use, it was simply advising the Department of Corrections staff about possible conditions in which that use could be particularly harmful. The policies also stated that there was scant scientific literature on this topic and that it was difficult actually gauge risk in the jail context. Much later, I also learned that despite my naïve efforts to protect the patients from being sprayed, that the correctional officers might utilize other equally harmful methods to achieve behavioral compliance, such as possibly beating or physically assaulting the patient to comply.
After a patient was sprayed with OC spray, the regulations in place stated that correctional officers had to offer the patient a medical visit. Many of them refused or did not want to get checked out, however many others wanted to be seen. There was not much we could do from our cubicles to ameliorate the effects of the capsaicin spray in their eyes or on their face. They had usually already been encouraged to wash their eyes and skin with water and to change the clothes they had been wearing. There was nothing we did or could prescribe to lessen the burning sensations of the spray, which is why many of them did not bother trying to get assessed.
As a staff physician, another part of the roles I performed including assessing either patients or correctional officers after what were known as “Uses of Force.” The “DOC Use of Force” included any incident that involved an altercation between an inmate and a correction officer. The “Use of Force” was often but not exclusively conducted by correctional officers or a special squad of officers named the “Probe Team” who suited up in riot gear and batons. When entering or exiting the building, I would sometimes see a “Probe Team” preparing to engage with a detained person, with one correctional officer holding up a handheld camcorder as was legally required. However, my role entailed examining a patient after a so-called “use of force” and filling out the necessary injury report. Sometimes, I might be additionally asked to examine one or more correctional officers who participated in the “use of force” incident.Fieldnote excerpt: When I saw the DOC [Department of Corrections] officer for "assault" I'd already seen the inmate, a very young man, who got sprayed, and got scratched up. Seeing as patients both the person in power and knowing that he is going to get money from his insurance company for assaulting this young man, who got all scratched up on the ground, it just made me sick. My face is furrowed… they want me to like smile or joke or whatever with them, no. It makes me sick to be complicit in the endeavor. The suffering of the torturer vs the one in power is not the same.
Essentially this officer is here to get his [insurance] reimbursement. As clinicians we used to have to see them so they could file to get occupational hazard insurance. I knew one CO who got $1100 a year for filing for “assault” from inmate patients. The ways in which I was complicit in bureaucratic or actual violence, literal profit off distress, are mind-blowing.
Filling out two kinds of forms in the aftermath of these incidents was confusing. One was a form to document where and whether there were any injuries on the incarcerated patient. It included a map of the body and the clinician was asked to put an X over where there were injuries incurred and describe the injuries in detail and any further necessary follow-up (x-rays, emergency room evaluation). I was told that these injury reports were evaluated by medical and security leadership to ensure that certain kinds of assault, for example, blows to the head, were not frequently occurring.
The other form that was often requested was by the correctional officers for reimbursement for their occupational hazards. This required logging into a separate medical documentation system for the officers and logging one of often numerous encounters they had been involved with over months and years, then printing them out documentation of the encounter so they could submit to their insurance company for monetary compensation for their occupational hazard.
In the aftermath of the uses of force, I felt complicit being forced to see both the patients who had been injured and the officers who were usually not injured, as if it was a level playing field. Refusal to participate in filling out the injury report forms—requiring a map of the body with my assessment of the injuries—was simply not done. At times, I could refuse to see the officers if I felt like resisting them, begging off that I was taking care of patients or otherwise busy, but they would often insistently hang around the cubicle and refuse to leave.
My clinical assessment of both parties—the incarcerated patient and the correctional officer—became a bureaucratic tool that could be variably wielded. And the act of sitting with them serially and examining one, then the other, as well as the assessment I was asked in each case to perform, was altered by the power differential. For my patients, I did the best I could. I cleaned and disinfected wounds, offered pain medication, or sleeping or anxiety medication. I was attentive to filling out the injury form in handwriting that was as legible as possible. But often there was little I could do. For the officers, I checked off that their skin or clothing was sprayed with capsaicin spray or saliva and they were encouraged to wash it off with soap and water. I often checked off on their forms that they had no hazardous exposure at all. Despite similar incident through-lines, this occurred several times a week, and each time I performed these assessments was a variation on a theme, with the incarcerated person much more likely to sustain any serious injury than the officer.
As a physician-ethnographer tasked with filling out these forms, I struggled with a fractured habitus compromised of shifting identities and roles. Bourdieu interpreted habitus as a socially inculcated, mental and embodied disposition classifying how one exists in the world (Bourdieu 1984), and in this context of disparate training philosophies, methods and practices, I felt three selves at play in this interaction. My physician self was often foregrounded. The work required was demanding, immediate, and time-consuming. The paperwork was rushed as officers or others waited impatiently for them to be filled out by hand. My ethnographer self was witnessing, observing, critical of my physician self’s complicity in a sociopolitical institution perpetuating inequality and violence. It accused my physician self as being what Foucault had called “the immediate anatomist of pain: wardens, doctors, chaplains, psychiatrists, psychologists, educationalists; by their very presence near the prisoner, they sing the praises that the law needs” (Foucault 1995: 11). And my third self, a moral self, responding affectively and instinctively as a human being reacting to the harms done by and to others, was simply aggrieved and upset. This self was no less a part of the three, however was often relegated to the realm of affect and vague unease, to the unceasing frown on my brow or the pit in my stomach as I filled out the form.
I worried that the forms could not account for the ongoing harms of the constant injuries between incarcerated patients and correctional officers. I hated filling them out although I did them diligently. However, upon reflection and greater contextualization, I realized that the paper injury reports were potentially doing something much more subversive in some ways, documenting what Homer Venters, the former chief medical officer of Rikers’ Correctional Health Services, has called the “epidemic of brutality in the NYC jail system” (Venters 2019, 36).
As Venters argued, the medical documentation was necessary to counteract the myriad of ways that systematic underreporting of these injuries occurred, and to call for outside investigators to look into patterns of beatings, abuses, particularly to the head and face, or injuries incurred even after patients were already restrained. Many times, both officers and incarcerated people took advantage purposefully of areas that were known to be out of view of surveillance cameras to beat others (or cameras were deliberately covered over with toilet paper or toothpaste). While the documentation seemed onerous to me, the statistics and documentation were vital to not erasing the grievous injuries that otherwise went underreported or under-analyzed by health department or external committees: “Part of our responsibility … was to point out how the jail system creates its own set of health risks for the people who pass through and to measure the uneven ways in which these risks are distributed across our patients” (2019: 42). The documents were collated for review by external jail oversight board known as the New York City Board of Correction, whose mandate was to oversee and improve policies and practices at the jail.
However, social scientists have noted that statistics and quantitative data, just as much as ethnography or archival work, tell specific narratives, often of those who wield the most power and social influence. Anthropologists have documented how metrics within the field of global health have been used to tell certain partial narratives that contribute to certain social outcomes (Adams 2016; Nguyen 2010; Biruk 2018). In the case of the jail injury reports, as onerous, partial and pain-staking as they were for me to fill out, they were one, albeit partial, attempt to account for suffering and create the metrics for advocacy and structural change. This is particularly relevant in the carceral setting in that that the stories of incarcerated patients are often disbelieved, ignored or distorted by the security apparatus. This phenomenon—that some stories and some data count more than others, such as physician notes versus patient narratives in the case of refugees experiencing trauma—has been observed by physician-anthropologists elsewhere (Fassin and Rechtman 2009).
One counterpart to the way in which these acts of violence become flattened into statistics is the knowledge production performed by ethnography. After I filled out the injury reports, I thought about the ways in which injury has been theorized and harkened back to the pit in my stomach. Laurence Ralph, in his ethnography working with people disabled by gang violence in Chicago, conceived of injury not only as an adjustment to the physical reality, but also representing potential transformation in spiritual, social, economic, and moral realms (Ralph 2014). How and in what ways did the injury evaluations and reports that I performed and the injuries sustained by my incarcerated patients change them and change me? Of course, it disillusioned me from the work, as I spent hours within a shift on paperwork rather than the provision of thorough, evidence-based medical care.
I believe that the triadic interactions of myself, an incarcerated patient, and a correctional officer, mattered far beyond the forms and the profit off distress. It seemed more important than simply witnessing or enacting bureaucratic necessity. Rather, the ethnographic attention to the incident revealed the culture of everyday violence and the ongoing injuries to the bodies and souls of incarcerated people that was made possible by the carceral system itself and by the sociopolitical system that accepted the ongoing existence of Rikers Island itself. It made me yearn for an abolitionist future, one where these interactions were not quotidian and acceptable.
Threats of Violence in the Infirmary
Even within the clinic itself, there was a non-zero risk of violence enacted upon healthcare workers. On one of the first days where I was starting to practice alone and had finished shadowing other clinicians, the central administrators requested that I leave the building I had trained in and “float” me to another building. Earlier that week at the other building, one of the frontline healthcare staff had been taken hostage by a patient for several hours (both individuals emerged physically unharmed). The staff had all been discussing what made that particular building a site of increased risk of violence and harm to staff, including factors such as prevalence of serious mental illness, solitary confinement and low staffing. I refused to go over to that building, not only because I didn’t have a car and I had never worked in that facility before, but also because I simply did not feel set up to be safe. Motivated largely by fear, and the feeling of self-preservation, I said no to participating in the suggested float and the medical director of the building allowed me to stay working in AMKC.
The configurations of economic, political, and criminal legal conditions that created risk environments for violence cannot be ignored. In some ways, it makes logical sense for an incarcerated person to access whatever means they can to feel seen or have their needs met. Often times, the usual access to medical, via what is known as a “sick slip,” or a piece of paper bearing symptoms triaged by nurses, are repeatedly ignored (Friedman, Burr, and Sufrin 2021). In this case, the act of taking a clinical staff member hostage does effectively upend the “jailer-jailed” paradigm. The person can no longer be ignored. The hostage situation, however, penetrated my psyche as well as the collective psyche of the other healthcare providers. In addition, earlier in the year, one of the clinicians had been strangled by a patient. There was, as expected, significant ongoing fear and outrage among the medical staff. Some other staff even confided to me they felt that the correctional officers deliberately were not keeping them safe, in order to justify their increased presence and increased staffing.
Several months later, I was confronted with the threat of violence while working an overnight shift. Around 5:30am, I was asked to see an incarcerated man who was complaining of stomach pains. I had been instructed that on the “night tour” shift that medical staff were required to see any patients who felt they were sick as a reason that they did not want to be transported to court that morning. As one of the other doctors told me, the only condition in which I was told that they could avoid going to court was if they were so sick they were in need of urgent hospitalization and we were calling 911 for an ambulance. Effectively, I was told to make everyone “fit for court.”
As I wrote in my fieldnotes:Fieldnote Excerpt: I also got threatened and backed up against the gate by a patient whom I made fit for court. He was a middle-aged man, and he was complaining of a hernia in his stomach that made lying on his back or sitting uncomfortable, he said he had had it for two to three months but he literally had never seen sick call for it. My job in the morning is to try to troubleshoot things for them and attend to their medical conditions but inevitably I have to make them fit for court. I offered him pain medication and then maybe a surgical consult for his hernia, and pts [patients] just don't understand, you can't just all of a sudden see a surgeon, and not to go court. He was like I don't even care about my court thing, it's small and stupid, but he just didn't want to sit all day in shackles. I try to be sympathetic but then I told the officer I was done with him, and honestly I didn't even see or feel a hernia, and then I was telling the officer to take him away, and he started to cuss me out. I backed up and the officer was standing between us and he was like, I'm going to punch your face. I was like, we've dealt with your issue. And there are like 15 ppl watching by the way, and I was like well, call 311 on me [make a complaint to the city], and he's like, no I'm just going to beat you the fuck up..”
Upon analyzing my fieldnotes, it was clear that only several months into socialization in the carceral infirmary space, that I had already assumed some of the clinical brusqueness that I saw in many of my colleagues. I had responded flatly and coldly to the patient to make a formal complaint to the city. But I couldn’t help wondering about the myriad of complex dynamics including disparate age, gender, ethnicity, and power that could have been part of this interaction. Importantly, I refused to recognize and respond to his symptoms in the moment. I dismissed him, summarily, to go to court, and in doing so dismissed his suffering and recognition of his humanity.
Escalating the encounter into one of potential violence—at least with threats and words—was one way of demonstrating his potential power and his disagreement with the conclusion of our encounter. This resonates with ethnographic work of Philippe Bourgois with Puerto Rican people in East Harlem, where violence becomes a means of asserting local dignity and respect within a devastatingly crushing neoliberal system that does not afford recognition to many of its most structurally vulnerable people (Bourgois 2003).
So the threat of violence was what was necessary for him at that moment, not the threat of making a civil complaint on the phone against me several hours later in private. And my actions and inactions—to determine if he was to go to court or to go to the hospital—had direct consequences on his life that day. What was disturbing, however, was the routine way in which the incident erupted and then receded. No one came to talk to me about the incident, about the threat of violence, and everyone went on with their quiet early morning routines, wrapping up the night’s work and preparing to leave within the next hour. This was just another form of everyday, quotidian violence knit into the work of the jail physician.
Kleinman has insisted in his previous work that “experience-near” ethnography is a means of making sense of human experience. He posits that we must understand what is at stake for each individual within “local moral worlds” as they play their part within social hierarchies and relationships. He and Joan Kleinman were critical of the involvement that individuals in both professionalized anthropology and medicine had to play in structural oppression, transforming illness experience into a diagnostic code (Kleinman and Kleinman 1991). They argue instead for an interpersonal form of subjectivity and the value of presenting interpersonal experience captured within ethnography, biography, and the arts. It is through the ethnographic method and analysis that I sought to make sense of my physician duties and also make sense of the violence.
However, sometimes the jail infirmary had an air of the truly absurd about it that broke us all out of conventional roles and relationships. The circular nature of witnessing and violence, as well as the complicated nature of purported victims and perpetrators across a very uneven social hierarchy, was mystifying. One incident demonstrates this absurd reality, where I mused, does the doctor make it worse?Fieldnote excerpt:
I arrive and the clinic is in a state of chaos, but I don't see the [other] doctor. I have no idea what is going on. It turns out [Dr. Smith]*2 called 911 for a patient who was probably drunk and assaultive off of “hooch” [home brewed alcohol]. I had seen this particular patient only three days ago, and gave him some ibuprofen. When I walked by him on my way into the clinic I had no idea it was him. There was a huge standoff and he had been refusing to go with the paramedics for like 1.5 hours. They were going to have to use a “Use of Force” against him. UGH!!! So this squad team comes up with their riot gear on and I hunker into the treatment room with the nurses. They tell me that I don't want to be out there to get injured, and I don't want to be out there to be seen on the camera to have to file a Nunez, which is a form you have to do if you see a Use of Force against a patient. I feel as if there is a moment in which I lost that I could have turned it around but by the time I recognized that I had seen the patient before, he was threatening to punch people.
The patient had come to file a Confidential Allegation, but was drunk, and refused vital signs with [Dr. Smith] so [Dr. Smith] punished him and called 911. Then Dr. Smith left and I'm like, damn, I'm having to deal with this mess. He's [the patient] cussing and he's like wait ‘til I come back here, they're not going to do anything to me, which certainly probably was somewhat true. Ugh. They made me go out of the treatment room to examine him and fill out an injury report after a Use of Force. He had been given versed [a benzodiazepine] so was sleepy, but seemed okay, not visibly injured.
So I did not actually witness the violence that took place against the patient in the infirmary space, although I was adjacent to the Use of Force. The nurses, in fact, had purposely corralled me out of the main clinic area into a side nursing room where they were sitting to avoid the space where the Use of Force was taking place, both for practical reasons, but also for obviating the systems in place intended to protect patients from Uses of Force. My fieldnotes reflect a brief moment of possibility, the potential of an intervention, however foreclosed. Even if I had tried to intervene, given my relative newness to the clinic, and the personalities and the power dynamics within the clinic itself, it was unlikely I would have had success in de-escalation.
I read my fieldnotes again. “[Dr. Smith] punished him and called 911.” How could it not be an absurd reality? Again, everyone seemed to be acting out predetermined roles: both the doctor and the patient knowing that he doesn’t need to go the hospital, that he’s just drunk, that “they won’t do anything to me” (at the hospital emergency room), but each are doing what they must do. The patient, resisting vital signs. The doctor, punishing him for resisting vital signs.
Upon further reflection, I realized what was actually one of the most disturbing elements about the incident is that the patient had come to file a Confidential Allegation, which is a formal complaint to medical staff of physical, verbal or sexual abuse, assault or harassment, to an individual by another incarcerated person or a staff member. It was in and itself a tragic outcome, additionally more so if he was unable to file this serious report of harm.
On top of that, I felt guilty about my inaction, my lack of witnessing. The nurses had pulled me away on purpose. This was to avoid having to fill out Nunez paperwork, which I had never completed before. On Rikers Island, the legal proceedings Nunez and the United States v. City of New York in 2011 led to the formation of a group known as the Nunez Commission. The Nunez Commission was a group of outside observers that became required after 11 plaintiffs were beaten by officers at Rikers Island. The lawsuit that led to the formation of the group asserted that Rikers Island leadership tolerated “a culture of routine and institutionalized staff violence against inmates”(Weiser 2012). What Nunez meant for staff, including healthcare workers, was mandatory reporting to try to create conditions of increased safety and transparency at Rikers.
However, the medical staff, including myself, had ducked that responsibility, and they often tried to avoid witnessing. None of us witnessed the Use of Force, subverting the existence of Nunez intended for us to document and witness violence. In this case, medical staff (Dr. Smith) had actually participated in the calling for the Use of Force, which resulted in chemical sedation, possible trauma, and a visit to the Emergency Room for the patient. Did the doctor make it worse? How had we been a part of aiding the security apparatus rather than enacting clinical duties of health and healing that we all purported to do?
Incarceration as Violence Itself: The Arrival of COVID-19
The arrival of COVID-19 onto this already fraught and tense delivery of healthcare within the jail system presented an additional layer of risk and potential harm to incarcerated individuals. A notable fact of jails, from a contagion and infection control perspective, is that the staff, including the correctional officers, who often have close and contained interaction in the air space of detained people inside, is that the officers and even healthcare staff posed risks to detained people by fact of their community interactions and engagements (Reinhart and Chen 2020). Incarcerated patients expressed to me their inability to control the behaviors and risks that non-incarcerated staff took.
COVID-19 fueled new anxiety and stress for the officers, medical staff, and patients alike. Buildings that had been previously closed as housing units in part of the consolidation and decarceration efforts were suddenly re-opened up to become COVID-19 isolation units. Entire housing units were put on quarantine, which made the people inside both anxious and aggressive, and disrupted the very flow and transport required for accessing healthcare services. Many of the officers refused to wear masks, although their union demanded and then took N95 masks from the health department of the city. All individuals, including correctional staff and healthcare providers, were seen as vectors of possible contagion from the outside world.
In April 2020, the court system had effectively been shut down. I saw many patients who were incarcerated on violations of parole (which in New York City, could be something as simple as not being at the address listed on parole at 7am on a Saturday). They were incarcerated for technical violations and were effectively unable to go home, since the courts were not essentially non-functional, keeping patients in a state of unending limbo and viral exposure.
The new infectious airborne threat of COVID-19 was another layer of possible harm, both psychic and physical, to patients. On one of my shifts, a patient with a certain high level of security designation, who also happened to have a serious cancer, was not allowed to be transferred to a health unit because his security status trumped his health risks. I tried to ensure at the minimum that the nurse could give him a mask. Security would not allow him to be transferred to the medical unit because of his high security classification.
The social relations were entirely upended by the presence and even the threat of the virus. Affiliated groups of incarcerated individuals, who belonged to established regional and national gangs, effectively “locked down” their housing units, patrolling who could enter and who could not, with threats of physical violence or coercion. I had one elderly patient with multiple medical co-morbidities return from the hospital to his housing unit telling me that he hadn’t eaten in four days because the “gang unit will jump someone [physically assault them] if they want to eat.” Individuals who belonged to gangs were prohibiting guards from bringing in food trays or allowing other incarcerated people to accept food trays. The gang leaders out of fear of contamination with the new virus were stopping the flow of community members like guards to the inside housing units where they felt they remained COVID-free.
COVID-19, therefore, represented environmental and structural violence on top of the quotidian, everyday violence such as getting pepper sprayed, as my patient’s comment demonstrated. Critically, the threat of the virus did represent for patients also a psychic harm, an unceasing anxiety. It was not only the emergence of a new viral pandemic, it was also an “epidemic of signification,” as Paula Treichler argued about the HIV/AIDS epidemic (Treichler 1999). In the jail, similar to the emergence of the HIV epidemic, the rapid transmission of illness caused by COVID-19 was not only biological in nature but also carried with it an explosion of meanings, many often based on emotions like fear, with such significations having a central role in social discourse and even institutional or public policy, no matter how biologically unfounded. The fight for masks and airspace within the jail were just beginning. The terror was real, as we had no treatments and barely any tests for an entirely new viral entity. This was stressful to patients and providers alike, within a carceral institution in which many incarcerated people lacked access to accurate news or means to prevent or treat the virus.
As a physician, I believed the best thing for these patients beginning to be released was that they could go home. Nothing seemed safer than going home. I could not write a prescription for their release, however, one year into the pandemic, the Chief Medical Officer of Correctional Health Services, Dr. Ross MacDonald, did in fact write an impassioned plea to the New York City Council in September 2021 for de-carceration as a means to decrease in-custody COVID-19 deaths (MacDonald 2021; Barnert et al. 2021).
The COVID-19 pandemic in carceral institutions signified structural violence and social suffering en masse, with the jail and prison systems suffering disproportionate amounts of harms and deaths (Kim et al. 2022; Toblin and Hagan 2021; Saloner et al. 2020). It was one additional vector of structural harms to burden those already sickest and most structurally marginalized.
My reflections on the environment of trauma and violence in carceral institutions, as well as my own role in the violence of care within these spaces, made me realize that I was marked as an agent of possible harm and violence despite my efforts to provide healing, patient accompaniment, or excellent care in spaces of violence. The question remained for me as it has for others who work within oppressive socio-structural political environments: can I both participate in a structurally violent apparatus and also work toward structural change within or against this system?
What was essential to my ability to undertake this work was a deep commitment to envisioning a world where all people, especially the most structurally vulnerable, could access the primary care, addiction and mental health care, that they needed where they live, without bars and without mandates. It was a calling to do so at the exhortations of one of my mentors, Paul Farmer, whose careful delineation of structural violence as a necessary theoretical lens had, in part, led me to the jail infirmary to study, work, observe, and write. He implored us all to use ethnography as a means to attend to social suffering: “The task at hand, if this silence is to be broken, is to identify the forces conspiring to promote suffering, with the understanding that these are differentially weighted in different settings” (Farmer 2005).
Having trained as both a physician and an anthropologist allowed to me to approach unpacking and undoing structural violence and the specific violence of carceral care from very different angles. As opposed to many of my colleagues in jail medicine, having an ethnographic stance, social medicine theories and a philosophy of critical inquiry allowed me to both document and assess the settings and systems in which this violence occurs.
In addition, quantitative data can lack the ability to comprehensively tell stories and contribute to a deeper understanding of why things happen. Ethnography, however, can produce knowledge that elucidate social forces at-play. In this instance, I used my clinical work as a physician combined with the analytic lens of anthropology to advocate and call for decarceration during COVID-19 in local newspapers to maximally improve the health of people at risk of harms or death related to incarceration (Thill and Sue 2021). Through these observations and this dual training, I sought to draw attention to issues that could be considered violations of human rights and lack of proper jail oversight and to call for person-centered care at Rikers and beyond. This is part of reinforcing a larger community call to both improve the material conditions that detained people currently face as well as to work toward a world of abolition, where individuals can access the care that they need in their own communities (Davis 2003; Kaba and Ritchie 2022).
Claire Wendland proposes that the uniqueness of dual training of a physician anthropologist allows individuals to “illustrate the connections between microscopic harm and macrosocial arrangements” (Wendland 2019). Remaining committed to the dual subjectivities and the fractured habitus of being both a physician and an ethnographer, in the tradition of critical medical anthropology, requires the ability to zoom out and examine structural socioeconomic forces of oppression in which we all participate (Singer and Baer 2018). It is a call to Farmer’s “pragmatic solidarity” (Farmer 2005) to using our dual training in medicine and the social sciences as a basis for data and compassion driven action.
In our current era of COVID-19 and the refutation of scientific evidence, ethnography proves to be a powerful and necessary means of gathering socially resonant data. Our writing and our work, therefore, can be means to contextualize both the individual experiences of patients as well as those of physicians, in order to illuminate the structural forces that disproportionately burden the poor, drawing attention to the failing healthcare and social apparatus in which we operate as citizens and laborers.
There is a new generation of critically engaged physician-anthropologists that are using this engaged witnessing to engage in pragmatic solidarity and advocacy for structural change as well as challenge pedagogical and theoretical approaches. They use experience-near clinical cases of individuals to highlight structural forces of oppression and poor health. Hansen and Metzl, as leading physicians-anthropologists, have formulated theories of structural competency and structural vulnerability that explicitly help trainees to understand how social and structural forces, not simply individual behaviors, influence and condition poor health outcomes (Metzl and Hansen 2014; Quesada et al. 2011; Bourgois et al. 2017).
Holmes has used his work and his embodied experience to document and advocate for the Mexican migrants who break their bodies picking fruit for pennies to satisfy demands of U.S. consumption (Holmes 2013). Sufrin has used her work and her “proximity near” experiences to advocate against the practice of shackling women during labor (Sufrin 2018). As the paper in this issue by Levenson and Samra illustrates, physicians and academics can use their social and political privilege to assist abolitionist community members to effectively call for stopping jail expansion. Many are teaching these concepts and engaging in advocacy and social activism within their institutions, to new medical and social science trainees, and within their communities more generally.
Because of the social platform and credibility of physician-anthropologists, many are positioned to be able to partner with their communities to conduct meaningful research with and for their communities and advocate successfully for direct changes to the upstream social structural determinants of health on issues like structural racism, enduring poverty, and ongoing violence. While recognizing fractured habitus and varying levels of complicity in structurally violent systems, they are able to engage with trainees about the messy systems in which we live and work within medicine and the academy, conduct relevant and impactful research, and use ethnographic and/or clinical narrative to inform policy and social action. However, future work should prioritize centering and partnering with people most directly affected and the ongoing work; one significant limitation of this case study is the elision of their voices.
This case study is intended to highlight how ethnographic and clinical work can be enlisted to improve socio-structural conditions from a myriad of angles, as was the case with my provision of medical care at Rikers Island. Rather than elide the messiness of both ethnographic and clinical endeavors across fractured subjectivities within biosocial engagements, it is critical to recognize them as a space of fruitful tension. Witnessing, combined with a critical theoretical lens, is necessary if not sufficient, for our pressing social conditions. Our work seeks to both engage in a new intersubjectivity of care and transform observations into meaningful and sustained social action.
Acknowledgements
I would like to acknowledge Liza Buchbinder, Bonnie Wong, Rebecca Newmark, Seth Holmes, Philippe Bourgois, for their dedication and support through this project conceptualization and manuscript preparation. I’d like to acknowledge and thank clinicians and jail medical leadership for their ongoing dedication to providing compassionate and evidence-based care for detained people.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declarations
Conflict of interest
The author declares she has no conflicts of interest.
Ethical Approval
This article does not contain any studies with human participants performed by any of the authors.
1 See Yale University IRB statement on case studies: https://hipaa.yale.edu/sites/default/files/files/Case%20Reports%20and%20Patient%20Privacy.pdf
2 Names have been anonymized here for privacy and composite cases have been utilized.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Adams Vincanne Metrics: What Counts in Global Health. Critical Global Health—Evidence, Efficacy, Ethnography 2016 Durham Duke University Press
Barnert Elizabeth Kwan Ada Williams Brie Ten Urgent Priorities Based on Lessons Learned From More Than a Half Million Known COVID-19 Cases in US Prisons American Journal of Public Health 2021 111 6 1099 1105 10.2105/AJPH.2021.306221 33856887
Binswanger I.A. Krueger P.M. Steiner J.F. Prevalence of Chronic Medical Conditions among Jail and Prison Inmates in the USA Compared with the General Population Journal of Epidemiology and Community Health 2009 63 11 912 919 10.1136/jech.2009.090662 19648129
Biruk, Crystal 2018 Cooking Data: Culture and Politics in an African Research World. Duke University Press. https://library.oapen.org/handle/20.500.12657/30253.
Bourdieu, Pierre 1984 Distinction: A Social Critique of the Judgement of Taste. Harvard University Press.
Bourgois, Philippe 2003 In Search of Respect: Selling Crack in El Barrio. 2nd ed. Structural Analysis in the Social Sciences. Cambridge; New York: Cambridge University Press.
Bourgois Philippe Holmes Seth M. Sue Kim Quesada James Structural Vulnerability: Operationalizing the Concept to Address Health Disparities in Clinical Care Academic Medicine : Journal of the Association of American Medical Colleges 2017 92 3 299 307 10.1097/ACM.0000000000001294 27415443
Davis, Angela 2003 Are Prisons Obsolete?: NY: Seven Stories Press. https://www.amazon.com/Are-Prisons-Obsolete-Angela-Davis/dp/1583225811.
Eisenberg-Guyot, Nadja 2021 New York City Has a Jail Problem. Current Affairs, March 15, 2021. https://www.currentaffairs.org/2021/03/new-york-city-has-a-jail-problem.
Farmer, Paul 2005 Pathologies of Power: Health, Human Rights, and the New War on the Poor, With a New Preface by the Author.
Fassin, Didier, and Richard Rechtman 2009 The Empire of Trauma: An Inquiry into the Condition of Victimhood. Princeton; Oxford Princeto University Press.
Fazel Seena Yoon Isabel A. Hayes Adrian J. Substance Use Disorders in Prisoners: An Updated Systematic Review and Meta-regression Analysis in Recently Incarcerated Men and Women Addiction (abingdon, England) 2017 112 10 1725 1739 10.1111/add.13877 28543749
Foucault Michel Discipline and Punish: The Birth of the Prison. 2nd Vintage 1995 Books New York Vintage Books
Friedman, Emmeline, Eliza Burr, and Carolyn Sufrin 2021 Seeking Recognition through Carceral Health Care Bureaucracy: Analysis of Medical Care Request Forms in a County Jail. Social Science & Medicine (1982) 291 (December): 114485. 10.1016/j.socscimed.2021.114485.
Good Byron J. Medicine 1994 Rationality and Experience An Anthropological Perspective. Cambridge University Press
Holmes Seth M. Fresh Fruit, Broken Bodies: Migrant Farmworkers in the United States 2013 Berkeley University of California Press
Kaba, Mariame, and Andrea Ritchie 2022 No More Police: A Case for Abolition. New York: New Press. https://thenewpress.com/books/no-more-police.
Kim Hannah Hughes Emily Cavanagh Alice Norris Emily Gao Angela Bondy Susan J. McLeod Katherine E. Kanagalingam Tharsan Kouyoumdjian Fiona G. The Health Impacts of the COVID-19 Pandemic on Adults Who Experience Imprisonment Globally: A Mixed Methods Systematic Review PloS One 2022 17 5 e0268866 10.1371/journal.pone.0268866 35594288
Kleinman Arthur Kleinman Joan Suffering and Its Professional Transformation: Toward an Ethnography of Interpersonal Experience Culture, Medicine and Psychiatry 1991 15 275 301 10.1007/BF00046540 1935180
MacDonald, Ross 2021 City Council Letter. September 10, 2021. https://www.ny1.com/content/dam/News/static/nyc/pdfs/RM-city-council-letter-9-10-21.pdf.
MacDonald Ross Kaba Fatos Rosner Zachary Vise Allison Weiss David Brittner Mindy Skerker Molly Dickey Nathaniel Venters Homer The Rikers Island Hot Spotters: Defining the Needs of the Most Frequently Incarcerated American Journal of Public Health 2015 105 11 2262 2268 10.2105/AJPH.2015.302785 26378829
MacDonald Ross Parsons Amanda Venters Homer D. The Triple Aims of Correctional Health: Patient Safety, Population Health, and Human Rights Journal of Health Care for the Poor and Underserved 2013 24 3 1226 1234 10.1353/hpu.2013.0142 23974393
Marshall, Thurgood, and Supreme Court of the United States 1976 U.S. Reports: Estelle v. Gamble, 429 U.S. 97. U.S. Reports: Retrieved from the Library of Congress, <www.loc.gov/item/usrep429097/>.
Massoglia, Michael, and Brianna Remster 2019 Linkages Between Incarceration and Health. Public Health Reports 134 (1): 8S-14S. 10.1177/0033354919826563.
Metzl Jonathan M. Hansen Helena Structural Competency: Theorizing a New Medical Engagement with Stigma and Inequality Social Science & Medicine Structural Stigma and Population Health 2014 103 February 126 133 10.1016/j.socscimed.2013.06.032
Nguyen Vinh-Kim The Republic of Therapy : Triage and Sovereignty in West Africa’s Time of AIDS 2010 Durham NC Duke University Press
NYC Press Office 2020 City Jail Population Drops Below 4,000 for First Time Since 1946. The Official Website of the City of New York. April 21, 2020. http://www1.nyc.gov/office-of-the-mayor/news/278-20/city-jail-population-drops-below-4-000-first-time-since-1946.
Quesada James Hart Laurie K. Bourgois Philippe Structural Vulnerability and Health: Latino Migrant Laborers in the United States Medical Anthropology 2011 30 4 339 362 10.1080/01459740.2011.576725 21777121
Rakia, Raven 2016 A Sinking Jail: The Environmental Disaster That Is Rikers Island. Grist. March 15, 2016. https://grist.org/justice/a-sinking-jail-the-environmental-disaster-that-is-rikers-island/.
Ralph, Laurence 2014 Renegade Dreams: Living through Injury in Gangland Chicago. Illustrated edition. Chicago ; London: University of Chicago Press.
Reinhart Eric Chen Daniel L. Incarceration And Its Disseminations: COVID-19 Pandemic Lessons From Chicago’s Cook County Jail Health Affairs 2020 39 8 1412 1418 10.1377/hlthaff.2020.00652 32496864
Rivera, Hon Carlina 2022 Report to the Committee on Finance and the Committee on Criminal Justice on the Fiscal 2023 Executive Plan and the Fiscal 2023 Executive Capital Commitment Plan (Department of Corrections DOC), May, 12.
Saloner Brendan Parish Kalind Ward Julie A. DiLaura Grace Dolovich Sharon COVID-19 Cases and Deaths in Federal and State Prisons JAMA 2020 324 6 602 603 10.1001/jama.2020.12528 32639537
Scheper-Hughes, Nancy, and Philippe Bourgois 2004 Introduction: Making Sense of Violence.” In Violence in War and Peace: An Anthology:Violence in War and Peace: An Anthology, 1–31. Malden MA: Blackwell Pub.
Singer, Merrill, and Hans A. Baer 2018 Critical Medical Anthropology. First issued in paperback. Critical Approaches in the Health Social Sciences Series. Boca Raton London New York: CRC Press.
Steadman Henry J. Osher Fred C. Robbins Pamela Clark Case Brian Samuels Steven Prevalence of Serious Mental Illness Among Jail Inmates Psychiatric Services 2009 60 6 761 765 10.1176/ps.2009.60.6.761 19487344
Sue, Kimberly 2019 Getting Wrecked: Women, Incarceration, and the American Opioid Crisis. California Series in Public Anthropology. Oakland, California: University of California Press.
Sufrin Carolyn ‘Doctor, Why Didn’t You Adopt My Baby?’ Observant Participation, Care, and the Simultaneous Practice of Medicine and Anthropology Culture, Medicine, and Psychiatry 2015 39 4 614 633 10.1007/s11013-015-9435-x 25697337
Sufrin Carolyn Jailcare: Finding the Safety Net for Women Behind Bars 2017 Oakland, California University of California Press
Sufrin Carolyn Making Mothers in Jail: Carceral Reproduction of Normative Motherhood Reproductive Biomedicine & Society Online 2018 7 November 55 65 10.1016/j.rbms.2018.10.018 30740546
Thill, Zoey, and Kimberly Sue 2021 Health and Safety for Everyone: Former City Jail Physicians on the Crisis at Rikers and Beyond. Gotham Gazette. October 13, 2021. https://www.gothamgazette.com/130-opinion/130-opinion/10825-city-jail-physicians-crisis-rikers-island.
Toblin Robin L. Hagan Liesl M. COVID-19 Case and Mortality Rates in the Federal Bureau of Prisons American Journal of Preventive Medicine 2021 61 1 120 123 10.1016/j.amepre.2021.01.019 33781619
Treichler Paula A. How to Have Theory in an Epidemic : Cultural Chronicles of AIDS 1999 Durham Duke University Press
Venters Homer Life and Death in Rikers Island 2019 1 Baltimore Johns Hopkins University Press
Weiser, Benjamin 2012 Lawsuit Accuses City’s Jails of Condoning Inmate Abuse. The New York Times, May 30, 2012, sec. New York. https://www.nytimes.com/2012/05/30/nyregion/suit-says-new-york-citys-jails-condone-guards-beatings-of-inmates.html.
Wendland Claire L. Physician Anthropologists Annual Review of Anthropology 2019 48 1 187 205 10.1146/annurev-anthro-102218-011338
| 36445550 | PMC9707201 | NO-CC CODE | 2022-12-01 23:20:21 | no | Cult Med Psychiatry. 2022 Nov 29;:1-20 | utf-8 | Cult Med Psychiatry | 2,022 | 10.1007/s11013-022-09812-2 | oa_other |
==== Front
Bone Marrow Transplant
Bone Marrow Transplant
Bone Marrow Transplantation
0268-3369
1476-5365
Nature Publishing Group UK London
1881
10.1038/s41409-022-01881-6
Correspondence
Steroid-free first line treatment of moderate and severe chronic GVHD: a survey from the Transplant Complications Working Party of the EBMT
http://orcid.org/0000-0002-4332-0114
Moiseev Ivan [email protected]
1
Ambron Pascale 2
Badoglio Manuela 2
http://orcid.org/0000-0002-3829-0088
Peczynski Christophe 2
http://orcid.org/0000-0003-3858-8180
Basak Grzegorz 3
http://orcid.org/0000-0001-7025-1735
Koenecke Christian 4
http://orcid.org/0000-0002-7568-8239
Schoemans Helene 56
http://orcid.org/0000-0003-4876-802X
Penack Olaf 7
http://orcid.org/0000-0001-9458-8025
Peric Zinaida 89
1 grid.412460.5 RM Gorbacheva Research Institute, Pavlov University, Lva Tolstogo St. 6/8, Saint Petersburg, Russian Federation
2 grid.462844.8 0000 0001 2308 1657 EBMT Paris Study Office; Department of Haematology, Saint Antoine Hospital; INSERM UMR 938, Sorbonne University, Paris, France
3 grid.13339.3b 0000000113287408 Department of Hematology, Oncology and Internal Medicine, University Clinical Center of the Medical University of Warsaw, Warsaw, Poland
4 grid.10423.34 0000 0000 9529 9877 Department of Hematology, Hemostasis, Oncology and Stem Cell Transplantation, Hannover Medical School, Hanover, Germany
5 grid.5596.f 0000 0001 0668 7884 Department of Hematology, University Hospitals Leuven, KU Leuven, Leuven, Belgium
6 grid.5596.f 0000 0001 0668 7884 Department of Public Health and Primary Care, ACCENT VV, KU Leuven-University of Leuven, Leuven, Belgium
7 grid.6363.0 0000 0001 2218 4662 Department of Hematology, Oncology and Tumorimmunology, Charité-Universitätsmedizin Berlin, Berlin, Germany
8 grid.4808.4 0000 0001 0657 4636 School of Medicine, University of Zagreb, Zagreb, Croatia
9 grid.412688.1 0000 0004 0397 9648 University Hospital Centre Zagreb, Zagreb, Croatia
29 11 2022
13
22 9 2022
15 11 2022
17 11 2022
© The Author(s), under exclusive licence to Springer Nature Limited 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Subject terms
Stem-cell therapies
Haematological cancer
==== Body
pmcTo the Editor:
Corticosteroids were established as the standard first line systemic treatment of chronic graft-versus-host disease (cGVHD) more than four decades ago [1]. Later attempts to improve the outcomes by using corticosteroids in combination with other immunosuppressive treatments demonstrated no improvement in survival compared to corticosteroids alone [2]. On the other hand, long-term administration of corticosteroids increases the risk of life-threatening infectious complications, leads to Cushing syndrome, bone loss, diabetes, avascular necrosis, myodystrophy, dyslipidemia, hypertension, suppressed growth in children, cataract, glaucoma and chronic adrenal insufficiency. All these complications create a significant healthcare burden on transplant centers and local providers [3]. Furthermore it was demonstrated that less than a half of patients have complete resolution of all cGVHD manifestations after systemic treatment, but rather either reduce or completely discontinue systemic immunosuppressive therapy with topically manageable residual cGVHD symptoms [4]. These two later factors have driven transplantation physicians to explore immunosuppressive approaches with reduced number of complications compared to corticosteroids. These alterations in clinical practice were further encouraged by introduction of new GVHD prophylaxis regimens, including ones without calcineurin inhibitors [5], and effective second-line options for cGVHD treatment that allow tapering and discontinuation of corticosteroids [6, 7]. Nonetheless, no prospective or large retrospective studies are published using these strategies in the first line systemic treatment of moderate and severe cGVHD by National Cacner Institute definitions. Thus a survey was conducted (Supplementary text S1) among all member centers of European Society for Blood and Marrow Transplantation (EBMT) performing allogeneic stem cell transplantation (HSCT) evaluating if steroid-free regimens are used in clinical practice. Statistical methods are presented in supplementary text S2.
Among 327 allogeneic EBMT centers, 102 (31%) centers from 27 countries responded to the questionnaire and 53 (51.9%) confirmed using steroid-free first line treatments in moderate and severe cGVHD (Fig. 1a). There was no association between using this approach and transplant center activity (p = 0.95) or JACIE accreditation status (p = 0.48). Among centers using steroid-free treatments, 75% use them in rare clinical situations, 19% in specific clinical situations and 6% for majority of patients. Most commonly used treatments included calcineurin inhibitors (CNIs) (71.4%), ruxolitinib (61.2%) and extracorporeal photopheresis (57.1%). A lower proportion of centers reported using mycophenolate mofetil (MMF) (32.7%), sirolimus (22.5%), rituximab (16.3%), imatinib (12.2%), methotrexate (12.2%), ibrutinib (8.2%) and other tyrosine kinase inhibitors (8.2%). No other steroid-free approaches were reported (Fig. 1b). The following approaches were reported by the centers as second line when remaining within steroid-free strategy: ruxolitinib (44.9%), extracorporeal photopheresis (14.3%) and calcineurin inhibitors (14.3%). Other regimens were also rarely used as a second choice of treatment (Supplementary Tables S1, S2).Fig. 1 Key information from the survey.
a Percentage of centers reporting use of steroid-free first line treatment of moderate and severe cGVHD. b Percentage of centers reporting specific steroid-free first line treatments. More than one answer was possible. c Percentage of centers reporting clinical indications for steroid-free first line treatments. More than one answer was possible. *Low lymphocyte count, abnormal glucose levels; ** steroid myopathy, osteonecrosis.
The majority of centers (59.2%) reported that steroid-free treatment was used only in a minority of cGVHD patients (defined by questionnaire as <10%), while 12 centers use this strategy in 10–20% of patients, 4 centers in 20–30%, 3 centers in 30–50% and only one in the majority of patients. The choice of steroid-free regimen was rarely dictated by underlying disease. Only 3 centers prefer to use this approach in myelofibrosis and one in non-hodgkin lymphomas. The rest reported that the decision is driven by other clinical factors, predominantly either ongoing severe infectious complications (63.3%) or concern for potential infectious complications (59.2%). Among infectious complications the responders highlighted the significance of fungal (44.9%), viral infections (44.9%) and to a lesser extent bacterial infections (28.6%) when choosing steroid-free approach. Another significant factor to avoid corticosteroids was presence of severe comorbidities (44.9%), including diabetes, severe osteoporosis, avascular osteonecrosis, preexisting myopathy, sarcopenia, poor performance status, obesity and metabolic syndrome, hypertension and uncontrolled psychiatric disorder. Several centers reported that neutropenia (2.0%), lymphopenia (10.2%) and patient preference (10.2%) might drive the choice of treatment (Fig. 1c). Underlying malignancy status was considered as additional factor when choosing steroid-free treatment by 53.1% of centers. Previous hematological relapse (8.16%), minimal residual disease (10.2%) or both (34.7%) were considered when choosing steroid-free approach. Also 22.45% considered mixed chimerism affecting their judgment regarding cGVHD steroid-free treatment. On the contrary, nor risk of COVID-19 (93.9%), neither active COVID-19 infection (87.8%) significantly affected clinical decision towards steroid-free regimens. The respondents commented that usually they postpone treatment until clearance of COVID-19.
Manifestations of cGVHD were also considered when choosing steroid-free first line treatment. The majority of centers use this approach in patients with moderate disease (65.3%), 30.6% in both moderate and severe, and 4.08% in certain forms of severe cGVHD. Centers preferably use steroid-free treatment in presence of skin involvement without scleroderma (63.3%) or with scleroderma (49.0%), involvement of oral mucosa (46.9%), eyes (40.8%), liver (32.7%), gastrointestinal (30.6%), joints (26.5%), lungs (20.4%) and genitalia (16.3%), which reflects the usual organ involvement in moderate and severe cGVHD. On the other hand, in a number of centers lung (66.0%), gastrointestinal (46.8%) and hepatic (25.5%) cGVHD is considered a contraindication to steroid-free approach. The other organ manifestations of cGVHD were considered a contraindication by smaller number of centers (Supplementary Table S3).
The majority of centers used National Institute of Health criteria to assess response (59.6%), however a significant proportion defined “clinical improvement” as a sign of response (40.4%). Major criteria for changing therapy were no response by 4 weeks (in 57.1% off centers), by 8 weeks (in 18.37%), 12 weeks (in 10.2%) and 24 weeks (in one center). Only 20.4% of centers indicated absence of complete response by 6–24 months as a sign of treatment failure. On the other hand, the majority of respondents indicated that progressive disease at any time point is an indication for different therapy. Addition of either standard-dose (≥1 mg/kg, 36.7%), or low-dose corticosteroids (<1 mg/kg, 46.9%) was indicated as the most common practice in case of first line failure. Only 16.3% remain within steroid-free strategy after first line failure and use ECP, combination of ECP and ruxolitinib, combination of rituximab, ECP and TKIs, combinations with MMF, methotrexate and CNIs. Majority of centers (87.8%) continue the initial steroid-free regimen while adding second line treatment.
The survey identified that more than half of the centers do use steroid-free regimens for cGVHD outside of clinical recommendations. Emergence of these practices are likely related to the changing landscape of either GVHD prophylaxis [5] and second-line GVHD treatment [6]. We identified that most commonly used steroid-free treatments involve CNIs, ECP and ruxolitinib. These methods were never studied in the randomized studies against corticosteroids, but all of them demonstrated steroid-sparing potential and allowed to discontinue steroids completely in a proportion of patients [7–9].
We observed that steroid-free practice is typically applied to less than 20% of patients with certain comorbidities and previous complications of HSCT. These results are in line with a recent multicenter study from the USA where 18% of patients were treated with steroid-free first line regimen [10]. Given the significant number of patients already treated with this approach our study warrants boosting research within the EBMT community both in the retrospective and prospective settings and define the optimal strategy for specific chronic GVHD manifestations. On the other hand, the study raises a concern about the extent of steroid-free treatment practices without existing evidence on outcomes and understanding of a clinical benefit to patients. Currently the response rate with this approach is unknown and might be relatively low. Twenty eight centers that responded to the survey agreed to participate in a retrospective analysis, which will identify safety and efficacy of first line steroid-free approach in moderate and severe chronic GVHD. The results of the present survey can provide guidance for further studies.
Supplementary information
Supplementary materials
Supplementary information
The online version contains supplementary material available at 10.1038/s41409-022-01881-6.
Acknowledgements
We thank all the centers that showed interest in this problem and responded to the survey.
Author contributions
IM: conceptualization, methodology, project administration, writing - original draft, visualization; PA: investigation, data curation; MB: methodology, validation; CP: dormal analysis; GB: methodology, writing - review & editing; CK: methodology, writing - review & editing; HS: methodology, writing – review & editing; OP: methodology, writing – review & editing; ZP: methodology, writing – review & editing, supervision.
Data availability
For primary data please contact EBMT study office, [email protected].
Competing interests
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Sullivan KM Shulman HM Storb R Weiden PL Witherspoon RP McDonald GB Chronic graft-versus-host disease in 52 patients: adverse natural course and successful treatment with combination immunosuppression Blood. 1981 57 267 76 10.1182/blood.V57.2.267.267 7004534
2. Flowers ME Martin PJ How we treat chronic graft-versus-host disease Blood. 2015 125 606 15 10.1182/blood-2014-08-551994 25398933
3. Fuji S Byrne M Nagler A Mohty M Savani BN How we can mitigate the side effects associated with systemic glucocorticoid after allogeneic hematopoietic cell transplantation Bone Marrow Transplant 2021 56 1248 56 10.1038/s41409-020-01205-6 33514922
4. Lee SJ Nguyen TD Onstad L Bar M Krakow EF Salit RB Success of immunosuppressive treatments in patients with chronic graft-versus-host disease Biol Blood Marrow Transplant 2018 24 555 62 10.1016/j.bbmt.2017.10.042 29133250
5. Kanakry CG Bolaños-Meade J Kasamon YL Zahurak M Durakovic N Furlong T Low immunosuppressive burden after HLA-matched related or unrelated BMT using posttransplantation cyclophosphamide Blood. 2017 129 1389 93 10.1182/blood-2016-09-737825 28049637
6. Zeiser R Polverelli N Ram R Hashmi SK Chakraverty R Middeke JM REACH3 investigators. ruxolitinib for glucocorticoid-refractory chronic graft-versus-host disease N Engl J Med 2021 385 228 38 10.1056/NEJMoa2033122 34260836
7. Cutler C Lee SJ Arai S Rotta M Zoghi B Lazaryan A Belumosudil for chronic graft-versus-host disease after 2 or more prior lines of therapy: the ROCKstar Study Blood 2021 138 2278 89 10.1182/blood.2021012021 34265047
8. Koc S Leisenring W Flowers ME Anasetti C Deeg HJ Nash RA Therapy for chronic graft-versus-host disease: a randomized trial comparing cyclosporine plus prednisone versus prednisone alone Blood. 2002 100 48 51 10.1182/blood.v100.1.48 12070007
9. Jagasia M Scheid C Socié G Ayuk FA Tischer J Donato ML Randomized controlled study of ECP with methoxsalen as first-line treatment of patients with moderate to severe cGVHD Blood Adv 2019 3 2218 29 10.1182/bloodadvances.2019000145 31332045
10. Pidala J Onstad L Martin PJ Hamilton BK Cutler C Kitko CL Initial therapy for chronic graft-versus-host disease: analysis of practice variation and failure-free survival Blood Adv 2021 5 4549 59 10.1182/bloodadvances.2021005286 34535015
| 36446899 | PMC9707203 | NO-CC CODE | 2022-12-01 23:20:21 | no | Bone Marrow Transplant. 2022 Nov 29;:1-3 | utf-8 | Bone Marrow Transplant | 2,022 | 10.1038/s41409-022-01881-6 | oa_other |
==== Front
Program Comput Soft
Programming and Computer Software
0361-7688
1608-3261
Pleiades Publishing Moscow
3685
10.1134/S0361768822060044
Article
Hidden Monitoring Based on Keystroke Dynamics in Online Examination System
https://orcid.org/0000-0003-4473-528X
Kochegurova E. A. [email protected]
https://orcid.org/0000-0001-8852-0737
Zateev R. P. [email protected]
grid.27736.37 0000 0000 9321 1499 National Research Tomsk Polytechnic University, pr. Lenina 30, 634050 Tomsk, Russia
28 11 2022
2022
48 6 385398
22 6 2022
4 7 2022
6 7 2022
© Pleiades Publishing, Ltd. 2022, ISSN 0361-7688, Programming and Computer Software, 2022, Vol. 48, No. 6, pp. 385–398. © Pleiades Publishing, Ltd., 2022.Russian Text © The Author(s), 2022, published in Programmirovanie, 2022, Vol. 48, No. 6.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The COVID-19 pandemic has accelerated the development of distance learning technologies, where online tests and exams play an important role. In online testing, the detection of various types of academic fraud, including cases where the examinee is falsely represented by another person, is of particular importance. Continuous biometric (behavioral) authentication can be a solution to counter unauthorized access. This paper proposes an authentication technology based on keystroke dynamics and hidden monitoring. An application for collecting and updating keystroke dynamics profiles of domain users and their continuous authentication is developed. The efficiency of reducing the dimension of the keystroke feature space based on the frequency of alphabetic letters is demonstrated. Results of popular performance metrics (FAR, FRR, ERR, ROC, and DET) are significantly improved already when evaluating only metric distances. For instance, ERR is reduced from 10.1% to 0.79%, which is comparable to the estimates provided by the kNN method for its optimal parameters.
issue-copyright-statement© Pleiades Publishing, Ltd. 2022
==== Body
pmcINTRODUCTION
The COVID-19 pandemic has had a significant impact on almost all spheres of everyday life. To reduce the infection rate, many companies and organizations transferred their employees to remote jobs and implemented other restrictive measures. Thereby, they accelerated the progress in digitalization, distance learning, telemedicine, online commerce, etc.
However, the positive technological impetus to the development of the IT industry has inevitably led to the growth of cybercrime. In the year the pandemic began (2020), a sharp increase in the number of cybercrimes, both throughout the world and in Russia, was recorded. According to Data Economy, an autonomous non-profit organization, in 2020, 363 thousand cybercrimes were reported in Russia, which is 77% more than in the previous year [1]. The statistics published by InfoWatch also indicates, in addition to the increase in the number of attacks by hacker groups, the emergence of new risks associated with remote work [2]. In 2021, a sharp increase in the number of intentional information leaks (82% of the total number) and leaks resulted from actions of cybercriminals (up to 63%) was recorded worldwide. Russia’s share in the number of global information leaks is quite impressive and amounts to 16.9%.
According to Ivideon, the number of cyber attacks in 2021, as compared to the previous year, increased by 40% worldwide and by 54% in Russia.
Even though information security has always been one of the highest priorities for any organization, this problem becomes especially important under the current conditions.
The education sector is one of the most vulnerable sectors in terms of cybersecurity, which is due to the mentality of students and circulation of large amounts of confidential information. This makes universities attractive targets for hackers. In September 2021 alone, there were more than 10 incidents involving ransomware that targeted personal data of students and teachers [3].
On the other hand, the pandemic has had a significant impact on the traditional system of higher and school education. According to UNESCO [4], the crisis has affected almost 1.6 billion students in more than 190 countries on all continents. The closure of schools and other educational institutions affected 94% students worldwide. Even though the number of partially or completely closed educational organizations decreased to 7.5% in September 2021 [5], the forms and methods of education have significantly transformed towards distance learning. However, universities are still not ready for full-scale implementation of distance learning. In the context of higher education, online learning is often reduced to webinars, video/audio conferences, and use of e-learning platforms (Moodle, WebTutor, and iSpring Learn).
Online tests and exams are of great importance for e-learning. Tests allow the teacher to receive feedback from each student, estimate the academic progress, and improve the learning process. However, students sometimes resort to a number of academic fraud methods during online tests, including impersonation [6, 7]. Currently, there are no efficient methods for detecting this type of fraud.
ACADEMIC INTEGRITY OF ONLINE LEARNING
Academic integrity is a problem not only for the education sector but also for the whole society. Academic integrity violations negatively affect the educational environment, skills acquired by students, and the overall image of the university.
The global transition of higher education to online technologies inevitably activated many traditional forms of academic fraud: plagiarism, falsification of results, unauthorized collaboration, impersonation, etc. In addition to organizational means used to counteract these types of fraud (the university code of conduct for students and administrative sanctions for academic integrity violations), there are also methodological recommendations for teachers. For online classes, it is recommended to
– increase the individuality and specificity of test assignments;
– reduce the factological part of assignments and augment their conceptual part to develop thinking skills;
– increase the number of synchronous checks of answers in online mode;
– increase the number of small and simple assignments and use free forms, e.g., an essay, for large assignments.
The proposed measures [8] partially reduce the number of academic integrity violations; however, they significantly increase the workload on the teacher.
Continuous authentication by hidden monitoring can be a solution to counter unauthorized cooperation and impersonation.
KEYSTROKE AUTHENTICATION AND IDENTIFICATION
To protect a computer system from unauthorized access, two-step verification is generally used:
– primary identification: establishing the identity of a user, i.e., confirming the legitimacy of an authorized user;
– dynamic (continuous) authentication: the continuous confirmation of the identity of the legitimate user.
Authentication Methods
Authentication is a process of comparing the data provided by the user with the data stored in a directory services database. If these data match, then the user is given access to protected resources; otherwise, access is denied [9].
There are several methods for user authentication (see Fig. 1). These methods can be divided into three main groups based on the following paradigms [10]:
Fig. 1. Authentication methods.
– something you know (e.g., password or PIN): knowledge-based authentication;
– something you have (e.g., token or smart card): attribute-based authentication;
– something you are, i.e., physiological and behavioral biometrics.
The “something you are” paradigm uses biometric features: physiological (fingerprint, face, iris, etc.) and/or behavioral (handwriting or keystroke patterns, gait, etc.) [19].
User recognition based on keystroke dynamics is quite attractive to organizations because it has lower implementation cost in comparison to other biometric methods, as it does not require any additional hardware. In addition to a standard keyboard, a highly efficient software application is required. While being rather accurate, this method is also capable of recognizing users in hidden mode, which is comfortable and does not distract users. Individual keystrokes, rhythm, and typing speed determine the keystroke dynamics (KD) of a user. All behavioral characteristics, including the KD, can gradually change with time; however, the probability of stealing or imitating these data is significantly lower than that for physiological data. That is why the KD can be useful for identification and authentication [11, 12]. The advantages, disadvantages, and examples of the authentication methods considered above are summarized in Table 1.
Table 1. Characteristics of authentication methods
Method Advantages Disadvantages Examples
Password 1. Easy to implement
2. Unambiguous recognition
1. Can be forgotten or stolen 1. Password
2. PIN
Attribute 1. Easy to implement
2. Low cost
1. Can be lost or stolen 1. Key
2. Smartcard
3. Token
Biometrics 1. Uniqueness
2. Impossible to forget/lose
1. Implementation cost
2. Data variability regardless of a person
1. Fingerprint
2. Voice
3. Keystroke pattern
Authentication Modes
Hidden continuous monitoring is the most reasonable (for recognition systems) and comfortable (for users) identification method. According to the global security survey published by IBM (2018), 44% of the respondents considered biometrics to be the most secure authentication method, while 65% found that biometrics facilitates the authentication process [13].
The KD, as a biometric behavioral characteristic, is dynamic in nature. It includes constant and random components. The constant component is due to human physiology, i.e., the user’s ability and skills in working with a keyboard. The random component depends on the psycho-emotional state of the user.
For the recognition, the dynamic behavioral features of the KD are more complex than the physiological ones. However, it is the behavioral features that are difficult to fake [10], which improves the effectiveness of impersonation detection.
Depending on the type of text (fixed or free), two main authentication modes can be distinguished [14]:
– static authentication (primary or by event);
– dynamic authentication (continuous).
It is continuous authentication that makes it possible to organize hidden identity verification of users during the entire session in any application. The recognition system captures keystrokes and compares them with an available template [6, 11, 17–19].
Static authentication can supplement the initial login procedure, or it can be activated when suspecting an intrusion [15, 16].
Both static and continuous authentications provide the second level of security when the user is already logged in. Meanwhile, dynamic authentication is aimed at continuous verification of the users or their psycho-emotional states.
This study is devoted to user authentication by continuous monitoring of KD features in online exam applications. More specifically, we focus on the following problems:
– collecting and updating KD samples for domain users;
– extending well-known static identification approaches to the case of dynamic identity recognition based on free and long texts;
– reducing the dimension of the space of extracted keystroke features to improve the selective properties of KD samples and the efficiency of continuous authentication.
Authentication Lifecycle
KD-based continuous authentication has a registration stage and an authentication stage (see Fig. 2).
Fig. 2. Continuous authentication lifecycle.
At the registration stage, the system receives keystroke data. Then, keystroke features (hold time, flight time, etc.) are extracted, a keystroke profile (template) is formed or modified in the database, and user authentication is carried out.
Thus, the lifecycle of continuous authentication includes four main steps.
I. Collection of KD data is a continuous process when the user works with a keyboard in any application. In Windows, the message interception mechanism is used. The Windows hook makes it possible to detect any keystroke event [20]. The OS captures the ANSI code and timestamps of key presses (Down/Press) and key releases (Up/Release). The accuracy of keystroke measurements is up to milliseconds.
II. Extraction of classification features. The raw keystroke data must be cleaned of outliers and invalid values; in some cases, they need to be normalized. Based on these data, more significant KD features (typing rhythm, typing speed, flight time, etc.) can be obtained, which reflect the unique behavioral characteristics of the user. There are many KD features with the most popular ones among researchers being digrams (digraphs): timestamps of two key states [17, 21–23]. The main KD features are as follows:
– hold (dwell) time;
– flight time;
– typing speed;
– number of typing errors;
– degree of arrhythmy when typing;
– specific use of service keys.
Figure 3 shows some of the most commonly used time and frequency features.
Fig. 3. Keystrokes in the down time/up time notation.
– DU: dwell time (DT) is a time interval between a key press (down, press) and a key release (up, release).
– UD: flight time (F-RP) is a time interval between the release of the previous key and the press of the next key.
– UU, F-PP or DD, F-RR are intervals between pressing or releasing one key and pressing or releasing the next key, respectively.
Subsystems for preprocessing timestamp data and extracting KD features form a set of desired features for each keystroke. Using this set, a keystroke profile (template) of the user is generated and placed in the database.
As a result of KD feature extraction, we obtain a bank of profiles, which requires adaptation. Being a biometric behavioral characteristic, the keystroke profile is variable and needs to be updated. For this purpose, the sliding window technique can be employed [49]. The bank of profiles is used to train the classifier and at the stage of identity recognition.
III. User recognition.
To protect a computer system from unauthorized access, two-step verification is generally implemented:
– primary identification: the identification of the user;
– dynamic (continuous) authentication: the continuous confirmation of the identity of the legitimate user.
Authentication is a task of classifying users registered in the system.
The basic methods and algorithms for user classification (recognition) are the same for static and continuous (dynamic) authentication. They can be divided into three groups:
– statistical;
– based on closeness estimates;
– machine learning.
The use of these methods and the main representatives of each group are considered in detail in [20].
Historically, the works on keystroke dynamics recognition dealt with predefined and structured texts, i.e., with static authentication. According to different authors [22, 24], the number of studies devoted to continuous authentication is still small and does not exceed 10–15% of all works on keystroke recognition. The first fruitful investigation of continuous authentication is the work by Gunetti, published in 2005, where recognition accuracy exceeded 90% [25]. For comparison, the accuracy of the very first recognition methods on free texts (Monrose, 1997) was 23%.
The reviews on keystroke recognition published in the last decade made it possible to generalize the data on the effectiveness of continuous authentication (the main results are presented in Table 2). In addition to the references, the table includes the classification parameter, recognition method, and performance indicators. The data were borrowed from our own studies [20, 26] or adapted from the reviews [17, 22, 24, 27–33].
Table 2. Dynamic identification methods
Year Work, author Parameter Method Efficiency
2005 [25] Gunetti FT Distance, R/A FAR-0.005%,
FRR-5%
2010 [32] Shimshon Clustering FAR 3.47%
FRR 0%
2011 [33] Messerman Statistical FAR-2.02%,
FRR-1.84%
2011 [37] Solami Clustering Accuracy 100%
2013 [27] Alsultan digraph Fusion FAR-21%, FRR-17%
2014 [35] Ahmed digraph Neuronal network FAR-0.015%, FRR-4.82%
2014 [39] Antal DT, FT Statistical
Support vector machine
Neural network
Decision tree
Accuracy 93.04%
2014 [40] Locklear Statistical EER
4.55–13.37%
2015 [41] Kang DT, FT Clustering,
distance
EER 3.8%
2015 [42] Matsubara digraph, DT Distance Accuracy 99%
2016 [23] Morales digraph, n-graph kNN,
distance
Accuracy 90%
2017 [31] Alsultan digraph, DT Support vector machine FAR 0.169,
FRR 0.423
2017 [36] Goodkind contextual features Naive Bayes Accuracy 82.2%
2017 [30] Ali kNN EER 3.7%
2021 [34] Chang DT, FT CNN-GRU Accuracy 99%,
EER 0.0690
Finally, we should note the conditional nature of KD-based approaches to user recognition. These approaches generally include a model and training methods; however, their combinations can be different.
IV. Decision making on user’s legitimacy.
This step fully depends on a particular applied problem and recognition performance indicators.
In the case of dynamic identification, the main goal of continuous monitoring is to provide permanent access to network resources for a user registered in the domain, as well as to prevent access for an unregistered user.
Generally, the probabilities of the corresponding false events are estimated: a false denial of access for a registered user and a false provision of access for an unregistered user. By analogy with estimates in statistical radio engineering, the following errors are most commonly used in keystroke dynamics studies [17, 20, 21, 36, 38].
– False rejection rate (FRR), which is a type I error that estimates the false denial of access to a legitimate (registered) user:
1 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$FRR = \frac{{FR}}{{TA + FA + TR + FR}}.$$\end{document}
– False acceptance rate (FAR), which is a type II error that estimates the false provision of access to an illegitimate user:
2 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$FRR = \frac{{FA}}{{TA + FA + TR + FR}}.$$\end{document}
Here,
– true accept (TA) is the number of valid accepts of legitimate users;
– true reject (TR) is the number of valid rejects of illegitimate users;
– false accept (FA) is the number of invalid accepts of illegitimate users;
– false reject (FR) is the number of invalid rejects of legitimate users.
The denominators of (1) and (2) contain the total number of attempts.
FRR and FAR depend on the adjustable threshold or sensitivity of the algorithm (see Fig. 4).
Fig. 4. Performance indicators for keystroke-based identification.
FRR, FAR, and ERR are sufficient for making a decision on the acceptance or rejection of the user. If the goal of the monitoring system is to provide a high level of security, then low thresholds should be used, which correspond to a large percentage of false rejects (FRR). Large FRR values provide higher security and more difficult access to the system for everyone, both insiders and outsiders. For a high threshold and low sensitivity of recognition algorithms, access becomes easier; however, FAR and the number of falsely accepted users increase. The tradeoff between FRR and FAR has to be determined individually in each case.
Another popular indicator that does not depend on the threshold is the equal error rate (EER). EER values correspond to equal values of FRR and FAR, which makes EER a universal performance indicator for any authentication system.
These three metrics (FRR, FAR, and EER) are most widely used for decision making in monitoring and authentication systems. Studies on keystroke dynamics also frequently employ the receiver operating characteristic (ROC), which is a ratio between TA and FA for different thresholds. ROC reflects the limiting capabilities of algorithms, which is especially important when investigating various classifiers.
HIDDEN MONITORING IN ONLINE EXAMINATION SYSTEMS
The COVID-19 pandemic has accelerated the introduction of the blended learning model, and many universities have implemented learning management systems. The main contribution to the final assessment of students' performance in these systems is made by online exams and tests. However, it is these forms of online learning that are most prone to academic fraud because the online environment allows students to work with little or no control. According to some researchers [43–45], the success of online exam systems is due to the use of biometric control and continuous monitoring during online exams. Keystroke dynamics makes it possible to organize the hidden monitoring of the identity of a legitimate student (who has passed primary identification) in a mode comfortable for him or her.
The main problems of keystroke dynamics—lifecycle, components, and efficiency—are discussed in Section 3. A specific use of these technologies is considered in Sections 4.1–4.4.
Authentication Lifecycle
The principles of continuous authentication outlined above are used to confirm the legitimacy of a student in the process of online testing.
The capabilities of continuous authentication in solving this problem are investigated by computational experiment. The experiment is conducted in accordance with the block diagram of the system shown in Fig. 5, which includes three main subsystems:
Fig. 5. Continuous authentication architecture (block diagram).
– registration;
– adaptation or template updating;
– user authentication.
In the case of distance learning, each subsystem operates in continuous mode. As shown in Fig. 5, the common steps of registration and authentication are data collection, data processing, and keystroke feature extraction.
4.1.1. Data collection in the experiment. The development of a continuous authentication system and its performance testing are carried out using some keystroke dataset, statistical or locally collected one.
There are several public datasets based on fixed texts (Buffalo and BiosecurID) and free texts (Clarkson II and Villani-2010) [18, 23, 46, 47]. Some fixed datasets (CMU and WEBGREYC) [46] form templates by using login and password, which is not acceptable for dynamic authentication. In addition, these datasets do not include templates for Russian texts. When recognizing Russian users, English-language templates are of interest only for primary identification and static authentication based on password data.
In this work, local data collection is carried out using a special application developed for Windows. The application has client–server architecture. On the client side, keystroke data are collected locally in the background.
The target audience of this study is the university domain whose users have above-average computer skills. Using the keyboard input interception mechanism implemented in Windows, the so-called hook traps, all keyboard events are recorded. For each event, an array is formed that contains the following data:
– user login in the domain;
– key code;
– event (key pressed/released);
– timestamp.
The collected data are represented in the text format as follows (see Fig. 6).
Fig. 6.
The other functions of the application are implemented on the server. This reduces the risk of stealing the keystroke data from less secure client machines.
4.1.2. Data preprocessing. The raw data collected during the user session are preprocessed in the system to eliminate outliers, unreasonable hold times (30–200 ms), unpaired events, and short sessions.
The most important part of the preprocessing is the choice of the duration of the session. Mathematically, this corresponds to the choice of time windows into which the flow of keyboard events is partitioned. The sizes of the windows can differ at the stages of registration and authentication. There are various methods used for this purpose [18]:
– partitioning based on the duration of the user session or on the number of clicks;
– sliding windows.
This study uses a sliding window of 500 characters in continuous authentication mode. The size of the data sample must be large enough to ensure the representativeness of individual characters in the window, as well as guarantee that their estimates are statistically significant and unbiased. However, it is well known that the keystroke pattern tends to change due to the psycho-emotional component. That is why the time window should be sliding. The sliding window is also opposed by the possibility of reverse biometrics: the possibility of compromising long unprotected templates. By combining these two methods, we select the minimum sample size for the consistent estimate of the average values for Russian letters.
4.1.3. Extraction of KD features and generation of keystroke templates. The dimension of the resulting feature space in the continuous authentication problem is quite high. This is due to the size of the Russian alphabet and the large number of KD features.
The analysis of the researches devoted to keystroke features showed that the most popular ones are DU and UD (dwell time and flight time): according to [12, 20], the frequency of using each of them ranges from 30% to 40% in applied research.
In our study, the keystroke data from the last session of each user are sent to the server in the text format.
Then, the statistical characteristics DU and UD are computed for each letter during a user session. The result is an updated keystroke template, adapted after the last session and saved in a JSON file.
4.1.4. Vector parameter generation. The keystroke template stores the processed and reliable data on the keystroke dynamics of users in the last session in terms of the selected KD features (average dwell values for each key).
To improve the informativity and reliability of the generated templates, additional metrics and tools can be used to derive an aggregate KD indicator.
Dimensionality reduction based on extraction of stable features.
Reducing the dimension of the feature space is an important problem for keystroke dynamics, especially in the context of distance education and online testing, where the confirmation of the legitimacy of a student must be carried out as quickly as possible to detect impersonation.
On the one hand, the extraction of stable KD features contributes to the reduction of the feature space. On the other hand, it improves the selective properties of the templates.
When extracting stable features, we can improve the informativity of the KD features themselves [49]. The idea of these methods is to narrow timestamp ranges based on some rule, most often, by using statistical criteria. In this case, the narrowing technique can be applied to individual sessions, individual users, etc.
In addition, to improve the informativity of KD features, heuristic methods can be employed. In this study, this method is associated with the frequency of alphabetic letters in texts. According to the Russian National Corpus (https://ruscorpora.ru), the frequency of Russian letters (expressed in %) decreases from 10.985 (letter О) to 0.9965 (letter Х) and 0.0375 (letter Ъ) (see Fig. 7a). Therefore, to obtain consistent estimates for all Russian letters, we need a very large sliding window, which is unacceptable in online authentication. A way to reduce the size of the window is to use a letter frequency threshold as a stable feature. For the threshold selected in this study (0.55), this means that the user template does not include the rarely used letters Ц, Щ, Э, Ф, Ъ, and Ё (see Fig. 7b).
Fig. 7. Frequencies of Russian letters.
The KD feature vector includes the features (the average DU values of the last session) of 27 Russian letters with the weight coefficients corresponding to the frequency of these letters in texts (see Fig. 7b).
4.1.5. Recognition of legitimate users. Static authentication and dynamic authentication are both aimed at confirming the legitimacy of the user who has already passed the initial registration. The difference is only in the type of the generated text. In the case of dynamic (continuous) authentication, the user creates free text in any OS application at any time; in the case of static authentication, the text and time of its generation are predetermined by the security system.
It should be noted that the confirmation of the user’s legitimacy based on his or her keystroke dynamics is a one-class classification problem. This problem is more complex than multiclass classification because the data of an illegitimate user are not available to the recognition system. The classifier is trained on objects of one class; in the process of testing, the algorithm determines whether a new object belongs to this class. The data of registered users are stored in the system in the form of dynamically updated templates. They can be considered objects of the same class (that of legitimate users).
In online learning systems, one-class classification can be carried out to recognize unregistered users, i.e., malefactors.
In online testing, one-class classification prevents impersonation of students.
Even though dynamic authentication is much more complex than static authentication, the recognition methods are similar. A brief review of the dynamic authentication methods developed over the last decade is presented in Table 3. The classification of recognition methods is quite conventional; however, it is possible to distinguish machine learning methods, statistical methods, and methods based on metric distances [20].
Table 3. Comparative analysis of performance indicators
distance method
Euclidean Manhattan kNN SVM
frequency frequency
– + – +
ERR, (0-100)% 10.8 0.99 10.1 0.79 0.54 3.72
threshold (FAR=FRR), ms 4.8 2.3 4.8 2.2 15
Accuracy, (0-100)% 90.87 99.20 83.41 99.4 99.45
Precision, (0-1) 0.83 0.98 0.73 0.98 0.98
Recall, (0-1) 0.71 0.83 0.74 0.94 0.98
In this work, the most popular recognition methods are chosen as classifiers in each of the three groups for the one-class classification problem:
– support vector machines (SVM) [17, 21, 34];
– k-nearest neighbors (kNN) algorithm [17, 21, 27, 34, 46];
– Euclidean and Manhattan metrics.
The basic idea of a one-class support vector machine (OCSVM) is to detect the boundaries of one class, rather than to separate objects of several classes as in the multiclass problem [48]. In addition, one-class classification is also known as an anomaly detection problem, which consists in finding patterns in data that do not conform to expected behavior [49], e.g., when the psycho-emotional state of the user changes under the influence of alcohol, etc.
The OCSVM maps feature vectors to a higher-dimensional space by using a kernel function. In the case of a radial basis function (RBF) kernel, there is a hyperplane that separates the majority of objects of a given class (in our case, legitimate users) from the origin. Exceptions are objects that lie closer to the origin than the hyperplane.
A one-class kNN (OCkNN) algorithm estimates the distance between class objects. A new object is considered an exception if the majority of the objects (e.g., their pth part) of the entire class are at a distance exceeding D from the new object. The distance is computed using a selected metric in the feature space (p and D are the parameters of the method).
The OCSVM and OCkNN algorithms are quite easy to implement. Their main requirement is the representativeness of the source dataset.
RESULTS
The main processes of the continuous authentication lifecycle in online learning are data registration and student authentication (see Fig. 2). In the process of registration, the continuous collection of keystroke data with subsequent extraction of keystroke features is carried out (Fig. 5).
The templates (profiles) of domain users are dynamic; they are formed by continuous monitoring of keystroke dynamics and do not depend on any particular application. The keystroke data for each domain user are collected in the sliding window and the template is formed using Russian and English letters. The study was carried out in the domain of a national Russian university. The window size was 500 clicks, upon accumulation of which the data were transferred to the server component of the program for preprocessing and KD feature extraction. As the number of clicks grew, the window was shifted and the next set of data was transmitted to the server, thus enabling continuous data registration in the current user session.
The server component of the program computes the average dwell times in the current session for each character and updates the template stored in the bank.
For each user, the bank stores templates for the last 10 sessions, which makes it possible to track changes in keystroke dynamics, which are due to fatigue, psycho-emotional state, etc.
The identification capabilities of templates generated for different users can be estimated visually and statistically. Figure 8 shows templates of six random users of the university domain for Russian letters. The Y-axis represents the dwell time (in ms). The individual lines depict the typing rhythm and determine the keystroke pattern of each user.
Fig. 8. Visualization of domain user templates.
Typing rhythms and speeds differ for different users, which results in different keystroke dynamics. This difference of KD features is also confirmed statistically, e.g., by the type and parameters of the distribution density (see Fig. 9).
Fig. 9. Distribution density for the KD parameters.
For instance, the distribution for User3 is characterized by a small dispersion and average typing speed (expectation), which is in complete agreement with Fig. 8 and is confirmed by small deviations between the mode and median of the series (large and small dashed lines).
Figures 8 and 9 show that the average values for User4 and User1 are approximately equal. However, the distribution for User1 is heavy-tailed. The mode and median for this series differ significantly from one another, as in the case of User06. For a representative sample and high typing speed, this can be a discriminative feature of an altered psycho-emotional state.
User5 also has a high typing speed. In this case, the distribution density is bimodal, which corresponds to good typing skills. For this user, there are typing overlaps when the next key is pressed while the previous one is not released.
The next stage of recognition is the confirmation of the legitimacy of the student. For this purpose, the template of the registered user from the bank is compared with the current template (see Fig. 5). When confirming the legitimacy of the user, there are two possible outcomes:
– the template of the last session matches the template from the bank for the registered user;
– the templates do not match.
The pattern matching is analyzed using SVM, kNN, Euclidean metric, and Manhattan metric. The choice of these methods is substantiated in Section 4.1.5.
The most important characteristic for any method is the decision threshold. The threshold is selected (assigned) by the security system based on priorities of tasks. Small thresholds correspond to small differences between the basic and current templates and complicate access to the corporate network for everyone, including legitimate users. Large thresholds (low sensitivity) correspond to easy access for everyone. Type I and II errors for the Manhattan metric are shown in Fig. 10 depending on threshold values.
Fig. 10. User recognition performance.
In Fig. 10b, FAR, FRR, and EER are plotted taking into account the letter frequency; in Fig. 10a, without taking it into account.
It can be seen that there is a significant decrease in the errors, by an order of magnitude on average at different threshold values. For instance, on the data from Table 3, these values are 10.85 and 0.995 for the Euclidean metric. In Table 3, for all selected methods, the EER values are given for a threshold at which FAR = FRR.
The authentication features based on the Manhattan and Euclidean metrics are almost identical with respect to the frequency of alphabetic letters. However, they have certain differences when the frequency is not taken into account. The kNN method has a slightly better performance, ERR = 0.545, at a slightly higher threshold 15 ms. However, this requires a time-consuming selection of algorithm parameters and does not guarantee that parameter adjustment would not be necessary under real-world conditions. The lack of optimal parameter values explains the low authentication results of the SVM method.
A confusion matrix is also a useful tool for analyzing the events that occur when recognizing legitimate and illegitimate users (see Table 4). The intersections of its rows and columns contain true (T) or false (F) outcomes of recognition: accept (A) or reject (R). The columns correspond to the recognition results, while the rows correspond to the real users.
Table 4. Confusion matrix
Recognized user
legitimate illegitimate
Actual user legitimate TA FR
illegitimate FA TR
The performance indicators shown in Table 3 are computed as follows:
3 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Accuracy = \frac{{TA + TR}}{{TA + FA + TR + FR}}.$$\end{document}
4 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Precision = \frac{{TA}}{{TA + FA}}.$$\end{document}
5 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Recall = \frac{{TA}}{{TA + FR}}.$$\end{document}
These three indicators reflect the quality of authentication of legitimate users from different angles.
Accuracy estimates the correct accepts and rejects for all possible legitimate and illegitimate users.
Precision is the ratio of correctly accepted users to all users accepted by the system.
Recall reflects the number of accepted users out of all legitimate ones. Recall is also referred to as sensitivity of a recognition model.
Accuracy is usually expressed as a percentage, while Precision and Recall vary from 0 to 1.
The last tools for authentication quality assessment are the DET and ROC curves shown in Fig. 11.
Fig. 11. User recognition performance.
These curves confirm the similar performance of the kNN method and the methods that are based on the similarity of templates and take the letter frequency into account.
CONCLUSIONS
In this paper, we have considered the approach to confirming the legitimacy of students in the process of distance learning by hidden monitoring of KD features.
It has been found that the hidden identification of students is possible on the basis of continuous monitoring of keystroke data for any software application. As a result, the following conclusions can be made.
1. Correction (adaptation) of KD samples by using a sliding window is required, which allows one to dynamically track changes in the user’s keystroke pattern and psycho-emotional state.
2. Selection of stable keystroke features contributes to the dimensionality reduction of the KD feature space, which improves selective properties of the templates.
3. In this study, as a stable feature, we have used the frequency of alphabetic letters in texts according to the Russian National Corpus (https://ruscorpora.ru). The exclusion of six letters with frequencies below 0.5% from the templates has led to a significant change in all performance indicators. For instance, ERR has decreased by an order of magnitude, on average from 105 to 15, while Accuracy, Precision, and Recall have increased by 6–135 on average, reaching 98% for both the Euclidean and Manhattan metrics.
4. The kNN method has showed slightly better results for optimal parameters: ERR = 0.545 (versus 0.795), Accuracy = 99%, and Precision = 0.98.
5. Combining simple distance estimation algorithms with letter frequency leads to the absence of complex parameter optimization procedures, as in machine learning methods (kNN), while providing similar performance.
With the recognition performance being a key factor in continuous authentication, parameter adaptation significantly reduces the overall efficiency of user validation.
Translated by Yu. Kornienko
==== Refs
REFERENCES
1 Klyuchevskaya, N., Information security and COVID-19: Recommendations for businesses and citizens. https://www.garant.ru/article/1421147.
2 Analytics of the information security industry. https://www.infowatch.ru/analytics/analitik.
3 Hackers attack universities and colleges: Leak digest. https://www.infowatch.ru/analytics/daydzhesty-i-obzory/khakery-atakuyut-universitety-i-kolledzhi-daydzhest-utechek.
4 Education: From the closure of educational institutions to the resumption of their work. https://ru.unesco.org/covid19/educationresponse.
5 Education: From destruction to recovery. https://ru.unesco.org/covid19/educationresponse.
6 Fenu G. Marras M. Boratto L. A multi-biometric system for continuous student authentication in e-learning platforms Pattern Recognit. Lett. 2018 113 83 92 10.1016/j.patrec.2017.03.027
7 Ngqondi, T., Maoneke, P.B., and Mauwa, L., A secure online exams conceptual framework for South African universities, Soc. Sci. Humanit. Open, 2021, vol. 3, no. 1, p. 100132. 10.1016/j.ssaho.2021.100132
8 Lee, C., How to maintain academic integrity in distance learning. https://www.turnitin.com/ru/blog/kak-podderzhivat-akademicheskuyu-chestnost-pri-distantsionnom-obuchenii.
9 Al-Naji F.H. Zagrouba R. A survey on continuous authentication methods in Internet of Things environment Comput. Commun. 2020 163 109 133 10.1016/j.comcom.2020.09.006
10 Dasgupta D. Roy A. Nag A. Advances in User Authentication 2017
11 Stylios I. Kokolakis S. Thanou O. Chatzis S. Behavioral devices: A survey Inf. Fusion 2021 66 76 99 10.1016/j.inffus.2020.08.021
12 Toosi R. Akhaee M.A. Time-frequency analysis of keystroke dynamics for user authentication Future Gener. Comput. Syst. 2021 115 438 447 10.1016/j.future.2020.09.027
13 IBM Security, Future of identity study. https://www.ibm.com/downloads/cas/QRBY08NO.
14 Hazan, I., Margalit, O., and Rokach, L., Supporting unknown number of users in keystroke dynamics models, Knowl.-Based Syst., 2021, vol. 221, p. 106982. 10.1016/j.knosys.2021.106982
15 Parkinson S. Khan S. Crampton A. Xu Q. Xie W. Liu N. Dakin K. Password policy characteristics and keystroke biometric authentication IET Biometrics 2021 10 163 178 10.1049/bme2.12017
16 Kochegurova, E.A., Gorokhova, E.S., and Mozgaleva, A.I., Development of the keystroke dynamics recognition system, J. Phys. Conf. Ser., 2017, vol. 803, p. 012073. 10.1088/1742-6596/803/1/012073
17 Kim J. Kim H. Kang P. Keystroke dynamics-based user authentication using freely typed text based on user-adaptive feature extraction and novelty detection Appl. Soft Comput. 2018 62 1077 1087 10.1016/j.asoc.2017.09.045
18 Lu X. Zhang S. Hui P. Lio P. Continuous authentication by free-text keystroke based on CNN and RNN Comput. Secur. 2020 96 01861 10.1016/j.cose.2020.101861
19 Dargan, S. and Kumar, M., A comprehensive survey on the biometric recognition systems based on physiological and behavioral modalities, Expert Syst. Appl., 2020, vol. 143, p. 113114. 10.1016/j.eswa.2019.113114
20 Kochegurova E.A. Martynova Y.A. Aspects of continuous user identification based on free texts and hidden monitoring Program. Comput. Software 2020 46 12 24 10.1134/S036176882001003X
21 Zaidi, A.Z., Chong, C.Y., Jin, Z., Parthiban, R., and Sadiq, A.S., Touch-based continuous mobile device authentication: State-of-the-art, challenges and opportunities, J. Network Comput. Appl., 2021, vol. 191, p. 103162. 10.1016/j.jnca.2021.103162
22 Teh P.S. Teoh A.B.J. Yue S. A survey of keystroke dynamics biometrics Sci. World J. 2013 2013 1 24 10.1155/2013/408280
23 Morales, A., Fierrez, J., Tolosana, R., Ortega-Garcia, J., Galbally, J., Gomez-Barrero, M., Anjos, A., and Marcel, S., KBOC: Keystroke biometrics ongoing competition, Proc. IEEE 8th Int. Conf. Biometrics Theory, Applications and Systems (BTAS), 2016. 10.1109/BTAS.2016.7791180
24 Pisani P.H. Lorena A.C. A systematic review on keystroke dynamics J. Braz. Comput. Soc. 2013 19 573 587 10.1007/s13173-013-0117-7
25 Gunetti D. Picardi C. Keystroke analysis of free text ACM Trans. Inf. Syst. Secur. 2005 8 312 347 10.1145/1085126.1085129
26 Kochegurova E. Luneva E. Gorokhova E. On continuous user authentication via hidden free-text based monitoring Adv. Intell. Syst. Comput. 2019 875 66 75 10.1007/978-3-030-01821-4_8
27 Alsultan A. Warwick K. Keystroke dynamics authentication: A survey of free-text methods Int. J. Comput. Sci. 2013 10 1 10
28 Mondal S. Bours P. A study on continuous authentication using a combination of keystroke and mouse biometrics Neurocomput. 2017 230 1 22 10.1016/j.neucom.2016.11.031
29 Zhong Y. Deng Y. A survey on keystroke dynamics biometrics: Approaches, advances, and evaluations Recent Adv. User Authentication Using Keystroke Dyn. Biom. 2015 2 1 22 10.15579/gcsr.vol2.ch1
30 Ali M.L. Monaco J.V. Tappert C.C. Keystroke biometric systems for user authentication J. Sign. Process. Syst. 2017 86 175 190 10.1007/s11265-016-1114-9
31 Alsultan A. Warwick K. Wei H. Non-conventional keystroke dynamics for user authentication Pattern Recognit. Lett. 2017 89 53 59 10.1016/j.patrec.2017.02.010
32 Shimshon, T., Moskovitch, R., Rokach, L., and Elovici, Y., Continuous verification using keystroke dynamics, Proc. Int. Conf. Computational Intelligence and Security (CIS), 2010, pp. 411–415. 10.1109/CIS.2010.95
33 Messerman, T., Mustafic, T., Camtepe, S.A., and Albayrak, S., Continuous and non-intrusive identity verification in real-time environments based on free-text keystroke dynamics, Proc. Int. Jt. Conf. Biometrics (JCB), 2011, pp. 1–8. 10.1109/IJCB.2011.6117552
34 Chang, H.C., Li, J., Wu, C., and Stamp, M., Machine learning and deep learning for fixed-text keystroke dynamics, 2021. 10.48550/arXiv.2107.07409
35 Ahmed A.A. Traore I. Biometric recognition based on free-text keystroke dynamics Cybern. IEEE Trans. 2014 44 458 472 10.1109/TCYB.2013.2257745
36 Goodkind A. Brizan D.G. Rosenberg A. Utilizing overt and latent linguistic structure to improve keystroke-based authentication Image Vision Comput. 2017 58 230 238 10.1016/j.ima-vis.2016.06.003
37 Al Solami, E., Boyd, C., Clark, A., and Ahmed, I., User-representative feature selection for keystroke dynamics, Proc. 5th Int. Conf. Network and System Security (NSS), 2011, pp. 229–233. 10.1109/IC-NSS.2011.6060005
38 Eberz, S., Rasmussen, K.B., Lenders, V., and Martinovic, I., Evaluating behavioral biometrics for continuous authentication: Challenges and metrics, Proc. ACM Asia Conf. Computer and Communications Security (ASIA CCS), 2017, pp. 386–399. 10.1145/3052973.3053032
39 Antal M. Szabo L.Z. Laszlo I. Keystroke dynamics on Android platform Procedia Technol. 2015 19 820 826 10.1016/j.protcy.2015.02.118
40 Locklear, H., Govindarajan, S., and Sitova, Z., Continuous authentication with cognition-centric text production and revision features, Proc. IEEE/IAPR Int. Jt. Conf. Biometrics (IJCB), 2014. 10.1109/BTAS.2014.6996227
41 Kang P. Cho S. Keystroke dynamics-based user authentication using long and free text strings from various input devices Inf. Sci. 2015 308 72 93 10.1016/j.ins.2014.08.070
42 Matsubara Y. Samura T. Nishimura H. Keyboard dependency of personal identification performance by keystroke dynamics in free text typing J. Inf. Secur. 2015 6 229 240 10.4236/jis.2015.63023
43 Wang, X., Yan, Z., Zhang, R., and Zhang, P., Attacks and defenses in user authentication systems: A survey, J. Network Comput. Appl., 2021, vol. 188, p. 103080. 10.1016/j.jnca.2021.103080
44 Muzaffar A.W. Tahir M. Anwar M.W. Chaudry Q. Mir S.R. Rasheed Y. A systematic review of online exams solutions in e-learning: Techniques, tools, and global adoption IEEE Access 2021 9 32689 32712 10.1109/ACCESS.2021.3060192
45 Jagadamba, G., Sheeba, R., Brinda, K.N., Rohini, K.C., and Pratik, S.K., Adaptive e-learning authentication and monitoring, Proc. 2nd Int. Conf. Innovative Mechanisms for Industry Applications (ICIMIA), 2020, pp. 277–283. 10.1109/ACIMIA48430.2020.9074955
46 Iapa, A. and Cretu, V., Shared data set for free-text keystroke dynamics authentication algorithms, 2021. 10.20944/preprints202105.0255.v1
47 González, N., Calot, E.P., Ierache, J.S., and Hasperué, W., On the shape of timings distributions in free-text keystroke dynamics profiles, Heliyon, 2021, vol. 7, no. 11, p. e08413. 10.1016/j.heliyon.2021.e08413
48 Mhenni A. Cherrier E. Rosenberger C. Essoukri Ben Amara N. Analysis of Doddington zoo classification for user dependent template update: Application to keystroke dynamics recognition Future Gener. Comput. Syst. 2019 97 210 218 10.1016/j.future.2019.02.039
49 Kazachuk, M.A., Dynamic user authentication based on keystroke dynamics analysis, Cand. Sci. (Phys.–Math.) Dissertation, Moscow, 2019.
50 Alpar, O., Biometric keystroke barcoding: A next-gen authentication framework, Expert Syst. Appl., 2021, vol. 177, p. 114980. 10.1016/j.eswa.2021.114980
51 Yang Y. Guo B. Wang Z. Li M. Yu Z. Zhou X. BehaveSense: Continuous authentication for security-sensitive mobile apps using behavioral biometrics Ad Hoc Networks 2019 84 9 18 10.1016/j.adhoc.2018.09.015
| 0 | PMC9707207 | NO-CC CODE | 2022-12-01 23:20:21 | no | Program Comput Soft. 2022 Nov 28; 48(6):385-398 | utf-8 | null | null | null | oa_other |
==== Front
J Racial Ethn Health Disparities
J Racial Ethn Health Disparities
Journal of Racial and Ethnic Health Disparities
2197-3792
2196-8837
Springer International Publishing Cham
36441494
1457
10.1007/s40615-022-01457-6
Article
Revisiting the Black-White Mental Health Paradox During the Coronavirus Pandemic
http://orcid.org/0000-0002-7077-6184
LaMotte Megan E. [email protected]
1
http://orcid.org/0000-0003-1738-9887
Elliott Marta 2
http://orcid.org/0000-0002-8036-6672
Mouzon Dawne M. 3
1 grid.266818.3 0000 0004 1936 914X Interdisciplinary Social Psychology Program, University of Nevada, Reno, 1664 N Virginia St, Reno, NV 89557 USA
2 grid.266818.3 0000 0004 1936 914X Department of Sociology, University of Nevada, Reno, Reno, NV USA
3 grid.430387.b 0000 0004 1936 8796 Department of Sociology, Rutgers, The State University of New Jersey, New Brunswick, NJ USA
28 11 2022
114
21 9 2022
3 11 2022
7 11 2022
© W. Montague Cobb-NMA Health Institute 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Black Americans have lower rates of depression and anxiety than Whites, despite greater exposure to stressors known to negatively impact mental health, characterized as the Black-White mental health paradox. This study revisited the paradox during the coronavirus pandemic. Drawing on stress process theory, minority stress theory, and the rejection-identification model of discrimination, in-group identity, and well-being, we analyzed original survey data from a quota sample of African American and White adults (N = 594). The survey included a range of stressors and coping resources, including those relevant to the pandemic (e.g., COVID-19 illness) and race (e.g., witnessing anti-Black police violence). Results indicate that despite African Americans’ greater exposure and vulnerability to racial discrimination, the Black-White mental health paradox holds, owing in part to protective effects of African American’s higher self-esteem. Directions for future exploration of the paradox are presented based on this study’s findings.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40615-022-01457-6.
Keywords
Race
Disparities
Mental health
Stressors
Coping resources
==== Body
pmcResearchers consistently find that despite an excess of exposure to stressors known to be associated with poor mental health, Black Americans tend to have as good if not better mental health as Whites [1, 2], a pattern known as the Black-White mental health paradox [3]. We investigate the paradox during a particularly stressful social context, 1 year into the global coronavirus pandemic. Our study contributes to understanding the Black-White mental health paradox by incorporating pandemic-specific stressors, and by testing multiple theoretically derived mechanisms that link stressors and coping resources to mental health. We also measure multiple outcomes including psychological distress, alternative measures of depression and anxiety, and self-reported somatic symptoms, an idiom of suffering that may be more typical of Black Americans than White Americans. In the following pages, we review evidence of the paradox, theoretical perspectives relevant to it, and research that attempts to explain it, all of which inform the research questions we test in this paper.
Background
Black Americans suffer from worse physical health outcomes than White Americans including higher rates of mortality [4], cardiovascular disease [5], stroke [6], and some cancers [7, 8]. Despite these health disparities, Black Americans’ mental health seems to fare equal or better than Whites. Nationally representative surveys conducted in the USA consistently find that Black Americans are significantly less likely to meet criteria for depression or anxiety disorders than White Americans [1, 2, 9–11] whereas racial differences in psychological distress are less clear. In a review of studies, Barnes and Bates [12] found that Black people reported higher levels of distress than Whites in 42 of 45 analyses, although most differences were not significant, the operationalization of distress varied, and most did not adjust for racial differences in education, employment, or income. Studies that do adjust for socioeconomic indicators find that Black adults have overall lower rates of distress than White adults [13], although low socioeconomic status (SES) Black adults have higher rates of distress than comparable White adults [14, 15], and increases in SES have diminishing returns to mental health for Black adults relative to White adults [16]. Studies that dichotomize the distress continuum tend to find higher rates of serious psychological distress (Kessler-6 ≥ 13) among Black people as compared to White people [17, 18].
According to stress process theory, stressors create stress that can manifest in mental health problems, but access to resources like social support, mastery, and self-esteem can directly improve health and reduce the impact of stressors (i.e., stress buffering) [19, 20]. The minority stress model [21] builds upon the stress process model by focusing on the roles of minority-specific stressors (e.g., racism and discrimination), in addition to minority-specific resources (e.g., racial identity). It posits that characteristics of identity may moderate the stress-mental health relationship [21]. For example, a strong racial identity may either buffer stress by deflecting the injury generated by racism or exacerbate the effects of stressors by rendering a person more vulnerable to minority-related stressors. In contrast, the rejection-identification model treats racial group identity as a mediator, wherein exposure to minority-group stress like racism increases in-group identification and sense of belonging, which in turn improve mental health [22].
Drawing on these theories, we explore four theoretical mechanisms that may explain the Black-White mental health paradox: differential exposure to stressors, differential access to resources, differential vulnerability to stressors, and differential efficacy of resources. The first mechanism, differential exposure to stressors, is unlikely to explain the paradox because Black Americans tend to experience more life event, financial, and race-related stressors than Whites [23, 24]. Moreover, they have been disproportionately impacted by infection with COVID-19, loss of loved ones to it [25–27], financial effects of the pandemic [28], and unlawful killings by the police [29].
Alternatively, the second theoretical mechanism, differential access to resources, suggests that Black people may have greater access to coping resources than Whites, which protects their mental health. Several coping resources have been tested in this regard, with mixed findings. Mouzon [30–32] found that while Black people often had greater social resources than White people (e.g., family interaction, fictive kin ties, support to/from church members, and religious involvement), none accounted for Black-White differences in mental disorders. Thomas Tobin, Erving, and Barve [33] found that Black people had greater family social support than White people, but Louie et al.’s [34] analyses of the same data found that Black people’s higher social support did not explain their mental health advantage. Past research consistently documents higher levels of religiosity among Black Americans than Whites [32, 34, 35] and although some researchers find religiosity accounts for a substantial portion of the paradox [34], most find that any differences in religious involvement do not fully account for better mental health among Black adults [32, 34].
In terms of personal resources, research consistently finds that Black Americans have higher self-esteem than Whites [33, 34, 36]. Moreover, Louie et al. [34] found that self-esteem mediated the race-mental health relationship accounting for 38% of Black-White differences in mental disorder and 79% of the Black-White difference in depressive symptoms. Regarding mastery, several studies found Black adults had significantly lower levels of mastery than White adults [37–40].
The rejection-identification model proposes that when exposure to discrimination leads members of minoritized groups to identify more closely with one another, the resulting in-group identity protects their mental health [22]. Black people have demonstrated greater “positive collective racial identity” than Whites [41], and studies have found that racial identity, self-esteem, and mental health are significantly inter-related among Black Americans [22, 42]. However, studies tend not to compare identity between Black and White people, perhaps because Black people’s identity has developed in response to historical and contemporary racism [43], whereas for White people, identity has been implicitly positive based on White privilege.
The third potential mechanism, differential vulnerability to stressors, might explain the Black-White mental health paradox if the effects of certain stressors have a greater negative impact on the mental health of White people than Black people, resulting in a Black mental health advantage despite greater stress exposure [44]. Recent evidence from the Health and Retirement Survey finds that despite greater chronic stress exposure across five life domains, older Black adults are less likely than older White adults to rank these experiences as stressful [45]. However, it is possible Black people are more vulnerable to stressors related to their race such as racial discrimination and exposure to racial violence than White people. In a national race-comparative study, Eichstaedt et al. [46] found that anger, sadness, depression and anxiety increased more among Black Americans than White Americans following the death of George Floyd in late May of 2020.
Lastly, the Black-White mental health paradox could be due to coping resources being more effective at protecting mental health for Black people relative to White people (differential efficacy of resources). Racial identity has been shown to buffer the negative mental health effects of discrimination among Black people [47], but it has also been found to exacerbate them [48]. Morin and Midlarsky [49] found that mastery was equally effective in reducing distress for Black and White adults. In contrast, Oates and Goode [50] found that mastery reduced distress more for White than for Black adults, whereas public religiosity (e.g., service attendance) reduced distress for Black adults only. There is also research supporting differential stress-buffering of religious resources by race among older adults [51].
Thus far, our theoretical exploration assumed the mental health paradox is objectively real; however, some scholars speculate that survey research does not accurately reflect the mental health of Black Americans, and that the apparent paradox is owing to methodological artifacts such as measurement and sampling biases. For example, racial minorities throughout the world, including Black Americans, often perceive mental illness as more stigmatized than racial majorities [52–54], and perceived stigma could cause Black people to downplay their mental health symptoms [3]. Additionally, measurement items used to categorize respondents as depressed in survey research focus on psychological symptoms (e.g., sad mood and anhedonia), but Black people may experience depression more somatically via bodily pains, resulting in underdiagnosis [12], an alternative we explore below.
The Current Study
This study tests research questions derived from the four theoretical mechanisms described above in comparing the determinants of distress, somatic symptoms, depression, and anxiety between Black and White Americans using original survey data collected 1 year into the pandemic. Figure 1 depicts a mediation model that illustrates our first two research questions, which concern how differential exposure to stressors (path a) and access to resources (path b) may mediate any direct association we find between race and mental health (path e) via indirect pathways (a × c and b × d).RQ1: Differential exposure: Are there racial differences in exposure to stressors that indirectly explain racial differences in mental health? (paths a x c)
RQ2: Differential access: Are there racial differences in access to resources that indirectly explain racial differences in mental health? (paths b x d)
Fig. 1 Mediator model: differential exposure to stressors and differential access to resources
We expect that Black participants will be exposed to more pandemic-related and race-related stressors than White participants, which in turn will predict worse mental health for Black participants relative to Whites (RQ1). We also expect that Black participants will have greater self-esteem, racial identity, and religiosity than White participants (RQ2), resource differences that will indirectly predict better mental health outcomes for Black versus White participants.
In Fig. 2, path a represents the expected negative effect of exposure to stressors on mental health, whereas path b represents the expected positive effect of resources. Paths c and d correspond to moderating mechanisms and illustrate our third and fourth research questions:RQ3: Differential vulnerability: Are there racial differences in the impact of stressors on mental health? (path c)
RQ4: Differential efficacy: Are there racial differences in the impact of resources on mental health? (path d)
Fig. 2 Moderator model: differential vulnerability to stressors and differential efficacy of resources by race
We expect Black participants will be more vulnerable to race-related stressors than White participants given that systemic racism disadvantages Black people, whereas Whites enjoy racial privilege. We also expect racial identity and religiosity will be more protective of mental health for Black participants than Whites and will attenuate the negative effects of stressors on mental health more for Black participants than for Whites.
Methods
This study analyzes online survey data collected from a quota sample of Black and White Americans throughout the USA between March 16 and April 17, 2021. The survey was programmed in Qualtrics and presented to research participants via Prolific Academia (www.prolific.co), a research recruitment website that facilitates quota sampling from respondents who tend to be less familiar with the purpose of research and less dishonest than Mturk workers [50]. Prolific recruits their users through social media advertising, a user referral scheme, and word-of mouth.
Respondents and Procedures
The sample included 297 African American respondents and 297 White respondents (total N = 594). We did not collect data from other Black-identifying groups (e.g., Afro-Caribbeans) and non-native Black people due to considerable between-group differences in mental health [56]; therefore, we refer to respondents as African Americans rather than Black Americans. Respondents had to be 18 or older, and we set quotas for quartiles of household income based on national estimates of the household income distribution for African Americans versus Whites to ensure our data captured variation in SES. The quartiles were (1) less than $10,000 to $29,999, (2) $30,000 to $59,999, (3) $60,000 to $99,999, and (4) $100,000 or more, and the quotas we set for each racial group were based on Current Population Survey household income data for African Americans and White Americans separately [57]. Potential respondents were invited to complete the survey on a first come-first serve basis. There were three attention-check items in the survey and four respondents were excluded owing to failing them. Respondents were paid $9.54 per hour and the survey took about 10 min.
Measures
The survey included existing validated measures and items designed to capture exposure to stressors specific to 2020 such as infection with the COVID-19 virus and witnessing police violence against Black people. Data were collected before COVID-19 vaccines were available to all US adults, so we did not inquire upon vaccination status. We provide a brief overview of the scales below and more detailed information including full lists of survey items in Online Supplemental File 1.
Stressors
Exposure to general stressors was measured by asking about events that occurred since February 2020 when the USA declared a public health emergency including a serious health condition (other than COVID-19), loss of a close family member or friend, serious relationship conflict, relationship dissolution, serious injury, legal troubles, or crime victimization. The events were summed and capped at two or more, which was the 90th percentile. Racial discrimination was measured with the 9-item Everyday Discrimination Scale (α = 0.95) [58]. The original scale did not link discriminatory acts to characteristics such as race or gender, whereas we reworded the question stem to make race salient: “In your day-to-day life, how often do any of the following things happen to you because you are [Black/White]?” Witnessing violence against Black people was measured by 3 items (α = 0.85) modeled after Tynes et al. [59] that asked respondents: “Since February 2020, how often have you seen images, videos, or in-person, Black people who were…,” followed by (1) arrested or detained by law enforcement officers, (2) shot or killed by law enforcement officers, (3) attacked by people who were not law enforcement.
We also included a series of measures relating to COVID-19. COVID-19 infection was measured by asking respondents if they had ever contracted COVID-19 and if so, what best characterized their symptom experience, resulting in a variable ranging from no known infection (0) to infection with severe symptoms (4). COVID-19’s impact on one’s social network was measured by asking how many close friends or family had been seriously ill with COVID-19 and how many died from COVID-19 illness. Because each distribution was positively skewed, we created two dichotomous variables where 1 = knew at least one person seriously ill with COVID-19, and 1 = knew at least one person who died from COVID-19 versus 0 = otherwise. Exposure to secondary pandemic-related stressors (caused by the pandemic but not the virus itself) was measured by asking if any of the following occurred since February 2020: began working or attending school from home, had a pay cut or reduction in work hours, job loss, unemployment, financial hardship, housing unaffordability, homelessness, or bankruptcy/major debt. The count of events was capped at four or more, which was the 90th percentile.
Resources
Self-esteem was measured with the 10-item Rosenberg Self-Esteem Scale (α = 0.93) [60]. Mastery was measured through the 7-item Mastery Scale (α = 0.84) [61]. Emotional social support was measured with two items (α = 0.88) from Ross and Mirowsky [62] that ask participants how much they agreed that: “I have someone I can really talk to” and “I have someone I can turn to for support and understanding when things get rough.” The scale from low (1) to high (5) social support was negatively skewed so we dichotomized social support such that 1 = high social support (agreed or strongly agreed to the statements) and 0 = low social support (neutral or disagreed with statements).
Religiosity was measured with two items (α = 0.81): importance of religious or spiritual beliefs, ranging from not at all important (1) to extremely important (4), and religious service attendance, ranging from never (0) to several times a week (5). The two items were standardized and averaged owing to different response options. Racial identity was measured with the 8-item racial centrality subscale (α = 0.92) of the Multidimensional Inventory of Black Identity (MIBI) [43], with “White” substituted for White respondents. We also measured access to health care based on the expectation that differential health care access would be relevant to COVID-19 infection and to mental health. Three items measured how easy or difficult it is for respondents to access (1) health care in general, (2) COVID-19-related health care, and (3) COVID-19 testing, each measured on a scale from very difficult (1) to very easy (5).
Mental Health Outcomes
We measured four mental health outcomes: psychological distress, somatic symptom burden, and prior diagnoses of depression, and anxiety. Distress was measured by the 6-item Kessler Distress Scale (α = 0.90) [63]. We capped the summed scores at 13 or more out of 18 because that represented the 85th percentile and is the recommended cutoff [63]. Somatic symptoms were measured by the 8-item Somatic Symptom Scale-8 (α = 0.82) (SSS8) [64] to account for the possibility that Black people express reactions to stressors more somatically than psychologically [12]. We capped the summed scores at 16 or more, the 90th percentile, which indicates “very high” somatic symptom burden [64]. Depression and anxiety were measured by self-reports rather than diagnostic inventories. Specifically, we asked respondents if they had been “diagnosed with and/or received mental health treatment for” depression or anxiety in the past 2 years so as not to omit cases who were diagnosed before the onset of the pandemic. If they said yes, we asked whether they thought they currently had the condition.Depression and anxiety were coded as binary variables wherein respondents who had either never been diagnosed or who had been diagnosed in the past 2 years but did not believe they still had depression or anxiety were coded zero, and those who had been diagnosed and indicated they still had the condition were coded as one.
Control variables were measured as follows: age (years), education level (dummy for 1 = bachelor’s degree or more), total household income (1 = < $10,000–13 = > $150,000; see Supplemental File 2 for full ordinal scale), household size (number of residents), gender (dummy for 1 = female or transwomen), sexual orientation (1 = not heterosexual), relationship status (1 = in a relationship), work status (1 = working outside of the home), and political orientation (1 = very liberal–5 = very conservative).
Analytic Strategy
Analyses were conducted in IBM SPSS Statistics 28 (IBM Corp., Armonk, N.Y., USA). We conducted independent samples t tests to detect significant differences between African Americans and Whites in means on ordinal, interval, and ratio variables and proportions on nominal variables. We used ordinary least squares (OLS) regression to analyze the effects of race, stressors, resources, and covariates on distress and somatic symptoms, and binary logistic regression to analyze their effects on depression and anxiety. The impact of COVID-19 on family and friends and access to COVID-related health care and testing are not included in our results because they were invariably non-significant. All regressions presented passed assumption checks and details of those analyses are in Online Supplemental File 3.
We explored RQ1 (differential exposure) and RQ2 (differential access) by conducting mediation analyses with the Process macro for SPSS [65] to estimate all indirect effects of race on outcomes via differential exposure to stressors and differential access to resources simultaneously, net of controls. The continuous version of social support was used in the mediation analysis, as required by the Process macro; otherwise, the binary version was used because it was skewed. Indirect effects were deemed significant if 95% confidence intervals estimated from 10,000 bootstrap samples did not contain zero. We explored RQ3 (differential vulnerability) and RQ4 (differential vulnerability) by adding two-way interaction terms between race and each stressor and resource to the OLS and logistic regression models, first testing interaction terms individually and subsequently testing significant interaction terms together. We also explored interactions between race and socioeconomic status (income and education) considering the literature on diminishing returns, but none were significant. The final models include only those interaction terms that were simultaneously significant.
Results
Table 1 presents the descriptive statistics for the overall sample and by race. Notably, all racial comparisons in mental health outcomes in this sample are consistent with the Black-White mental health paradox, reflecting equal or better mental health among African Americans than White Americans. For example, African Americans and Whites did not significantly differ in average distress symptoms, somatic symptoms or in proportions of those diagnosed with depression. However, African Americans were less likely to have been diagnosed with anxiety than were Whites, and among those who were diagnosed with depression or anxiety, African Americans were less likely to report they still had either condition.Table 1 Descriptive statistics overall and by race with tests of mean and proportional differences
Total African Am White
(N = 594) (n = 297) (n = 297)
Variables Mean(sd)/prop Min Max Mean(sd)/prop Mean(sd)/prop
Outcomes
Distress 6.48(4.62) 0.00 13.00 6.24(4.75) 6.72(4.49)
Somatic symptoms 7.06(5.10) 1.00 16.00 6.72(5.11) 7.40(5.07)
Depression (1 = diagnosed) 0.23 0.00 1.00 0.21 0.26
Depression (1 = diagnosed and ongoing) 0.16 0.00 1.00 0.13* 0.19*
Anxiety (1 = diagnosed) 0.28 0.00 1.00 0.22** 0.34**
Anxiety (1 = diagnosed and ongoing) 0.23 0.00 1.00 0.17*** 0.29***
Stressors
General stressors 0.70(0.78) 0.00 2.00 0.70(0.80) 0.69(0.76)
Racial discrimination 2.05(1.08) 1.00 5.89 2.61***(1.14) 1.48***(0.63)
Witnessing violence against Blacks 8.81(2.74) 3.00 15.00 8.93(3.07) 8.69(2.36)
Secondary pandemic stressors 1.71(1.39) 0.00 4.00 1.91***(1.46) 1.50***(1.29)
COVID-19 infection 0.34(0.86) 0.00 4.00 0.27*(0.75) 0.41*(0.95)
Knew someone seriously ill with COVID-19 (1) 0.33 0.00 1.00 0.35 0.30
Knew someone who died of COVID-19 (1) 0.14 0.00 1.00 0.14 0.14
Resources
Self-esteem 3.75(0.97) 1.00 5.00 3.84*(0.97) 3.67*(0.97)
Mastery 3.57(0.83) 1.00 5.00 3.56(0.83) 3.57(0.83)
Emotional social support 4.20(1.04) 1.00 5.00 4.09*(1.13) 4.31*(0.93)
Emotional social support (1 = high) 0.78 0.74* 0.82*
Religiosity 0(1) − 1.41 1.30 0.29***(0.93) − 0.29***(0.98)
Racial identity 3.05(1.17) 1.00 5.00 3.78***(0.95) 2.32***(0.88)
General health care access 4.05(1.01) 1.00 5.00 4.02(0.97) 4.08(1.05)
COVID-19 health care access 3.92(0.99) 1.00 5.00 3.92(0.93) 3.93(1.04)
COVID-19 testing access 3.97(0.97) 1.00 5.00 3.90(0.99) 4.05(0.96)
Controls
Age 35.84(12.57) 18.00 78.00 33.32***(11.04) 38.36***(13.48)
18–24 0.20 0.00 1.00 0.22 0.17
25–34 0.36 0.00 1.00 0.42*** 0.29***
35–54 0.34 0.00 1.00 0.30* 0.38*
55 and over 0.11 0.00 1.00 0.06*** 0.16***
Education (1 = bachelor’s degree or more) 0.55 0.00 1.00 0.52 0.59
Total household income (13 = > $150,000) 7.52(4.04) 1.00 13.00 6.60***(3.97) 8.44***(3.91)
Household size 2.95(1.43) 1.00 9.00 3.13**(1.57) 2.78**(1.27)
Gender (1 = female) 0.51 0.00 1.00 0.50 0.52
Sexual orientation (1 = not heterosexual) 0.15 0.00 1.00 0.14 0.15
Relationship status (1 = in a relationship) 0.60 0.00 1.00 0.52*** 0.68***
Work status (1 = works outside the home) 0.40 0.00 1.00 0.40 0.39
Political orientation (5 = very conservative) 2.42(1.20) 1.00 5.00 2.45(1.17) 2.39(1.24)
*p < .05, **p < .01, ***p < .001
In terms of stressors, African Americans experienced more racial discrimination and secondary pandemic stressors than Whites, whereas Whites reported more COVID-19 infection than African Americans. Regarding resources, self-esteem, religiosity, and racial identity were significantly higher among African Americans, whereas social support was higher among Whites. No other stressors or resources differed significantly between racial groups. On average, the sample was about 35 years old, and the majority had at least a bachelor’s degree, were female, heterosexual, had a total household income between $50,000 and $69,999, had about three people living in their household, were in a relationship, and worked from home. The sample leaned slightly liberal in political orientation. African Americans were significantly younger, of lower income, had more people living in their households and less likely to be in a relationship than Whites. There were no differences in education, gender, sexual orientation, work status, or political orientation by race.
RQ 1 and 2: Differential Exposure to Stressors and Access to Resources
We tested indirect effects of race on distress, somatic symptoms, depression, and anxiety to determine whether differential exposure to stressors (e.g., discrimination) and differential access to resources (e.g., self-esteem, racial identity) explain racial differences in mental health. Table 2 shows the significant indirect effects. There was an indirect effect of racial discrimination on distress and somatic symptoms, indicating that all else being equal, being African American was associated with increased distress and somatic symptoms because African Americans reported substantially more racial discrimination than Whites (RQ1). However, these indirect effects on distress and somatic symptoms were counterbalanced by the negative indirect effects of self-esteem, suggesting that African Americans tend to be less distressed and have fewer somatic symptoms than Whites because they have greater self-esteem. In contrast, there were no indirect effects of racial discrimination on self-reported depression or anxiety diagnoses, but the indirect effects of self-esteem on depression and anxiety remained.Table 2 Unstandardized indirect effects of race on mental health through racial discrimination, self-esteem
Distress Som. symptoms Depression Anxiety
n = 594 n = 594 n = 584 n = 582
Mediator variables b 95% CI b 95% CI b 95% CI b 95% CI
Racial discrimination 1.08 (0.67, 1.53) 1.16 (0.62, 1.77) –- –- –- –-
Self-esteem − 0.90 (− 1.34, − 0.51) − 0.62 (− 0.99, − 0.31) − 0.19 (− 0.39, − 0.05) − 0.21 (− 0.41, − 0.08)
Dashes are in place of non-significant coefficients. Indirect effects of all other stressors and resources were tested, but yielded non-significant results across all outcomes so they are omitted from the table. Results are adjusted for covariates
RQ 3 and 4: Differential Vulnerability to Stressors’ Effects and Efficacy of Resources
Table 3 presents the linear regressions of distress and somatic symptoms, indicating that all else being equal, African Americans had lower distress and fewer somatic symptoms than Whites. General stressors and racial discrimination were positively associated with distress and somatic symptoms. Self-esteem and mastery were negatively associated with distress and somatic symptoms. Social support was positively associated with somatic symptoms in the multivariate model, despite being associated with fewer somatic symptoms in the bivariate analysis (Supplemental File 2). This was due to negative suppression, which happens when controlling for a third variable reverses the sign of another predictor variable [66]. In this case, only when self-esteem was in the regression did social support have a positive coefficient rather than the expected negative coefficient. Since social support and self-esteem were positively correlated, the suppressor effect represents the predicted value of somatic symptoms in the unusual case that a person has high social support, but very low self-esteem. All interactions between race and each stressor, and race and each resource were non-significant, suggesting that the differential vulnerability and efficacy mechanisms do not explain racial differences in distress or somatic symptoms.Table 3 Distress and somatic symptoms regressed on race, stressors, psychosocial resources, and covariates (N = 594)
Distress Somatic Symptoms
Model 1 Model 2
Variables b b
Constant 17.27*** 11.09***
Race (1 = African American) − 1.30*** − 1.65**
General stressors 0.64*** 1.32***
Racial discrimination 0.65*** 0.84***
Witnessing violence against Blacks 0.08 0.05
Secondary pandemic stressors 0.07 0.27
COVID-19 infection 0.02 0.01
Self-esteem − 2.40*** − 1.50***
Mastery − 0.89*** − 1.02***
Emotional social support (1 = high) 0.35 1.05*
Religiosity 0.09 0.04
Racial identity 0.03 − 0.01
General health care access − 0.06 − 0.01
Age (reference group = 18–24)
25–34 0.03 1.33*
35–54 − 1.19** 1.05
55 and over − 2.35*** 0.60
Education (1 = bachelor's degree or more) 0.04 − 0.67
Total household income 0.01 − 0.01
Household size 0.13 0.30
Gender (1 = female) 0.77** 1.47***
Sexual orientation (1 = not heterosexual) − 0.16 0.08
Relationship status (1 = in a relationship) − 0.18 − 0.31
Work status (1 = works outside the home) − 0.45 0.23
Political orientation (5 = conservative) − 0.27* − 0.08
Adjusted R-squared 0.57 0.34
*p < .05, **p < .01, ***p < .001
Table 4 presents logistic regressions of depression and anxiety, presenting direct effects in Models 3a and 4a and interactive effects in Models 3b and 4b. African Americans had lower odds of self-reported depression and anxiety diagnoses than Whites. General stressors were positively associated with both outcomes, whereas witnessing violence against Black people was positively associated with depression only. Self-esteem was negatively associated with depression and anxiety, but social support was positively associated with anxiety (owing to the suppressor effect involving self-esteem). Racial identity was positively associated with depression.Table 4 Depression and anxiety regressed on race, stressors, resources, and covariates
Depression (N = 584) Anxiety (N = 582)
Model 3a Model 3b Model 4a Model 4b
Variables b OR b OR b OR b OR
Constant − 1.14 0.32 1.03 2.79 − 2.20* 0.11 − 3.99** 0.02
Race (1 = African American) − 1.75*** 0.17 − 6.79*** 0.00 − 1.42*** 0.24 1.04 2.82
General stressors 0.76*** 2.14 0.82*** 2.27 0.56*** 1.75 0.56*** 1.76
Racial discrimination 0.27 1.31 − 0.68 0.51 0.18 1.20 0.25 1.29
Witnessing violence against Blacks 0.14* 1.15 0.17* 1.18 0.04 1.04 0.19* 1.21
Secondary pandemic stressors − 0.05 0.96 − 0.08 0.92 − 0.02 0.98 − 0.04 0.96
COVID-19 infection 0.17 1.18 0.52** 1.69 − 0.09 0.92 − 0.06 0.94
Self-esteem − 0.58** 0.56 − 0.91*** 0.40 − 0.61*** 0.54 − 0.62*** 0.54
Mastery − 0.36 0.70 − 0.44 0.64 − 0.09 0.91 − 0.11 0.90
Emotional social support (1 = high) 0.31 1.36 0.33 1.40 1.04** 2.83 1.06** 2.87
Religiosity − 0.16 0.85 − 0.11 0.90 − 0.23 0.79 − 0.25 0.78
Racial identity 0.43* 1.54 0.42* 1.53 0.21 1.24 0.29 1.33
General health care access 0.13 1.13 0.13 1.14 0.24 1.27 0.26 1.30
Age (reference group = 18–24)
25–34 0.86 2.37 1.07* 2.91 1.06** 2.88 1.14** 3.12
35–54 0.42 1.53 0.55 1.73 0.62 1.85 0.62 1.86
55 and over 0.87 2.40 0.93 2.54 0.31 1.36 0.38 1.47
Education (1 = bachelor’s degree or more) − 0.22 0.80 − 0.24 0.79 − 0.34 0.71 − 0.40 0.67
Total household income − 0.04 0.97 − 0.03 0.97 − 0.04 0.96 − 0.04 0.96
Household size − 0.24 0.79 − 0.30* 0.74 0.09 1.09 0.09 1.09
Gender (1 = female) 0.73* 2.08 0.70* 2.02 1.04*** 2.84 1.06*** 2.87
Sexual orientation (1 = not heterosexual) 0.56 1.76 0.77* 2.17 0.82* 2.26 0.84* 2.31
Relationship status (1 = in a relationship) − 0.52 0.60 − 0.54* 0.58 − 0.18 0.84 − 0.22 0.80
Work status (1 = works outside the home) − 0.60 0.55 − 0.68* 0.51 − 0.52* 0.60 − 0.54* 0.58
Political orientation (5 = conservative) − 0.38* 0.68 − 0.34* 0.71 − 0.19 0.83 − 0.16 0.85
Differential vulnerability to stressors
Afr. Am. × general stressors –- –- –- –-
Afr. Am. × discrimination 1.36** 3.89 –- –-
Afr. Am. × witnessing violence against Blacks –- –- − 0.29** 0.75
Afr. Am. × secondary pandemic stressors –- –- –- –-
Afr. Am. × COVID-19 infection − 1.32** 0.27 –- –-
Differential efficacy of resources
Afr. Am. × self-esteem 0.72* 2.06 –- –-
Afr. Am. × mastery –- –- –- –-
Afr. Am. × emotional social support –- –- –- –-
Afr. Am. × religiosity –- –- –- –-
Afr. Am. × racial identity –- –- –- –-
Afr. Am. × health care access –- –- –- –-
Nagelkerke R-squared 0.38 0.44 0.33 0.34
*p < .05, **p < .01, ***p < .001
There were two significant interactions representing differential vulnerability to stressors and one representing differential efficacy of resources in the depression model (Model 3b). Race moderated the effects of discrimination (Fig. 3a), COVID-19 infection (Fig. 3b), and self-esteem (Fig. 3c). Racial discrimination increased the odds of depression for African Americans (b = 0.68, p < 0.001), but not Whites (b = − 0.68, p = 0.11), and COVID-19 infection increased the odds of depression for Whites (b = 0.52, p = 0.01), but decreased the odds for African Americans (b = − 0.80, p = 0.04). Self-esteem decreased the odds of depression for Whites (b = − 0.91, p < 0.001), but not for African Americans (b = − 0.19, p = 0.46). In the regression for anxiety (Model 4b), there was one significant interaction representing differential vulnerability to witnessing violence against Black people. Witnessing violence against Black people increased the odds of anxiety for Whites (b = 0.19, p = 0.01) but was not associated with anxiety for African Americans (b = − 0.09, p = 0.18) (Fig. 3d).Fig. 3 Plotted interactions for (a) differential vulnerability to discrimination, (b) differential vulnerability to COVID-19, (c) differential efficacy of self-esteem, and (d) differential vulnerability to witnessing violence against Black people
Discussion
Consistent with past research on the Black-White mental health paradox, and in contrast to physical health disparities that disproportionately affect African Americans, this study found that African Americans had equal if not better mental health than White Americans. We found no differences in levels of distress or somatic symptoms between groups, and when controlling for stressors, resources, and covariates, African Americans had significantly lower levels of distress and somatic symptoms. The absence of elevated distress among African Americans is consistent with some studies [13] but contradicts studies that measure distress as a binary variable [12, 17, 18]. The lack of elevated somatic symptoms among African Americans is a novel finding that does not support the argument that the paradox is an artifact of how mental health is measured [12], although African Americans may under-report them.
We also found that African Americans were less likely than Whites to report a diagnosis of depression or anxiety and to think they still had either condition. Despite using a different method of measuring depression and anxiety that requires access to health care to have received a diagnosis, our findings are consistent with psychiatric epidemiological studies of nationally representative samples [1, 2, 9–11] that use algorithms to diagnose survey respondents, regardless of whether they have sought treatment. While these findings support the paradox, they could also reflect several biases. First, mental illness stigma among African Americans, which is common among racial minorities [52–54], may discourage them from seeking treatment or identifying symptoms as depression or anxiety [67]. Second, even when African Americans do seek help, mental health providers have the tendency to diagnose them with more severe psychotic disorders than Whites such as bipolar and schizophrenia [68, 69]. Ultimately, all measures of mental illness, whether categorical or continuous, self-reported or diagnosed by a clinician, are limited by the absence of objective measures or biomarkers of specific psychiatric disorders [70], and depend instead upon the presentation and interpretation of symptoms and behaviors. Despite this inherent limitation, the Black-White mental health paradox is consistent across different types of measures of mental illness, including the self-report measures used in this paper, the symptom algorithms utilized in psychiatric epidemiology surveys [1, 2, 9–11], and clinical data [71].
As expected, we found no support that racial differences in mental health outcomes were attributed to Whites being exposed to more stressors than African Americans (RQ1). Rather, African Americans were more exposed to racial discrimination than Whites, which indirectly increased their distress and somatic symptoms. In fact, if it were not for the role of resources, African Americans might well have reported worse mental health relative to Whites owing to racism. However, differential access to self-esteem partially explained the paradoxical differences in mental health (RQ2). Self-esteem was higher among African Americans than Whites, consistent with past studies [33, 34, 36], which indirectly reduced distress, somatic symptoms, depression, and anxiety. The counterbalancing between the effects of racial discrimination and self-esteem on distress and somatic symptoms may account for the lack of mean differences between racial groups in these outcomes, as Louie and Wheaton found in their study of adolescents [72].
Regarding differential vulnerability (RQ3), the findings differed by race depending on the stressor in question. African Americans were more vulnerable to the effects of racial discrimination, as it was associated with depression for African Americans, but not for Whites. However, contrary to our expectations and prior research by Eichstaedt et al. [46], Whites appeared more vulnerable to witnessing violence against Black people as it was associated with increased anxiety for Whites only. This unexpected finding could reflect Whites’ lack of life experience with racial discrimination, discomfort with and unpreparedness for the graphic footage of racial violence such as the videotaped murder of George Floyd, an African American man, by a White police officer. It could also reflect the process of rejection identification wherein exposure to oppression of one’s racial group increases racial identity which in turn protects the mental health of minority group members [22]. However, we did not find any protective effects of racial identity, which is the key mediator in the rejection-identification model so this framework must be more thoroughly investigated.
Whites’ mental health also appeared more vulnerable to COVID-19 infection and illness, which was associated with increased odds of depression for Whites and decreased odds for African Americans. This divergent pattern could be an artifact of limitations in our data. Namely, only 14 Whites (5%) had depression and COVID-19, and even fewer African Americans (4, or 1%) had both. Second, depression may have preceded infection, given that pre-existing mental illness may increase susceptibility to COVID-19 illness [73].
Self-esteem was the only resource that was differentially efficacious (RQ4), but not for African Americans, as expected. Rather, it reduced the odds of depression for Whites only, even though African Americans reported higher self-esteem, and it indirectly protected their mental health from the ill effects of discrimination. This finding could simply reflect the low probability of depression for African Americans across all levels of self-esteem, which is clear in Fig. 3c.
This study is limited by relying on cross-sectional data, which is especially relevant for the analyses of depression and anxiety that could have existed prior to the stressors we measured during the pandemic. Additionally, while our quota sample mirrors the distributions of household income in the US population by race, the sample is not nationally representative. Unlike the general US population, the majority in both White and African American groups were college educated and worked from home. The inconsistencies in representativeness may explain why the data did not capture the known disproportionate impact of COVID-19 illness on African Americans, but did capture the disproportionate secondary impacts of the pandemic, which are more closely related to income. Future research on the Black-White mental health paradox should collect panel data from a probability sample to track changes in stressors, resources, and mental health with measures of multiple dimensions of racial identity. Longitudinal panel data including these constructs could be used to ascertain how aspects of racial identity develop in response to racism and influence psychological responses to future racism. Additionally, they could demonstrate whether stress “immunity” develops over time in response to cumulative exposure to racism, which could explain why witnessing violence against Black people was only associated with anxiety for White Americans.
In conclusion, despite the disproportionate impact of the global pandemic on African Americans and heightened attention to systemic racism, this study finds that the Black-White mental health paradox persists. Although racial discrimination continues to put African American mental health at risk, self-esteem appears to counteract racism’s pernicious effects, such that African Americans are no more distressed or in physical pain than Whites, whereas Whites are more likely to have depression and anxiety. Nonetheless, the struggle among African Americans to protect their mental health in response to racism continues and should not be minimized.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 19 KB)
Supplementary file2 (DOCX 28 KB)
Supplementary file3 (DOCX 13 KB)
Author Contribution
All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Megan LaMotte and Marta Elliott. The first draft of the manuscript was written by Megan LaMotte and Marta Elliott, and all authors participated in revising the first draft and all subsequent drafts of the manuscript. All authors read and approved the final manuscript.
Funding
This study was funded internally by the School for Social Research and Justice Studies at the University of Nevada, Reno.
Data Availability
For requests to view data and/or research materials, email the corresponding author.
Code Availability
For requests to view code used in data analysis, email the corresponding author.
Declarations
Ethics Approval
This study was granted exemption from review by the University of Nevada, Reno Institutional Review Board.
Consent to Participate
Informed consent was obtained from all respondents included in the study.
Consent for Publication
Not applicable.
Competing Interests
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Breslau J Aguilar-Gaxiola S Kendler KS Su M Williams DR Kessler RC Specifying race-ethnic differences in risk for psychiatric disorder in a USA national sample Psychol Med 2006 36 1 57 68 10.1017/S0033291705006161 16202191
2. Hasin DS Goodwin RD Stinson FS Grant BF Epidemiology of major depressive disorder: results from the National Epidemiologic Survey on alcoholism and related conditions Arch Gen Psychiatry 2005 62 10 1097 1106 10.1001/archpsyc.62.10.1097 16203955
3. Keyes CLM The Black-White paradox in health: flourishing in the face of social inequality and discrimination J Pers 2009 77 6 1677 1706 10.1111/j.1467-6494.2009.00597.x 19796064
4. Curtin SC, Arias E. Mortality trends by race and ethnicity among adults aged 25 and over, 2000–2017. NCHS Data Brief, no 342. Hyattsville, MD: National Center for Health Statistics. 2019. https://www.cdc.gov/nchs/products/databriefs/db342.htm.
5. Leigh JA Alvarez M Rodriguez CJ Ethnic minorities and coronary heart disease: an update and future directions Curr Atheroscler Rep 2016 18 2 9 10.1007/s11883-016-0559-4 26792015
6. Graham G Disparities in cardiovascular disease risk in the United States Curr Cardiol Rev 2015 11 3 238 245 10.2174/1573403X11666141122220003 25418513
7. Baquet CR Mishra SI Commiskey P Ellison GL DeShields M Breast cancer epidemiology in blacks and whites: disparities in incidence, mortality, survival rates and histology J Natl Med Assoc 2008 100 5 480 488 18507200
8. Du XL Meyer TE Franzini L Meta-analysis of racial disparities in survival in association with socioeconomic status among men and women with colon cancer Cancer 2007 109 11 2161 2170 10.1002/cncr.22664 17455219
9. Blazer DG Kessler RC McGonagle KA Swartz MS The prevalence and distribution of major depression in a national community sample: the National Comorbidity Survey Am J Psychiatry 1994 151 7 979 986 10.1176/ajp.151.7.979 8010383
10. Weissman MM Leaf PJ Tischler GL Blazer DG Karno M Bruce ML Florio LP Affective disorders in five United States communities Psychol Med 1988 18 1 141 153 10.1017/S0033291700001975 3363034
11. Williams DR Gonzalez HM Neighbors H Neese R Abelson JM Sweetman J Jackson JS Prevalence and distribution of major depressive disorder in African Americans, Caribbean Blacks, and Non-Hispanic Whites: results from the National Survey of American Life Arch Gen Psychiatry 2007 64 305 315 10.1001/archpsyc.64.3.305 17339519
12. Barnes DM Bates LM Do racial patterns in psychological distress shed light on the Black-White depression paradox? A systematic review Soc Psychiatry and Psychiatr Epidemiol 2017 52 8 913 928 10.1007/s00127-017-1394-9 28555381
13. Bratter JL Eschbach K Race/ethnic differences in nonspecific psychological distress: evidence from the National Health Interview Survey Soc Sci Q 2005 86 3 620 644 10.1111/j.0038-4941.2005.00321.x
14. Kessler RC Neighbors HW A new perspective on the relationships among race, social class, and psychological distress J Health Soc Behav 1986 27 2 107 115 10.2307/2136310 3734380
15. Nuru-Jeter A Williams CT LaVeist TA A methodological note on modeling the effects of race: the case of psychological distress Stress Health 2008 24 5 337 350 10.1002/smi.1215
16. Assari S Lapeyrouse LM Neighbors HW Income and self-rated mental health: Diminished returns for high income Black Americans Behav Sci 2018 8 5 50 10.3390/bs8050050 29772799
17. Reeves WC, Strine TW, Pratt LA, Thompson W, Ahluwalia I, Dhingra SS, McKnight-Eily LR, Harrison L, D’Angelo DV, Williams L, Morrow B, Gould D, Safran MA. Mental illness surveillance among adults in the United States. In: Morbidity and Mortality Weekly Report. Centers for Disease Control. 2011. https://www.cdc.gov/mmwr/preview/mmwrhtml/su6003a1.htm
18. Roxburgh S Untangling inequalities: gender, race, and socioeconomic differences in depression Sociol Forum 2009 24 2 357 381 10.1111/j.1573-7861.2009.01103.x
19. Pearlin LI Menaghan EG Lieberman MA Mullan JT The stress process J Health Soc Behav 1981 22 4 337 356 10.2307/2136676 7320473
20. Turner RJ Understanding health disparities: the relevance of the stress process model Soc Ment Health 2013 3 3 170 186 10.1177/2156869313488121
21. Meyer IH Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: conceptual issues and research evidence Psychol Bull 2003 129 5 674 697 10.1037/0033-2909.129.5.674 12956539
22. Branscombe NR Schmitt MT Harvey RD Perceiving pervasive discrimination among African Americans: implications for group identification and well-being J Pers Soc Psychol 1999 77 1 135 149 10.1037/0022-3514.77.1.135
23. Lee RT Perez AD Boykin CM Mendoza-Denton R On the prevalence of racial discrimination in the United States PLoS ONE 2019 10.1371/journal.pone.0210698 31891641
24. Williams DR Hogue C Hargraves M Scott-Collins K Race, stress, and mental health: findings from the Commonwealth Minority Health Survey Minority health in America: findings and policy implications from the Commonwealth Fund Minority Health Survey 2000 Baltimore, MD Johns Hopkins University Press 209 243
25. CDC. Risk for COVID-19 infection, hospitalization, and death by race/ethnicity. Centers for Disease Control and Prevention. 2020. (https://www.cdc.gov/coronavirus/2019-ncov/covid-data/investigations-discovery/hospitalization-death-by-race-ethnicity.html). Accessed 19 Oct 2021.
26. Goldstein A, Guskin E. Almost one-third of Black Americans know someone who died of Covid-19, survey shows. Washington Post. 2020. https://www.washingtonpost.com/health/almost-one-third-of-black-americans-know-someone-who-died-of-covid-19-survey-shows/2020/06/25/3ec1d4b2-b563-11ea-aca5-ebb63d27e1ff_story.html. Accessed 17 Jul 2020
27. Stafford K, Fingerhut H. Black Americans most likely to know a COVID-19 victim. AP News. 2020. https://apnews.com/article/52ed0842bd17102560e5d896be79d38c
28. Enriquez D Goldstein A COVID-19’s socioeconomic impact on low-income benefit recipients: early evidence from tracking surveys Socius 2020 6 1 17 10.1177/2378023120970794
29. Fagan J Campbell AD Race and reasonableness in police killings Boston University Law Review 2020 100 951
30. Mouzon DM Can family relationships explain the race paradox in mental health? J Marriage Fam 2013 75 2 470 485 10.1111/jomf.12006
31. Mouzon DM Relationships of choice: can friendships or fictive kinships explain the race paradox in mental health? Soc Sci Res 2014 44 32 43 10.1016/j.ssresearch.2013.10.007 24468432
32. Mouzon DMM Religious involvement and the Black-White paradox in mental health Race Soc Probl 2017 9 1 63 78 10.1007/s12552-017-9198-9
33. Thomas Tobin CS Erving CL Brave A Race and SES differences in psychosocial resources: implications for social stress theory Soc Psychol Q 2021 84 1 1 25 10.1177/0190272520961379
34. Louie P Upenieks L Erving CL Thomas Tobin CS Do racial differences in coping resources explain the Black-White paradox in mental health? A test of multiple mechanisms J Health Soc Behav 2021 00 1 16
35. Chatters LM Taylor RJ Bullard KM Jackson JS Race and ethnic differences in religious involvement: African Americans, Caribbean Blacks, and Non-Hispanic Whites Ethn Racial Stud 2009 32 1143 1163 10.1080/01419870802334531 20975850
36. Twenge JM Crocker J Race and self-esteem: meta-analyses comparing Whites, Blacks, Hispanics, Asians, and American Indians and comment on Gray-Little and Hafdahl (2000) Psychol Bull 2002 128 3 371 408 10.1037/0033-2909.128.3.371 12002695
37. Assari S Race, sense of control over life, and short-term risk of mortality among older adults in the United States Arch Med Sci 2017 13 5 1233 1240 10.5114/aoms.2016.59740 28883866
38. Bruce MA Thornton MC It’s my world? Exploring Black and White perceptions of personal control Sociol Q 2004 45 3 597 612 10.1111/j.1533-8525.2004.tb02305.x
39. Hill-Joseph EA Coping while Black: chronic illness, mastery, and the Black-White health paradox J Racial Ethn Health Disparities 2019 6 5 935 943 10.1007/s40615-019-00594-9 31054142
40. Shaw BA Krause N Exploring race variations in aging and personal control J Gerontol 2001 56 2 S119 124 10.1093/geronb/56.2.S119
41. Crocker J Luhtanen R Blaine B Broadnax S Collective self-esteem and psychological well-being among White, Black, and Asian college students Pers Soc Psychol Bull 1994 20 5 503 513 10.1177/0146167294205007
42. Ida AK Christie-Mizell CA Racial group identity, psychosocial resources, and depressive symptoms: exploring ethnic heterogeneity among Black Americans Sociol Focus 2012 45 1 23 10.1080/00380237.2012.630900
43. Sellers RM Smith MA Shelton JN Rowley SAJ Chavous TM Multidimensional model of racial identity: a reconceptualization of African American racial identity Pers Soc Psychol Rev 1998 2 1 18 39 10.1207/s15327957pspr0201_2 15647149
44. Sternthal MJ Slopen N Williams DR Racial disparities in health: How much does stress really matter? Du Bois Rev 2011 8 1 95 113 10.1017/S1742058X11000087 29887911
45. Brown LL Abrams LR Mitchell UA Ailshire JA Measuring more than exposure: does stress appraisal matter for Black-White differences in anxiety and depressive symptoms among older adults? Innov Aging 2020 4 5 1 15 10.1093/geroni/igaa040
46. Eichstaedt JC Sherman GT Giorgi S Roberts SO Reynolds ME Ungar LH Guntuku SC The emotional and mental health impact of the murder of George Floyd on the US population Proc Natl Acad Sci 2021 10.1073/pnas.2109139118 34544875
47. Chae DH Lincoln KD Jackson JS Discrimination, attribution, and racial group identification: implications for psychological distress among Black Americans in the National Survey of American Life (2001–2003) Am J Orthopsychiatry 2011 81 4 498 506 10.1111/j.1939-0025.2011.01122.x 21977935
48. Burrow AL Ong AD Racial identity as a moderator of daily exposure and reactivity to racial discrimination Self and Identity 2010 9 4 383 402 10.1080/15298860903192496
49. Morin RT Midlarsky E Social support, mastery, and psychological distress in Black and White older adults Int J Aging Hum Dev 2016 82 2–3 209 228 10.1177/0091415015627161 26834090
50. Oates GL Goode J Racial differences in effects of religiosity and mastery on psychological distress: evidence from national longitudinal data Soc Ment Health 2013 3 1 40 58 10.1177/2156869312455930 23762783
51. Bierman A Does religion buffer the effects of discrimination on mental health? Differing effects by race J Sci Study Relig 2006 45 4 551 565 10.1111/j.1468-5906.2006.00327.x
52. Anglin DM Link BG Phelan JC Racial differences in stigmatizing attitudes toward people with mental illness Psychiat Serv 2006 57 6 6 10.1176/ps.2006.57.6.857
53. Corrigan PW Watson AC The stigma of psychiatric disorders and the gender, ethnicity, and education of the perceiver Community Ment Health J 2007 43 5 439 458 10.1007/s10597-007-9084-9 17876705
54. Eylem O De Wit L Van Straten A Steubl L Melissourgaki Z Topgüloǧlu Danlşman G De Vries R Kerkhof AJFM Bhui K Cuijpers P Stigma for common mental disorders in racial minorities and majorities: a systematic review and meta-analysis BMC Public Health 2020 20 1 1 20 31898494
55. Peer E Brandimarte L Samat S Acquisti A Beyond the Turk: alternative platforms for crowdsourcing behavioral research J Exp Soc Psychol 2017 70 153 163 10.1016/j.jesp.2017.01.006
56. Williams DR Haile R Gonzalez HM Neighbors H Baser R Jackson JS The mental health of Black Caribbean immigrants: results from the National Survey of American Life Am J Public Health 2007 97 1 52 59 10.2105/AJPH.2006.088211 17138909
57. Flood A, King M, Rodgers R, Ruggles S, Warren JR. Current population survey: version 7.0 . IPUMS. 2020. 10.18128/D030.V7.0
58. Williams DR Yu Y Jackson JS Anderson NB Racial differences in physical and mental health: socio-economic status, stress and discrimination J Health Psychol 1997 2 3 335 351 10.1177/135910539700200305 22013026
59. Tynes BM Willis HA Stewart AM Hamilton MW Race-related traumatic events online and mental health among adolescents of color J Adolesc Health 2019 65 3 371 377 10.1016/j.jadohealth.2019.03.006 31196779
60. Rosenberg M Society and the adolescent self-image 1965 Princeton, N.J Princeton University Press
61. Pearlin LI Schooler C The structure of coping J Health Soc Behav 1978 19 1 2 21 10.2307/2136319 649936
62. Ross CE Mirowsky J Explaining the social patterns of depression: control and problem solving or support and talking? J Health Soc Behav 1989 30 2 206 219 10.2307/2137014 2738367
63. Kessler RC Barker PR Colpe LJ Epstein JF Gfroerer JC Hiripi E Howes MJ Normand SLT Manderscheid RW Walters EE Zaslavsky AM Screening for serious mental illness in the general population Arch Gen Psychiatry 2003 60 2 184 189 10.1001/archpsyc.60.2.184 12578436
64. Gierk B Kohlmann S Kroenke K Spangenberg L Zenger M Brahler E Lowe B The Somatic Symptom Scale (SSS-8): a brief measure of somatic symptom burden JAMA Internal Med 2014 174 3 399 407 10.1001/jamainternmed.2013.12179 24276929
65. Hayes AF Introduction to mediation, moderation, and conditional process analysis: a regression-based approach 2013 New York, NY, US Guilford Press
66. Martinez Gutierrez N Cribbie R Incidence and interpretation of statistical suppression in psychological research Can J of Behav Sci 2021 53 480 488 10.1037/cbs0000267
67. Clement S Schauman O Graham T Maggioni F Evans-Lacko S Bezborodovs N Morgan C Rüsch N Brown JSL Thornicroft G What is the impact of mental health-related stigma on help-seeking? A systematic review of quantitative and qualitative studies Psychol Med 2015 45 1 11 27 10.1017/S0033291714000129 24569086
68. Nazroo JY Bhui KS Rhodes J Where next for understanding race/ethnic inequalities in severe mental illness? Structural, interpersonal and institutional racism Sociol Health Illn 2020 42 2 262 276 10.1111/1467-9566.13001 31562655
69. Neighbors HW Jackson JS Campbell L Williams D The influence of racial factors on psychiatric diagnosis: a review and suggestions for research Community Ment Health J 1989 25 4 301 311 10.1007/BF00755677 2697490
70. Kirkpatrick RH Munoz DP Khalid-Khan S Booij L Methodological and clinical challenges associated with biomarkers for psychiatric disease: a scoping review J Psychiatr Res 2021 143 572 579 10.1016/j.jpsychires.2020.11.023 33221025
71. Sclar DA Robison LM Skaer TL Ethnicity/race and the diagnosis of depression and use of antidepressants by adults in the United States Int Clin Psychopharmacol 2008 23 106 109 10.1097/YIC.0b013e3282f2b3dd 18301125
72. Louie P Wheaton B The Black-White paradox revisited: understanding the role of counterbalancing mechanisms during adolescence J Health Soc Behav 2019 60 2 169 187 10.1177/0022146519845069 31072135
73. Taquet M Luciano S Geddes JR Harrison PJ Bidirectional associations between COVID-19 and psychiatric disorder: retrospective cohort studies of 62354 COVID-19 cases in the USA Lancet Psychiatry 2021 8 2 130 140 10.1016/S2215-0366(20)30462-4 33181098
| 36441494 | PMC9707209 | NO-CC CODE | 2022-12-01 23:20:21 | no | J Racial Ethn Health Disparities. 2022 Nov 28;:1-14 | utf-8 | J Racial Ethn Health Disparities | 2,022 | 10.1007/s40615-022-01457-6 | oa_other |
==== Front
Jindal Global Law Review
Jindal Global Law Review
0975-2498
2364-4869
Springer India New Delhi
178
10.1007/s41020-022-00178-5
Article
Reforms in legal aid and awareness with regard to the aged in India: A case for an inclusive approach
http://orcid.org/0000-0002-8317-9303
K Rajashree [email protected]
1
Singai Chetan [email protected]
2
1 grid.448707.8 Karnataka State Law University, Bengaluru, India
2 grid.464941.a Ramaiah University of Applied Sciences, Bengaluru, India
29 11 2022
124
7 11 2022
© The Author(s), under exclusive licence to O.P. Jindal Global University (JGU) 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
With the increase in the ageing population in India and globally, there is a need to understand the specific challenges faced by the aged from the perspective of legal impediments for securing their well-being. Laws and legal institutions need to be simplified while improving the overall accessibility to justice for senior citizens. There is a need to usher in systemic changes to protect senior citizens through an inclusive legal system.
This article attempts to understand the challenges faced by senior citizens and traces how legal aid and awareness can be an effective tool to remedy some of the issues they confront. It begins by offering a descriptive analysis of the relevant Law Commission reports to understand the aspirations that are laid down with reference to granting legal aid and awareness. Such an analysis was sought so as to provide inputs to the semi-structured interviews that were conducted with senior citizens who were exposed to certain vulnerabilities because of lack of accessibility of legal aid/awareness.
The insights from the interviews provide evidence of the experiences of senior citizens in engaging with the law and legal institutions. These experiences aid in presenting the gaps between the recommendations of the Law Commission reports and the realities on the ground. To this end, the article examines the role and relevance of legal aid for the aged while highlighting the inability of the system to respond to some of the legal needs of the elderly and how these needs can be served by building a robust and rigorous legal aid and awareness mechanism.
Keywords
Senior citizens
Aged
Legal aid
Legal awareness
Inclusive
==== Body
pmcIntroduction
One of the many constants of life is ageing, which is central to maintain the harmony of life. In our Indian society, from time immemorial, the aged have been revered, they have been constantly sought, for giving advice and making decisions. These notions have been changing with the passage of time and as such there has been a paradigm shift in such practices. The ageing population have largely been the victims of such changes in the larger socio-economic milieu. The aged, who are an important section of the society are unfortunately the most vulnerable when it comes to getting access to law and legal institutions. The pandemic has further exposed the need for access to law, the legal systems and services which are crucial aspects for ensuring a dignified life for the aged and the elderly. Such vulnerabilities are further compounded by limited interventions by the state.
In the recent past, there have been significant efforts towards according greater protection to the aged. There are dynamic changes within the family, necessitated by radical alterations to the institution, more so from a socio-economic perspective. This has pushed the aged to the brink of negligence, abuse, and suicide. The traditional perspective of living with and caring for the aged has undergone tremendous changes, and the aged are now perceived as a liability. The disconnect between the younger population and the aged has been widening with time. This necessitates a well-thought-out agenda and approach for addressing the innumerable challenges the aged encounter from a legal perspective. The traditional idea of caring for the aged, considered not only as a social but also a moral obligation, has been further consolidated and strengthened by the government through a series of interventions, especially in the last few decades.
The aged form a vulnerable group of the population due to their physical and mental dependency. Protecting the aged is not only a moral obligation on the part of the younger generation but also a constitutional obligation cast on the State under Part-IV of the Directive Principles of State Policy.1 The provision casts responsibility on the state to secure the rights of the aged who belong to that category of citizens who are considered unwanted. Also referred to as ‘senior citizens’ or ‘elder persons’, they form a considerable portion of the population, locally as well as globally. The National Policy on Senior Citizens 2011 has largely indicated that there shall be an expansion of this group between 2000 and 2050.2 According to the report, the demographic profile of those aged above 60 years is expected to increase by 326 per cent and those above 80 years by 700 per cent.3 India’s life expectancy increased to 69.7 years in 2020 and is projected at 75.9 years by 2050, with 8 per cent of the population being elderly.4 Globally, the United Nations World Assembly study conducted in the year 2002 had various recommendations for the aged, including the need to enable a supportive environment, provide legal advice, overcome legal barriers to accessing healthcare services, accord legal protection to elderly women, and strengthen the existing legal systems to eliminate abuse.5
The elderly need access to healthcare, housing, safety-security, and various welfare measures. Additionally, atrocities and abuse against them have increased manifold. The National Crime Record Bureau data between 2014 and 2016 indicates a steady increase in crimes against the elderly from 18,174 to 21,410 reported cases.6 The situation is further exacerbated in such cases where there is a need for legal advice and access to legal institutions.7 An increase in life expectancy due to advancements in science and technology and a simultaneous increase in neglect and abuse of the elderly have necessitated state interventions to assure them a dignified life as enshrined in Article 21 of the Constitution.8
With the expansion of the ambit of Article 21 of the Constitution, the judiciary has brought within its purview various facets of the right to live with human dignity,9 the right to access health and medical care,10 right to a dignified life,11 and governmental responsibility in providing medical aid,12 which are aligned to Article 25 of the Universal Declaration of Human Rights.13 Consequentially, the state had introduced the Maintenance and Welfare of Parents and Senior Citizens Act 2007, which provides relief to senior citizens who are largely neglected by their children. Such interventions by the state are rather limited and ad hoc.
COVID-19, an unprecedented pandemic, has created personal, financial, and social chaos for the aged. Specifically, abandonment, lack of accessibility, and added to this their inability to use technology have led to their further isolation. With the society adapting itself to the new norms post-pandemic, the aged have been at the crossroads, with little or no legal information on knowledge of specific laws, provisions, schemes, and general lack of awareness on the availability of free legal aid. Due to the inability to connect with their immediate family members, and loss of financial support and social life, the aged have limited or no points of access to seek legal advice and recourse. Across the globe, we have witnessed many frontline workers, individuals, police, and non-governmental organisations (NGOs), stepping in to provide relief to the aged in the form of medical and food relief. But there are no specific instances reported where senior citizens are recipients of legal awareness and legal aid, with particular reference to India. It is equally true that the ageing Indian population are economically and socially dependent. With limited social security schemes, the old-age dependency ratio, which was 9.8 per cent in 2020, is likely to increase to 19.8 per cent by 2050.14 It is time that all senior citizens be covered under the ambit of a separate class/category of citizens who should be made eligible to seek free legal aid and awareness, thereby ensuring legal security.
The Global Study on Legal Aid, jointly prepared by the United Nations Development Programme and the United Nations Office on Drugs and Crime, propagates the idea that in an endeavour to ensure equal access to justice, the rights of the elderly need to be protected.15 As per the report, such measures should address the special needs of those groups, including gender-sensitive issues, and incorporate age-appropriate measures. The report also suggests that the elderly are specifically disadvantaged, especially while availing legal assistance and accessing the courts.16 The report is critical from the perspective of non-availability of free legal aid and services to the specific population consisting of senior citizens who are vulnerable. Hence, the report demands interventions from the state for the welfare and security of senior citizens. 17
The Centre for Human Rights, National University of Advanced Legal Studies, in its report on the elderly has explicitly brought to the forefront the challenges faced by senior citizens and rendered suggestions to cope with specific issues.18 The project report reflects that one of the main problems faced by the ageing population is the lack of appropriate legal awareness.19 Hence there is a need for state intervention to effect appropriate changes to the existing legal framework and introduce policy changes to provide free legal aid and create legal awareness for the aged. This is crucial in order to build a just and inclusive society aligned with Sustainable Development Goal (SDG) 16.20 The SDG’s targets include the promotion of law at national and international levels to ensure access to justice for all. The SDG also intends to achieve effective, accountable, and transparent institutions to ensure ‘inclusive’ decision making at all levels.21 Hence the provision of legal aid is absolutely important for the well-being of senior citizens. The provision of legal counselling is also an important element for sound social, emotional, and physical well-being.22 Against the backdrop of these goals, the present article explores the role and relevance of legal aid and awareness through various Law Commission reports and key documents on legal aid. This analysis has provided the impetus to understand the ground-level realities, which, backed by empirical insights, serve as the basis to render evidence-based recommendations.
Methodology
The study proceeds by first undertaking a review of the literature highlighting the barriers that senior citizens face and the need for formulating inclusive policies and legislation for them. Next, we examine various Law Commission reports, specifically focusing on what they aspire to achieve for senior citizens through policies and legislations. This sets the context for verifying the validity of these goals ‘in the field’ through specific case studies. The fifth section of the article uses case studies to illustrate and analyse the issues associated with senior citizens. Interviews were conducted with senior citizens in old age homes, in urban and rural areas, and those who had sought maintenance before Maintenance Tribunals. In addition, semi-structured interviews were conducted with advocates, judicial officers, civic activists, and administrative officers experienced in issues relating to access to legal aid for the aged.
Information for developing specific case studies was collated through semi-structured, in-depth interviews. Interviews were conducted in-person, via telephone, and via video conferencing in some instances.23 The consent of the respondents was taken prior to the interviews. The authors made handwritten notes wherever the respondents expressed scepticism with regard to the interviews being recorded, while some of the respondents consented to recording the interviews. The respondents were identified and selected through a purposive sampling method, as a cross-section of them were required to have an in-depth understanding of the issues faced by senior citizens. The researchers identified patterns through the interviews, which helped them to categorise three important themes––financial constraints, lack of speedy disposal of cases, and lack of legal awareness. The interviews of the respondents have been presented in the form of narratives, and the researchers have ensured confidentiality through the method of coding. The interviews of senior citizens are coded as RSSC01, 02, and so forth. Similarly, civil activists are assigned the code RSCA01, 02, etc., while administrative officers, advocates, and district magistrates are assigned the codes RSAO01, RSADV01, and RSDM01 respectively. Based on the inputs from the interviews, the article offers recommendations to make the law and legal systems accessible and inclusive for the aged.
Review of literature
COVID-19 has had a serious and pervasive effect on the judiciary, and India is no exception to this. In the midst of the global pandemic, when every state institution was vehemently trying to adapt itself, the judiciary and the judicial systems were, and to a certain extent still are, struggling with the need to adapt to new norms of hearing, representation, and maintenance of safety protocols.24 Despite effecting changes in response to the pandemic, legal institutions and the judiciary have had their fair share of struggles.25 One of the major impediments in this regard pertains to lack of access to justice, since a large segment of the society could not access the courts virtually, which left them out of the legal processes.26 Although the pandemic elevated the Indian judiciary in terms of adapting itself to the vast infrastructural changes, it did leave out a majority of the vulnerable sections of the society from the judicial purview.27
The aged have been stripped of their right to live with dignity due to the generation gap, economic reasons, abandonment due to migration, and many other factors. A review of the laws enacted specifically for the aged shows that there exist some major deficits in framing policies, legislations, and the justice delivery system. A closer look reveals that there is a dearth of specialised legislations for senior citizens. It is only the Maintenance and Welfare of Parents and Senior Citizens Act 2007 which is available for their benefit. Unlike other vulnerable segments of society, such as women and the underprivileged, the aged do not have adequate protection, which strips them of their rights.28 The aged have been repeatedly subjected to atrocities, and studies have hinted that lack of awareness and guidance pose a major impediment in the enforcement of their legal rights.29 The literature is replete with studies that show that the aged need to be protected as they are vulnerable. They play a very active role in the familial setup, yet suffer from loneliness, isolation, and lack of social recognition.30
Policies for the aged in India are influenced by global discourse and international initiatives such as the Macau Plan of Action on Ageing for Asia and the Pacific and the Madrid International Plan of Action on Ageing.31 The Macau Plan of Action focuses on four main areas for the ageing population, viz., understanding the challenges they face, the preparation towards harnessing the ageing population, development of infrastructural requirements, and delivery of essential services.32 On similar lines, the Second World Assembly conducted by the United Nations in Spain, through its Political Declaration and Madrid International Plan of Action on Ageing, drew the attention of the global community towards the development of senior citizens, health-related aspects, and enabling a supportive environment.33 However, these discourses do not focus on the need for an inclusive approach with regard to accessibility of law and legal institutions for senior citizens.
There has been much discussion about access to healthcare and social security measures for the aged, but a grey area exists in terms of providing access to law and legal institutions. Despite the enactment of certain welfare legislations, the ground-level reality reflects a stark contrast while implementing these provisions for the aged.34 With widespread financial, verbal, and social neglect of the aged, it is all the more imperative for the state to provide them with free legal aid and awareness. Effective laws and implementation of the legal system with special reference to the aged are negligible in the Indian context.35
The aged are subject to a variety of abuse, such as physical, mental, and financial abuse. Studies have indicated that the majority of the aged are financially dependent on their children. Added to this are the health problems, which make them susceptible to mistreatment and denial of the basic necessities of life.36 Financial autonomy is a key aspect of the ability of aged persons to access the courts of law during their twilight years. Lack of a continuous source of financial support, and material exploitation, amount to denial of financial accessibility to elders, which includes illegal and improper use of their funds, properties, and assets,37 leaving them vulnerable. Such instances validate the need for providing them with free legal aid and awareness. Elder abuse and mistreatment is today a public health problem that is bound to multiply in the coming years.38 The increase in the adult population necessitates more interventionist policies to curb elderly abuse and protect adults.39 Hence, there is a growing need to provide the aged with legal aid and awareness that will facilitate them in leading a life of dignity by helping them overcome abuse through systematic and systemic changes.
Most often, abusers of senior citizens are family members, which leads to under-reporting of such instances. This is mostly due to dependency, fear of loss of reputation, and incidental consequences of reporting within the family.40 Addressing elder abuse and neglect not only requires appropriate legal interventions but also legal remedies through state action. Across the globe, the need for state preparation to deal with the ageing population and in order to meet their concomitant needs is often discussed in relation to the financial burden that it can have on the state.41 It is important that we treat the aged as resources and prepare the nation for the challenges that they face. Social inclusion of the aged in the domestic legislative and policy spectrum is being advocated across the globe,42 and it is time that India too adopts such policies for the aged into its legal sphere.
Arguments are advanced that the elderly poor should be amongst the first to benefit from the legal system and need to be provided free access to the legal system. With rising costs and recession, especially due to the pandemic, the pressure on aged people to seek legal help is increasing. For the poor elderly, access to legal advice could also mean access to income, food, shelter, and medicine.43
Legal aid to the aged through the lens of law commission reports and key documents on legal aid
Article 39A, inserted into the Indian Constitution through the 42nd Amendment, casts responsibility on the state to render free legal aid to the poor and underprivileged.44 Pursuant to this, various schemes and legislations have been framed in order to give effect to this mandate.45 Access to free legal aid is no longer considered as charity, but more as a basic human right that every individual is entitled to.46 Mandating free legal aid ensures that an individual is able to exercise their right to claim free legal aid and awareness with dignity.47 Despite the mandate, however, access to free legal aid and to courts of law is still an aspiration.48 The judiciary too has stressed the need for free legal aid through its various judgments in the past.49 Further, there has been very limited understanding in terms of rendering legal aid to the aged. Hence, this section explores how the Law Commission reports, over time, have made recommendations and rendered suggestions for the welfare of the aged/senior citizens. Some of the key documents pertaining to legal aid to senior citizens are mapped (See Table 1), and a descriptive analysis of the same is pursued in this section.
Table 1 List of law commission reports regarding legal aid in India
Sl. No Title of Report Year
1. Report of the Committee on Legal Aid and Legal Advice in the State of Bombay 1950
2. 14th Law Commission of India Report on Reforms of Judicial Administration 1958
3. 48th Law Commission Report 1972
4. Report of the Expert Committee on Legal Aid, Processual Justice to the People 1974
5. 128th Law Commission Report on the Cost of Litigation 1988
6. 131st Law Commission Report on the Role of the Legal Profession in Administration of Justice 1988
On 23 March 1949, the Bombay Legal Aid Committee was constituted under the aegis of Honourable Justice N H Bhagwati to examine the need for providing free legal aid in India.50 The Report drew the attention towards a similar committee that was constituted under Lord Rushcliffe by Lord Chancellor Viscount Simon in order to make recommendations with regard to the then prevalent legal aid practices in England and Wales. The Bombay committee Report while citing the Rushcliffe report recommended expanding the ambit of legal aid to a wider section of the population who did not have access to legal help. In fact, the Rushcliffe report had cast the responsibility on the state while stressing the need for the legal profession to provide free legal aid to the poor and the deserving.51 On similar lines, The Bombay Legal aid report, while recognising the need for equal access to legal aid and advice, specifically brought the backward communities under its ambit through the means test—a test by which a person’s financial position which fell below the prescribed level was made eligible to seek free legal aid.52 The report was crucial, as the Indian government, acting on the suggestion of the Honourable Secretary of the Bombay Legal Aid Society, through a circular invited the views and opinions of all the provincial governments to make further provisions for expanding the scheme of legal aid to a much wider target audience.53 The resolution passed was placed before the Bombay Legislative Council in September 1947 to constitute a committee to look into the aspect of providing free legal aid. The excerpt of the resolution is as follows:This Council recommends to the Government to appoint a Committee whether it is desirable to give legal aid at Government cost to the poor and other deserving people who have to come before the courts of law, and if so what steps (including legislation) should be taken for giving such aid.54
The suggestion was again placed before the Bombay Legislative Council in another Resolution in January 1949 by J H Shamsuddin, the then MLA in the Bombay Legislative Assembly with the intention of providing assistance to the deserving in all civil and criminal matters.55
The report was important from the legal aid perspective but did not specifically target the ageing population except to observe that there was a need to bring many more classes of people within the ambit of legal aid in the future. The proposals made in the resolution were no doubt accepted by the Government of Bombay,56 but the terms of reference set out for the Committee included only granting free legal aid for the poor and persons with limited means in all civil and criminal cases. The report was indeed elaborate while laying out the provision of legal aid at district and taluka57 levels, to be administered in all civil, criminal, administrative, and tribunal matters. The report stressed the need for specifically providing legal aid to the backward classes by applying the ‘means test’ to determine their eligibility.58
The report no doubt was limited, but it set the context for the future course of action to be taken with regard to granting legal aid to a wide range of categories of people. The report was the first of its kind in India, but it set the basis for the First Law Commission constituted under the aegis of M C Setalvad. The 14th Report submitted by the First Law Commission credited the Bombay Legal Aid Society for drawing the attention of the government towards the crucial task of rendering legal aid.59 The Bombay Legal Aid Committee Report has also been highlighted as one of the most informed and detailed studies conducted in the area of legal aid. But sadly, its recommendations were not taken up on a priority basis in comparison to the Five-Year Plans. Based on the report, the Government of West Bengal set up a committee under the aegis of Sir Arthur Trevor Harris, the then Chief Justice of the High Court of Calcutta, to render recommendations for providing free legal aid to the ‘indigent’ population.60
The First Law Commission’s 14th Report was brought out under the chairmanship of MC Setalvad with an intention to reformulate various central legislations.61 Its report was the first comprehensive enquiry into the legal system prevalent in India after independence. The idea was to readjust the machinery of law and justice to meet the changing and growing needs of society. The report stressed the need for securing equality before the law and equal protection of the laws, and that people needed to have equal opportunity to access the courts and be adequately represented. The report was of the opinion that legal aid should also be inclusive of legal advice along the lines practised in Canada.62 The report also drew attention towards the legal aid report brought forth in United Kingdom by the committee set up by Lord Chancellor in 1944, which had recommended that a wider income-group of people required to be included within the ambit of free legal aid, and the term ‘poor people’ needed to be replaced by ‘assisted person’.63 The 14th Report, highlighted the responsibility of the state with regard to providing legal aid in the following words:Legal aid is a service that the modern welfare state owes to its citizen. It is a part of the protection of the citizens’ individuality which in our modern conception of the relation between the citizen and the state can be claimed by those citizens who are too weak to protect themselves. Just as the modern state tries to protect the poorer classes against the common dangers of life such as employment, disease, old age, social oppression, etc., so it should protect them when legal difficulties arise.64
The report opined that though the state might not be responsible for the outbreak of epidemics, for old age, or economic crisis, it had an inherent responsibility towards the law. The report was progressive and placed responsibility on the judiciary, legal profession, and the government to enact and implement laws that advanced social justice goals. The report was progressive in not only recommending including within the sweep of legal aid the marginalised sections, but also adopting an inclusive approach to treat legal aid on par with social insurance schemes for the aged.65 The recommendations made in the 14th Law Commission Report were fully incorporated in the Legal Aid and Advice Act 1949. Until the recommendations made by the report were sought to be fully incorporated into suitable legislations and schemes, legal aid was being rendered through some voluntary organisations in the provinces of Bombay, Calcutta, and Bangalore.66
The 6th Law Commission brought out the 48th Report under the chairmanship of P B Gajendragadkar at the instance of the Ministry of Home Affairs to bring reforms in the Code of Criminal Procedure (CrPC).67 The report, among other issues focused on providing free legal aid to women and children, especially those seeking maintenance under Section 488 of the CrPC. The report highlighted the importance of free legal aid to the indigent, with special reference to criminal cases. It highlighted that this was a basic human right requirement stipulated under Articles 14 and 21 of the Constitution, the denial of which would be tantamount to denial of access to justice. The report also highlighted the fact that many deserving cases were left out for want of legal aid.68 The report was crucial as it sought to introduce legal aid aspects into family matters.69 But it did not highlight aspects of legal aid in maintenance issues pertaining to senior citizens. Only some of the recommendations of the report were incorporated into the Criminal Law (Amendment) Act 1972.70
The Report of the Expert Committee on Legal Aid, Processual Justice to the People (1973), chaired by Justice V R Krishna Iyer,71 was path-breaking as it recognised the need for legal aid to the indigent and disadvantaged sections of the society not only through state interventions but also highlighted the role of law schools and law students to provide free legal aid.72 The Committee was of the opinion that legal aid was a product of the legal reform movement in India, and hence it needed to contribute to the corpus juris of the ones who truly deserved legal aid.73 According to the report, the winds of change were inevitable through law reforms that could be ‘informed, purposeful and well-conceived’.74 The report also highlighted the need for legal aid for the geographically disadvantaged, villagers, agricultural labourers, industrial workers, women, children, Harijans,75 prisoners, and the minorities. But the report did not categorically include the aged. The committee opined that legal aid and legal awareness were in consonance with each other, and that it was important that the truly deserving should be entitled so that they were not played out at the hands of their opponents.
In Chap. 8 of the report, reference has been made to the need for public assistance to the aged, old age being considered as a case of ‘undeserved want’ as defined under Article 41 of the Constitution. The report highlighted the need for securing state interventions in the form of beneficial legislation and legal reforms.
The report was commendable as it included various classes of people within its sweep to be made eligible for free legal aid. The inclusion of law students as integral members of schemes to provide legal aid was recommended in this report. The Report had highlighted the inclusion of law students in legal aid schemes formulated by Gujarat Legal Aid Committee and Tamil Nadu Legal Aid Committee.76 This was one of the first detailed reports, post India’s independence, to deal exclusively with legal aid aspects at various stages along with the states’ financial implications.77 It was instrumental in paving the way for the establishment of the Legal Services Authority (LSA) to exclusively deal with and undertake legal aid activities at the national, state, and district levels.78
The 128th Law Commission Report on the Cost of Litigation addressed the issue of the rising cost of litigation.79 The intention of assigning the task to the Law Commission was to report on ‘the cost of litigation with a view to lessening the burden on the litigants.’80 The report highlighted that one of the major roadblocks in creating access to justice was the high cost of litigation. It stated the need for the state to bear the cost of litigation by rendering free legal aid to the deserving.81 The report was particularly important as it recommended that the legal aid scheme be extended wherever possible. Though it did not explicitly make provisions for senior citizens, it emphasised the responsibility of the state aligned to Article 39A of the Constitution. The report highlighted that economic constraints cannot be the reason for denying access to justice.82
The 131st Law Commission Report on the Role of the Legal Profession in Administration of Justice (1988) prepared under DA Desai highlighted the need for providing legal aid as per the constitutional mandate of Article 39A.83 The administration of justice is strengthened by providing free legal aid to the deserving. The report stressed the need for securing ‘equal opportunities to all litigants in search of justice.’84 It hinted that those who were socially disabled should not be denied access to the courts of law.
Most of the State LSA have been involved in spreading awareness to the aged through distribution of printed booklets pertaining to their rights. Nevertheless, the LSA has not considered including senior citizens as a category of individuals to be entitled to free legal aid and awareness.
It would be pertinent to note that there are very limited policies for the elderly to ensure them effective access to justice and the laws. It is interesting that it is the Bill on Maintenance and Welfare of Parents and Senior Citizens 2019, to a certain extent, which mentions the need for providing more accessible and age friendly environment and other public facilities along with helpline services for senior citizens.85 There has been sporadic effort in helping the elderly through government interventions in matters such as health and medicine, pensions, and housing,86 while largely leaving them out of the purview of easy access to courts of law with a view to ensuring equity. Hence it is imperative that the government formulate effective strategies and make suitable inclusions for the ageing population to be entitled to free legal aid.
Legal aid to the aged: Case studies
The case studies presented in this section examine how senior citizens are challenged in today’s context. In order to understand the same, in-depth interviews comprising a cross-section of senior citizens were conducted. The interviews of the senior citizens are presented in the form of narratives along with expert opinions of administrative officers, civil activists, advocates, and judicial officers. An admixture of interviews of senior citizens from old age homes, single senior citizens, and senior citizens from rural and urban backgrounds has been presented herewith. The section puts forth the various vulnerabilities that senior citizens face, such as financial inability, uncertain judicial outcomes, physical and mental fragility, and, largely, the Indian social setup that deters them from approaching and seeking redressal from the courts of law. The article provides the impetus for the government to take note of the legal inadequacies with respect to senior citizens and undertake effective steps to address their issues.
Economic constraints
Economic poverty of the aged is one of the reasons why they face marginalisation. Sociologically, retirement and loss of income are major barriers for senior citizens. Further, studies have shown that adult children are still largely dependent on senior citizens financially.87 In India, a vast majority of senior citizens are left out of the purview of receiving pension, thus making them rely on inconsistent sources of financial income to support themselves in their later years. In this regard, an elderly widowed woman was of the following opinion:I, being the second wife of my husband, do not have children of my own. Elder wife’s children forced me to give away all my property, jewellery, silver coins, and shares of my late husband. Currently, I have moved out of the house and am staying in a rented apartment. Although I receive a pension, I am barely able to pay rent and manage my daily affairs and medicines. I was sceptical of approaching the courts as I would not have received any maintenance as I draw pension every month. Moreover, I did not want to approach the court as I did not have much faith in the law and legal systems. (RSSC01)
Senior citizens’ inability to earn, added to financial instability, is a major cause of concern for them in approaching the courts of law. They are severely handicapped because of the financial uncertainties they need to face which they are not prepared for.
Such instances are not uncommon even amongst the urban affluent senior citizens. Financial constraint is a source of challenge to even senior citizens with a sound financial background. From some of the respondents’ interview, it became clearer that parents from well-to-do backgrounds have been immediately thrown out of the house once all the property has been transferred in the children’s name. Hence, another respondent who runs an old age home had the following to say:Most of the senior citizens have been defrauded by their children. In one such instance, senior citizens came through an NGO that provided them with legal aid. Their son used to run a chit-fund in their name and had disappeared with all the money. They were aged around 90 and above and could barely hear. The NGO had stopped providing them with legal help. There were eight criminal charges of fraud that were levelled against them. The family is well known in Bangalore as they had donated large swathes of land for the construction of temples. Hence they reach out to us and don’t have much awareness about the law. It is mostly the NGOs and police authorities that help them during such instances. (RSCA01)
Such instances reaffirm the need for financial succour from the state. The respondent in the above instance elaborated the need for the state to step in to help senior citizens deal with legal processes and institutions in an appropriate manner. The respondent further elaborated that senior citizens are unable to independently handle the situation and state intervention is paramount in such instances.
There is a need for financial support not only at the grassroots level, but also at various stages of the litigation process. It is not uncommon for senior citizens to receive legal help and remedy in the initial stages under the Maintenance Tribunal.88 Powers are vested with the tribunals to cancel the deed that transferred the rights to the children when they did not take up the responsibility of looking after their aged parents. But the issue is problematic, when appealed because that increases the financial burden of the senior citizens. Lack of financial support remains a major roadblock in such instances. An advocate who dealt with one such matter made the following observation in this regard:A senior citizen aged around 75 had a son aged 45 who abandoned his father after all the money and property were transferred in the son’s name. The father approached the Maintenance Tribunal. The Assistant Commissioner passed an order cancelling the gift deed made by the father to the son. The same was challenged by the son before the Hon’ble High Court. The client was referred to him via another advocate colleague. These senior citizens do not have much idea and are completely dependent on their advocates for guidance and support. Although the appeal was dismissed and upheld the Assistant Commissioner’s order, the senior citizen was put in much peril. (RSADV01)
Free legal aid and assistance ought to be given to the aged at various levels. Senior citizens can be taken advantage of when they rely completely on advocates. The above was corroborated by an Assistant Commissioner, who is also vested with the powers to act as the Presiding Officer of the Maintenance Tribunal, in the following words:There is absolutely no interference by lawyers when the matters are before the Maintenance Tribunal. We try to ensure that 99 per cent of the cases are in favour of senior citizens, but despite that, we have witnessed that the majority of the children approach the courts after the cancellation of the registered deeds that were made by the aged parents in their favour. (RSAC01)
Hence, it is important to ensure that in such cases, where children appeal the matter, senior citizens are not incapacitated from taking the matter forward due to lack of financial support. Legal aid is also crucial in appellate matters, in order to ensure that they are not misled by some unscrupulous members of the legal profession.
With the changing social structures of the family set-up, there have been drastic changes in the relationship between parents and children. In some instances, the children are preoccupied with earning for the family while leaving the aged at home to take care of legal matters. One of the senior citizens expressed his helplessness in the following words:I am a 70-year-old man, a widower, with a petition between my brothers and I for sharing of property. The property is still in the name of my late father and hence [I] approached the court in Chikkaballapur. I myself have been running in and out of the courts for the past five years and have been looking for a compromise now as none of my children are interested in my affairs. I have failing health and mostly stay in temples. I have no one to even accompany me to visit the courts as well. Advocates do not say anything, and we are left at their mercy in our twilight years. (RSSC02)
The changing dynamics of the familial set-up have further isolated senior citizens. As a consequence, there is a genuine need for legal assistance.
The problem is even more persistent in semi-urban and rural areas, compounded by the fact that the elderly need to travel to talukas for hearing of the petitions and to the nearest towns and cities, especially in appellate matters. In such cases, the senior citizens are dependent on the younger generation for commuting as well as for financial considerations to visit the courts for the hearings. Narrating one such instance, a senior citizen recounted his ordeal as follows:I am a 65-year-old and have three cases pending before various forums in Chikkaballapur and Gauribidanur. One case pertains to Easementary rights, which has been going on since 2017, and at the evidence stage; the second case pertains to a property dispute between my wife with her siblings; and the third case pertains to the division of property between my brothers and me. My son resides in Bangalore while my daughter is married. It is becoming expensive by the day to deal with all three cases. (RSSC03)
The above instance connotes the need for an active response from the state with regard to allocating fixed financial resources, to entail the class of senior citizens to be made eligible for availing of free legal aid and services. The problem of financial assistance is pervasive when it comes to senior citizens. Hence it is time that the state becomes serious with regard to the issue to ensure access to justice.
Judicial intervention: Lack of early disposal of cases
The Indian judiciary is saddled with cases, and despite so many years of independence, it grapples with mounting numbers of cases. The judiciary is facing an uphill task in responding to the increasing volume of cases.89 India’s Law Commission reports have time and again highlighted this issue since the year 1964. The surging cases and the inability of the judiciary to dispose of the matter has shaken the trust of the common man.90 In particular, civil justice in India is very slow, and the rate of case disposal is still very low.91 Such delays have a visible impact, especially on senior citizens, as evidenced below.I did not feel like approaching the court to fight for my rights after my son, daughter-in-law, and two daughters did not take care of me. I did not have much faith in the judiciary. I did consult my niece, who is a lawyer, and she too said that there would be a delay in the courts. I did not want to risk at this age approaching a court. My niece did help me with all the legal advice, but for how long could I be dependent on her, even if I approached the court. Maybe I would have been comfortable if I had received any help from the government. (RSSC03)
Time is a crucial factor for senior citizens, whereby they look for early disposal of cases. They are mostly at a stage where they wish to secure the rest of their life and live comfortably. They do not wish to spend the rest of their years in a court of law. Most of the interviewees strongly believed that they were better off without approaching the court and that it was a futile exercise. Some of them opined that they were uncertain if they would even be alive by the time the judgment was delivered by the courts.
Respondent RSSC03 said that since he suffered from depression from the fact that he was living alone, he could not bear the thought of approaching the court and falling victim in the hands of the law and legal system. Inordinate delay in the court procedures has compelled senior citizens at various stages to suffer silently rather than approach a court of law.
A social activist, involved in ensuring that senior citizens are rehabilitated in old age homes, observed that senior citizens in cases of abuse reach out to the Old Age Help Line, where they are counselled and sent to organisations that take care of them. The National Helpline for Senior Citizens is an initiative of the Ministry of Social Justice and Empowerment, Government of India.92 There is unwillingness on their part to approach the court as they are scared and do not believe that much can be achieved by approaching the court of law. The respondent had even filed a public interest litigation on behalf of senior citizens. He made the following statement, requesting anonymity:We do not have much trust in the system. We cannot convince the senior citizens at this stage to approach a court of law. There have also been instances where some have even tried to take advantage of their age and approach the court for personal benefits. It is a hassle for the senior citizens at this age as there is an inordinate delay in the justice delivery systems. They are at a stage wondering how they will live, and it would be difficult and time-consuming. Most of the time, we are with them, but who will be with them to attend to the cases and take them forward. Most of the time, we try to effect a compromise between the parents and the children. (RSCA02)
Delay in court procedures and lack of any specific intervention from the law and legal processes, force senior citizens to suffer in isolation. Even if they do approach the courts, they do not have appropriate legal guidance to manoeuvre the situation appropriately.
Government administrative officials who are given charge of the Maintenance Tribunal said that the procedure before the Maintenance Tribunal is very simple as there are not many formalities involved. The concerned officer stated that they receive two to three such cases that come before them every month on average, which is treated on a priority and deal with the cases immediately. The officers are empathetic; despite this, senior citizens feel that they are not attended to on a timely basis. The administrative officer opined that:As an Executive Magistrate and an Assistant Commissioner, I try my best to hear the cases. But the senior citizens’ cases are also taken up along with many other public hearings. Our schedules are tight, and many senior citizens are reluctant to attend the cases along with the others. There is no fixed date and hearing, and sometimes, they just leave. We try to ensure that the matter is disposed of in 60 days; despite that, in some instances, it can go up to 120 days. (RSEM02)
Legal institutions need to be inclusive of senior citizens. Provisions ought to be made to expedite cases of senior citizens. As court procedures are cumbersome, the senior citizens need to be heard on separate days and times. The law and the institutions need to be responsive and undertake such appropriate measures to aid in early disposal of cases.
The situation became all the more problematic with the pandemic; senior citizens could not be called to the court due to the restrictions imposed in response to the pandemic. At the same time, senior citizens’ matters could not be taken up, as they were unfamiliar with online court processes and mechanisms. In this regard, a senior citizen from a taluka responded with regard to the status of the cases during the pandemic:During the pandemic, dates would mostly be given by clerks; either the judges were getting transferred, or they were in training. One case has been pending since 2017, another since 2009, and the property dispute has been pending since the year 1989. After 60 years, most of us suffer from some or the other ailment. Since we could not step into the courts, we got ‘Corona dates’, there were hardly any judges or lawyers in the court during the pandemic, and we did not know how to proceed as our advocates would not say anything to us. (RSSC04)
The above situations have led to inordinate delay in the court processes, especially the ones that involved senior citizens. Respondent RSEM03, an officer, observed that for some time, senior citizens did face ‘exclusion’ during the pandemic. On similar lines, a senior advocate practising in the High Court of Karnataka made the following statement:In my so many years of practice, what I have experienced is that early hearing on a priority basis is one of the ways for early disposal of cases for senior citizens. But in reality, that is not the case. The implementation is lacking in this regard, which puts the senior citizens at a disadvantage. I have been to the National Institute of Mental Health and Sciences (NIMHANS) and KIDWAI Memorial Institute of Oncology Cancer Research and Training Centre, where I have rendered pro-bono legal advice to many patients and their relatives. There is a need to build a sense of trust in the judicial system. (RSAD02)
Hence the law and the legal system need to respond to the aged in a systematic and time-bound manner. In order to ensure early disposal of cases and build trust towards law amongst senior citizens, interventions from the legal fraternity through legal aid and awareness are all the more pertinent.
Lack of legal awareness
Legal awareness is of paramount importance to the general public. Legal awareness is essentially the awareness of citizens’ rights and obligations, knowledge of the existing legislations, and, at the same time, the ability to deploy this knowledge in exercising and defending their rights through legal mechanisms.93 There is a need to educate the senior citizens to help them understand what law is, and how it functions and works for their benefit. Legal consciousness is an imperative tool to help senior citizens to help themselves. A senior citizen in an old age home had the following observation to make:My wife and I were abandoned; we reached out to the Old Age Help Line through a neighbour, where we were counselled and sent to this NGO. We are now confused and unable to comprehend and process the whole situation. Hence it is difficult for us to take a sudden decision. We are victims of abuse and wish we were educated about the law. (RSSC05)
Most of the respondents lamented that they were not legally conscious and wished they were educated about the laws and the legal institutions. Similar response was espoused by RSSC01. The respondent regretted that she had no idea of the law. She expressed the view that she was not aware of the provisions of the Hindu Succession (Amendment) Act, 2005 which considered women as coparceners, until recently.
She expressed with anguish that despite being educated, she did not know ‘how’ and ‘where’ to approach [the law] while the brothers took away all her ancestral property. She expressed that ‘I would have been comfortable if we had received some knowledge of the law.’ She said that lack of legal awareness is a major problem, especially for senior citizens.
The issue is persistent, especially among senior citizens in rural areas, as they are absolutely unaware of the law and the legal systems. There is a sense of apprehension, especially when it comes to law and advocates. They are too scared to even speak up with their advocates about the progress of the case. They spoke to us in hushed tones and were scared that someone would hear and report it to their advocates and the courts. There was large-scale ignorance amongst the senior citizens, and they looked upon the judicial process as elite. In this regard, it is pertinent to highlight how scared a senior citizen was when he made the following statement:We just blindly listen to what the lawyers tell us. They hush us up when we ask them any question. We have left everything in the hands of the advocate. We approach the local panchayat members, who refer us to an advocate as we don’t have much choice in this regard. I sometimes feel that advocates purchase a settlement with the opposite party for a lesser price, and we have no option but to concede to the same. Most of the time, the advocates don’t respond to our questions, and we are left totally helpless. (RSSC04)
Large-scale ignorance of law was a major impediment to senior citizens, who expressed a lot of fear of the system. An administrative officer, while highlighting the issue pertaining to legal awareness, made the following statement:The Act pertaining to the Maintenance of Senior Citizens is there since 2007. We are now in 2022, yet people are not aware of the Act. We are trying to educate people as well as the police. The police, too, treat this as a regular civil dispute. The civil courts have no jurisdiction in such matters. The people are influenced by local leaders and gram panchayat members who act on their advice. (RSAC01)
These situations highlight the fact that due to a lack of legal awareness, most senior citizens are victims of misinformation and miscommunication. In order to remedy the situation, there is a need for proper legal awareness through appropriate authorities. A judicial official said that it is pertinent to involve the LSA in this regard. He commented:The LSA needs to be active. They need to give suggestions, especially as to how senior citizens need to be made aware of various legal procedures and how to approach the authorities. The LSA functions need to include appropriate provisions to render legal aid and awareness to senior citizens. (RSDM04)
The officer further lamented that he had never heard of the LSA referring any senior citizens’ cases, to the Tribunals or to the courts, to take up such matters. The need of the hour is an effective legal system, which can be achieved by rendering effective legal awareness.
Recommendations
Through this study, the researchers intend to pursuade the relevant authorities to take note of the difficulties faced by senior citizens and undertake appropriate measures to ease the law and legal processes for senior citizens. Thus, in this section, a course of action is recommended that can be considered by the concerned authorities. The mental faculty, financial instability, and educational and social backwardness of senior citizens fuel the need to implement the following recommendations:
Section 12 of the Legal Services Authority Act states the criteria for receiving legal services, which include members of Scheduled Castes, Scheduled Tribes, women, children, victims of human trafficking, persons with special needs, industrial workers, victims of mass disaster, persons in custody, and people with low annual income. Suitable amendments ought to be made to the Legal Services Authorities Act 1987 to make senior citizens eligible for free legal aid and services. The various Law Commission reports have time and again suggested including more deserving people within the ambit of free legal aid. It is time that suitable additions be made to Section 12 of the Legal Services Authority Act.
The National Legal Services Authority (NALSA) formulated a Legal Services Scheme in 2016 to ensure the welfare of senior citizens. One of the preliminary objectives of the scheme is to provide legal awareness to senior citizens. The objectives also aim to establish legal aid clinics at all levels, including universities. The State Legal Services Authority has to implement the same by conducting dissemination activities of its policies, schemes, and programmes to senior citizens as well as government functionaries. There is a need to give effect to its objectives in its true spirit by conducting adequate legal awareness programmes/workshops specifically targeting senior citizens on issues such as maintenance, civil rights, elder abuse, property rights, etc., which are still lacking. The awareness programmes should also include taking note of simple problems such as non-availability of Aadhaar cards, ration cards, and other important documents. Once the problems are noted, it is equally important that the problems be forwarded to the relevant authorities and compliance ensured. It is important that the senior citizens are empowered to act on the information provided to them.
Under the Maintenance Act 2007, the Maintenance Tribunals are presided over by administrative officers. These administrative officers have been assigned numerous responsibilities, including: managing governmental responsibilities in terms of formulation, implementation, and monitoring of policies in consultation with the various departments—financial management, disaster relief, and many others. Hence, the administrative officers have been accorded immense responsibilities. There is a need for appointment of a separate judicial officer to deal with senior citizens maintenance cases, as the administrative officers may not have the requisite expertise to act as judicial officers. This approach ensures early and speedy disposal of senior citizens’ cases on priority.
Senior citizens face innumerable challenges especially from the health perspective. In order to accommodate them and to ensure a fair opportunity to access justice, it is pertinent to allocate separate dates and times of hearing to senior citizens. The cases are not to be treated as regular cases and need to be accorded priority. This is to be done keeping in mind the health and psychological limitations faced by them.
Part IV of the Bar Council of India Rules on Legal Education (2008) makes the establishment of legal aid clinics in law schools mandatory. This is an important prerequisite for law schools for establishment as well as for the purpose of accreditation. At the same time, NALSA has formulated the National Legal Services Authority (Legal Services Clinics in Universities, Law Schools, and Other Universities) Scheme in 2013. The scheme is pursued with the intention of sensitising the students towards the needs of marginalised and vulnerable sections of society through awareness programmes such as camps, seminars, street plays, and counselling services. This provides an opportunity for the LSA to utilise the services of law students as well train them in the required skill-set necessary for the profession. In order to achieve this, the LSA should work in tandem with NGOs and law schools’ legal aid clinics at the district and taluka levels to help them in creating awareness and providing legal aid to senior citizens.
The police are the first point of contact for any information that citizens seek. Hence, it is pertinent that they are guided accordingly. In most instances, when senior citizens approach them, the police treat the matters as regular matters. Hence, police officials need to be trained and made aware so as to not treat senior citizens’ maintenance cases as regular civil suits.
Senior citizens need to be guided towards accessing legal institutions. Hence, if consultation centres are established in certain designated police stations, it would help senior citizens who lack knowledge and awareness of the law to undertake appropriate steps or decide their future course of action.
Separate infrastructure facilities should be established for senior citizens in the court premises at the district and taluka levels. The facilities can include separate enclosures or waiting lounges, seating spaces, basic healthcare facilities, essential goods, and ramps.
In most cases, especially in rural and semi-urban areas, there is a general fear amongst the public with regard to approaching the courts and legal institutions. In some instances, the aged are misled by people who do not have adequate information. Hence it would be appropriate to appoint special officers or train the existing panchayat officers by equipping them with the right information to guide the senior citizens to appropriate forums and oversee the implementation of the above requirements.
The present Senior Citizens Act 2007 does not contain specific provision for early disposal of cases. Hence, if specific provisions to expedite the cases are incorporated, it can ensure early disposal of maintenance cases. This casts responsibility on the administrative officers to engage with senior citizens on a priority basis.
Declarations
Conflicts of Interest
The authors have no conflicts of interest to declare that are relevant to the content of this article.
1 Constitution of India 1950, Article 41: ‘The State shall, within the limits of its economic capacity and development, make effective provision for securing the right to work, to education and to public assistance in cases of unemployment, old age, sickness and disablement, and in other cases of undeserved want.’
2 National Policy on Senior Citizens 2011, 2. https://socialjustice.gov.in/writereaddata/UploadFile/dnpsc.pdf. Accessed 15 August 2022.
3 Ibid.
4 Leyanna Susan George, ‘Ageing in India: An Intricate Phenomenon’ in Helaine Seline (ed), Aging across Cultures: Growing Old in the Non-Western World (Springer 2021) 165–179.
5 United Nations Department of Economic and Social Affairs on Ageing, Report of the Second World Assembly on Ageing, A/CONF.197/9 (2002).
6 National Crime Records Bureau, Crime in India (2019) Chap. 6A, Table 6A.1. https://ncrb.gov.in/sites/default/files/crime_in_india_table_additional_table_chapter_reports/Table%206A.1.pdf. Accessed 05 February 2022.
7 Alan Gutterman, 'Older Persons Access to Justice’ (2022) (Oakland CA: Older Persons’ Rights Project, 2022) 1–2.
8 Anuradha Thakur, ‘Care of Senior Citizens and the Role of the State’ (2008) 43(17) Economic and Political Weekly 11.
9 Sunil Batra v Delhi Administration (1978) 4 SCC 409.
10 Consumer Education and Research Centre v Union of India AIR 1995 SC 636.
11 Francis Coralie Mullin v Union Territory of Delhi (1981) 1 SCC 608 [Bhagwati PN J].
12 Paschim Banga Khet Mazdoor Samity v State of West Bengal (1996) 4 SCC 37 [Agrawal SC J].
13 Universal Declaration of Human Rights, Article 25(1): ‘Everyone has the right to a standard of living adequate for the health and well-being of himself and of his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control.’
14 S Vijay Kumar, ‘Social Security for the Elderly in India’ (2001) 31(4) Social Change 21.
15 United Nations Office on Drugs and Crime, Global Study on Legal Aid––Global Report (2016) 73. The study provides insights into the current state of legal aid practices across the globe. The report also highlights specific challenges and priority areas of legal aid delivery.
16 Ibid. 75, 76, 128.
17 Ibid. 7.
18 National Human Rights Commission, Human Rights of the Elderly Persons: Law, Policies and Implementation: A Study with Special Reference to Kerala (An initiative by the Centre for Human Rights, National University of Advanced Legal Studies) (2018). https://nhrc.nic.in/sites/default/files/Project%20Report%20on%20the%20Rights%20of%20the%20Elderly_NUALS.pdf. Accessed 05 March 2022.
19 Ibid.
20 UN Sustainable Development Goal 16: ‘Promote peaceful and inclusive societies for sustainable development, provide access to justice for all and build effective, accountable, and inclusive institutions at all levels.’ https://sdgs.un.org/goals/goal16. Accessed 22 August 2022.
21 United Nations Department of Economic and Social Affairs, (The Sustainable Development Goals Report, 7 July 2022) https://unstats.un.org/sdgs/report/2022/The-Sustainable-Development-Goals-Report-2022.pdf. Accessed 22 August 2022.
22 Marie Weil, ‘Involvement of Senior Citizens in Needs Assessment and Service Planning’ (1982) 11(2–3) Non-profit and Voluntary Sector Quarterly 59, 64.
23 The interviews were conducted between the duration of January 2022 to May 2022.
24 Kumar Vaibhav, ‘Access to Justice during COVID-19 Pandemic: An Indian Perspective’ (Asia Pacific Regional Forum Publications, 12 August 2021). https://www.ibanet.org/article/F1A1E3A6-675F-455B-B190-038306399CBB. Accessed 23 October 2022.
25 Joanne Harris, ‘Access to Justice: India Leads Post COVID Shifts in Courts’ Use of Technology’ (International Bar Association, 12 October 2022). https://www.ibanet.org/Access-to-justice-India-virtual-courts. Accessed 23 October 2022.
26 Ajay Pal Singh, ‘Virtual Courts during the Pandemic: A Critical Exposition’ (2021) 24 Supremo Amicus 452.
27 Pragya Jha and J Nayak, ‘COVID-19: The Catalyst to an E-Judiciary’ (2020) 18 Supremo Amicus 644.
28 Animesh Puneet Gupta and Adhiraj Bhandari, ‘The Experienced, the Aged, the Ignored: A Review of Senior Citizens and the Elderly Laws in India’ (2018) 6 Supremo Amicus 1.
29 Niharika Thakur, ‘Senior Citizens: Our Rights or Duties?’ (2018) 4 Supremo Amicus 274.
30 SM Ayoob, ‘Senior Citizens and Their Roles in Family and Household’ (2020) 13 Journal of Politics and Law 32.
31 Rama Baru, Madhurima Nundy, Ramila Bisht, et al., ‘Integration of Health Services for Older Persons in Urban India’ (2021) 56(37) Economic and Political Weekly 36.
32 United Nations Economic and Social Commission for Asia and the Pacific, Guidelines on the Implementation of the Macao Plan of Action on Ageing for Asia and the Pacific (7 April 2000). https://digitallibrary.un.org/record/433301?ln=en. Accessed 1 November 2022.
33 United Nations, Political Declaration and Madrid International Plan of Action on Ageing, Second World Assembly on Ageing, Madrid, Spain, 08–12 April 2002 (New York 2002). https://www.un.org/esa/socdev/documents/ageing/MIPAA/political-declaration-en.pdf. Accessed 31 July 2022.
34 Jyotsna M Kalavar, Duvvuru Jamuna, and Farida Kassim Ejaz, ‘Elder Abuse in India: Extrapolating from the Experiences of Seniors in India’s “Pay and Stay” Homes’ (2013) 25(1) Journal of Elder Abuse & Neglect 3.
35 S Bhattacharya and S Bhattacharya, ‘What Happens to the “Hand That Rocked the Cradle”? A Study of Elderly Abuse in India’ (2014) 16(3) Journal of Adult Protection 166.
36 Srinivasan Chokkanathan and Alex EY Lee, ‘Elder Mistreatment in Urban India: A Community Based Study’ (2005) 17(2) Journal of Elder Abuse & Neglect 45.
37 Namkee G Choi and James Mayer, ‘Elder Abuse, Neglect, and Exploitation’ (2000) 33(2) Journal of Gerontological Social Work 5.
38 Elsie Yan, ‘Elder Abuse: A Public Health Issue’ (2019) 59(3) The Gerontologist 598.
39 Gwendolyn Fearing, Christine L Sheppard, Lynn McDonald, et al., ‘A Systematic Review on Community-Based Interventions for Elder Abuse and Neglect’ (2017) 29(2–3) Journal of Elder Abuse & Neglect 102.
40 Lynette Joubert and Sonia Posenelli, ‘Responding to a “Window of Opportunity”: The Detection and Management of Aged Abuse in an Acute and Subacute Health Care Setting’ (2009) 48(7) Social Work in Health Care 702.
41 Ernest Wing-Tak Chui, ‘Caring for Our Seniors––Private Issue or Public? The Asian Scene’ (2012) 5(1) Journal of Asian Public Policy 51.
42 Nils Dahl, ‘Social Inclusion of Senior Citizens in Japan: An Investigation into the “Community-Based Integrated Care System”’ (2018) 30(1) Contemporary Japan 43.
43 Burton D Fretz, ‘Legal Services and the Elderly Poor’ (1984) 8(3) Generations: Journal of the American Society on Aging 14.
44 Constitution of India 1950, Article 39A: ‘The State shall secure that the operation of the legal system promotes justice, on a basis of equal opportunity, and shall, in particular, provide free legal aid, by suitable legislation or schemes or in any other way, to ensure that opportunities for securing justice are not denied to any citizen by reason of economic or other disabilities.’
45 Schemes such as NALSA (Legal Services to Senior Citizens) Scheme, 2016; Legal Services to Victims of Acid Attacks Scheme, 2016; Protection and Enforcement of Tribal Rights Scheme, and many more are formulated by NALSA.
46 Chetan Kaushik, ‘Free Legal Aid: A Human Right, Not “Charity”’ (2020) 3(4) International Journal of Law Management & Humanities 1821, 1826.
47 Isabell Manoj and Angel Shaji, ‘Is Free Legal Aid Available and Beneficial for the Targeted Community’ (2022) 5(2) International Journal of Law Management & Humanities 543.
48 Andrew Higgins, ‘Legal Aid and Access to Justice in England and India’ (2014) 26(1) National Law School of India Review 13.
49 See, e.g., Hussainara Khatoon and Ors v Secretary, State of Bihar (1980) 1 SCC 98 [Bhagwati PN J].
50 Legal Department for the Government of Bombay, Report of the Committee on Legal Aid and Legal Advice. Under Government Resolution No 3157 (White Paper, Cm 1[(21]), 2009) part III (viii) 2.
51 Ibid. 1.
52 Ibid. 11.
53 No D.10/46-C & G (Judl.), 1946, cited in Report of the Committee on Legal Aid and Legal Advice (n 50) 2.
54 Report of the Committee on Legal Aid and Legal Advice (n 50) citing the Bombay Legislative Council resolution 2.
55 Ibid.
56 Ibid. 2, 3.
57 An administrative unit in India below a district. Also called ‘tehsil’ in north India.
58 Ibid. 30–39.
59 Law Commission of India, Reforms of Judicial Administration (Law Com No 14, 1958) 587.
60 Ibid. 15, 16, 588.
61 Ibid. 587–624.
62 Ibid. 590.
63 Ibid. 591.
64 Ibid. 593.
65 Ibid. 594.
66 Ibid. 592.
67 Law Commission of India, Some Questions under the Criminal Procedure Code (Law Com No 48, 1972) 23–28.
68 Ibid. 9.
69 Ibid. 22, 23.
70 The Code of Criminal Procedure 1973.
71 Government of India, Report of the Expert Committee on Legal Aid: Processual Justice to People (May 1973).
72 Ibid. 155–164.
73 Ibid. 165.
74 Ibid. 166.
75 Members of the hereditary Hindu group who were formerly considered inferior and who had the lowest social status.
76 Report of the Expert Committee on Legal Aid (n 71) 155, 156.
77 Ibid. 274.
78 Ibid. Chapter 19, 273.
79 Law Commission of India, Cost of Litigation (Law Com No. 128, 1986).
80 Ibid. 1.
81 Ibid. 3–6.
82 Ibid. 6.
83 Law Commission of India, Report on the Role of Legal Profession in Administration of Justice (31 August 1988).
84 Ibid. 13.
85 Section 25 of the Bill seeks to insert a new section 23A for ‘Other Welfare Measures for Senior Citizens’ after Section 23 of the principal Act, Maintenance and Welfare of Parents and Senior Citizens Act, 2007.
86 Aparajita Chattopadhyay, ‘Population Policy for the Aged in India’ (2004) 39(43) Economic and Political Weekly 4694.
87 A Justine Kristel Villegas, ‘Greying Matters: Aging in Contemporary Philippine Society’ (2014) 62(1) Philippine Sociological Review 139.
88 The Maintenance Tribunal is established under section 7, by the respective state governments under the Maintenance and Welfare of Parents and Senior Citizens Act, 2007. The Tribunal is presided over by an officer not below the rank of a Sub-Divisional officer of a state. Under Section 9, the Tribunal is vested with the power to order maintenance in the form of a monthly allowance for senior citizens whose children have either neglected or refused to take care of them.
89 Gurrashmeet Singh, ‘Delay in Justice Delivery System in India’ (2021) 1 Jus Corpus Law Journal 634.
90 Jeffrey Falt, ‘Congestion and Delay in Asia’s Courts’ (1985) 4(1–2) UCLA Pacific Basin Law Journal 90.
91 Krishna Agrawal and Neha Dixit, ‘Civil Justice in India’ (2016) 3(4) BRICS Law Journal 71.
92 National Implementing Agency (an empowered Committee) set up by the Ministry of Social Justice and Empowerment along with NISD (National Institute for Social Defence) is currently hosting the Elder Line in collaboration with the state governments. Operationally, the state agencies will be working closely with all state departments and district functionaries to deliver services. This is a collaborative approach between MOSJE, State departments, Identified state agency, NGOs, local senior citizens group, voluntary groups etc. For more information see PIB Delhi, ‘National Helpline for Senior Citizens’ (27 September 2021). https://pib.gov.in/PressReleasePage.aspx?PRID=1759358#:~:text=Elder%20Line%2014567%20is%20a,of%20life%20of%20senior%20citizens. Accessed 2 November 2022.
93 Oleksandr V Skrypniuk, Nataliya M Onishchenko, Natalia M Parkhomenko, ‘Awareness in Law as Strategical Direction of Legal Policy’ (2019) 10(5) Journal of Advanced Research in Law and Economics 1534.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
| 0 | PMC9707211 | NO-CC CODE | 2022-12-01 23:20:21 | no | Jindal Global Law Review. 2022 Nov 29;:1-24 | utf-8 | null | null | null | oa_other |
==== Front
Appl Psychophysiol Biofeedback
Appl Psychophysiol Biofeedback
Applied Psychophysiology and Biofeedback
1090-0586
1573-3270
Springer US New York
9570
10.1007/s10484-022-09570-2
Article
Effects of Audio Brain Entrainment on Korean People with Mild Insomnia
http://orcid.org/0000-0001-5473-9597
Ho Chee Meng Benjamin [email protected]
1
Jeong Hana 12
Lim Young-Hyun 3
Park Se Jin [email protected]
1
1 AI Research Group, Sewon Intelligence, Sejong-daero, Jung-gu, Seoul, Korea
2 grid.411143.2 0000 0000 8674 9741 Department of Physical Therapy, Konyang University, Gwanjeo-Dong, Seo-Gu, Daejeon, Korea
3 GEOMC Co. Ltd, Siheung-dong, Sujeong-gu, Gyeonggi-do Korea
29 11 2022
110
4 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Sleep health has become an important healthy lifestyle. Research has shown that almost one-fifth of the Korean adult population does not have sufficient sleep. The lack of sleep is associated with significant medical, psychological, social, and economic issues. People are not only yearning for sufficient sleep but the quality of sleep as well. Usually, the obvious choice will be the use of pharmaceuticals however, these often have various side effects, and the lasting use of these medications could become a concern. Therefore, new non-drug alternatives are sought after. Audio brain entrainment is a procedure that modules neural activities by synchronizing brainwave frequency with pulse tones. By producing frequency tones for the deep sleep stage, it promotes a good night’s sleep. In this paper, we developed a pillow integrated with the audio speakers that produce alpha and theta beats that should help improve sleep. Sleep polysomnography was performed on 10 people to compare the effects of the audio stimulus. Initial results showed a positive effect on sleep onset latency, indicating that sleep induction happened. This noninvasive stimulation technique can be a promising candidate for wearable bioelectronics medicine and further neuroscience research.
Keywords
Sleep
Brain entrainment
Insomnia
Sleep treatment
GEOMC Co. Ltdip2020-0021
==== Body
pmcIntroduction
With humans sleeping for almost one-third of their life, sleep health has become an essential component of a healthy lifestyle (Aminoff et al., 2011; Perry et al., 2013). The timing, duration, and quality of sleep are important indicators of health, and this is further emphasized during the Covid-19 pandemic (Chaput et al., 2018; Xiao et al., 2020a, b). While a good night’s sleep may provide mental and physical recovery (Bonnar et al., 2018; Kim, 2009), sleep disorders are associated with significant medical, psychological, and social disturbances (Cho et al., 2019; Leggett et al., 2017; Vgontzas & Kales, 1999). Different studies had associated people with sleeping disorders with a higher risk of mortality, diabetes mellitus, hypertension, cardiovascular disease, coronary heart disease, and obesity (Im & Kim, 2017; Itani et al., 2017; Jike et al., 2018). For these reasons, good sleep is crucial to a healthy life.
In the United States, more than a third of American adults do not get enough sleep (Perry et al., 2013). However, insufficient sleep is not exclusively a US problem but affects most developed countries (Hafner et al., 2016). According to a report by the Organization for Economic Cooperation and Development countries (OECD), the average sleep time in South Korea was lower than the OECD average (South Korea: 462 min per day; OECD average: 502 min per day) (Chapple & Ladaique, 2009; Kim et al., 2018). Treatment for sleep disorders usually involves pharmacotherapeutics or hypnotics however there are some concerns about the side effects or lasting use of these medications (Lie et al., 2015; Pagel & Parnes, 2001). Possible side effects include daytime drowsiness, cognitive impairments, reduced motor coordination risk of abuse, and long-term dependence (Frandsen et al., 2014; Kripke et al., 2012). Therefore, the demand for low-cost, non-pharmaceutical alternative aids has grown.
In Korea, non-pharmaceutical such as herbal medicine, meditation, acupuncture, exercise, etc. are used (Kwon et al., 2016). One alternative method is the use of audio stimulation (Jespersen et al., 2015; Lai & Good, 2006). Trials have shown that music can be a self-help intervention to improve sleep (Chen et al., 2014; Cho et al., 2019; Feng et al., 2018; Jespersen et al., 2015; Lai & Good, 2006) and a Google search on ‘music’ and ‘sleep’ reveals a huge market promoting its sleep-inducing properties (Jespersen et al., 2015). Besides music, audio pulses can be applied to elicit the brain’s frequency following response, encouraging the brainwaves to align to the frequency of a given beat (Huang & Charyton, 2008). Brainwave entrainment is a procedure that modules neural activities by synchronizing brainwave frequency with that of the stimuli (Dickson & Schubert, 2019). There are four common brainwave bandwidths (delta, theta, alpha, and beta) and the related mental activities have been studied intensely (Thompson & Thompson, 2003). For example, delta frequency (0.5–4 Hz) is dominant during the deep sleep state while theta frequencies (5–8 Hz) are observed in the drowsy or relaxed state. In addition, alpha frequencies (9–13 Hz) are present in various mental activity such as mediation, creativity, calmness, etc. and beta frequencies (13–30 Hz) occurs before reactive psychomotor actions such as intellectual activity and emotional and anxious states. (Tang et al., 2016).
In this paper, we developed a pillow integrated with the audio speakers that produced specifically binaural beats from 2 different frequencies. We hypothesized that the audio stimulation will facilitate EEG slowing (via feedback, entrainment, conditioning, etc.) which may improve sleep induction and maintenance. This study will examine the feasibility of a 30 min administered pulse tone stimulation for sleep promotion in Korean adults with insomnia in a controlled environment. Sleep polysomnography was performed on 10 Korean individuals and sleep data were collected over 2 different nights for everyone. Each night, participants were either exposed to the audio stimulation or not (control). As sound is usually combined with light to create an audio-visual stimulation, a saliva test for melatonin is performed to see if sound alone may affect the circadian rhythm cycle. A survey was also conducted to see the viability of the pillow as a commercial product.
Method
Participants
This sleep study was conducted at the Korea Institute of Standards and Science (KRISS) and people were recruited from the Chungcheongnam-do region.
The Insomnia Severity Index (ISI) was used to determine the people with insomnia (Bastien et al., 2001). This index is a brief self-report questionnaire that measures the patient’s perception of insomnia severity. It has been translated and validated in various languages, demonstrating adequate psychometric properties for several age groups (Cho et al., 2014). The Korean version of the Insomnia Severity Index (ISI-K) was verified (Cho et al., 2014) and applied in this study. Points of 0–7 are interpreted as no clinical insomnia, 8–14 as mild clinical insomnia, 15–21 as moderate insomnia, and 22–28 as severe insomnia. In this experiment, subjects with 8 or more points through self-examination were selected.
Participants who met the criteria were tested after obtaining consent and explaining the purpose, benefits, and risks of the study through the approval of the Ethics Review Committee of the KRISS (KRISS-IRB-2020-5). The day before the experiment, participants were prohibited from drinking alcohol, caffeine, or smoking which are known substances that may influence sleep (Zunhammer et al., 2014).
Experimental Materials (Stimuli)
In this study, a pillow with speakers that generates sound for brain entrainment was designed and fabricated. The pillow material is made of foam and the height of the pillow can be divided into three layers (Fig. 1). Each participant was allowed to select their preferred height. The height chosen was recorded and used for the two nights. For audio brain entrainment, the speaker generated synchronized pulsed tones at an alpha or theta frequency which was provided by GEOMC (Tracy et al., 2007). The sound is a combination of alpha and theta frequencies (4–12 Hz range), beginning with alpha (9–12 Hz), then theta (4–8 Hz), and then ending in alpha. These pulse tones aim to induce changes in the brain waves to functionally activate the brain. The sound comprises the pulse tones and natural sounds (i.e., wave sound, stream sound, rain sound) to optimize sleep induction, and is designed to be applied for 30 min before sleep to induce deep sleep and to improve the sleep quality. The volume of the tone pulses and the background nature sounds were adjusted according to the participant preference while the rate of pulsing tones was fixed.
Fig. 1 A Picture of the pillow with speakers embedded. B The sectioning of the pillow. Layers were either added or removed according to the us
Experimental Procedures
Experiments were conducted in the lab sleep chamber at KRISS where the temperature, humidity, and illumination were all calibrated and fixed. The laboratory temperature was determined to be at 21.57 ± 0.86℃ and humidity at 55.31 ± 5.89% R.H. To provide a constant sleep environment for all subjects, sleep time for both days was approximately 7 h with environmental factors such as attire and bed location kept constant.
For this single group cross experimental study, the experiments were conducted two times with each participant (application of sound and the control). Each participant was measured on two different nights and was subjected to a random allocation of treatment (Sound application or no sound application) to see if the pulse tone improved the sleep. All sensors were to be connected 30 min before the participant’s sleep time. Once the sensors are attached, the participant will lie down on the pillow and be subjected to either the sound application or no sound. The sound will be applied for 30 min and for the control, the participant will be on the pillow without activation of the sound.
Sleep Polysomnography
Sleep polysomnography is an examination to detect and record various biological signals, electroencephalogram (EEG), electrooculogram (EOG), electromyogram (EMG), airflow signals, and respiratory effort signals that appear during sleep (Ryu & Choi, 2019). This can be defined as a test that detects and records various biological signals such as oxygen saturation, electrocardiogram (ECG), body position, and snoring to determine sleep states or diagnosis of sleep disorders.
The HANDYEEG SystemPLUS Evolution (Micromed S.p.A., Italy) is a portable device that reads sleep conditions by recording biological signals such as EEG, EOG, EMG, and ECG and was used in this study (Ryu & Choi, 2019). The location for the EEG sensors attachment is Fp1, Fp2, C3, C4, O1, O2, and the EOG sensors are PNG1 and PNG2 where the channels are 1 cm up and down from the outer eye angle (Fig. 2A and B). The ECG sensor is attached to the ECG1 channel 5 cm below the Jugular notch and inside the fourth and fifth ribs (Fig. 2C) For the EMG measurement, the electrodes were attached to the skin over both the mentalis muscle of the lower lip (Fig. 2A).
Fig. 2 A Front position of the EEG, EOG & EMG sensors. B Back position of the EEG sensors. C Chest position for the ECG sensor
After attaching the various sensors, the experiment commences in the self-governing sleeping room. The data collected were processed and analyzed accordingly. The SystemPLUS Evaluation data collection software was used, and the data analysis was performed twice by two experts (manual reading every 30 s) with the REM brant Analysis Manager. The contents for sleep scoring data of the polysomnography test include light-out clock time, lights on clock time, total sleep time (TST), and total recording time. Data were further processed to get Total recording time (TRT), sleep latency (SL), wake after sleep onset (WASO), sleep efficiency (SE), duration of each sleep phase (Time in each stage), and the percentage of each sleep stage (Ryu & Choi, 2019).
Saliva Test
The amount of melatonin was compared and verified through a saliva test. The saliva test was performed 20 min after the MC Square system was applied or 10 min before sleeping, and immediately after waking up. The saliva samples were collected by chewing on a cotton swab for about 3 min, after which the subject placed the swab into a sterile plastic tube. The tubes were stored in the refrigerator at 4 °C and delivered to the SQLAB institution for analysis. Samples were serial diluted and melatonin was determined by the high sensitivity enzyme immunoassay kit ELISA (Salimetrics Cat No: 1-3402), with an inter-assay coefficient of variation at 3.2% and intra-assay coefficient of variation at 1.2%.
Subjective Evaluation (Survey)
A questionnaire was used for a subjective pillow evaluation, and the satisfaction level was determined according to the questions. The questionnaire comprised questions that were answered on 5-grade verbal ratings of overall satisfaction (very dissatisfied to very satisfied), level of comfortability on the neck (very uncomfortable to very comfortable), the height of the pillow (very low to very high), the width of the pillow (very narrow to very wide), length of the pillow (very short to long), cushion of the pillow (very soft to very hard), the temperature of the pillow (very cold to very hot), the humidity of the pillow (moist to dry), easiness of head-turning (not at all to very easy). The questionnaire was performed after the participant woke up during the first experiment. Results were then tabulated and converted to a percentage.
Data Analysis
The general characteristics of the subjects and ISI scores were expressed as mean and standard deviation. The Shapiro-Wilk test was performed to determine whether data is a normal distribution and results of the sleep polysomnography results were compared using the Mann-Whitney U test as some data did not have a normal distribution. All statistical analyses were performed using the SPSS ver. 20.0 (IBM Corp., Armonk, NY, USA) and a p-value of < 0.05 was considered statistically significant.
Results
Participants
15 People met the criterion initially but 5 of them had to be excluded as they were receiving medical treatment which could affect the study. A final group of 10 people proceeded with the sleep experiment. Three men and seven women participated with an average height of 164.7 cm and an average weight of 60.8 kg (Table 1). The average age of the participants was 31 years old with an ISI (score) of 9.37 ± 2.43 which classified them to have a mild case of insomnia.
Table 1 General characteristics of subjects
Variable Group
Age (year) 31.30 ± 10.70
Height (cm) 164.70 ± 8.87
Weight (kg) 60.80 ± 9.88
ISI (score) 9.37 ± 2.43
ISI insomnia severity index
Sleep Polysomnography
Sleep polysomnography results for both application of sound and the control were analyzed and presented in Fig. 3. For this experiment, the following measurements were used: TST, SE, SL, and WASO. In addition, the total sleep time percentages of each stage (Stage 1, Stage 2, Stage 3, and REM stage were also broken down to observe any possible irregularities.Fig. 3 There were no significant differences between experimental conditions on the PSG variables but there is a significant difference for sleep latency (SL). p-Values are for overall differences between conditions and evaluated on degrees of freedom. Data represent mean+-S.E.M.
For the sleep quality results (Fig. 3), there was a significant decrease in SL from 24.65 min to 9.7 min when the participants were exposed to the sound. TST and SE showed a slight increase while WASO had a decrease when compared to the control. However, while these results showed improvement in sleep quality, there was no significant difference when compared to the controls and more samples will be needed. As brain entrainment uses similar frequencies to synchronize brain activity, we looked at the different sleep stage percentages to see if there are any changes to the user’s sleep pattern (Fig. 4). When the sound is applied, there was a decrease in stage 1 and stage 2 percentages from 71.63 to 68.92% and 16.63–14.92% respectively. However, there was an increase in stage 3 and REM sleep from 8.73 to 12.46% and 3.04–3.69% when compared to the control. However, as there were no significant differences for all stages in our experiments, more data needs to be collected to give us a better understanding of how the different sound frequencies may play a part in sleep.Fig. 4 The overall average of subjects’ sleep architecture from both nights. There were no significant differences between experimental conditions and the control for all stages of sleep. Data represent mean+- S.E.M
Saliva Test
We performed a melatonin test to observe what will happen to melatonin levels when the audio stimulus was only applied (Fig. 5). The results of the saliva test before sleep showed melatonin amounts of an average of 16.66 pg/ml (sound applied) and melatonin amounts averaged 17.03 pg/ml (control). After waking up, the average amount of melatonin was 17.87 pg/ml and if the sound was not applied, the average amount of melatonin was 19.80 pg/ml. Both results are not significantly different indicating that melatonin levels with and without the sound application were quite similar.Fig. 5 Amount of melatonin before sleep and after sleep of the 10 participants when MC Square system is applied and not applied. Data represent mean+-S.E.M
Survey
A subjective evaluation survey was conducted to evaluate the sleep pillow satisfaction with the sound system. The resulting values were summarized in Fig. 6. The overall satisfaction of the system was high with, 20% who are in the middle and 80% were satisfied. In terms of neck support, 55% find it comfortable while the rest is just acceptable. For the width (67% neutral), height (100% neutral), length (89% neutral), and thickness (88.9% neutral) of the pillow, most users found it to be acceptable which means that sleep was not affected by the pillow conditions. The temperature of the pillow is about 50% slightly warm and 50% no difference. About 78% of the users found the humidity of the pillow to be reasonable while the other 22% either found it slightly wet or slightly dry. When the question “Is it easy to turn your head around?“ was asked, 20% found it neutral while 80% found it to be easy. This shows that the speakers do not affect the movement of the user during sleep. 70% of the participant will consider using the pillow again. The overall results of the customer survey showed that the design of the pillow is widely accepted by most participants. While the survey is encouraging, other ergonomic designs will be needed to improve the pillow.Fig. 6 A summary of the survey
Discussion
Sleep polysomnography popularly known as a ‘Sleep study’ has been used for decades to diagnose and evaluate the severity of sleep disorders (Shrivastava et al., 2014). While there is no standardization of the reporting process, most sleep reports are based on certain sleep characteristics that provide quantitative information regarding the patient’s sleep and its deviation from their normal pattern (Shrivastava et al., 2014).
The TST of all participants was determined to gauge how long they slept. The total sleep time refers to the total amount of time for all sleep stages (Shrivastava et al., 2014). A low total sleep time may indicate that the patient slept for an insufficient period due to non-medical/non-physiological reasons, certain medical or sleep disorders, or because of the effect of medications. For this experiment, the time measured was set when the moment stage 1 of the sleep cycle appears and ends just before the wake-up stage. While there is a slight increase in TST when the sound is applied, it was not determined to be significantly different when compared to the control. The average TST was about 344.15 min or 5.7 h when used with the smart pillow, which is still below the recommended levels (7–9 h) by the National Sleep Foundation (Kim et al., 2019). In Korea, researchers found that about 16.7% of Koreans were sleeping for less than 5 h (Kim et al., 2019). This was supported by the ISI index as well.
Furthermore, the overall quality of sleep various parameters such as SE, SL, and WASO were also analyzed. While the duration of sleep is important, research showed that the quality of sleep is more important (Shrivastava et al., 2014). SE is an important parameter for sleep studies as it gives an overall sense of how well the patient slept. It looks at the total time in bed which was spent in sleep. It is calculated as a percentage of Total Sleep Time (TST)/Total Recording Time (TRT). SE mean was about 84% for all users before the brain entrainment was introduced and when the pulse tones were applied, efficiency rose to 88%. There is an approximately 4% increase in SE when the smart pillow was applied. Sleep latency is the duration time between when the light is turned off (lights out) to the time the patient falls asleep based on the reading of the machine (Shrivastava et al., 2014). The MC Square system showed an average sleep latency period of 9.70 min and when compared to the control sleep latency period of 24.65 min, showed an improvement in the participants falling asleep faster. Wake after sleep onset (WASO) refers to the period of wakefulness occurring after defined sleep onset. WASO results showed a reduced time when the MC system is applied. Participants’ WASO showed a slight decrease of 2 min when exposed to the system, indicating that the GEOMC system may improve the overall sleep quality by reducing the time to go to bed and go back to sleep faster went woken up. While the SE and WASO were not significant, sleep onset latency showed that there is a significant decrease. From these 4 characteristics, we can conclude that while the smart pillow may not increase the duration of sleep, brain entrainment may have a positive impact on sleep latency. This was also reported by Lazic et al. (Lazic & Ogilvie, 2007) that sleep onset latencies tended to be shorter in the tone condition when compared to music or the control.
Normal human sleep is entered through non-REM sleep, which is composed of three stages (I, II, III) ranging from the lightest to the deepest sleep, and REM sleep. Typically, non-REM and REM sleep alternate in cycles across sleep. Stage I sleep is the transition from wakefulness to deeper sleep and is the lightest stage of sleep. Stage II sleep constitutes up to 50% of total sleep time and is a true physiologic stage of sleep. Stages III and IV sleep are the deepest sleep and most restorative for body function. During the different sleep stages, different frequencies of brain waves are active (Carskadon & Dement, 2005). As brain entrainment uses similar frequencies to synchronize brain activity, we looked at the different sleep stage percentages to see if there are any changes to the user’s sleep pattern when the tones are applied. Initial results showed that there may be a positive effect on sleep as there is an increase in both deep sleep and REM sleep (Rama et al., 2005) which explains some of the profound consequences of insufficient sleep on thinking, emotions, and physical health. However, as there were no significant differences for all stages in our experiments, more data needs to be collected to give us a better understanding.
Melatonin is a reliable indicator of circadian rhythm and has been frequently used in sleep studies (Kazemi et al., 2018). Previous studies have shown a decrease in daytime melatonin levels when audiovisual stimulation was used (Tang et al., 2016). As most brain entrainment experiments with the audio stimulus are combined with visual stimulation, we wanted to determine if audio stimulus may have a role in melatonin levels. Melatonin concentration before sleeping and immediately after waking up was measured to determine if there was any effect on circadian rhythm by the pulsing tones. From the experiment, we showed that light stimulus rather than sound stimulus may be the main initiator for melatonin production during brain entrainment for sleep.
While the survey indicates a positive experience when using the pillow, there are other possible factors to take into consideration for future prototypes to ensure a better sleep quality too (Son et al., 2020):The consideration of other materials to ensure better comfortability, cushioning, and neck support.
Depending on the environment, temperature and humidity need to be considered as well.
Allowance for movement as people may not be sleeping in the same position (supine, side, or prone, altered by having an arm under a pillow or folding the pillow) all the time.
Limitation
The initial results have demonstrated that audio brain entrainment alone can improve sleep quality, especially for sleep latency. This has also been seen in other studies (Tang et al., 2016; Tang et al., 2015) with different groups of people showing positive effects of audio brain entrainment. However, more experiments need to be done to elucidate exactly how the pulse tones affect the brain. While the results are promising, future studies will be required. Due to the covid pandemic, a small sample size was obtained, and additional precautions were taken to prevent the spread (the use of disinfectant and changing of bed sheets) may create a non-constant environment for the user. As results indicate an effect on the participants with mild insomnia, this research can be expanded to different age groups (i.e. young adults or elderly) or people with different insomnia severity who are either not suitable or reduce their consumption of hypnotic drugs. Future experiments can be performed with a sleep diary and longer trials if there are any long-term effects with larger sample size. The mechanism of audio-visual entrainment and how the neurological response relative to the outcome measures (i.e., sleep) remains to be fully demonstrated. Future studies may consider using neuroimaging to describe the concurrent brain activity changes in responding to the audio-visual entrainment as well.
Conclusion
In this paper, we have shown that a combination of pulse tone at alpha and theta frequencies may have a positive effect on people with mild insomnia. While the pillow design may need further improvements, initial results showed that an audio stimulation system decreases sleep onset latency. Time spent at sleep stages 1 and 2 slightly decreased while time in stage 3 and REM sleep increased. As those results were not significant, more experiments must be performed. Audio stimulation does not affect the melatonin levels. Considering that treatment options for insomnia are somewhat limited, pulse tones could serve as an initial intervention in a stepped-care approach for the management of insomnia in Korea. Therefore, audio brain entrainment pillows can be an easy and user-friendly application treatment for sleep disorders. This noninvasive stimulation technique is a promising candidate for wearable electronics and further neuroscience research.
Acknowledgements
This study was financially supported by GEOMC Co. Ltd (ip2020-0021, Title: mc square: a study on the evaluation of sleep quality improvement), and the authors gratefully acknowledge the Korea Research Institute of Standards and Science (KRISS), Korea for the support of this work.
Declarations
Conflict of interest
The authors declare that this study received funding from GEOMC Co. Ltd. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Aminoff, M. J., Boller, F., & Swaab, D. F. (2011). We spend about one-third of our life either sleeping or attempting to do so. Handbook of clinical neurology/edited by P.J. Vinken and G.W. Bruyn, 98, vii. 10.1016/B978-0-444-52006-7.00047-2
Bastien CH Vallières A Morin CM Validation of the insomnia severity index as an outcome measure for insomnia research Sleep Medicine 2001 2 4 297 307 10.1016/S1389-9457(00)00065-4 11438246
Bonnar D Bartel K Kakoschke N Lang C Sleep interventions designed to improve athletic performance and recovery: A systematic review of current approaches Sports Medicine 2018 48 3 683 703 10.1007/s40279-017-0832-x 29352373
Carskadon MA Dement WC Normal human sleep: An overview Principles and Practice of Sleep Medicine 2005 4 1 13 23 10.1016/j.mcna.2004.01.001
Chapple, S., & Ladaique, M. (2009). Society at a glance 2009: OECD social indicators. Organisation for Economic Co-operation and Development. 10.1787/soc_glance-2008-en.
Chaput JP Dutil C Sampasa-Kanyinga H Sleeping hours: What is the ideal number and how does age impact this? Nature and Science of Sleep 2018 10 421 430 10.2147/NSS.S163071 30568521
Chen CK Pei YC Chen NH Huang LT Chou SW Wu KP Ko PC Wong AM Wu CK Sedative music facilitates deep sleep in young adults The Journal of Alternative and Complementary Medicine 2014 20 4 312 317 10.1089/acm.2012.0050 23663079
Cho JHJ Olmstead R Choi H Carrillo C Seeman TE Irwin MR Associations of objective versus subjective social isolation with sleep disturbance, depression, and fatigue in community-dwelling older adults Aging & Mental Health 2019 23 9 1130 1138 10.1080/13607863.2018.1481928 30284454
Cho YW Song ML Morin CM Validation of a Korean version of the insomnia severity index Journal of Clinical Neurology 2014 10 3 210 215 10.3988/jcn.2014.10.3.210 25045373
Dickson GT Schubert E How does music aid sleep? Literature review Sleep Medicine 2019 63 142 150 10.1016/j.sleep.2019.05.016 31655374
Feng F Zhang Y Hou J Cai J Jiang Q Li X Zhao Q Li B Can music improve sleep quality in adults with primary insomnia? A systematic review and network meta-analysis International Journal of Nursing Studies 2018 77 189 196 10.1016/j.ijnurstu.2017.10.011 29100201
Frandsen R Baandrup L Kjellberg J Ibsen R Jennum P Increased all-cause mortality with psychotropic medication in Parkinson’s disease and controls: A national register-based study Parkinsonism & Related Disorders 2014 20 11 1124 1128 10.1016/j.parkreldis.2014.07.012 25164488
Hafner M Stepanek M Taylor J Troxel WM Van Stolk C Why sleep matters - the economic costs of insufficient sleep: A cross-country comparative analysis RAND Corporation 2016 10.7249/RR1791
Huang TL Charyton C A comprehensive review of the psychological effects of brainwave entrainment Alternative Therapies in Health and Medicine 2008 14 5 38 50 18780583
Im E Kim GS Relationship between sleep duration and Framingham cardiovascular risk score and prevalence of cardiovascular disease in Koreans Medicine 2017 96 37 e7744 e7744 10.1097/MD.0000000000007744 28906359
Itani O Jike M Watanabe N Kaneita Y Short sleep duration and health outcomes: A systematic review, meta-analysis, and meta-regression Sleep Medicine 2017 32 246 256 10.1016/j.sleep.2016.08.006 27743803
Jespersen KV Koenig J Jennum P Vuust P Music for insomnia in adults Cochrane Database of Systematic Reviews 2015 10.1002/14651858.CD010459.pub2 26270746
Jike M Itani O Watanabe N Buysse DJ Kaneita Y Long sleep duration and health outcomes: A systematic review, meta-analysis and meta-regression Sleep Medicine Reviews 2018 39 25 36 10.1016/j.smrv.2017.06.011 28890167
Kazemi R Motamedzade M Golmohammadi R Mokarami H Hemmatjo R Heidarimoghadam R Field study of effects of night shifts on cognitive performance, salivary melatonin, and sleep Safety and Health at Work 2018 9 2 203 209 10.1016/j.shaw.2017.07.007 29928535
Kim HJ Oh SY Joo JH Choi DW Park EC The relationship between sleep duration and perceived stress: Findings from the 2017 community health survey in Korea International Journal of Environmental Research and Public Health 2019 16 17 3208 10.3390/ijerph15122648 31484289
Kim SJ Pharmacotherapy of insomnia Journal of the Korean Medical Association 2009 52 7 719 726 10.5124/jkma.2009.52.7.719
Kim SJ Han KT Jang SY Yoo KB Kim SJ The association between migraine and types of sleep disorder International Journal of Environmental Research and Public Health 2018 15 12 2648 10.3390/ijerph15122648 30486273
Kripke DF Langer RD Kline LE Hypnotics’ association with mortality or cancer: A matched cohort study BMJ Open 2012 2 1 e000850 10.1136/bmjopen-2012-000850 22371848
Kwon CY Suh HW Choi EJ Chung SY Kim JW Complementary and alternative medicine in clinical practice guideline for insomnia Korean Society of Oriental Neuropsychiatry 2016 27 4 235 248 10.7231/JON.2016.27.4.235
Lai HL Good M Music improves sleep quality in older adults Journal of Advanced Nursing 2006 53 1 134 144 10.1111/j.1365-2648.2004.03281.x 16422710
Lazic SE Ogilvie RD Lack of efficacy of music to improve sleep: A polysomnographic and quantitative EEG analysis International Journal of Psychophysiology 2007 63 3 232 239 10.1016/j.ijpsycho.2006.10.004 17123654
Leggett A Assari S Burgard S Zivin K The effect of sleep disturbance on the association between chronic medical conditions and depressive symptoms over time Longitudinal and Life Course Studies: International Journal 2017 8 2 138 151 10.14301/llcs.v8i2.433 28966664
Lie, J. D., Tu, K. N., Shen, D. D., & Wong, B. M. (2015). Pharmacological treatment of insomnia. P & T: a peer-reviewed journal for formulary management, 40(11), 759–771. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4634348/
Pagel JF Parnes BL Medications for the treatment of sleep disorders: An overview Primary Care Companion to the Journal of Clinical Psychiatry 2001 3 3 118 125 10.4088/pcc.v03n0303 15014609
Perry GS Patil SP Presley-Cantrell LR Raising awareness of sleep as a healthy behavior Preventing Chronic Disease 2013 10 E133 10.5888/pcd10.130081 23928458
Rama, A. N., Cho, S. C., & Kushida, C. A. (2005). Normal human sleep. In Sleep: A comprehensive handbook (pp. 1–9). 10.1002/0471751723.ch1
Ryu G Choi JH Interpretation of polysomnography Journal of Clinical Otolaryngology Head and Neck Surgery 2019 30 1 12 19 10.35420/jcohns.2019.30.1.12
Shrivastava D Jung S Saadat M Sirohi R Crewson K How to interpret the results of a sleep study Journal of Community Hospital Internal Medicine Perspectives 2014 4 5 24983 24983 10.3402/jchimp.v4.24983 25432643
Son J Jung S Song H Kim J Bang S Bahn S A survey of Koreans on sleep habits and sleeping symptoms relating to pillow comfort and support International Journal of Environmental Research and Public Health 2020 17 1 302 10.3390/ijerph17010302 31906363
Tang HYJ Riegel B McCurry SM Vitiello MV Open-loop audio-visual stimulation (AVS): A useful tool for management of insomnia? Applied Psychophysiology and Biofeedback 2016 41 1 39 46 10.1007/s10484-015-9308-7 26294268
Tang HYJ Vitiello MV Perlis M Riegel B Open-loop neurofeedback audiovisual stimulation: A pilot study of its potential for sleep induction in older adults Applied Psychophysiology and Biofeedback 2015 40 3 183 188 10.1007/s10484-015-9285-x 25931250
Thompson, M., & Thompson, L. (2003). The neurofeedback book: An introduction to basic concepts in applied pychophysiology.
Tracy JI Ahmed N Khan W Sperling MR A test of the efficacy of the MC Square device for improving verbal memory, learning and attention International Journal of Learning Technology 2007 3 2 183 202 10.1504/IJLT.2007.014844
Vgontzas M Kales M Sleep and its disorders Annual Review of Medicine 1999 50 1 387 400 10.1146/annurev.med.50.1.387 10073285
Xiao H Zhang Y Kong D Li S Yang N The effects of social support on sleep quality of medical staff treating patients with coronavirus disease 2019 (COVID-19) in january and february 2020 in China Medical Science Monitor: International Medical Journal of Experimental and Clinical Research 2020 26 e923549 e923549 10.12659/MSM.923549 32132521
Xiao H Zhang Y Kong D Li S Yang N Social capital and sleep quality in individuals who self-isolated for 14 days during the coronavirus disease 2019 (COVID-19) outbreak in january 2020 in China Medical Science Monitor: International Medical Journal of Experimental and Clinical Research 2020 26 e923921 e923921 10.12659/MSM.923921 32194290
Zunhammer M Eichhammer P Busch V Sleep quality during exam stress: The role of alcohol, caffeine and nicotine PloS One 2014 9 10 e109490 10.1371/journal.pone.0109490 25279939
| 36445544 | PMC9707213 | NO-CC CODE | 2022-12-01 23:20:21 | no | Appl Psychophysiol Biofeedback. 2022 Nov 29;:1-10 | utf-8 | Appl Psychophysiol Biofeedback | 2,022 | 10.1007/s10484-022-09570-2 | oa_other |
==== Front
Appl Psychophysiol Biofeedback
Appl Psychophysiol Biofeedback
Applied Psychophysiology and Biofeedback
1090-0586
1573-3270
Springer US New York
9570
10.1007/s10484-022-09570-2
Article
Effects of Audio Brain Entrainment on Korean People with Mild Insomnia
http://orcid.org/0000-0001-5473-9597
Ho Chee Meng Benjamin [email protected]
1
Jeong Hana 12
Lim Young-Hyun 3
Park Se Jin [email protected]
1
1 AI Research Group, Sewon Intelligence, Sejong-daero, Jung-gu, Seoul, Korea
2 grid.411143.2 0000 0000 8674 9741 Department of Physical Therapy, Konyang University, Gwanjeo-Dong, Seo-Gu, Daejeon, Korea
3 GEOMC Co. Ltd, Siheung-dong, Sujeong-gu, Gyeonggi-do Korea
29 11 2022
110
4 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Sleep health has become an important healthy lifestyle. Research has shown that almost one-fifth of the Korean adult population does not have sufficient sleep. The lack of sleep is associated with significant medical, psychological, social, and economic issues. People are not only yearning for sufficient sleep but the quality of sleep as well. Usually, the obvious choice will be the use of pharmaceuticals however, these often have various side effects, and the lasting use of these medications could become a concern. Therefore, new non-drug alternatives are sought after. Audio brain entrainment is a procedure that modules neural activities by synchronizing brainwave frequency with pulse tones. By producing frequency tones for the deep sleep stage, it promotes a good night’s sleep. In this paper, we developed a pillow integrated with the audio speakers that produce alpha and theta beats that should help improve sleep. Sleep polysomnography was performed on 10 people to compare the effects of the audio stimulus. Initial results showed a positive effect on sleep onset latency, indicating that sleep induction happened. This noninvasive stimulation technique can be a promising candidate for wearable bioelectronics medicine and further neuroscience research.
Keywords
Sleep
Brain entrainment
Insomnia
Sleep treatment
GEOMC Co. Ltdip2020-0021
==== Body
pmcIntroduction
With humans sleeping for almost one-third of their life, sleep health has become an essential component of a healthy lifestyle (Aminoff et al., 2011; Perry et al., 2013). The timing, duration, and quality of sleep are important indicators of health, and this is further emphasized during the Covid-19 pandemic (Chaput et al., 2018; Xiao et al., 2020a, b). While a good night’s sleep may provide mental and physical recovery (Bonnar et al., 2018; Kim, 2009), sleep disorders are associated with significant medical, psychological, and social disturbances (Cho et al., 2019; Leggett et al., 2017; Vgontzas & Kales, 1999). Different studies had associated people with sleeping disorders with a higher risk of mortality, diabetes mellitus, hypertension, cardiovascular disease, coronary heart disease, and obesity (Im & Kim, 2017; Itani et al., 2017; Jike et al., 2018). For these reasons, good sleep is crucial to a healthy life.
In the United States, more than a third of American adults do not get enough sleep (Perry et al., 2013). However, insufficient sleep is not exclusively a US problem but affects most developed countries (Hafner et al., 2016). According to a report by the Organization for Economic Cooperation and Development countries (OECD), the average sleep time in South Korea was lower than the OECD average (South Korea: 462 min per day; OECD average: 502 min per day) (Chapple & Ladaique, 2009; Kim et al., 2018). Treatment for sleep disorders usually involves pharmacotherapeutics or hypnotics however there are some concerns about the side effects or lasting use of these medications (Lie et al., 2015; Pagel & Parnes, 2001). Possible side effects include daytime drowsiness, cognitive impairments, reduced motor coordination risk of abuse, and long-term dependence (Frandsen et al., 2014; Kripke et al., 2012). Therefore, the demand for low-cost, non-pharmaceutical alternative aids has grown.
In Korea, non-pharmaceutical such as herbal medicine, meditation, acupuncture, exercise, etc. are used (Kwon et al., 2016). One alternative method is the use of audio stimulation (Jespersen et al., 2015; Lai & Good, 2006). Trials have shown that music can be a self-help intervention to improve sleep (Chen et al., 2014; Cho et al., 2019; Feng et al., 2018; Jespersen et al., 2015; Lai & Good, 2006) and a Google search on ‘music’ and ‘sleep’ reveals a huge market promoting its sleep-inducing properties (Jespersen et al., 2015). Besides music, audio pulses can be applied to elicit the brain’s frequency following response, encouraging the brainwaves to align to the frequency of a given beat (Huang & Charyton, 2008). Brainwave entrainment is a procedure that modules neural activities by synchronizing brainwave frequency with that of the stimuli (Dickson & Schubert, 2019). There are four common brainwave bandwidths (delta, theta, alpha, and beta) and the related mental activities have been studied intensely (Thompson & Thompson, 2003). For example, delta frequency (0.5–4 Hz) is dominant during the deep sleep state while theta frequencies (5–8 Hz) are observed in the drowsy or relaxed state. In addition, alpha frequencies (9–13 Hz) are present in various mental activity such as mediation, creativity, calmness, etc. and beta frequencies (13–30 Hz) occurs before reactive psychomotor actions such as intellectual activity and emotional and anxious states. (Tang et al., 2016).
In this paper, we developed a pillow integrated with the audio speakers that produced specifically binaural beats from 2 different frequencies. We hypothesized that the audio stimulation will facilitate EEG slowing (via feedback, entrainment, conditioning, etc.) which may improve sleep induction and maintenance. This study will examine the feasibility of a 30 min administered pulse tone stimulation for sleep promotion in Korean adults with insomnia in a controlled environment. Sleep polysomnography was performed on 10 Korean individuals and sleep data were collected over 2 different nights for everyone. Each night, participants were either exposed to the audio stimulation or not (control). As sound is usually combined with light to create an audio-visual stimulation, a saliva test for melatonin is performed to see if sound alone may affect the circadian rhythm cycle. A survey was also conducted to see the viability of the pillow as a commercial product.
Method
Participants
This sleep study was conducted at the Korea Institute of Standards and Science (KRISS) and people were recruited from the Chungcheongnam-do region.
The Insomnia Severity Index (ISI) was used to determine the people with insomnia (Bastien et al., 2001). This index is a brief self-report questionnaire that measures the patient’s perception of insomnia severity. It has been translated and validated in various languages, demonstrating adequate psychometric properties for several age groups (Cho et al., 2014). The Korean version of the Insomnia Severity Index (ISI-K) was verified (Cho et al., 2014) and applied in this study. Points of 0–7 are interpreted as no clinical insomnia, 8–14 as mild clinical insomnia, 15–21 as moderate insomnia, and 22–28 as severe insomnia. In this experiment, subjects with 8 or more points through self-examination were selected.
Participants who met the criteria were tested after obtaining consent and explaining the purpose, benefits, and risks of the study through the approval of the Ethics Review Committee of the KRISS (KRISS-IRB-2020-5). The day before the experiment, participants were prohibited from drinking alcohol, caffeine, or smoking which are known substances that may influence sleep (Zunhammer et al., 2014).
Experimental Materials (Stimuli)
In this study, a pillow with speakers that generates sound for brain entrainment was designed and fabricated. The pillow material is made of foam and the height of the pillow can be divided into three layers (Fig. 1). Each participant was allowed to select their preferred height. The height chosen was recorded and used for the two nights. For audio brain entrainment, the speaker generated synchronized pulsed tones at an alpha or theta frequency which was provided by GEOMC (Tracy et al., 2007). The sound is a combination of alpha and theta frequencies (4–12 Hz range), beginning with alpha (9–12 Hz), then theta (4–8 Hz), and then ending in alpha. These pulse tones aim to induce changes in the brain waves to functionally activate the brain. The sound comprises the pulse tones and natural sounds (i.e., wave sound, stream sound, rain sound) to optimize sleep induction, and is designed to be applied for 30 min before sleep to induce deep sleep and to improve the sleep quality. The volume of the tone pulses and the background nature sounds were adjusted according to the participant preference while the rate of pulsing tones was fixed.
Fig. 1 A Picture of the pillow with speakers embedded. B The sectioning of the pillow. Layers were either added or removed according to the us
Experimental Procedures
Experiments were conducted in the lab sleep chamber at KRISS where the temperature, humidity, and illumination were all calibrated and fixed. The laboratory temperature was determined to be at 21.57 ± 0.86℃ and humidity at 55.31 ± 5.89% R.H. To provide a constant sleep environment for all subjects, sleep time for both days was approximately 7 h with environmental factors such as attire and bed location kept constant.
For this single group cross experimental study, the experiments were conducted two times with each participant (application of sound and the control). Each participant was measured on two different nights and was subjected to a random allocation of treatment (Sound application or no sound application) to see if the pulse tone improved the sleep. All sensors were to be connected 30 min before the participant’s sleep time. Once the sensors are attached, the participant will lie down on the pillow and be subjected to either the sound application or no sound. The sound will be applied for 30 min and for the control, the participant will be on the pillow without activation of the sound.
Sleep Polysomnography
Sleep polysomnography is an examination to detect and record various biological signals, electroencephalogram (EEG), electrooculogram (EOG), electromyogram (EMG), airflow signals, and respiratory effort signals that appear during sleep (Ryu & Choi, 2019). This can be defined as a test that detects and records various biological signals such as oxygen saturation, electrocardiogram (ECG), body position, and snoring to determine sleep states or diagnosis of sleep disorders.
The HANDYEEG SystemPLUS Evolution (Micromed S.p.A., Italy) is a portable device that reads sleep conditions by recording biological signals such as EEG, EOG, EMG, and ECG and was used in this study (Ryu & Choi, 2019). The location for the EEG sensors attachment is Fp1, Fp2, C3, C4, O1, O2, and the EOG sensors are PNG1 and PNG2 where the channels are 1 cm up and down from the outer eye angle (Fig. 2A and B). The ECG sensor is attached to the ECG1 channel 5 cm below the Jugular notch and inside the fourth and fifth ribs (Fig. 2C) For the EMG measurement, the electrodes were attached to the skin over both the mentalis muscle of the lower lip (Fig. 2A).
Fig. 2 A Front position of the EEG, EOG & EMG sensors. B Back position of the EEG sensors. C Chest position for the ECG sensor
After attaching the various sensors, the experiment commences in the self-governing sleeping room. The data collected were processed and analyzed accordingly. The SystemPLUS Evaluation data collection software was used, and the data analysis was performed twice by two experts (manual reading every 30 s) with the REM brant Analysis Manager. The contents for sleep scoring data of the polysomnography test include light-out clock time, lights on clock time, total sleep time (TST), and total recording time. Data were further processed to get Total recording time (TRT), sleep latency (SL), wake after sleep onset (WASO), sleep efficiency (SE), duration of each sleep phase (Time in each stage), and the percentage of each sleep stage (Ryu & Choi, 2019).
Saliva Test
The amount of melatonin was compared and verified through a saliva test. The saliva test was performed 20 min after the MC Square system was applied or 10 min before sleeping, and immediately after waking up. The saliva samples were collected by chewing on a cotton swab for about 3 min, after which the subject placed the swab into a sterile plastic tube. The tubes were stored in the refrigerator at 4 °C and delivered to the SQLAB institution for analysis. Samples were serial diluted and melatonin was determined by the high sensitivity enzyme immunoassay kit ELISA (Salimetrics Cat No: 1-3402), with an inter-assay coefficient of variation at 3.2% and intra-assay coefficient of variation at 1.2%.
Subjective Evaluation (Survey)
A questionnaire was used for a subjective pillow evaluation, and the satisfaction level was determined according to the questions. The questionnaire comprised questions that were answered on 5-grade verbal ratings of overall satisfaction (very dissatisfied to very satisfied), level of comfortability on the neck (very uncomfortable to very comfortable), the height of the pillow (very low to very high), the width of the pillow (very narrow to very wide), length of the pillow (very short to long), cushion of the pillow (very soft to very hard), the temperature of the pillow (very cold to very hot), the humidity of the pillow (moist to dry), easiness of head-turning (not at all to very easy). The questionnaire was performed after the participant woke up during the first experiment. Results were then tabulated and converted to a percentage.
Data Analysis
The general characteristics of the subjects and ISI scores were expressed as mean and standard deviation. The Shapiro-Wilk test was performed to determine whether data is a normal distribution and results of the sleep polysomnography results were compared using the Mann-Whitney U test as some data did not have a normal distribution. All statistical analyses were performed using the SPSS ver. 20.0 (IBM Corp., Armonk, NY, USA) and a p-value of < 0.05 was considered statistically significant.
Results
Participants
15 People met the criterion initially but 5 of them had to be excluded as they were receiving medical treatment which could affect the study. A final group of 10 people proceeded with the sleep experiment. Three men and seven women participated with an average height of 164.7 cm and an average weight of 60.8 kg (Table 1). The average age of the participants was 31 years old with an ISI (score) of 9.37 ± 2.43 which classified them to have a mild case of insomnia.
Table 1 General characteristics of subjects
Variable Group
Age (year) 31.30 ± 10.70
Height (cm) 164.70 ± 8.87
Weight (kg) 60.80 ± 9.88
ISI (score) 9.37 ± 2.43
ISI insomnia severity index
Sleep Polysomnography
Sleep polysomnography results for both application of sound and the control were analyzed and presented in Fig. 3. For this experiment, the following measurements were used: TST, SE, SL, and WASO. In addition, the total sleep time percentages of each stage (Stage 1, Stage 2, Stage 3, and REM stage were also broken down to observe any possible irregularities.Fig. 3 There were no significant differences between experimental conditions on the PSG variables but there is a significant difference for sleep latency (SL). p-Values are for overall differences between conditions and evaluated on degrees of freedom. Data represent mean+-S.E.M.
For the sleep quality results (Fig. 3), there was a significant decrease in SL from 24.65 min to 9.7 min when the participants were exposed to the sound. TST and SE showed a slight increase while WASO had a decrease when compared to the control. However, while these results showed improvement in sleep quality, there was no significant difference when compared to the controls and more samples will be needed. As brain entrainment uses similar frequencies to synchronize brain activity, we looked at the different sleep stage percentages to see if there are any changes to the user’s sleep pattern (Fig. 4). When the sound is applied, there was a decrease in stage 1 and stage 2 percentages from 71.63 to 68.92% and 16.63–14.92% respectively. However, there was an increase in stage 3 and REM sleep from 8.73 to 12.46% and 3.04–3.69% when compared to the control. However, as there were no significant differences for all stages in our experiments, more data needs to be collected to give us a better understanding of how the different sound frequencies may play a part in sleep.Fig. 4 The overall average of subjects’ sleep architecture from both nights. There were no significant differences between experimental conditions and the control for all stages of sleep. Data represent mean+- S.E.M
Saliva Test
We performed a melatonin test to observe what will happen to melatonin levels when the audio stimulus was only applied (Fig. 5). The results of the saliva test before sleep showed melatonin amounts of an average of 16.66 pg/ml (sound applied) and melatonin amounts averaged 17.03 pg/ml (control). After waking up, the average amount of melatonin was 17.87 pg/ml and if the sound was not applied, the average amount of melatonin was 19.80 pg/ml. Both results are not significantly different indicating that melatonin levels with and without the sound application were quite similar.Fig. 5 Amount of melatonin before sleep and after sleep of the 10 participants when MC Square system is applied and not applied. Data represent mean+-S.E.M
Survey
A subjective evaluation survey was conducted to evaluate the sleep pillow satisfaction with the sound system. The resulting values were summarized in Fig. 6. The overall satisfaction of the system was high with, 20% who are in the middle and 80% were satisfied. In terms of neck support, 55% find it comfortable while the rest is just acceptable. For the width (67% neutral), height (100% neutral), length (89% neutral), and thickness (88.9% neutral) of the pillow, most users found it to be acceptable which means that sleep was not affected by the pillow conditions. The temperature of the pillow is about 50% slightly warm and 50% no difference. About 78% of the users found the humidity of the pillow to be reasonable while the other 22% either found it slightly wet or slightly dry. When the question “Is it easy to turn your head around?“ was asked, 20% found it neutral while 80% found it to be easy. This shows that the speakers do not affect the movement of the user during sleep. 70% of the participant will consider using the pillow again. The overall results of the customer survey showed that the design of the pillow is widely accepted by most participants. While the survey is encouraging, other ergonomic designs will be needed to improve the pillow.Fig. 6 A summary of the survey
Discussion
Sleep polysomnography popularly known as a ‘Sleep study’ has been used for decades to diagnose and evaluate the severity of sleep disorders (Shrivastava et al., 2014). While there is no standardization of the reporting process, most sleep reports are based on certain sleep characteristics that provide quantitative information regarding the patient’s sleep and its deviation from their normal pattern (Shrivastava et al., 2014).
The TST of all participants was determined to gauge how long they slept. The total sleep time refers to the total amount of time for all sleep stages (Shrivastava et al., 2014). A low total sleep time may indicate that the patient slept for an insufficient period due to non-medical/non-physiological reasons, certain medical or sleep disorders, or because of the effect of medications. For this experiment, the time measured was set when the moment stage 1 of the sleep cycle appears and ends just before the wake-up stage. While there is a slight increase in TST when the sound is applied, it was not determined to be significantly different when compared to the control. The average TST was about 344.15 min or 5.7 h when used with the smart pillow, which is still below the recommended levels (7–9 h) by the National Sleep Foundation (Kim et al., 2019). In Korea, researchers found that about 16.7% of Koreans were sleeping for less than 5 h (Kim et al., 2019). This was supported by the ISI index as well.
Furthermore, the overall quality of sleep various parameters such as SE, SL, and WASO were also analyzed. While the duration of sleep is important, research showed that the quality of sleep is more important (Shrivastava et al., 2014). SE is an important parameter for sleep studies as it gives an overall sense of how well the patient slept. It looks at the total time in bed which was spent in sleep. It is calculated as a percentage of Total Sleep Time (TST)/Total Recording Time (TRT). SE mean was about 84% for all users before the brain entrainment was introduced and when the pulse tones were applied, efficiency rose to 88%. There is an approximately 4% increase in SE when the smart pillow was applied. Sleep latency is the duration time between when the light is turned off (lights out) to the time the patient falls asleep based on the reading of the machine (Shrivastava et al., 2014). The MC Square system showed an average sleep latency period of 9.70 min and when compared to the control sleep latency period of 24.65 min, showed an improvement in the participants falling asleep faster. Wake after sleep onset (WASO) refers to the period of wakefulness occurring after defined sleep onset. WASO results showed a reduced time when the MC system is applied. Participants’ WASO showed a slight decrease of 2 min when exposed to the system, indicating that the GEOMC system may improve the overall sleep quality by reducing the time to go to bed and go back to sleep faster went woken up. While the SE and WASO were not significant, sleep onset latency showed that there is a significant decrease. From these 4 characteristics, we can conclude that while the smart pillow may not increase the duration of sleep, brain entrainment may have a positive impact on sleep latency. This was also reported by Lazic et al. (Lazic & Ogilvie, 2007) that sleep onset latencies tended to be shorter in the tone condition when compared to music or the control.
Normal human sleep is entered through non-REM sleep, which is composed of three stages (I, II, III) ranging from the lightest to the deepest sleep, and REM sleep. Typically, non-REM and REM sleep alternate in cycles across sleep. Stage I sleep is the transition from wakefulness to deeper sleep and is the lightest stage of sleep. Stage II sleep constitutes up to 50% of total sleep time and is a true physiologic stage of sleep. Stages III and IV sleep are the deepest sleep and most restorative for body function. During the different sleep stages, different frequencies of brain waves are active (Carskadon & Dement, 2005). As brain entrainment uses similar frequencies to synchronize brain activity, we looked at the different sleep stage percentages to see if there are any changes to the user’s sleep pattern when the tones are applied. Initial results showed that there may be a positive effect on sleep as there is an increase in both deep sleep and REM sleep (Rama et al., 2005) which explains some of the profound consequences of insufficient sleep on thinking, emotions, and physical health. However, as there were no significant differences for all stages in our experiments, more data needs to be collected to give us a better understanding.
Melatonin is a reliable indicator of circadian rhythm and has been frequently used in sleep studies (Kazemi et al., 2018). Previous studies have shown a decrease in daytime melatonin levels when audiovisual stimulation was used (Tang et al., 2016). As most brain entrainment experiments with the audio stimulus are combined with visual stimulation, we wanted to determine if audio stimulus may have a role in melatonin levels. Melatonin concentration before sleeping and immediately after waking up was measured to determine if there was any effect on circadian rhythm by the pulsing tones. From the experiment, we showed that light stimulus rather than sound stimulus may be the main initiator for melatonin production during brain entrainment for sleep.
While the survey indicates a positive experience when using the pillow, there are other possible factors to take into consideration for future prototypes to ensure a better sleep quality too (Son et al., 2020):The consideration of other materials to ensure better comfortability, cushioning, and neck support.
Depending on the environment, temperature and humidity need to be considered as well.
Allowance for movement as people may not be sleeping in the same position (supine, side, or prone, altered by having an arm under a pillow or folding the pillow) all the time.
Limitation
The initial results have demonstrated that audio brain entrainment alone can improve sleep quality, especially for sleep latency. This has also been seen in other studies (Tang et al., 2016; Tang et al., 2015) with different groups of people showing positive effects of audio brain entrainment. However, more experiments need to be done to elucidate exactly how the pulse tones affect the brain. While the results are promising, future studies will be required. Due to the covid pandemic, a small sample size was obtained, and additional precautions were taken to prevent the spread (the use of disinfectant and changing of bed sheets) may create a non-constant environment for the user. As results indicate an effect on the participants with mild insomnia, this research can be expanded to different age groups (i.e. young adults or elderly) or people with different insomnia severity who are either not suitable or reduce their consumption of hypnotic drugs. Future experiments can be performed with a sleep diary and longer trials if there are any long-term effects with larger sample size. The mechanism of audio-visual entrainment and how the neurological response relative to the outcome measures (i.e., sleep) remains to be fully demonstrated. Future studies may consider using neuroimaging to describe the concurrent brain activity changes in responding to the audio-visual entrainment as well.
Conclusion
In this paper, we have shown that a combination of pulse tone at alpha and theta frequencies may have a positive effect on people with mild insomnia. While the pillow design may need further improvements, initial results showed that an audio stimulation system decreases sleep onset latency. Time spent at sleep stages 1 and 2 slightly decreased while time in stage 3 and REM sleep increased. As those results were not significant, more experiments must be performed. Audio stimulation does not affect the melatonin levels. Considering that treatment options for insomnia are somewhat limited, pulse tones could serve as an initial intervention in a stepped-care approach for the management of insomnia in Korea. Therefore, audio brain entrainment pillows can be an easy and user-friendly application treatment for sleep disorders. This noninvasive stimulation technique is a promising candidate for wearable electronics and further neuroscience research.
Acknowledgements
This study was financially supported by GEOMC Co. Ltd (ip2020-0021, Title: mc square: a study on the evaluation of sleep quality improvement), and the authors gratefully acknowledge the Korea Research Institute of Standards and Science (KRISS), Korea for the support of this work.
Declarations
Conflict of interest
The authors declare that this study received funding from GEOMC Co. Ltd. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Aminoff, M. J., Boller, F., & Swaab, D. F. (2011). We spend about one-third of our life either sleeping or attempting to do so. Handbook of clinical neurology/edited by P.J. Vinken and G.W. Bruyn, 98, vii. 10.1016/B978-0-444-52006-7.00047-2
Bastien CH Vallières A Morin CM Validation of the insomnia severity index as an outcome measure for insomnia research Sleep Medicine 2001 2 4 297 307 10.1016/S1389-9457(00)00065-4 11438246
Bonnar D Bartel K Kakoschke N Lang C Sleep interventions designed to improve athletic performance and recovery: A systematic review of current approaches Sports Medicine 2018 48 3 683 703 10.1007/s40279-017-0832-x 29352373
Carskadon MA Dement WC Normal human sleep: An overview Principles and Practice of Sleep Medicine 2005 4 1 13 23 10.1016/j.mcna.2004.01.001
Chapple, S., & Ladaique, M. (2009). Society at a glance 2009: OECD social indicators. Organisation for Economic Co-operation and Development. 10.1787/soc_glance-2008-en.
Chaput JP Dutil C Sampasa-Kanyinga H Sleeping hours: What is the ideal number and how does age impact this? Nature and Science of Sleep 2018 10 421 430 10.2147/NSS.S163071 30568521
Chen CK Pei YC Chen NH Huang LT Chou SW Wu KP Ko PC Wong AM Wu CK Sedative music facilitates deep sleep in young adults The Journal of Alternative and Complementary Medicine 2014 20 4 312 317 10.1089/acm.2012.0050 23663079
Cho JHJ Olmstead R Choi H Carrillo C Seeman TE Irwin MR Associations of objective versus subjective social isolation with sleep disturbance, depression, and fatigue in community-dwelling older adults Aging & Mental Health 2019 23 9 1130 1138 10.1080/13607863.2018.1481928 30284454
Cho YW Song ML Morin CM Validation of a Korean version of the insomnia severity index Journal of Clinical Neurology 2014 10 3 210 215 10.3988/jcn.2014.10.3.210 25045373
Dickson GT Schubert E How does music aid sleep? Literature review Sleep Medicine 2019 63 142 150 10.1016/j.sleep.2019.05.016 31655374
Feng F Zhang Y Hou J Cai J Jiang Q Li X Zhao Q Li B Can music improve sleep quality in adults with primary insomnia? A systematic review and network meta-analysis International Journal of Nursing Studies 2018 77 189 196 10.1016/j.ijnurstu.2017.10.011 29100201
Frandsen R Baandrup L Kjellberg J Ibsen R Jennum P Increased all-cause mortality with psychotropic medication in Parkinson’s disease and controls: A national register-based study Parkinsonism & Related Disorders 2014 20 11 1124 1128 10.1016/j.parkreldis.2014.07.012 25164488
Hafner M Stepanek M Taylor J Troxel WM Van Stolk C Why sleep matters - the economic costs of insufficient sleep: A cross-country comparative analysis RAND Corporation 2016 10.7249/RR1791
Huang TL Charyton C A comprehensive review of the psychological effects of brainwave entrainment Alternative Therapies in Health and Medicine 2008 14 5 38 50 18780583
Im E Kim GS Relationship between sleep duration and Framingham cardiovascular risk score and prevalence of cardiovascular disease in Koreans Medicine 2017 96 37 e7744 e7744 10.1097/MD.0000000000007744 28906359
Itani O Jike M Watanabe N Kaneita Y Short sleep duration and health outcomes: A systematic review, meta-analysis, and meta-regression Sleep Medicine 2017 32 246 256 10.1016/j.sleep.2016.08.006 27743803
Jespersen KV Koenig J Jennum P Vuust P Music for insomnia in adults Cochrane Database of Systematic Reviews 2015 10.1002/14651858.CD010459.pub2 26270746
Jike M Itani O Watanabe N Buysse DJ Kaneita Y Long sleep duration and health outcomes: A systematic review, meta-analysis and meta-regression Sleep Medicine Reviews 2018 39 25 36 10.1016/j.smrv.2017.06.011 28890167
Kazemi R Motamedzade M Golmohammadi R Mokarami H Hemmatjo R Heidarimoghadam R Field study of effects of night shifts on cognitive performance, salivary melatonin, and sleep Safety and Health at Work 2018 9 2 203 209 10.1016/j.shaw.2017.07.007 29928535
Kim HJ Oh SY Joo JH Choi DW Park EC The relationship between sleep duration and perceived stress: Findings from the 2017 community health survey in Korea International Journal of Environmental Research and Public Health 2019 16 17 3208 10.3390/ijerph15122648 31484289
Kim SJ Pharmacotherapy of insomnia Journal of the Korean Medical Association 2009 52 7 719 726 10.5124/jkma.2009.52.7.719
Kim SJ Han KT Jang SY Yoo KB Kim SJ The association between migraine and types of sleep disorder International Journal of Environmental Research and Public Health 2018 15 12 2648 10.3390/ijerph15122648 30486273
Kripke DF Langer RD Kline LE Hypnotics’ association with mortality or cancer: A matched cohort study BMJ Open 2012 2 1 e000850 10.1136/bmjopen-2012-000850 22371848
Kwon CY Suh HW Choi EJ Chung SY Kim JW Complementary and alternative medicine in clinical practice guideline for insomnia Korean Society of Oriental Neuropsychiatry 2016 27 4 235 248 10.7231/JON.2016.27.4.235
Lai HL Good M Music improves sleep quality in older adults Journal of Advanced Nursing 2006 53 1 134 144 10.1111/j.1365-2648.2004.03281.x 16422710
Lazic SE Ogilvie RD Lack of efficacy of music to improve sleep: A polysomnographic and quantitative EEG analysis International Journal of Psychophysiology 2007 63 3 232 239 10.1016/j.ijpsycho.2006.10.004 17123654
Leggett A Assari S Burgard S Zivin K The effect of sleep disturbance on the association between chronic medical conditions and depressive symptoms over time Longitudinal and Life Course Studies: International Journal 2017 8 2 138 151 10.14301/llcs.v8i2.433 28966664
Lie, J. D., Tu, K. N., Shen, D. D., & Wong, B. M. (2015). Pharmacological treatment of insomnia. P & T: a peer-reviewed journal for formulary management, 40(11), 759–771. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4634348/
Pagel JF Parnes BL Medications for the treatment of sleep disorders: An overview Primary Care Companion to the Journal of Clinical Psychiatry 2001 3 3 118 125 10.4088/pcc.v03n0303 15014609
Perry GS Patil SP Presley-Cantrell LR Raising awareness of sleep as a healthy behavior Preventing Chronic Disease 2013 10 E133 10.5888/pcd10.130081 23928458
Rama, A. N., Cho, S. C., & Kushida, C. A. (2005). Normal human sleep. In Sleep: A comprehensive handbook (pp. 1–9). 10.1002/0471751723.ch1
Ryu G Choi JH Interpretation of polysomnography Journal of Clinical Otolaryngology Head and Neck Surgery 2019 30 1 12 19 10.35420/jcohns.2019.30.1.12
Shrivastava D Jung S Saadat M Sirohi R Crewson K How to interpret the results of a sleep study Journal of Community Hospital Internal Medicine Perspectives 2014 4 5 24983 24983 10.3402/jchimp.v4.24983 25432643
Son J Jung S Song H Kim J Bang S Bahn S A survey of Koreans on sleep habits and sleeping symptoms relating to pillow comfort and support International Journal of Environmental Research and Public Health 2020 17 1 302 10.3390/ijerph17010302 31906363
Tang HYJ Riegel B McCurry SM Vitiello MV Open-loop audio-visual stimulation (AVS): A useful tool for management of insomnia? Applied Psychophysiology and Biofeedback 2016 41 1 39 46 10.1007/s10484-015-9308-7 26294268
Tang HYJ Vitiello MV Perlis M Riegel B Open-loop neurofeedback audiovisual stimulation: A pilot study of its potential for sleep induction in older adults Applied Psychophysiology and Biofeedback 2015 40 3 183 188 10.1007/s10484-015-9285-x 25931250
Thompson, M., & Thompson, L. (2003). The neurofeedback book: An introduction to basic concepts in applied pychophysiology.
Tracy JI Ahmed N Khan W Sperling MR A test of the efficacy of the MC Square device for improving verbal memory, learning and attention International Journal of Learning Technology 2007 3 2 183 202 10.1504/IJLT.2007.014844
Vgontzas M Kales M Sleep and its disorders Annual Review of Medicine 1999 50 1 387 400 10.1146/annurev.med.50.1.387 10073285
Xiao H Zhang Y Kong D Li S Yang N The effects of social support on sleep quality of medical staff treating patients with coronavirus disease 2019 (COVID-19) in january and february 2020 in China Medical Science Monitor: International Medical Journal of Experimental and Clinical Research 2020 26 e923549 e923549 10.12659/MSM.923549 32132521
Xiao H Zhang Y Kong D Li S Yang N Social capital and sleep quality in individuals who self-isolated for 14 days during the coronavirus disease 2019 (COVID-19) outbreak in january 2020 in China Medical Science Monitor: International Medical Journal of Experimental and Clinical Research 2020 26 e923921 e923921 10.12659/MSM.923921 32194290
Zunhammer M Eichhammer P Busch V Sleep quality during exam stress: The role of alcohol, caffeine and nicotine PloS One 2014 9 10 e109490 10.1371/journal.pone.0109490 25279939
| 36467016 | PMC9707217 | NO-CC CODE | 2022-12-01 23:20:21 | no | Freie Zahnarzt. 2022 Nov 29; 66(12):8-11 | latin-1 | Freie Zahnarzt | 2,022 | 10.1007/s12614-022-1015-1 | oa_other |
==== Front
Aging Clin Exp Res
Aging Clin Exp Res
Aging Clinical and Experimental Research
1594-0667
1720-8319
Springer International Publishing Cham
2308
10.1007/s40520-022-02308-4
Original Article
Frailty and long-COVID: is COVID-19 responsible for a transition in frailty status among older adults who survived hospitalization for COVID-19?
Ferrara Maria Cristina 1
Zarcone Cristina 2
Tassistro Elena 3
Rebora Paola 3
Rossi Emanuela 3
Luppi Fabrizio 4
Foti Giuseppe 5
Squillace Nicola 6
Lettino Maddalena 7
Strepparava Maria Grazia 8
Bonfanti Paolo 9
http://orcid.org/0000-0001-5430-0947
Bellelli Giuseppe [email protected]
10
the STORM Long-COVID Team
1 grid.7563.7 0000 0001 2174 1754 School of Medicine and Surgery, University of Milano-Bicocca, Milan, Italy
2 Medicine Unit, ASST Valle Olona, Saronno, Italy
3 grid.7563.7 0000 0001 2174 1754 Bicocca Center of Bioinformatics, Biostatistics and Bioimaging, University of Milano-Bicocca, Milan, Italy
4 grid.415025.7 0000 0004 1756 8604 School of Medicine and Surgery, University of Milano-Bicocca, Respiratory Disease Unit, San Gerardo Hospital, ASST-Monza, Monza, Italy
5 grid.415025.7 0000 0004 1756 8604 School of Medicine and Surgery, University of Milano-Bicocca, Anesthesia and Intensive Care Unit, San Gerardo Hospital, ASST-Monza, Monza, Italy
6 grid.415025.7 0000 0004 1756 8604 Infectious Disease Unit, San Gerardo Hospital, ASST-Monza, Monza, Italy
7 grid.415025.7 0000 0004 1756 8604 Cardiology Division, San Gerardo Hospital, ASST-Monza, Monza, Italy
8 grid.7563.7 0000 0001 2174 1754 Clinical Psychology Unit, School of Medicine and Surgery, University of Milano-Bicocca, Monza, Italy
9 grid.415025.7 0000 0004 1756 8604 School of Medicine and Surgery, University of Milano-Bicocca, Infectious Disease Unit, San Gerardo Hospital, ASST-Monza, Monza, Italy
10 grid.415025.7 0000 0004 1756 8604 School of Medicine and Surgery, University of Milano-Bicocca, Acute Geriatric Unit, San Gerardo Hospital, ASST-Monza, Via Pergolesi 33, Monza, Italy
29 11 2022
17
7 9 2022
16 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Background
There is a paucity of knowledge about the effects of COronaVIrus Disease-19 (COVID-19) on long-term frailty development or progression over time.
Aim
This study aims to assess transitions in frailty status in older adults who survived hospitalization for COVID-19.
Methods
This is a longitudinal panel study. A multidisciplinary outpatient follow-up service was established since summer 2020, for the evaluation of individuals discharged alive, after hospitalization due to COVID-19. Frailty status was assessed in-hospital and at follow-up using the clinical frailty scale (CFS). Main patients’ characteristics, including health, functional, cognitive, and psychological status were collected.
Results
A total of 177 patients aged 65 years and older were evaluated until June 2022. They were predominantly male, with a median age of 70 (Q1–Q3 67–75) years and a median body mass index of 27.5 (Q1–Q3 24.9–30.6) kg/m2 at hospital admission. The median follow-up time was 6.3 (Q1–Q3 3.7–10.9) months. Sixty-one patients (34.5%) scored worse at CFS follow-up compared to hospital admission, and twenty-two patients (12.4%) became frail.
Discussion and conclusion
This study shows that one out of three older patients previously hospitalized for COVID-19 had an unfavorable transition in CFS score during a median follow-up of nearly 6 months. Specific interventions to prevent frailty development or progression should be considered for patients at risk. Further studies are required to confirm our findings.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40520-022-02308-4.
Keywords
Frailty
Geriatric assessment
Older
COVID-19
==== Body
pmcBackground
Studies have shown that COronaVIrus Disease-19 (COVID-19) can have prolonged effects over several months, a condition termed long-COVID syndrome [1]. Long-COVID syndrome may involve almost all body systems and encompasses several symptoms, such as fatigue, breathlessness, tachycardia, cognitive deficits, dysautonomia, depression, and hair loss [1].
Frailty is an age-related clinical syndrome, with decline in physiological capacity of several organ systems, characterized by increased susceptibility to sudden, disproportionate functional decline following stressor events [2]. There is wide consensus that addressing frailty is the best method to detect heterogeneity in risk of adverse outcomes among people of the same chronological age [2].
COVID-19 has disproportionately affected frail older people, with high rates of mortality and symptom persistence in the long term [3–5]. No studies, however, have assessed the long-term effect of COVID-19 on the transition in frailty syndrome over time after hospitalization.
This study aims to fill this gap, assessing a cohort of older patients who were hospitalized for COVID-19 during three pandemic waves (March–June 2020; September–December 2020; February–May 2021).
Methods
Patients and setting
This study was conducted in a 700-bed urban hospital of the Northern Italy, serving a population of about 870,000 inhabitants. During the first three pandemic waves, most acute wards of that hospital were transformed into COVID-19 wards to respond to the massive demand of infected people for hospitalization. All patients referred to the hospital were admitted directly to the acute wards after an initial evaluation by an emergency room physician and an infectious disease specialist. There were no criteria based on age for the admission to the acute geriatric ward or other medical wards. Shortly after the beginning of the first pandemic wave (March 2020), the Hospital Physician Directors' Board held daily meetings to monitor the situation and reorganize the patients’ care. It was during one of these meetings that they decided to establish a post-acute outpatient service for COVID-19 survivors, which opened in summer 2020.
The inclusion criteria for being admitted to the service were: (1) a previous hospitalization for COVID-19; (2) a negative real-time reverse transcriptase-polymerase chain reaction for SARS-CoV-2 at the follow-up visit, and (3) an informed consent signed by the patient.
This study was approved by the Monza-Brianza Ethics Committee (Clinical.Trial.gov Identifier: NCT04424992).
Multidisciplinary assessment
The follow-up consisted in a multidisciplinary team assessment, including specialists in infectious, respiratory, cardiac, hematology, intensive care, and geriatrics, respectively. Each assessment was carried out sequentially on the same day by the specialists, to avoid waste of time for the patient. Follow-up timing, originally set between 3 and 6 months from hospital discharge, has been delayed up to 11 months, due to the subsequent intermittent pandemic waves. All the enrolled patients already had the following data registered at hospital’s ward admission: socio-demographic data (gender, age, and cohabitation), body mass index (BMI), presence of the principal acute or chronic diseases (hypertension, ischemic heart disease, congestive heart failure, peripheral vascular disease, stroke or transient ischemic attack, diabetes, depression, osteoporosis, osteoarthritis, pulmonary diseases, renal failure, liver disease, hyperthyroidism or hypothyroidism, visual impairment, hearing impairment, dementia, solid neoplasm, hematological neoplasm, peptic ulcer, anemia, and rheumatological disease) and Charlson comorbidity index (CCI) [6].
Presence of frailty was assessed twice (at hospital admission and at follow-up visit), with the clinical frailty scale (CFS) [7], a nine-point scale based on the assessment of mobility, energy, physical activity, and function. A score of 1 indicates a very fit person, whereas 9 corresponds to a terminally ill person. The CFS has been widely used in intensive care units (ICUs) and in geriatric and medical wards with COVID-19 patients [5, 8]. At hospital admission, the CFS was ranked by an intensivist (for patients admitted to ICU) or a geriatrician (for patients admitted to medical wards), collecting information from patients or relatives. All the raters (either intensivists or geriatricians) have been previously trained in the use of CFS. The score was obtained after having gathered the patient’s clinical information, including autonomy in the basic and instrumental activities of daily living, cognitive and psychological status, and the presence of diseases, with reference to 2 weeks before hospital admission. At follow-up, patients also underwent the Montreal Cognitive Assessment (MoCA) [9] and the Short Physical Performance Battery (SPPB) [10] to better evaluate global cognitive and functional status; the CFS was ranked again by an intensivist or geriatrician according to the ward of admission whereas the SPPB and the MoCA scores were ranked by a geriatrician previously trained in their use.
Statistical analyses
Patient characteristics are described as median and first-third quartiles (Q1–Q3) for continuous variables and number (percentages) for categorical variables. Longitudinal transitions in CFS score from hospital admission to follow-up are presented as an alluvial plot. A multiple linear regression model was used to evaluate the association between CFS at follow-up and baseline characteristics: age, sex, CCI, and maximum ventilatory support during hospitalization, adjusting for CFS at baseline and follow-up time. Statistical analyses were performed with R 4.1.2 (http://www.R-roject.org).
Results
Between August 2020 and June 2022, 468 older (≥ 65 years) patients who were hospitalized with COVID-19 during the first 3 pandemic waves, underwent a complete geriatric assessment both at the hospital’s admissions and at the post-acute outpatient service. The demographic and clinical characteristics of our sample (N = 177) at hospital admission are shown in Table 1. The median age was 70 (Q1–Q3 67–75) years, patients were predominantly male (N = 118, 66.7%) and married or living maritally (N = 123, 74.1%). At baseline, most patients were either overweight or obese (median BMI 27.5, Q1–Q3 24.9–30.6), in good health (median CCI 3, Q1–Q3 2–4), and fit: in fact, 28 (15.8%) patients scored 1 on the CFS, 75 (42.4%) scored 2, and 48 (27.1%) scored 3. Only 26 patients scored 4 or more. The main diseases were hypertension (N = 97, 58.4%) and diabetes without complications (N = 31, 18.7%) while only one patient had a history of dementia. Most patients (79.8%) had bilateral pneumonia. The median length of hospitalization was 21 (Q1–Q3 14–34) days, and 38 patients (25.9%) were admitted to the ICU with a median length of stay of 14 (Q1–Q3 9–28) days. Figure 1 shows the alluvial plot of the longitudinal transitions in CFS score from hospitalization to follow-up (median follow-up time = 6.3, Q1–Q3 3.7–10.9 months), along with the numbers of each transition. A substantial number of patients scored worse (61 out of 177, 34.5%) at follow-up compared to hospital admission, and 22 patients with a CFS ≤ 3 at hospital admission became frail (CFS ≥ 4) at follow-up (12.4%). COVID-19-associated symptoms or signs reported by the patients who became newly frail at follow-up were motor deficits at discharge, and sleep and mood disorders at the follow-up (Supplementary Table S1). Female sex [95% CI (0.632; 0.133)] and tracheotomy/ECMO/intubation ([95% CI (0.429; 1.296)] were positively associated with higher CFS at follow-up, after adjusting for CFS at hospital admission, time of follow-up, and CCI (Table 2).Table 1 Demographic and clinical characteristics of the study population at hospital admission
Variable N = 177
Age (years), median (Q1–Q3) 70 (67–75)
Gender (males), N (%) 118 (66.7)
Living situation, N (%)
Alone/caregiver 17 (10.4)
With family members 145 (88.4)
Nursing home 2 (1.2)
Married/partner 123 (74.1)
Schooling, N (%)
≤ 8 years 96 (57.1)
9–13 years 45 (26.8)
> 13 years 27 (16.1)
BMI (kg/m2), median (Q1–Q3) 27.5 (24.9–30.6)
Chest radiography, N (%) (available in N = 139 subjects)
Bilateral pneumonia 111 (79.8)
Unilateral pneumonia 19 (13.7)
Acute respiratory distress syndrome, N (%) 63 (48.5)
Mild 9 (18.0)
Moderate 20 (40.0)
Severe 21 (42.0)
Maximum ventilatory support during stay, N (%)
Low flow O2 26 (17.8)
CPAP 85 (58.2)
Tracheotomy/ECMO/intubation 35 (24.0)
Maximum CRP [mg/dL], median (Q1–Q3) 10.5 (6.0–15.9)
Comorbidities, N (%) (available in N = 166 subjects)
Hypertension 97 (58.4)
Myocardial infarction 12 (7.2)
Congestive heart failure 3 (1.8)
Peripheral vascular disease 11 (6.6)
Stroke/transient ischemic attack 3 (1.8)
Diabetes mellitus, uncomplicated 31 (18.7)
Diabetes mellitus, end-organ damage 7 (4.2)
Depression 8 (4.8)
Osteoporosis 3 (1.8)
Osteoarthritis 5 (3.0)
COPD 9 (5.4)
Renal failure 8 (4.8)
Liver disease, mild 5 (3.0)
Liver disease, moderate to severe 4 (2.5)
Hyperthyroidism or hypothyroidism 12 (7.2)
Visual impairment 1 (0.6)
Hearing impairment 2 (1.2)
Dementia 1 (0.6)
Solid neoplasm 11 (6.6)
Hematological neoplasm 1 (0.6)
Peptic ulcer 2 (1.2)
Anemia 5 (3.0)
Rheumatological disease 5 (3.0)
Charlson comorbidity index, median (Q1–Q3) 3 (2–4)
CFS, N (%)
1—Very fit 28 (15.8)
2—Fit 75 (42.4)
3—Managing well 48 (27.1)
4—Living with very mild frailty 14 (7.9)
5—Living with mild frailty 10 (5.6)
6—Living with moderate frailty 2 (1.1)
Delirium, N (%) 6 (4.1)
Length of hospital stay (days), median (Q1–Q3) 21 (14–34)
Admitted to ICU, N (%) (available in N = 147 subjects) 38 (25.9)
Length of ICU stay (days), median (Q1–Q3) 14 (9–28)
Q quartile, BMI body mass index, CPAP continuous positive airway pressure therapy, ECMO extracorporeal membrane oxygenation, CRP C-reactive protein, COPD chronic obstructive pulmonary disease, CFS clinical frailty scale, ICU intensive care unit
Fig. 1 Alluvial plot and numbers of the longitudinal transitions in clinical frailty scale (CFS) score from hospital admission to follow-up. Note: CFS clinical frailty scale
Table 2 Multiple linear regression model on CFS at follow-up by CFS at baseline, follow-up time, age, sex, CCI, and maximum ventilatory support during hospitalization
Coefficient estimate 95% confidence interval p value
Intercept 0.756 (− 1.006; 2.518) 0.398
CFS at admission 0.903 (0.782; 1.025) < 0.001
Follow-up time 0.000 (− 0.001; 0.001) 0.925
Age − 0.004 (− 0.029; 0.020) 0.730
Males − 0.382 (− 0.632; − 0.133) 0.003
CCI 0.025 (− 0.026; 0.075) 0.333
Maximum ventilatory support during hospitalization
Low flow O2 Reference – –
CPAP 0.230 (− 0.122; 0.581) 0.199
Tracheotomy/ECMO/intubation 0.863 (0.429; 1.296) < 0.001
CFS clinical frailty scale, CCI Charlson comorbidity index, CPAP continuous positive airway pressure, ECMO extracorporeal membrane oxygenation
Supplementary Table S2 shows the MoCA and SPPB score at follow-up, corrected for age and level of education. MOCA score was abnormal (MoCA < 20) in 19 patients (12.8%) and borderline (MoCA 20–25) in 66 (44.6%) while 7 (4%) patients had a SPPB score suggesting overt disability (0–2), 15 (8.6%) moderate physical frailty, and 39 (22.3%) mild physical frailty.
Discussion
This study shows that one out of three older patients previously hospitalized for COVID-19 had an unfavorable transition in CFS score from hospitalization to follow-up, and about one out of eight patients became mildly or overtly frail. To the best of our knowledge, this is the first study evaluating the transition in frailty status among patients previously hospitalized for COVID-19, using a clinically based assessment.
A recent systematic review and meta-analysis found 16 studies focusing on frailty transitions in community-dweller adults [11]. All these studies categorized frailty according to the Fried’s criteria and the average follow-up period was 3.9 years. The pooled rates of frailty transitioning in this review were 4.5% (95% CI 3.2–6.1%) and 18.2% (95% CI 14.9–21.7%) among robust and pre-frail individuals, respectively, in a population that was on average older than ours [11]. Moreover, a study by Ahmad et al. [12] included a population which was more similar to ours with regard to the cohort's median age (70 years, comparable to our study), the overall prevalence of frailty at baseline (9.4% vs. 14.7% in our study), and the follow-up period (12 months vs. 6 months in our study). Again, frailty was evaluated in the Ahmad’s study according to the Fried’s criteria, while with the 9-item CFS in our study. Ahmad et al. found that 22.9% of participants worsened their frailty status at 12 months versus 34.5% in our study, and the observed 12-month frailty transitions from robust to frail and from pre-frail to frail were 2.9% and 8.9%, respectively [12].
Only one study, as far as we know, examined the transitioning to frailty during the COVID-19 pandemic [13]. Shinohara et al. explored frailty transition during three COVID-19 waves (from May to July 2020, from November 2020 to January 2021, and from May to July 2021) in 1953 community-dwelling older people (≥ 65 years) living in Japan, using the frailty screening index (FSI) [13]. Among 706 respondents, they found a 9.8% increase in frailty transition, which is slightly lower than ours [13]. Based on these findings, we can hypothesize that the proportion of frailty transitioning in our study is worse than in pre-COVID-19 general population. However, given the dynamic nature of frailty, we cannot exclude that some individuals who were labeled as frails at 6-month follow-up might have reverted to non-frail status in the following months.
A decline in immune function and the proinflammatory state are key components of frailty in COVID-19 patients [14]. A recent review suggests that hyper-inflammatory status, typical of SARS-CoV-2 infection, may exacerbate the immunosenescence process, promote endothelial damage, and lead to myofibrillar breakdown and muscle degradation through mitochondrial dysfunction and autophagy [15]. Therefore, it might be hypothesized that COVID-19 can induce frailty through an acute dysregulation of the immune response and the development of deep changes in patient’s body composition. The aftermath of these immunological phenomena, augmented by anosmia, ageusia, reduced food intake, and the lack of physical activity imposed by the national lockdown, may have further accelerated individual’s catabolism [15] and muscle wasting. Moreover, the association between ICU admission and frailty after COVID-19 is in line with these assumptions since ICU-acquired weakness is a well known and common neuromuscular complication of critical illnesses [16].
Our findings are in accordance with Shinohara’s study [13] and with another study showing that sarcopenia affected nearly one in five young and older COVID-19 survivors, being higher in patients with a longer hospital stay and lower in patients who were more physically active and had higher levels of serum albumin [17].
It is of interest that nearly 35% of patients had a follow-up SPPB score indicative of either disability or physical frailty and that 12% of patients had MoCA score indicative of cognitive impairment. There might be a relationship between COVID-19 and these findings, but the lack of data regarding the SPPB and MoCA scores at hospital admission prevents conclusions from being drawn about the timeframe for impairment onset.
Holistic management of COVID-19 patients appears essential to minimize the deleterious effect of this disease, preventing long-COVID-19 syndrome and the development or progression of frailty. To this aim, specific interventions including physical activity programs and/or nutritional support should be started as soon as possible in patients at risk.
A limitation of this study concerns the lack of a comparison group that prevents us to assess if transitions to frailty for non-COVID hospital admissions are substantially different from COVID group. However, this would have required a specific study design and fundings that we have not. Another limitation concerns the method of assessing CFS, which was obviously different from baseline to follow-up: CFS assessment on hospital admission was mainly based on telephonic interview, given the pandemic context, and this might be a source of bias, as we cannot exclude that grading of frailty levels may have differed between the two observations. A third limitation is that CFS have been ranked either by both intensivists and geriatricians, which may have partially biased the study results. However, it should be considered that all raters were familiar with the use of CFS, as this tool has been used in our hospital for many years, and the CFS inter-rater reliability is generally very good [18].
It should also be considered that the only method used to capture frailty was CFS. Indeed, at the time of the study, the electronic medical records of our hospital did not include information about the patient’s functional status and other domains required to retrospectively build up a frailty index according to the Rockwood’s approach. Furthermore, it was impossible to assess frailty according to the Fried’s phenotype in the pandemic context, since it requires the use of specific devices (such as handgrip strength) and pencil-and-paper tests which were not feasible in a work environment burdened by stress and high contagiousness. Lack of details on the level of physical activity and dietary changes occurring during the study period, may also be considered a limitation. Lastly, the sample size in this single-center study is relatively small and includes a quite selected population, since adhesion to the program was on a voluntary basis.
In conclusion, this study suggests that long-COVID may have influenced the transition in frailty status after a median follow-up of 6 months in a cohort of older patients who survived hospitalization for COVID-19. Knowledge in this field may be critical to promote specific interventions and prevent frailty development or progression for individuals at risk. Further studies are required to confirm our findings.
Supplementary Information
Below is the link to the electronic supplementary material.Table S1: COVID-19 associated symptoms or signs reported by the patients at discharge and at follow-up, according to the presence of new-onset frailty at follow-up. Table S2: Montreal Cognitive Assessment (MoCA) and Short Physical Performance Battery scores at follow-up assessment. Table S3: Full list of the members of the STORM Long-COVID Team. (DOCX 23 KB)
Acknowledgements
We are grateful to Drs. Silvia Mori for study support, and to Dr. Fiona Ecarnot and Dr. Kimberly Alison Katte for English editing.
Author contributions
MCF: data collection, study design, interpretation of data, draft and critical revision of the manuscript, final approval. CZ: data collection, final approval; ET: data analysis and interpretation, critical revision of the manuscript, final approval. PR: data analysis and interpretation, critical revision of the manuscript, final approval. ER: data analysis and interpretation, critical revision of the manuscript, final approval. FL: study design, interpretation, critical revision of the manuscript, final approval. GF: study design, interpretation, critical revision of the manuscript, final approval. NS: study design, interpretation, critical revision of the manuscript, final approval. ML: study design, interpretation, critical revision of the manuscript, final approval. MGS: study design, interpretation, critical revision of the manuscript, final approval. PB: study design, critical revision of the manuscript, final approval. GB: conceptualization, study design, interpretation of data, draft and critical revision of the manuscript, final approval.
Funding
The authors did not receive support from any organization for the submitted work.
Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Competing interests
All authors declare no conflict of interest.
Ethical standards
The study was compliant with the declaration of Helsinki and no identifiable personal data were used for this study.
Informed consent
Informed consent for participation in clinical studies was obtained from all patients.
Ethical approval
The study protocol was approved by the hospital Institutional Review Board.
Statement of human and animal rights
This study was approved by the Monza-Brianza Ethics Committee (Clinical.Trial.gov Identifier: NCT04424992).
Paolo Bonfanti for the STORM Long-COVID Team.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Lopez-Leon S Wegman-Ostrosky T Perelman C More than 50 long-term effects of COVID-19: a systematic review and meta-analysis Sci Rep 2021 11 16144 10.1038/s41598-021-95565-8 34373540
2. Dent E Martin FC Bergman H Management of frailty: opportunities, challenges, and future directions Lancet 2019 394 1376 1386 10.1016/S0140-6736(19)31785-4 31609229
3. Onder G Rezza G Brusaferro S Case-fatality rate and characteristics of patients dying in relation to COVID-19 in Italy JAMA 2020 323 1775 1776 10.1001/jama.2020.4683 32203977
4. Ward H Flower B Garcia PJ Global surveillance, research, and collaboration needed to improve understanding and management of long COVID Lancet 2021 398 2057 2059 10.1016/S0140-6736(21)02444-2 34774190
5. Rebora P Focà E Salvatori A The effect of frailty on in-hospital and medium-term mortality of patients with COronaVIrus Disease-19: the FRACOVID study Panminerva Med 2021 10.23736/S0031-0808.21.04506-7 34761887
6. Charlson ME Pompei P Ales KL A new method of classifying prognostic comorbidity in longitudinal studies: development and validation J Chronic Dis 1987 40 373 383 10.1016/0021-9681(87)90171-8 3558716
7. Rockwood K Song X MacKnight C A global clinical measure of fitness and frailty in elderly people CMAJ 2005 173 489 495 10.1503/cmaj.050051 16129869
8. Shears M Takaoka A Rochwerg B Assessing frailty in the intensive care unit: a reliability and validity study J Crit Care 2018 45 197 203 10.1016/j.jcrc.2018.02.004 29574340
9. Nasreddine ZS Phillips NA Bédirian V The montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment J Am Geriatr Soc 2005 53 695 699 10.1111/j.1532-5415.2005.53221.x 15817019
10. Guralnik JM Seeman TE Tinetti ME Validation and use of performance measures of functioning in a non-disabled older population: MacArthur studies of successful aging Aging (Milano) 1994 6 410 419 10.1007/BF03324272 7748914
11. Kojima G Taniguchi Y Iliffe S Transitions between frailty states among community-dwelling older people: a systematic review and meta-analysis Ageing Res Rev 2019 50 81 88 10.1016/j.arr.2019.01.010 30659942
12. Ahmad NS Hairi NN Said MA Prevalence, transitions and factors predicting transition between frailty states among rural community-dwelling older adults in Malaysia PLoS ONE 2018 5 e0206445 10.1371/journal.pone.0206445
13. Shinohara T Saida K Tanaka S Transition to frailty in older Japanese people during the coronavirus disease 2019 pandemic: a prospective cohort study Arch Gerontol Geriatr 2022 98 104562 10.1016/j.archger.2021.104562 34715458
14. Vellas C Delobel P de Souto BP Covid-19, virology and geroscience: a perspective J Nutr Health Aging 2020 24 685 691 10.1007/s12603-020-1416-2 32744561
15. Piotrowicz K Gąsowski J Michel JP Post-COVID-19 acute sarcopenia: physiopathology and management Aging Clin Exp Res 2021 33 2887 2898 10.1007/s40520-021-01942-8 34328636
16. Piva S, Fagoni N, Latronico N (2019) Intensive care unit-acquired weakness: unanswered questions and targets for future research. F1000Res 8:1000. 10.12688/f1000research.17376.1
17. Martone AM, Tosato M, Ciciarello F et al, Gemelli Against COVID-19 Post-Acute Care Team (2022) Sarcopenia as potential biological substrate of long COVID-19 syndrome: prevalence, clinical features, and risk factors. J Cachexia Sarcopenia Muscle. 10.1002/jcsm.12931
18. Lo AX Heinemann AW Gray E Inter-rater reliability of clinical frailty scores for older patients in the emergency department Acad Emerg Med 2021 2021 110 113 10.1111/acem.13953
| 36445566 | PMC9707219 | NO-CC CODE | 2022-12-01 23:20:21 | no | Aging Clin Exp Res. 2022 Nov 29;:1-7 | utf-8 | Aging Clin Exp Res | 2,022 | 10.1007/s40520-022-02308-4 | oa_other |
==== Front
Behav Res Methods
Behav Res Methods
Behavior Research Methods
1554-351X
1554-3528
Springer US New York
36443581
2022
10.3758/s13428-022-02022-z
Article
The Chinese Lexicon Project II: A megastudy of speeded naming performance for 25,000+ traditional Chinese two-character words
Tse Chi-Shing [email protected]
12
Chan Yuen-Lai 1
Yap Melvin J. 3
Tsang Ho Chung 1
1 grid.10784.3a 0000 0004 1937 0482 Department of Educational Psychology, The Chinese University of Hong Kong, Hong Kong, China
2 grid.10784.3a 0000 0004 1937 0482 Centre for Learning Sciences and Technologies, The Chinese University of Hong Kong, Hong Kong, New Territories China
3 grid.4280.e 0000 0001 2180 6431 Department of Psychology, National University of Singapore, Singapore, Singapore
28 11 2022
121
6 11 2022
© The Psychonomic Society, Inc. 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Using a megastudy approach, (Tse et al., 2017 Behavior Research Methods, 49, 1503–1519) established a large-scale repository of lexical variables and lexical decision responses for more than 25,000 traditional Chinese two-character words. In the current study, we expand their database by collecting norms for speeded naming reaction times (RTs) and accuracy rates, and compiling more lexical variables (e.g., phonological consistency and semantic neighborhood size). Following Tse et al.’s procedure, about 33 college-aged native Cantonese speakers in Hong Kong read aloud each word. We conducted item-level regression analyses to test the relative predictive power of orthographic variables (e.g., stroke count), phonological variables (e.g., phonological consistency), and semantic variables (e.g., semantic transparency) in naming performance. We also compared the effects of lexical variables on naming performance and Tse et al.’s lexical decision performance to examine the extent to which effects are task-specific or task-general. Freely accessible to the research community, this resource provides a valuable addition to other influential mega-databases, such as the English Lexicon Project (Balota et al., 2004 Journal of Experimental Psychology: General, 133, 283–316), and furthers our understanding of Chinese word recognition processes.
Keywords
Chinese
Two-character word
Naming
Megastudy
Visual word recognition
==== Body
pmcIn Chinese, characters are monosyllabic morphemes that are often combined to form a two-character word. According to the Institute of Language Teaching and Research (1986), 73.6% of modern Chinese words are two-character words (see Li et al., 2015, for more discussion), suggesting that the majority of Chinese words consist of two characters (Myers, 2006; Packard, 2000). Given this important feature of Chinese, Tse et al.’s (2017) megastudy focused on two-character words when developing a Chinese database of normed lexical decision (i.e., deciding whether a two-character string forms a Chinese word, e.g., 朋友friend, or a nonword, e.g., 形忌). In the present study, we expanded this database by norming speeded naming (i.e., reading aloud a Chinese word) performance, which has been considered one of the gold standards for assessing skilled readers’ word recognition abilities (e.g., Yap & Rickard Liow, 2016) and for developing computational models of lexical processing (e.g., Coltheart et al., 2001). Moreover, we compiled more lexical variables (e.g., phonological consistency) and evaluated the role of orthographic, phonological, and semantic variables in naming versus Tse et al.’s lexical decision performance. In our speeded naming task, Cantonese, rather than Mandarin, was used, as it is the native dialect of our participant population (i.e., Hong Kong, as also in our previous megastudy, Tse et al., 2017). Before elaborating on the details of our study, we first briefly introduce the megastudy approach and then selectively review naming studies of Chinese characters/words to motivate the use of that approach to test the predictions presented at the end of the Introduction section.
Megastudy approach in psycholinguistics
Using the megastudy approach in psycholinguistic studies, researchers compile the values of lexical variables for a very large pool of words and collect the normative reaction time (RT) and accuracy rates for these words in a lexical processing task. In contrast to the conventional factorial-design approach in which stimuli are selected based on a small set of criteria, the megastudy approach allows the language to define the stimuli (see Balota et al., 2013, and Tse et al., 2017, for reviews of problems associated with factorial-design experiments and how they can be addressed by the megastudy approach). After this, researchers conduct item-level multiple regression analyses on the normed data to test for the influence of lexical variables on word processing. The databases developed with this megastudy approach, often termed “Lexicon Project,” have been reported in American English (Balota et al., 2007), British English (Keuleers et al., 2012), single-character Chinese (Chang et al., 2016; Liu et al., 2007; Sze et al., 2014), two-character Chinese (Tsang et al., 2018; Tse et al., 2017), French (Ferrand et al., 2010), Dutch (Keuleers et al., 2010), Italian (Barca et al., 2002), and Malay (Yap et al., 2010). The item-level analyses help identify the proportion of unique variance of performance that lexical variables, whether continuous or categorical, explain after statistically controlling for the effect of other variables, and reveal their relative contribution in lexical processing (e.g., Sze et al., 2015). These analyses can also address other research questions, such as replicating theoretically important lexico-semantic effects (Balota et al., 2013), evaluating the impact of novel lexical variables (Kang et al., 2011; Yarkoni et al., 2008), assessing computational models (Perry et al., 2007), and revealing the role of individual differences in word recognition (Lim et al., 2020; Yap et al., 2012).
Tse et al. (2017) used the megastudy approach to establish a lexical decision database of RT and accuracy rates for more than 25,000 Chinese two-character words. Each of 594 native Cantonese-speaking participants made lexical decisions for about 1400 words and 1400 nonwords, presented over three sessions. Each word was normed based on about 33 responses. Item-level regression analyses were conducted to determine the influence of theoretically important lexical characteristics on lexical decision performance. Making use of these data, Tse and Yap (2018) compared the role of orthographic, phonological, and semantic variables in predicting lexical decision performance and found some theoretically driven interaction effects (e.g., word frequency × character frequency) that are not readily explainable by any model of Chinese word recognition. Lim et al. (2020) further demonstrated that the effects of some lexical variables (e.g., word frequency) on lexical decision could be moderated by participants’ lexical processing fluency. All these demonstrate the usefulness of a megastudy database for investigating the effects of lexical variables on performance in a lexical processing task.
Why investigate speeded naming in Chinese?
In the speeded naming task, participants read aloud a visually presented word as quickly and as accurately as they can. In the present study, we extended Tse et al.’s (2017) database by collecting participants’ normed naming responses for their large pool of 25,000+ two-character Chinese words. We also conducted analyses on both the normed lexical decision data and the naming data in Tse et al.’s database to test whether their reported lexical effects might reflect general word recognition processes or merely idiosyncratic task demands.
Lexical effect on speeded naming performance
As few studies have examined the speeded naming of two-character Chinese words, in the following review we also refer to the work reported on the speeded naming of single Chinese characters. For orthographic variables, Yu and Cao (1992, see also Liu et al., 2007) found that high-frequency characters (e.g., 花 flower) were named faster than low-frequency characters (e.g., 堃 compliance). Gao et al. (2016) reported that high-frequency words (e.g., 多謝 thank you) were named faster than low-frequency words (e.g., 睿哲 divinely wise). Number of strokes is often used to indicate visual complexity (e.g., Xing et al., 2004), wherein characters with more strokes are considered visually more complex (e.g., 人 human vs. 鬱 depressed). Shen and Zhu (1994, see also Leong et al., 1987; Liu et al., 2007) demonstrated that characters with fewer strokes were named faster than those with more strokes.
For phonological variables, Hue et al. (1992, see also Liu et al., 2007) showed that phonologically regular characters (a character is pronounced the same as its phonetic radical, e.g., 叮ding11 and 丁ding1) were named faster than irregular characters (a character and its phonetic radical do not share the pronunciation, e.g., 模mou4 and 莫mok6). Orthography-to-phonology mapping consistency refers to whether the mapping is systematic between a character’s phonetic radical and its pronunciation. For example, 換wun6 is a highly consistent character, as all other characters sharing 奐 as their phonetic radical are also pronounced as wun6, e.g., 喚 and 渙. In contrast, 板baan2 is less consistent, because characters with 反 as their phonetic radicals are pronounced differently, from 皈gwai1 to 飯faan6. Yang et al. (2009) showed that characters consistent with orthography-to-phonology mapping (e.g., 渙thaw out) were named faster than inconsistent characters (e.g., 飯rice). Phonological consistency refers to whether a character has one (consistent, e.g., 骨gwat1) or more than one (inconsistent, e.g., 重zung6 and cung5) pronunciation (Tan & Perfetti, 1999). Leong and Cheng (2003) found that words with consistent characters (e.g., 骨 in 骨頭 bone) were named faster than those with inconsistent characters (e.g., 重 in 體重 weight), whether the character was the first or second in the word. Finally, characters that share pronunciations with more other characters (i.e., high homophone density, e.g., 支) were named faster than those that share pronunciations with fewer other characters (i.e., low homophone density, e.g., 鏡) (e.g., Ziegler et al., 2000, but see the null effect of homophone density in Chang et al., 2016).
For semantic variables, characters that combine more readily with another character to form a word (i.e., larger neighborhood size, e.g., 花園 garden, 花盆 flowerpot, 花店 florist, 花叢 flower bush, 花市 flower market, 花椒 wild pepper, 花蜜 honey, 開花 to blossom, 花錢 to spend money, 花粉 pollen, 鮮花 fresh flower, etc., for 花) were named faster than those not as likely (i.e., smaller neighborhood size, e.g., only 鳳凰 phoenix for 凰) (e.g., Chang et al., 2016; Liu et al., 2007, see Baayen et al., 2006 for similar findings in English). Peng et al. (2003) reported that characters with more meanings (e.g., 行) were named faster than those with fewer meanings (e.g., 廚). Semantic transparency refers to whether words are semantically related (transparent, e.g., 黑板black-board [blackboard]) or unrelated to their characters (opaque, e.g., 東西east–west [thing]). It can be quantified by participants’ ratings for the semantic relatedness between a word and its characters (e.g., Tse et al., 2017). Although words with more semantically transparent characters yielded faster lexical decision performance than those with less semantically transparent characters (e.g., Myers et al., 2004; Tse et al., 2017), to our knowledge this variable has not been examined in the two-character Chinese word naming literature. In other languages like English, transparent words were reported to yield better lexical decision and naming performance than opaque words (e.g., Kim et al., 2019).
Most of the previous findings in Chinese character/word speeded naming have been based on factorial-design experiments, wherein a relatively small set of words constrained to be matched on various lexical characteristics were used. It is important to test whether these effects of lexical variables occur when a much larger pool of words is included in the megastudy approach. Assuming that the findings of naming single characters can be generalized to naming two-character words, we expect to replicate most, if not all, of the above findings.
The role of character variables in word processing
There has been debate on whether Chinese words are accessed holistically or analytically (by combining the characters) in the literature. Packard (1999) has argued that the access is via the whole word, as the large number of homophones and extent of semantic ambiguity in Chinese might make character processing too costly in terms of time and effort. However, this holistic view might be incompatible with findings that character variables, such as character frequency, in addition to word variables (e.g., word frequency), account for variance in lexical decision performance (e.g., Tse et al., 2017, see also the evidence of priming studies, e.g., Tsang & Chen, 2014). Moreover, Tse et al. showed that the effects of lexical variables on lexical decision performance were quite similar whether the characteristic was associated with the first or second character, suggesting that word processing may be parallel, with both characters being processed simultaneously. That said, it is not clear whether this parallel processing also occurs in naming. Specifically, the speeded naming task requires participants to read aloud the first and then second characters, and the voice key is triggered to record naming RT once they begin to pronounce the first character (i.e., speech onset). Unlike in English, where participants have to process the entire letter string to correctly pronounce the word, participants may not need to completely process the whole Chinese word before naming the first character of the word.2 If so, we expect that character variables, especially those of the first character, would be more predictive of naming performance than word variables, suggesting that the holistic versus analytic nature of Chinese word processing depends on task demands. To our knowledge, this issue has not been addressed in factorial-design experiments. The megastudy approach with multiple regression analyses could compare the predictive power of the lexical variables of the first and second characters so as to shed light on this issue.
The role of orthographic, phonological, and semantic variables in word processing
In contrast to the English language, an alphabetic writing system, Chinese language possesses a logographic system with far more opaque orthography; that is, there is no regular mapping between spelling and sound. While the phonetic radical of about 80% of Chinese characters provides pronunciation hints (Hoosain, 1992), this depends on orthography-to-phonology mapping consistency (i.e., whether a character’s pronunciation is shared with other Chinese characters possessing the same phonetic radical). Hence, it is possible that phonology plays a less influential role in Chinese, compared with English, word naming. On the other hand, Perfetti and Tan (1999) argued that Chinese word recognition results from a convergence of orthography, phonology, and semantics. Because orthography-to-phonology mapping (often in one-to-one relationship, e.g., 表 pronounced as biu2) is more reliable than orthography-to-semantics mapping (often in one-to-many relationship, e.g., 表 can mean watch, express, surface, or meter), phonology is the privileged constituent in word recognition. Inconsistent with this view, based on Tse et al.’s (2017) normed lexical decision data, Tse and Yap (2018, see also Sze et al., 2015) reported that orthographic and semantic variables, respectively, accounted for more variance than phonological variables. This suggests that phonological activation, compared with semantic activation, is less influential in a task which does not explicitly require the generation of word phonology and instead emphasizes the processing of stimulus familiarity (e.g., Balota & Chumbley, 1984). Evidence supporting the role of phonological information in Chinese word recognition has often been based on the speeded naming task (e.g., Perfetti & Tan, 1999), which places more emphasis on the generation of phonology. By identifying the proportion of variance in naming performance accounted for by orthographic, phonological, and semantic variables, we expect that phonological variables play a much larger role than semantic variables in naming performance.
In the few studies that have explored task-specific differences between Chinese lexical decision and naming (e.g., Gao et al., 2016; Li et al., 2017), word frequency was reported to produce a larger effect in the lexical decision task. Given the demands of speeded naming, when activating the whole word, participants might also sequentially activate the phonological information for the first and then the second character (e.g., Li et al.). This contrasts with the lexical decision task, in which participants can respond only by processing the whole word. The first-character-focused naming process might somehow play a role in reducing the word frequency effect in Chinese lexical processing. This is consistent with the view that, relative to naming, lexical decision emphasizes more frequency-based information in discriminating between word and nonword (e.g., Balota & Chumbley, 1984). In English, studies reported task-specific effects in lexical decision and naming (e.g., Andrews & Heathcote, 2001; Balota & Chumbley, 1985; Balota et al., 2004). For instance, there was an influence of semantic variables in lexical decision (e.g., Yap & Balota, 2009), but semantic effects in naming were smaller (as compared with lexical decision, e.g., Balota et al., 2004) or restricted to low-frequency irregular words (e.g., Cortese et al., 1997), so semantic variables might have a greater influence on lexical decision than on naming.
Most of the findings in Chinese have been based on factorial-design experiments and should thus be tested for generalizability using the megastudy approach (see, e.g., Balota et al., 2004, for similar analyses in English). By conducting the same analyses on the current naming data and Tse et al.’s lexical decision data, we compare the influences of these variables between the two tasks. We expect phonological variables to play a larger role in naming than in lexical decision, and semantic variables to play a larger role in lexical decision than in naming. For example, the first-character frequency effect would be larger in naming than in lexical decision, whereas the word frequency effect would be larger in lexical decision than in naming.
Present study
To recapitulate, we expanded Tse et al.’s (2017) database by norming speeded naming measures for more than 25,000 two-character Chinese words and compiling more lexical variables to test the role of orthographic, phonological, and semantic variables (see Table 1) in naming performance and to compare those with lexical decision performance in the same item-level multiple regression analyses. We have the following predictions.We replicate the lexical effects (e.g., word frequency and phonological consistency) of previous speeded naming studies in Chinese, even though most of them were based on single-character naming factorial-design experiments.
Given that participants serially activate individual character phonology when they begin to pronounce the two-character words, character variables (e.g., first-character frequency) will be more predictive of naming performance than word variables (e.g., word frequency). The first-character lexical variables will also be more predictive of naming performance than the second-character lexical variables.
By comparing lexical decision (Tse et al., 2017) and naming based on the proportion of variance explained in multiple regression analyses, we expect that phonological variables play a larger role than semantic variables in naming, and semantic variables play a larger role than phonological variables in lexical decision (i.e., similar to Tse & Yap’s, 2018, findings, based on Tse et al.’s, 2017, normed lexical decision data).
Table 1 Variables listed in the Excel .xlsx and Unicode .csv files
Column Variable name Definition
1 Word_Trad Chinese word in traditional characters
2 Word_Sim Chinese word in simplified characters
3 Ntrials Number of participants whose trials were sufficiently reliable to provide the RT for that item (maximum being 33)
4 Acc Mean accuracy rate for each word computed across participants
5 RT Mean RT for each word computed across participants
6 RT_SD Standard deviation of the RT for each word
7 zRT Mean of the standardized RT for each word computed across participants
8–9 Pronounciation_1 and Pronounciation_2 Phonological transcriptions/syllables (with lexical tone) of the first and second characters are Cantonese Pronunciations based on the Multi-function Chinese Character Database (see https://humanum.arts.cuhk.edu.hk/Lexis/lexi-can/ or https://humanum.arts.cuhk.edu.hk/Lexis/lexi-mf/ for more details)
10–22 Initial_phoneme_feature_1, …, Initial_phoneme_feature_13 The 13 dichotomous variables to code the initial phoneme of the first character on 13 features, in the following order: affricative, alveolar, bilabial, labiodentals, dental, fricative, glottal, liquid, stop, nasal, palatal, velar, and voiced (e.g., Spieler & Balota, 1997).
23–24 Stroke_1 and Stroke_2 Number of strokes of the first and second characters based on a pocket dictionary (Que, 2008)
25–26 Phoneme_1 and Phoneme_2 Number of phonemes of the first and second characters
27–28 Character_frequency_1 and Character_frequency_2 Character frequency of the first and second characters based on Cai and Brysbaert’s (2010) contextual diversity subtitle character frequency count
29 Word_frequency Word frequency based on Cai and Brysbaert’s (2010) contextual diversity subtitle word frequency
30–31 Regularity_1_dummy_contrast_1 and Regularity_2_dummy_contrast_1 Whether the character is pronounced the same as (i.e., regular) or different from (i.e., irregular) its phonetic radical, e.g., 叮 is regular, as it shares the same pronunciation as 丁. For “change of tone,” the character shares the same syllable but not the same tone with its phonetic radical (e.g., regularity of 情cing4 vs. 青cing1). Regular = 1, irregular = −0.5, change of tone = −0.5
32–33 Regularity_1_dummy_contrast_2 and Regularity_2_dummy_contrast_2 See above. Regular = 0, irregular = 0.5, change of tone = −0.5
34–35 OP_Consistency_1 and OP_Consistency_2 Orthography-to-phonology mapping consistency: the extent to which the mapping is systematic between a character’s phonetic radical and its pronunciation. For example, 換 wun6 is a character with high consistency, because all other characters sharing 奐 as their phonetic radical are also pronounced as wun6, e.g., 喚 and 渙. In contrast, 板 baan2 is less consistent, because characters with 反 as their phonetic radicals are pronounced differently, from 皈 gwai1 and 飯 faan6.
36–37 Phono_consistency_1 and Phono_consistency_2 Phonological consistency: whether a character has one (phonologically consistent) or more than one (phonologically inconsistent) pronunciation (as defined by Leong & Cheng, 2003)
38–39 Homophone_1 and Homophone_2 Homophone density of the first and second character: number of characters that share the same pronunciation provided in the Multi-function Chinese Character Database (http://humanum.arts.cuhk.edu.hk/Lexis/lexi-mf/)
40–41 Neighborhood_1 and Neighborhood_2 Neighborhood size of the first and second character: number of two-character words sharing the same character, regardless of the character position (Tsai et al., 2006)
42–43 Meaning_1 and Meaning_2 Number of meanings of the first and second characters: raw count of distinct conceptual representations a character has, i.e., how polysemous a character is (Tsang et al., 2018)
44–45 Transparency_1 and Transparency_2 Semantic transparency between first/second characters and word based on standardized means of participants’ semantic relatedness ratings for the first/second characters in respect to the word
Method
To ensure comparability, the participants, material, design, and procedure followed those in our lexical decision megastudy (Tse et al., 2017).
Participants
Two hundred and ninety-seven native Cantonese-speaking students from the Chinese University of Hong Kong (CUHK), with self-reported normal or corrected-to-normal vision, participated in our study. Data from 19 additional participants were replaced due to their failure to participate in all experimental sessions. Participants were asked to report their gender (191 female), age (M = 19.73, SD = 1.43), amount of time they read Chinese materials per week (M = 210.43 minutes, SD = 460.82), cumulative grade-point average (M = 3.23, SD = .34),3 self-rated knowledge in seven-point scales for traditional characters (M = 5.63, SD = .88) and spoken Cantonese (M = 5.99, SD = .90), and overall grades and sub-grades in Chinese language university entrance exam (overall: M = 4.64, SD = 1.22; reading comprehension: M = 4.82, SD = 1.52; writing: M = 4.28, SD = 1.56; listening and integrated skills: M = 5.19, SD = 1.41; speaking: M = 3.88, SD = 1.39).4 We did not impose any restriction on participants’ Chinese proficiency, as substantial variability in proficiency may allow an exploration of individual differences in language processing (e.g., Lim et al., 2020, see also Yap et al., 2012). Each participant was paid 300 HKD (~38 USD) for their voluntary participation.
Materials and design
We include the words that were normed for lexical decision performance in the Chinese Lexicon Project (see Tse et al., 2017, for the selection procedure). Stimuli were all traditional Chinese characters typically used in Hong Kong and should thus be familiar to participants. The lexical variables are listed in Table 1 (see Tse et al., 2017s, for how semantic transparency was normed). All values of these variables are listed in the .xlsx/.csv file available at: https://osf.io/vwnps. Using Tse et al.’s procedure (see also the English Lexicon Project of Balota et al., 2007), we divided 25,281 two-character Chinese words into nine lists of 2809 words each, with a restriction that the proportion of the words with a specific first character in each subset roughly reflected those in the whole word pool. We collected data from 297 participants (i.e., 9 lists × 33 participants) to obtain 33 naming responses for each word. This number of observations per word is comparable to the mean of other normed naming databases of other languages (e.g., 25 in Balota et al.; > 25 in Ferrand et al., 2010; 31 in Goh et al., 2020; 39 in Keuleers et al., 2010; 39 in Keuleers et al., 2012) and allows us to obtain stable estimates of each word’s mean RTs. Each participant read aloud 936–937 words in each of the three sessions and thus in total 2809 words. The presentation order of word sets across the three sessions were counterbalanced across participants.
Procedure
Informed consent was obtained at the beginning of the study. Each participant was tested in a quiet cubicle in three sessions separated by no more than a week. PC-compatible computers with E-Prime 2.0 (Schneider et al., 2001) were used to display the stimuli and collect RT and accuracy data. (The participants’ naming duration was also recorded, but we do not consider them in the current work.) Stimuli were in white and visually presented on a black background, one at a time, at the center of the screen. In each session, 936–937 words were divided into four blocks of 234–235 words. At the end of each block, participants received a self-paced rest break, so there were three rest breaks in each session. Participants took about 55 min to complete each session. The presentation order of the four blocks within each session were randomized for each participant. Within each block, the presentation order of the words was also randomized for each participant.
Each trial started with a 500-ms fixation point *** at the center of the screen, followed by a 120-ms blank-screen interstimulus interval. A two-character Chinese word in font size 36 and font type 標楷體 DFKai-SB was presented at the same location as the fixation point until the participant made a response. The physical size of the word was about 2.5 cm in height by 5 cm in width. Given that participants sat 55 cm away from the screen, the visual angle was about 1.56°. To ensure that the current data reflected the young adults’ naming responses to the script that was most familiar to them, we presented all stimuli in traditional characters, following Tse et al. (2017). The participant was instructed to read aloud the word in Cantonese (i.e., their native language) as quickly and as accurately as possible to trigger the voice key connected via the Chronos response box to the computer. The Chronset algorithm (Roux, Armstrong, & Carreiras, 2016) was used to ensure the precision of onset detection. The word remained on the screen for an additional 500 ms after voice onset and then disappeared. If no response was detected in 4000 ms, the word disappeared and then a verbal signal 太慢!Too Slow! appeared for 500 ms. At the end of each trial, a blank screen was displayed for 1000 ms, to serve as an intertrial interval. Following Balota et al.’s (2007) procedure, before the 1000-ms intertrial interval, participants were instructed to manually code, in a self-paced manner, their own responses as correct pronunciation, mispronunciation, or microphone error. Participants were informed about the importance and use of these coding options. At the start of a session, participants were presented 10 practice words to become familiar with the task. These words did not appear as experimental words in that session.
Results
A phonetic bias might occur for an articulatory reason (some phonemes take more time to initiate) or acoustic reason (some phonemes take more time for the voice key to detect) (e.g., Liu et al., 2007). Following previous studies (e.g., Balota et al., 2004), we controlled for these variables in our analyses by coding the following 13 characteristics of the initial phoneme of the first characters as 1 (presence) or 0 (absence): affricative, alveolar, bilabial, labiodentals, dental, fricative, glottal, liquid, stop, nasal, palatal, velar, and voiced.
Only responses for experimental words were analyzed. Based on 297 participants, the mean accuracy rate (i.e., percentage of correct responses after excluding trials with microphone error) was 96.40% (SD = 2.41%). Following previous studies (e.g., Sze et al., 2014; Tse et al., 2017), remaining responses faster than 200 ms or slower than 3000 ms were excluded (about 0.41% of the trials). The mean and SD were then computed for each participant’s word responses. Any correct word response 2.5 SD above or below their mean was regarded as an outlier and excluded (about 3.31% of the trials). The mean RT of the trimmed correct word trials was 659.53 ms (SD = 186.71 ms). As suggested by Faust et al. (1999) and following previous megastudies (e.g., Balota et al., 2007; Ferrand et al., 2010; Ferrand et al., 2018; Keuleers et al., 2012; Sze et al., 2014; Tse et al., 2017; Yap et al., 2010), these raw RTs were transformed into z scores for each participant, before averaging across participants for each word to yield individual word zRT values. This standardization controls for differences in overall RT and variability between participants, which then helps to reduce noise in the data. Previous works reported that zRT was more reliable than raw RT (e.g., Ferrand et al., 2010). The level of significance was set at .05.
To determine the reliability of our dependent measures (RT, zRT, and accuracy rates), we followed Tse et al. (2017) and computed the split-half correlation, with a correction for length (i.e., about 33 observations per word) with the Spearman-Brown formula (2 × r)/(1 + r). This reflects the proportion of variance in a variable that can be explained. The corrected correlations (rcorr) between the dependent measures computed on the first half (N = 16) of participants who saw the word and the dependent measures computed on the second half (N = 17) of participants who saw the word were .82, .88, and .86 for RT, zRT, and accuracy, respectively. The finding that the reliability of zRTs is higher than the reliability of the raw RTs is in line with Faust et al.’s (1999) view that taking away differences in overall RT and variability between participants may remove noise from the data and does not artificially reduce the variability of the words.
Following Tse et al. (2017), we excluded words that yielded lower than 70% accuracy rates (about 2.31%). We ran item-level multiple regression analyses analogous to those that Tse et al. did for the lexical decision data, with mean zRT, averaged across participants, as the dependent variable. For the accuracy data,5 since the values are bounded (from 0 to 100) and violate the assumptions of linear regression models (i.e., without boundaries), we first transformed each word’s mean accuracy rates to logit (i.e., unbounded score) and then ran the same item-level multiple regression analysis as that for mean zRT, using the binomial family under the generalized linear modeling function in R 4.1.3. Note that McFadden’s pseudo R2 values were computed for these accuracy analyses, which could not be directly compared with the R2 values for zRT. The values of the lexical variables as predictors were all mean-centered in all regression analyses (Jaccard & Turrisi, 2003). Table 1 presents all variables listed in the Excel .xlsx and Unicode .csv files, which are available at: https://osf.io/vwnps (Tse_et_al.xlsx and Tse_et_al.csv). Following previous megastudies (e.g., Tse et al., 2017), we entered the lexical variables in the following order: characteristics of initial phoneme of the first character (step 1), orthographic variables (number of strokes, the frequency of the first and second characters, and word frequency) (step 2), phonological variables (number of phonemes, homophone density, phonological consistency, phonological regularity,6 and orthography-to-phonology mapping consistency of the first and second characters) (step 3), and semantic variables (neighborhood size, number of meanings, and semantic transparency of the first and second characters) (step 4). After taking into account the missing values of some words, the following analyses were based on 19,888 words (about 78.67% of 25,281 words in total).
Analyses of naming performance
Tables 2 and 3 report the descriptive statistics of and inter-correlational matrices among these lexical variables and the two behavioral measures (zRT and accuracy). There was no multicollinearity problem for orthographic, phonological, and semantic variables, as indicated by their overall moderate- to low inter-correlations (see Table 3). The variance inflation factors were also generally low (< 1.94) in all these variables.7 The betas (standardized regression coefficients) of these lexical variables when they were first entered in the regression models and the R2 change (i.e., change in proportion of variance accounted for in zRT) or pseudo R2 change in accuracy at each step are reported in Table 4. The total proportions of variance accounted for were 51.6% and 19.1% for zRT and accuracy, respectively.Table 2 Descriptive statistics of the lexical variables and two behavioral measures (zRT and accuracy) involved in item-level regression analyses
Mean SD Range
Naming accuracy .98 .04 .70–1.00
Naming zRT −.036 .438 −1.02 to 2.54
Number of strokes C1 10.61 4.43 1–30
Number of strokes C2 10.62 4.48 1–33
Log frequency C1 3.36 .49 .00–3.80
Log frequency C2 3.41 .47 .00–3.80
Log frequency (word) 1.58 .81 .00–3.80
Number of phonemes C1 3.35 .72 1–6
Number of phonemes C2 3.32 .73 2–6
Homophone density C1 17.61 15.75 0–106
Homophone density C2 17.67 15.62 0–106
Phonological consistency C1 .59 .49 0–1
Phonological consistency C2 .58 .49 0–1
Phonological regularity C1 (dummy contrast 1) −.29 .52 −0.5 to 1
Phonological regularity C1 (dummy contrast 2) .33 .32 −0.5 to 0.5
Phonological regularity C2 (dummy contrast 1) −.27 .54 −0.5 to 1
Phonological regularity C2 (dummy contrast 2) .33 .32 −0.5 to 0.5
Orthography-to-phonology mapping consistency C1 .16 .37 0–1
Orthography-to-phonology mapping consistency C2 .17 .37 0–1
Neighborhood size C1 38.15 33.88 1–229
Neighborhood size C2 44.43 40.46 1–229
Number of meanings C1 5.16 3.27 1–21
Number of meanings C2 5.28 3.27 1–21
Semantic transparency C1 −.0089 .49 −2.14 to 1.61
Semantic transparency C2 −.0069 .51 −2.24 to 1.81
N = 19,888 (i.e., words with values in all available lexical variables). C1 = first character. C2 = second character. The three-level variable, phonological regularity, is coded as two dummy contrasts (regular = 1 0, irregular = −0.5 0.5, change of tone = −0.5 0.5)
Table 3 Inter-correlational matrices among these non-centered lexical variables and the two behavioral measures (zRT and accuracy) involved in item-level regression analyses (N = 19,888)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
1 Accuracy 1
2 zRT −.490** 1
3 Number of strokes C1 −.082** .251** 1
4 Number of strokes C2 −.058** .139** .019** 1
5 Log frequency C1 .307** −.499** −.262** −.019** 1
6 Log frequency C2 .266** −.318** −.010 −.267** .151** 1
7 Log frequency (word) .194** −.330** −.057** −.070** .245** .228** 1
8 Number of phonemes C1 .004 −.112** .074** .002 −.002 −.004 −.016* 1
9 Number of phonemes C2 −.018* .029** .010 .116** .003 −.063** −.042** .021** 1
10 Homophone density C1 .016* .081** .023** −.001 −.090** −.013 −.010 −.151** −.002 1
11 Homophone density C2 .008 .010 −.008 −.024** −.026** −.037** −.011 −.016* −.217** .025** 1
12 Phonological consistency C1 .041** .060** .062** .018** −.099** −.031** −.017* −.095** −.003 .064** .010 1
13 Phonological consistency C2 .045** .025** .003 .091** −.010 −.117** −.043** .000 −.024** −.004 .066** .022** 1
14 O-to-P mapping consistency C1 .001 .006 .022** .005 −.035** −.006 −.013 .015* .006 .057** .000 .027** .004 1
15 O-to-P mapping consistency C2 −.002 .014* .001 −.017* −.011 −.009 −.001 .000 .009 .008 .076** .010 .028** .006 1
16 Phonological regularity (contrast 1) C1 −.039** .092** .201** .022** −.205** −.050** −.029** −.017* .003 .197** .006 .071** −.003 .129** .006 1
17 Phonological regularity (contrast 2) C1 .030** −.078** −.256** −.010 .158** .019** .024** −.037** −.001 −.062** −.003 −.019** .005 −.038** .002 −.423** 1
18 Phonological regularity (contrast 1) C2 −.037** .069** .007 .230** −.042** −.219** −.043** .004 .016* .007 .159** .014* .112** .005 .114** .029** −.018* 1
19 Phonological regularity (contrast 2) C2 .018* −.042** −.002 −.263** .021** .153** .033** −.013 −.075** −.001 −.015* −.006 −.048** −.011 −.027** −.016* .024** −.436** 1
20 Neighborhood size C1 .156** −.348** −.346** .002 .562** .040** .108** .062** .022** −.124** .000 −.097** .009 .001 −.002 −.188** .185** .000 .005 1
21 Neighborhood size C2 .092** −.201** .006 −.385** .017* .538** .111** .001 −.091** −.016* .005 −.015* −.137** −.001 −.015* −.016* .003 −.190** .193** −.004 1
22 Number of meanings C1 .115** −.295** −.242** −.004 .477** .045** .093** .087** .010 −.111** −.024** −.230** −.008 .003 −.005 −.169** .169** −.020** .017* .570** .009 1
23 Number of meanings C2 .080** −.164** −.015* −.249** .045** .472** .101** .010 −.022** −.010 −.052** −.025** −.251** .002 −.006 −.020** .013 −.178** .171** .008 .567** .040** 1
24 Semantic transparency C1 −.056** .161** .207** −.028** −.238** −.010 −.016* −.003 .011 .036** .013 .092** −.009 .010 .019** .073** −.089** .017* .001 −.223** .069** −.330** .031** 1
25 Semantic transparency C2 −.060** .146** .031** .214** −.076** −.271** −.121** .025** .074** .014* −.053** .017* .104** .008 .004 .021** −.002 .062** −.103** −.056** −.327** −.031** −.365** .069** 1
*p < .05 (two-tailed). **p < .01 (two-tailed). C1 = first character. C2 = second character. O-to-P = orthography-to-phonology
Table 4 Standardized regression coefficients (beta values) on naming zRT and accuracy rate (N = 19,888)
zRT Accuracy
First character’s initial phoneme - affricative 0.182** 0.121
First character’s initial phoneme - alveolar NA1 0.113
First character’s initial phoneme - bilabial −0.032 0.198
First character’s initial phoneme - labiodentals −0.069** 0.004
First character’s initial phoneme - dental −0.005 −0.006
First character’s initial phoneme - fricative 0.240** 0.174
First character’s initial phoneme - glottal −0.151** NA2
First character’s initial phoneme - liquid −0.013 0.129
First character’s initial phoneme - stop −0.050 0.157
First character’s initial phoneme - nasal −0.053** 0.151
First character’s initial phoneme - palatal −0.011 0.477
First character’s initial phoneme - velar 0.052** 0.132
First character’s initial phoneme - voiced −0.072 −0.138
Step 1 - R2 change 0.103** 0.002
Number of strokes C1 0.152** −0.011
Number of strokes C2 0.064** < 0.001
Log frequency C1 −0.398** 0.642**
Log frequency C2 −0.207** 0.558**
Log frequency (word) −0.177** 0.289**
Step 2 - R2 change 0.379** 0.170
Number of phonemes C1 −0.143** 0.060
Number of phonemes C2 0.004 0.003
Homophone density C1 0.020** 0.005
Homophone density C2 −0.007 0.002
Phonological consistency C1 (1 = consistent) −0.035** 0.266**
Phonological consistency C2 (1 = consistent) −0.017** 0.272**
Orthography-to-phonology mapping consistency C1 (1 = consistent) 0.003 0.038
Orthography-to-phonology mapping consistency C2 (1 = consistent) 0.007 −0.018
Phonological regularity C1 (1 = regular, −0.5 = irregular, −0.5 = change of tone) −0.014* 0.053
Phonological regularity C1 (0 = regular, 0.5 = irregular, −0.5 = change of tone) 0.017** −0.034
Phonological regularity C2 (1 = regular, −0.5 = irregular, −0.5 = change of tone) −0.005 0.034
Phonological regularity C2 (0 = regular, 0.5 = irregular, −0.5 = change of tone) 0.010 −0.079
Step 3 - R2 change 0.021** 0.017
Neighborhood size C1 −0.110** 0.003
Neighborhood size C2 −0.068** −0.001
Number of meanings C1 −0.016* −0.006
Number of meanings C2 0.003 −0.005
Semantic transparency C1 0.028** −0.047
Semantic transparency C2 0.008 −0.003
Step 4 - R2 change 0.013** 0.002
*p < .05 (two-tailed). **p < .01 (two-tailed). C1 = first character. C2 = second character. NA1 = this variable was excluded due to multicollinearity problem. NA2 = the regression coefficient could not be estimated due to the problem of singularities. The R2 change in accuracy analyses was based on pseudo R2 values, so their statistical significance could not be determined, and they cannot be compared with the R2 values estimated in zRT analyses
Variance accounted for in naming zRT and accuracy
First, characteristics of the initial phoneme of the first character accounted for 10.3% and 0.2% of the variance in naming zRT and accuracy, respectively, suggesting that phonological coding at the first character’s phoneme level is a significant predictor of naming zRT, but much less so naming accuracy. This reflects that the coding of onsets may primarily influence response speed due to voice key sensitivity and articulation, rather than response accuracy. Thus, they were statistically controlled in the following analyses. Second, phonological variables accounted for much less variance in naming zRT than orthographic variables (2.1% vs. 37.9%) and slightly more variance in naming zRT than semantic variables (2.1% vs. 1.3%). A similar pattern was observed in naming accuracy. The modest amount of variance accounted for by phonological variables in Chinese naming relative to English naming (e.g., Balota et al., 2004) could be due to the deeper orthography in the Chinese language relative to the English language.
Predictions 1 and 2: Effect of lexical variables on naming zRT and accuracy
We test whether our findings replicate the lexical effects of previous speeded naming studies in Chinese, most of which were based on single-character naming factorial-design experiments. Our findings were generally consistent with those reported in previous works, although there are some exceptions, thereby being partially congruent with our Prediction 1.
For orthographic variables, naming responses were faster for words comprising characters with fewer strokes (e.g., 人) than those with more strokes (e.g., 體), consistent with previous findings (e.g., Shen & Zhu, 1994). The naming responses were faster and more accurate for higher-frequency words (e.g., 多謝 thank you), relative to lower-frequency words (e.g., 睿哲 divinely wise), and for those with higher-frequency characters (e.g., 花 flower), relative to those with lower-frequency characters (e.g., 堃 compliance). These findings are also consistent with earlier studies (e.g., Gao et al., 2016; Yu & Cao, 1992). Consistent with our Prediction 2, character variables (e.g., first-character frequency) were more predictive of naming performance, as reflected by larger regression coefficients (see Table 4), than word variables (e.g., word frequency). In addition, the effects of number of strokes and character frequency were larger for the first character than the second character.
For phonological variables, naming responses were faster and more accurate when the first or second character was phonologically consistent (e.g., 骨 is pronounced as gwat1 in all words with it, such as骨頭gwat1 tau4 [bone]), relative to phonologically inconsistent (e.g., 重 is pronounced as cung5 in 體重tai2 cung5 [weight], zung6 in 重要zung6 jiu3 [important], or cung4 in 重申cung4 san1 [reiterate]). This effect was in line with previous findings (e.g., Leong & Cheng, 2003), although the effect was similar8 for the first and second characters. The naming responses were faster when the first character contained more phonemes (e.g., 框 kwaang1), relative to fewer phonemes (e.g., 之 zi1), and when the first character was phonologically regular (i.e., pronounced the same as its phonetic radical) than when it shared the same syllable but not the same tone with its phonetic radical (i.e., change of tone) or was phonologically irregular (i.e., pronounced differently from its phonetic radical). The naming responses were slower when the first character was phonologically irregular (vs. change of tone). The phonological regularity effect was in line with Hue et al. (1992) and it occurred in the first, but not the second, character. The naming responses were slower when the first character was higher in homophone density (e.g., 支zi1, which is also the pronunciation of 袛, 諮, 孜, 蜘, 吱, 芝, 茲, 肢, 滋, 脂, 姿, 枝, 資, 知, 之, and 恣...) than lower in homophone density (e.g., 鏡 geng3, a pronunciation is not shared with other characters), contradicting the positive or null effect of homophone density reported in previous studies (e.g., Chang et al., 2016; Ziegler et al., 2000). The null effect of orthography-to-phonology mapping consistency was inconsistent with the facilitatory effect reported by Yang et al. (2009). Other than phonological consistency, all significant effects of phonological variables on naming were larger for the first character than the second character, partially consistent with our Prediction 2.
For semantic variables, naming responses were faster when the first character was associated with larger neighborhood size (e.g., 花 flower, relative to smaller neighborhood size, e.g., 凰 phoenix) or more meanings (e.g., 行, relative to fewer meaning, e.g., 廚). These are consistent with previous findings (e.g., Chang et al., 2016; Peng et al., 2003). In contrast, naming responses were slower when the first character was more transparent in meaning. All these effects occurred only for the first character or were larger for the first character than the second character, consistent with our Prediction 2.
Prediction 3: Comparison between lexical decision and naming
To test our Prediction 3, we compare the effect of lexical variables on naming versus lexical decision by combining Tse et al.’s (2017) normed lexical decision data with the current naming data. After excluding words that yielded lower than 70% accuracy in lexical decision or naming data, we ran analyses similar to those we did for naming data based on 18,736 words (about 74.11% of 25,281 words in total). The betas of these lexical variables when they were first entered in the regression models and the R2 change for zRT and pseudo R2 change for accuracy at each step are reported in Table 5. The total proportions of variance accounted for were 36.5% and 20.1% for lexical decision zRT and accuracy, respectively, and 51.8% and 19.1% for naming zRT and accuracy, respectively.Table 5 Standardized regression coefficients (beta values) on lexical decision and naming zRT and accuracy rate (N = 18,736)
Lexical Decision Naming
zRT Accuracy zRT Accuracy
First character’s initial phoneme - affricative 0.006 −0.101 0.191** 0.150
First character’s initial phoneme - alveolar NA1 0.049 NA1 0.115
First character’s initial phoneme - bilabial −0.009 0.010 −0.039* 0.256
First character’s initial phoneme - labiodentals 0.018* 0.005 −0.070** 0.026
First character’s initial phoneme - dental −0.024 < 0.001 −0.013 0.016
First character’s initial phoneme - fricative −0.020 −0.098 0.246** 0.223
First character’s initial phoneme - glottal 0.026** NA2 −0.153** NA2
First character’s initial phoneme - liquid 0.011 −0.106 −0.015 0.123
First character’s initial phoneme - stop 0.045 −0.117 −0.044 0.192
First character’s initial phoneme - nasal −0.021 −0.032 −0.060** 0.150
First character’s initial phoneme - palatal −0.034 0.026 −0.021 0.561
First character’s initial phoneme - velar −0.038* 0.127 0.049** 0.150
First character’s initial phoneme - voiced 0.052 −0.087 −0.061 −0.140
Step 1 - R2 change 0.002** 0.001 0.109** 0.003
Number of strokes C1 0.027** 0.011 0.156** −0.013
Number of strokes C2 0.026** 0.009 0.066** 0.003
Log frequency C1 −0.067** 0.028 −0.405** 0.675**
Log frequency C2 −0.065** 0.003 −0.210** 0.575**
Log frequency (word) −0.545** 0.625** −0.151** 0.236**
Step 2 - R2 change 0.349** 0.187 0.376** 0.169
Number of phonemes C1 −0.021** 0.022 −0.144** 0.062
Number of phonemes C2 −0.002 −0.002 0.005 −0.004
Homophone density C1 −0.019** < 0.001 0.020** 0.005
Homophone density C2 −0.015* < 0.001 −0.008 0.002
Phonological consistency C1 (1 = consistent) −0.029** 0.043 −0.029** 0.242*
Phonological consistency C2 (1 = consistent) −0.027** 0.053 −0.017** 0.260*
Orthography-to-phonology mapping consistency C1 (1 = consistent) 0.003 −0.003 0.003 0.034
Orthography-to-phonology mapping consistency C2 (1 = consistent) 0.010 0.003 0.005 −0.018
Phonological regularity C1 (1 = regular, −0.5 = irregular, −0.5 = change of tone) 0.005 < 0.001 −0.016** 0.074
Phonological regularity C1 (0 = regular, 0.5 = irregular, −0.5 = change of tone) 0.007 0.023 0.017** −0.042
Phonological regularity C2 (1 = regular, −0.5 = irregular, −0.5 = change of tone) −0.001 0.002 −0.003 0.024
Phonological regularity C2 (0 = regular, 0.5 = irregular, −0.5 = change of tone) 0.021** −0.069 0.010 −0.107
Step 3 - R2 change 0.003** 0.001 0.021** 0.016
Neighborhood size C1 −0.060** 0.001 −0.101** 0.003
Neighborhood size C2 −0.044** < 0.001 −0.068** −0.001
Number of meanings C1 0.014 −0.002 −0.015* −0.001
Number of meanings C2 0.037** −0.008 0.003 −0.001
Semantic transparency C1 −0.065** 0.185** 0.034** −0.092
Semantic transparency C2 −0.063** 0.162* 0.017** −0.026
Step 4 - R2 change 0.011** 0.012 0.012** 0.003
*p < .05 (two-tailed). **p < .01 (two-tailed). C1 = first character. C2 = second character. NA1 = this variable was excluded due to multicollinearity problem. NA2 = the regression coefficient could not be estimated due to the problem of singularities. The R2 change in accuracy analyses was based on pseudo R2 values, so their statistical significance could not be determined, and they cannot be compared with the R2 values estimated in zRT analyses
Variance accounted for in lexical decision and naming zRT and accuracy
Consistent with previous naming studies in other languages (e.g., Yap & Balota, 2009), the characteristics of the initial phoneme of the first character accounted for more variance in naming zRT than in lexical decision zRT (10.9% vs. 0.2%, respectively) but were quite similar in naming accuracy (0.3%) and lexical decision accuracy (0.1%).
To compare the variance explained by orthographic, phonological, and semantic variables in lexical decision and naming, orthographic variables accounted for slightly more variance in naming zRT (37.6%) than lexical decision zRT (34.9%), but the difference was the opposite for accuracy (16.9% vs. 18.7%, respectively). Consistent with our Prediction 3, phonological variables accounted for slightly more variance in naming zRT/accuracy than lexical decision zRT/accuracy (2.1%/1.6% vs. 0.3%/0.1%, respectively). Semantic variables accounted for a similar proportion of variance in naming and lexical decision zRT (about 1.1–1.2%), although they accounted for slightly more variance in lexical decision accuracy than naming accuracy (1.2% vs. 0.3%, respectively). The latter finding was in line with our Prediction 3 and consistent with the emphasis of lexical decision on semantic information for word/nonword discrimination. Nevertheless, it is important to note that the differences in the contribution of orthographic, phonological, and semantic variables between lexical decision and naming were smaller in Chinese word processing than English word processing (e.g., Yap & Balota, 2009), all in all suggesting that similar word recognition processes/mechanisms might be used to process two-characters Chinese words in the lexical decision and naming tasks. This was attributed to the deep orthography of Chinese language. The overall variance patterns of naming and lexical decision findings based on this subset (N = 18,736) were the same as those reported above and in Tse and Yap (2018), respectively, although the word pools were not perfectly overlapping across studies, and transformed accuracy rates (in logits), instead of raw accuracy rate as in Tse and Yap, were used in the current study.
Effect of lexical variables on lexical decision and naming zRT and accuracy
The overall pattern of naming results was similar to those reported above, except that now both semantically transparent first and second characters, not just the semantically transparent first character, predicted slower naming zRT. Still consistent with our Prediction 2, the effect of semantic transparency was larger for the first character than for the second character. For the lexical variables shared between the two studies, the pattern of lexical decision findings was similar to that reported in Tse and Yap (2018), except that the number of strokes, character frequency, and phonological consistency no longer predicted lexical decision accuracy. It is noteworthy that, unlike Tse and Yap’s results, there was no speed–accuracy trade-off for the effect of number of strokes on lexical decision performance in the current study.9
The comparison of the effects of lexical variables on lexical decision and naming showed the following findings (see Table 5).
First, we found that the character with more strokes slowed both lexical decision and naming zRT, although the effect was larger on naming than on lexical decision. Similarly, the facilitatory effect of character frequency was also larger on naming than on lexical decision. This effect was also larger in the first character than in the second character for naming, but the effect was of similar magnitude between characters for lexical decision. In contrast, the facilitatory effect of word frequency was larger on lexical decision than on naming. Similar results regarding word frequency were obtained in the English Lexicon Project (Balota et al., 2004).
Second, the facilitatory effect of the first character with more phonemes was larger for naming zRT than for lexical decision zRT. While the homophone density of the first and second characters sped up lexical decision responses, the homophone density of the first character, but not second character, slowed naming responses. The facilitatory effect of phonological consistency on accuracy was larger in naming than in lexical decision, although the effect was similar in size for lexical decision and naming zRT. There was a facilitatory effect of first-character phonological regularity on naming, but not on lexical decision. For the character that shared the same syllable but different tone with its phonetic radical (i.e., “change of tone”), we obtained faster naming zRT for these “change-of-tone” first characters (e.g., 胖bun6, which shares the same syllable but not the same tone with its phonetic radical 半bun3) than phonologically irregular first characters (e.g., 耀jiu6, which does not share pronunciation with its phonetic radical 翟zaak6). The lexical decision zRT was also faster for “change-of-tone” second character than phonologically irregular second character.
Finally, the facilitatory effect of first and second character’s neighborhood size was larger in naming zRT than in lexical decision zRT.10 The words with more first-character meaning sped up naming responses, whereas those with more second-character meaning slowed down lexical decision responses. Whereas the first and second character’s semantic transparency facilitated lexical decision zRT and accuracy, they slowed naming zRT (especially the first character’s one).
Discussion
The purpose of the current study was twofold. First, using a megastudy approach, we expanded Tse et al.’s (2017) database, which was developed for lexical decision performance, by norming speeded naming RT and accuracy rates for more than 25,000 traditional Chinese two-character words. We also compiled more lexical variables (e.g., phonological consistency and semantic neighborhood size). Second, we performed item-level multiple regression analyses to test the relative predictive power of orthographic variables (e.g., stroke count), phonological variables (e.g., phonological consistency), and semantic variables (e.g., semantic transparency) in naming and compared these with those in lexical decision, based on Tse et al.’s normed data, to determine whether specific lexical effects are task-specific or task-general. We had three major predictions. Prediction 1: We expected to replicate the lexical effects of previous Chinese speeded naming studies. Prediction 2: Given the serial nature of the word pronunciation, we expected that the character variables, especially the first-character variable, would be more predictive of naming performance than word variables. Prediction 3: In the comparison for the proportion variance explained by phonological and semantic variables between lexical decision and naming, we expected phonological variables to account for larger variance than semantic variables in naming, but expected the opposite pattern to occur in lexical decision. In the following, we discuss the findings in response to the above predictions as well as their general implications for Chinese lexical processing.
Prediction 1: Replication of benchmark findings in speeded naming
The current naming data generally replicate the standard lexical effects in Chinese naming literature. Naming performance, as indicated by faster zRT and/or higher accuracy, was facilitated as a function of word frequency (e.g., Gao et al., 2016), both characters’ character frequency (e.g., Yu & Cao, 1992), phonological consistency (e.g., Leong & Cheng, 2003), and neighborhood size (e.g., Chang et al., 2016), and the first character’s number of phonemes, number of meanings, and phonological regularity (e.g., Hue et al., 1992). In contrast, naming performance was inhibited as a function of number of strokes for both characters (e.g., Shen & Zhu, 1994) and as a function of homophone density and semantic transparency for the first character. The inhibitory effect of homophone density was inconsistent with that reported in previous studies (e.g., Chang et al., 2016; Ziegler et al., 2000). Also, unlike Yang et al. (2009), orthography-to-phonology mapping consistency did not predict naming performance.
It is noteworthy that most of the previous findings on Chinese speeded naming were based on single characters; that is, participants read aloud the single characters shown on the screen, rather than two-character words as in the present study. Psycholinguistic effects might not necessarily be generalizable from single-character studies to multiple-character studies. When the character appears in a word context, the lexical characteristics of word and/or another character might moderate the processing of the character being pronounced. For example, naming a character in a high-frequency word might be less affected by its lexical characteristics than when the same character appears in a low-frequency word, as word familiarity might facilitate the activation of the whole word and in turn its characters. On the other hand, given the sequential (i.e., first, then second character) processing implicated in word naming, the influence of the lexical characteristics of the second character might be preempted by that of the lexical characteristics of the first character and the whole word. These possibilities could be further investigated by comparing the lexical effects on lexical decision and naming in the isolated context versus word context. Moreover, in the regression analyses, the interaction terms could be added to test the moderating role of a specific word’s/character’s variables. Nevertheless, the results of the lexical effects in our normed naming data provide useful empirical benchmarks for the Chinese word processing literature.
Prediction 2: Predictive power of character versus word variables and the first-character versus second-character variables in speeded naming
By comparing the differences in the contribution of the lexical effect of the words and/or the first and second characters on speeded naming, we may determine whether the pronunciation of the Chinese words is activated holistically or through the combination of the characters, and whether the influence of the first character is stronger than that of the second character. For the effects of the character’s versus the word’s lexical variable, we compare the character and word variables at the same (orthographic) level, that is, the effect of character frequency versus the effect of word frequency. Consistent with our Prediction 2, the effect of word frequency was smaller than the effect of character frequency, especially the first character’s frequency, in both zRT and accuracy (see Table 4). For the effects of the first character’s versus second character’s lexical variable, we found that the significant lexical effects, with the exception of phonological consistency, of the first character on naming performance were stronger than those of the second character (see the regression coefficients in Table 4), again in line with our Prediction 2. These suggest that when participants read aloud a word, they first processed the first character, followed by the second character, rather than processing them simultaneously. Given that the voice key was triggered to record naming RT once participants began to pronounce the first character (i.e., speech onset), their naming responses may be more influenced by the lexical characteristics of the first character than those of the second character and the word. This result is consistent with the analytic view (e.g., Tse et al., 2017), but not the holistic view (e.g., Packard, 1999) of Chinese word processing. At the very least, these findings show that the holistic versus analytic nature of word processing is flexible and depends on task demands. This view is similar to word processing in English in that the meaning of a word is simultaneously and independently accessed via direct retrieval of a whole-word representation, which is more sensitive to word frequency, and a decomposition-then-composition process of constituent (character) representations, which is more sensitive to constituent variables (e.g., Baayen et al., 1997; Libben & Jarema, 2007). Other character-level and word-level lexical variables (e.g., word imageability) should be considered in future research to further test the holistic versus analytic nature of Chinese word processing.
Prediction 3: Predictive power of phonological versus semantic variables in speeded naming versus lexical decision
In the Chinese language, the orthography-to-phonology mapping is nearly deterministic (i.e., often in a one-to-one relationship, e.g., 表pronounced as biu2), whereas the orthography-to-semantics mapping is under-deterministic (i.e., often in a one-to-many relationship, e.g., 表can mean watch, express, surface, or meter). Perfetti and Tan (1998, 1999) argued that the one-to-one relation can be more quickly established in Chinese word recognition, so phonology should play a stronger role than semantics in Chinese word processing (see also Tan & Perfetti, 1999, for a model of visual recognition of two-character Chinese words). Tse and Yap (2018) analyzed Tse et al.’s (2017) normed lexical decision data and showed that orthographic and semantic variables, respectively, accounted for more variance than phonological variables, although some could point out that the lexical decision task does not require the generation of word phonology. Based on the R2 or pseudo R2 changes in each step (see Tables 4 and 5), orthographic variables (e.g., character and word frequency) were associated with the strongest predictive power on naming performance, followed by phonological variables; semantic variables accounted for the least variance in our analyses. Phonological variables did account for more variance than semantic variables in predicting naming performance, consistent with Tan and Perfetti’s view. However, the proportion of variance accounted for was still low for phonological variables (smaller than 2.5%) even when the naming task explicitly tapped character phonology, suggesting that the contribution of word phonology might not be as large as what Tan and Perfetti proposed.
Few studies have directly compared the effect of lexical variables on lexical decision versus naming for two-character Chinese words. Consistent with these previous findings (e.g., Gao et al., 2016; Li et al., 2017), we found that word frequency was a stronger predictor for lexical decision performance (−.545) than naming performance (−.151, see Table 5). This is compatible with the view that lexical decision emphasizes more frequency information, an index of whole-word familiarity, when participants discriminate words from nonwords (e.g., Balota & Chumbley, 1984). On the other hand, because participants read aloud the two-character word beginning from the first character, the first-character-focused naming process might reduce the influence of the word frequency effect in Chinese lexical processing. In fact, when comparing the effects of the first versus second character’s lexical variables on lexical decision and naming performance (see Table 5), we did find that the predictive power was stronger for the first character’s variable than the second character’s variable in naming zRT (e.g., −.405 vs. −.210 in character frequency, −.144 vs. .005 in the number of phonemes, and .034 vs. .017 in semantic transparency). In contrast, the difference in predictive power of the first versus second character’s variable was not as large in lexical decision (e.g., −.067 vs. −.065 in character frequency, −.021 vs. −.002 in the number of phonemes, and −.065 vs. −.063 in semantic transparency). The significant predictive power of character variables showed that words are processed not merely holistically (as reflected by word variables), but also analytically (as reflected by character variables). The different patterns of predictive power of the first versus second character’s orthographic, phonological, and semantic variables in the lexical decision and naming tasks suggest that the influence of first versus second character’s variables depends heavily on the task demand, regardless of the lexical characteristics of the characters/words.
The importance of task demands could be observed in the relative contribution of the phonological and semantic variables on naming and lexical decision performance. While orthographic variables accounted for much more variance than semantic and phonological variables in both lexical decision and naming performance, semantic variables accounted for more variance than phonological variables in lexical decision, and phonological variables accounted for more variance than semantic variables in naming. While these are consistent with our Prediction 3, semantic variables accounted for just slightly more variance in lexical decision accuracy (.012) than in naming accuracy (.003) but similar variance when the zRT measure was considered (.011 vs. .012) (see Table 5). This suggests that phonological variables clearly contributed more to naming performance than to lexical decision performance, but the contribution of semantic variables was quite similar in the performance of the two tasks. This was in contrast to those obtained in English, where the influence of semantic variables was more salient in lexical decision than in naming (e.g., Chang et al., 2019; Cortese & Khanna, 2007; Cortese et al., 1997; Yap & Balota, 2009). Semantic variables might have more influence on Chinese naming than on English naming because more than half of Chinese characters have no systematic orthography-to-phonology mappings (Chang et al., 2016), such that readers might rely on semantics when pronouncing Chinese characters (see also Chang & Lee, 2018).
Finally, it is noteworthy that there was a marked contrast in the role of semantic transparency in lexical decision versus naming performance. While semantic transparency facilitated lexical decision, the same variable, especially for the first character, had an inhibitory effect on naming. Following the general dual-route model of compound word processing (e.g., Libben & Jarema, 2007), the meaning of a word can be accessed via direct retrieval from semantic memory or via decomposition, with the meanings of its characters first activated and then combined to obtain the word meaning. The outputs from direct retrieval and decomposition routes are similar for transparent words (e.g., 花園garden). In contrast, for opaque words at least one of the characters is unrelated to the whole word, so there is a conflict between the two routes (e.g., combined meaning of the two unrelated characters, 花flower and 生grow versus meaning of the word, 花生peanut). In lexical decision, this conflict may trigger participants to do post-lexical checking to confirm the word lexicality (see, e.g., Balota & Chumbley, 1984), thereby slowing the word recognition process for opaque, relative to transparent, words and producing the semantic transparency effect (see, e.g., Kim et al., 2019). In naming, there was no need for post-lexical checking, as participants read aloud the word character-by-character, so the conflict in the meaning of whole word and its characters likely did not influence the naming responses. That being said, participants named opaque words faster than transparent words. This negative semantic transparency effect was unexpected and should be further examined in future research.
Comparison with Chang and Lee’s (2020) and Sun et al.’s (2018) findings
Before concluding the current study, two recent studies using the megastudy approach to examine the influence of lexical variables on lexical decision and naming of Chinese characters/words are worth discussing in the context of our current findings.
Using 3314 traditional Chinese single characters as their stimuli, Chang and Lee (2020) examined the predictive power of age of acquisition, as well as semantic variables, in naming and lexical decision of traditional Chinese characters. Contrary to the findings in English (e.g., Yap & Balota, 2009), they found that semantic variables had higher predictive power in naming than in lexical decision. These findings were also incongruent with our current findings that semantic variables had similar predictive power in naming zRT and lexical decision zRT. In fact, semantic variables even accounted for slightly more variance in lexical decision accuracy than in naming accuracy. However, it is noteworthy that phonological and semantic variables were defined differently between the two studies. Indeed, there were only two lexical variables in common in Chang and Lee and the current study.11 Character frequency, defined as an orthographic variable in the current study but a semantic variable in Chang and Lee, was more predictive of naming than lexical decision in both studies. Moreover, the facilitatory effect of character frequency was also consistently obtained in the two studies. On the other hand, we found that number of strokes was more predictive of naming than lexical decision, contrary to the absence of such task differences reported by Chang and Lee.
Apart from little overlap in the lexical variables in the analyses, there were a number of stimulus and procedural differences that might complicate the direct comparison of the findings of the two studies. First, whereas Chang and Lee (2020) used single characters as stimuli, we used two-character words. Second, their participants read aloud the characters in Mandarin, instead of Cantonese as in the current study. Third, phonological variables were quantified based on different dialects (Mandarin and Cantonese) in the two studies. Fourth, as mentioned above, most of the lexical variables were not shared between the two studies. For instance, concreteness, imageability, and age of acquisition were included in Chang and Lee but not in the current study, whereas semantic transparency and neighborhood size were included in the current study but not in Chang and Lee. Hence, future research can be directed at determining whether some of these differences contributed to the above discrepancies in the findings in the two studies.
Unlike Chang and Lee (2020), which focuses on the comparison between lexical decision and naming of single traditional characters, Sun, Hendrix, Ma, and Baayen (2018) included single-character, two-character, three-character, and four-character words in their Chinese lexical database for simplified Chinese and performed analyses on existing lexical decision and naming data in other megastudies (e.g., naming data in Sun, 2016, as cited in Sun et al., 2018 and lexical decision data in Tse et al., 2017 and Tsang et al., 2018). It is noteworthy that the lexical variables in Sun et al. were not the same as those in the current study, as they were based on Mandarin dialect and simplified-script characters, in contrast to Cantonese dialect and traditional-script characters in the current study. Given the difference in script, Sun et al. restricted the evaluation of their measures to the subset of the two-character words in Tse et al. (2017), for which the written form is identical in simplified and traditional Chinese, such that the number of words was reduced from 25,281 to 8005. This limits the number of words in their analyses. Sun et al.’s naming data were provided by a single participant and based on 25,935 two-character words. In the following, we compare the current findings with Sun et al.’s results on single characters and two-character words. We also consider their single-character data, as we did with Chang and Lee and other studies that did not use the megastudy approach.
For single-character lexical decision data, in line with the current study, Sun et al. (2018) reported the facilitatory effect of character frequency and inhibitory effect of number of strokes. For two-character lexical decision data, consistent with the current study, Sun et al. also found the facilitatory effect of word frequency and character frequency, with the former effect being larger than the latter. The effects of first-character and second-character frequency were also similar in magnitude. Although Sun et al. did not obtain a significant effect of stroke count, this could be due to the restricted set of stimuli that was used in their study and/or that the effect size of this variable was rather small, as indeed reflected by their relatively small regression coefficients in our analyses (see Table 5).
For single-character naming data, in line with the current findings, Sun et al. (2018) found a facilitatory effect of character frequency and inhibitory effect of number of strokes. For two-character naming data, Sun et al. also found a facilitatory effect of character frequency and word frequency. However, the effects of character frequency (beta = −.025 for the first character and −.012 for the second character) were similar to or even smaller than the effect of word frequency (beta = −.021), in contrast to the current study in which the effect of first-character or second-character frequency was larger than the effect of word frequency (see Tables 4 and 5). However, the finding that the effect of first-character frequency was stronger than that of second-character frequency was in line with our current findings. Consistent with the current study, Sun et al. also found that the naming responses were slowed with the number of strokes, with the effect being larger for the first characters than the second characters.12
Unlike other studies (including the current one), Sun et al. (2018) used information-theoretic measures to examine how the uncertainty of characters at the word level (e.g., entropy and conditional probability) influenced lexical decision and naming performance. They obtained a facilitatory effect of first-character entropy, suggesting that the greater uncertainty about the second character given the first character could trigger faster lexical decision and naming responses. Although this lexical decision finding was in contrast to the findings in the English language (e.g., Schmidtke et al., 2016; Hendrix et al., 2017), it highlights the importance of taking the combinatorial properties of characters into account when investigating lexical processing above the character level in future studies.
Conclusion and future directions
Motivated by the megastudy approach (e.g., Balota et al., 2013), the current extension of the Chinese Lexicon Project to speeded naming data represents an important addition to the different existing lexicon projects (e.g., English, Dutch, French, and Malay) being rapidly developed across the world and in line with the current research zeitgeist. Given that there are more native speakers of Chinese than any other language and that two-character words are the most common type of word encountered in Chinese reading, a two-character Chinese word speeded naming database can significantly contribute to research in psycholinguistics and other research domains using Chinese word stimuli. In the .xlsx/.csv files available at: https://osf.io/vwnps, we make the item-level data freely accessible to the research community, which allows researchers to search for naming RT, zRT, and accuracy rate for words and descriptive statistics of lexical variables (Table 1) for characters and words. This database serves as a critical resource of lexical characteristics and behavioral measures for future research. Researchers could perform item-level regression analyses to test the higher-order interactions among lexical variables (e.g., word frequency × character frequency interaction, to determine whether character frequency effect might be particularly salient for certain levels of word frequency), the evidence of which might be mixed in previous factorial-design experiments. This may address some theoretical issues in Chinese word processing, such as whether words and their characters are represented at the same level in the Chinese mental lexicon, whether words are accessed as a whole or via their characters, and the extent to which phonological and semantic variables interact to influence Chinese naming, which will have implications on the models of Chinese word recognition (see, e.g., Tse & Yap, 2018, for an example based on Tse et al.’s, 2017, normed lexical decision data).
Researchers could add other variables to our dataset to make it more comprehensive. For instance, previous studies showed that characters with higher phonological frequency (i.e., those with higher cumulative character frequency for their homophone mates) were named more slowly than those with lower phonological frequency (e.g., Ziegler et al., 2000, but see Chen et al., 2009). Chang and Lee (2020) showed that, after controlling for semantic variables, age of acquisition accounted for more variance in naming than in lexical decision for single characters. In addition, this variable interacted with phonological consistency in predicting character naming. Huang et al. (2006) showed that characters with higher summed frequency of all words in which a character occurred were named more slowly than those with lower summed frequency. It is important to test whether all these findings could be generalized in the current dataset (i.e., a large pool of two-character words with traditional script and Cantonese-speaking participants). The present large-scale data for various lexical variables and speeded naming performance could also facilitate the development of computational models of Chinese lexical processing (e.g., Shuai & Malins, 2017; Yang et al., 2009), with their computer simulation providing greater clarity and transparency than traditional descriptive models. Apart from addressing theoretical questions related to word recognition models, the current data, together with those developed by Tse et al. (2017) for lexical decision, could be regarded as normative data for traditional Chinese word recognition by native Cantonese-speaking university students and utilized to build Chinese proficiency tests. Using the findings from the French Lexicon Project (Ferrand et al., 2010), Brysbaert (2013) selected French words and nonwords varying in difficulty level and constructed a short French language proficiency test (LEXTALE-FR), which yielded good psychometric properties and can be used to assess French proficiency for both native and second language learners. Researchers could explore the possibility of developing a Chinese proficiency test with receptive (word recognition) and production (word pronunciation) components, based on lexical decision and speeded naming data, respectively. The large word pool in these databases would make it possible to construct multiple forms of tests that involved different sets of words and nonwords with similar difficulty levels, such that readers, be they first or second language learners, could be assessed without repeating the same set of words.
Open Practices Statement
The data are available at: https://osf.io/vwnps, and none of the experiment was preregistered.
This research was supported by the General Research Fund (#14600019) awarded by the Hong Kong Research Grants Council. We thank Chloe Chow for her help in the data collection.
1 All phonological transcriptions/syllables are Cantonese pronunciations based on the Multi-function Chinese Character Database (see https://humanum.arts.cuhk.edu.hk/Lexis/lexi-can/ or https://humanum.arts.cuhk.edu.hk/Lexis/lexi-mf/ for more details).
2 It is possible that when the word consists of at least one phonologically inconsistent character (i.e., character that can have more than one pronunciation), people might then need to process both characters in order to assign the correct pronunciation of that phonologically inconsistent character. For example, 重can be pronounced as cung4, cung5, or zung6. When pronouncing 重要, 重has to be pronounced as zung6. Thus, people need to process both characters 重and 要of the word 重要, in order to assign correct pronunciations of 重 (i.e., zung6, but not cung4 or cung5).
3 The mean and SD of cumulative grade-point average (max = 4) are based on 233 participants who were not first-year students and/or able to provide this information.
4 All but one of our participants (i.e., N = 296) took the Hong Kong Diploma of Secondary Education Examination (HKDSE) for university entrance, in which their performance is classified by a seven-point scale (5**, 5*, 5, 4, 3, 2, and 1, with the last two being the failing grades). We converted their top two grades 5** and 5* into 7 and 6, respectively, for their overall grades and sub-grades in the Chinese language university entrance exam. The overall grades were based on 294 participants, as two participants did not report that. The speaking sub-grades were based on 194 participants, because this subtest was canceled in 2020 and 2021 due to the COVID-19 pandemic.
5 We thank an anonymous reviewer for this suggestion.
6 Phonological regularity is a three-level variable, so it is coded as two dummy variables (regular = 1 0, irregular = −0.5 0.5, change of tone = −0.5 to 0.5; i.e., the character shares the same syllable but not the same tone with its phonetic radical, e.g., regularity of 情cing4 vs. 青cing1).
7 While some of the characteristics of the initial phoneme of the first character show relatively high variance inflation factors, after excluding those with higher than 10 in the analyses (i.e., affricative, alveolar, fricative, stop, and voiced), we obtained similar findings for all orthographic, phonological, and semantic variables reported in the main text. The only exception was that the second character’s phonological regularity dummy contrast (0 = regular, 0.5 = irregular, −0.5 = change of tone) approached significance (beta = .012), suggesting that zRT was slower when the second character was phonologically irregular (vs. change of tone). In short, we do not consider that the high variance inflation ratio in five of the first character’s initial phoneme variables distorted the pattern of major findings.
8 The larger effect of phonological consistency in zRT for the first character, beta = −.035, than the second character, beta = −.017, was somehow counteracted by the slightly larger effect in accuracy for the second character, beta = .272, than the first character, beta = .266.
9 These discrepancies were attributed to different dependent variables (logit-transformed accuracy rate in the current study versus raw accuracy rate in Tse & Yap, 2018) used in the analyses. Another difference was the inclusion of characteristics of the initial phoneme of the first character in the analyses. We reran another set of regression analyses, which was the same as reported in the main text but excluding the characteristics of the initial phoneme of the first character. The results of those analyses were qualitatively the same as those that included characteristics of the initial phoneme of the first character, except that lexical decision responses were slower when the first character consisted of a greater number of meanings.
10 Given the moderate correlation between neighborhood size and character frequency (+.562 and +.538, see Table 3), characters with more neighbors are often of high character frequency, so one could argue that neighborhood size might not be a pure semantic variable. In the current analyses, we found that first-character and second-character frequency and first-character and second-character neighborhood size accounted for unique variance in naming performance. Nonetheless, in future research it will be important to examine the interaction between neighborhood size and character frequency and test whether these two variables influence the common stage of Chinese word processing during naming, based on Sternberg’s (1967) additive factors logic (see Yap et al., 2008, for an example).
11 Phonological consistency was defined at the sublexical level in Chang and Lee (2020)—that is, the number of friends (characters sharing the same phonetic radical and pronunciation) divided by the total number of characters sharing the same phonetic radical—whereas in the current study, we define that as whether a character has one (phonologically consistent) or more than one (phonologically inconsistent) pronunciation. For phonological regularity, Chang and Lee defined the characters with unpronounceable phonetic radical as a separate group, whereas in the current study they were defined as irregular characters. They did not separate the characters that shared the same syllable but not the same tone with their phonetic radical (i.e., “change of tone” characters), as in the current research. Finally, although Chang and Lee defined semantic ambiguity as the number of meanings of a character, they quantified that based on subjective ratings rather than raw count of distinct conceptual representations of a character, as in the current study. Hence, we do not compare the findings for these variables across the two studies.
12 The orthography-to-phonology mapping measures in Sun et al. (2018) were not defined in the same way as in the current study, e.g., phonological consistency and orthography-to-phonology mapping consistency). Specifically, Sun et al. defined the “friend of a character” as an occurrence of the same character–pronunciation mapping in a different word. It might be a bit similar to the neighborhood size in the current study, although we count the occurrence of the character in different words, regardless of whether the character was pronounced similarly or differently. For instance, 重要, 重量, and 重陽 are all counted toward the neighborhood size, even though 重 is pronounced differently in these three words. Nevertheless, similar to the facilitatory effect of the first-character friends on single-character lexical decision and naming data of Sun et al. (i.e., faster responses for those with more first-character friends), we found that words with larger first-character neighborhood size yielded faster responses than those with smaller first-character neighborhood size.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Andrews S Heathcote A Distinguishing common and task-specific processes in word identification: A matter of some moment? Journal of Experimental Psychology: Learning, Memory, and Cognition 2001 27 514 540 11294447
Baayen RH Dijkstra T Schreuder R Singulars and plurals in Dutch: Evidence for a parallel dual route model Journal of Memory and Language 1997 37 94 117 10.1006/jmla.1997.2509
Baayen RH Feldman L Schreuder R Morphological influences on the recognition of monosyllabic monomorphemic words Journal of Memory and Language 2006 53 496 512
Balota DA Chumbley JI Are lexical decisions a good measure of lexical access? The role of word frequency in the neglected decision stage Journal of Experimental Psychology: Human Perception and Performance 1984 10 340 357 6242411
Balota DA Chumbley JI The locus of word-frequency effects in the pronunciation task: Lexical access and/or production? Journal of Memory & Language 1985 24 89 106 10.1016/0749-596X(85)90017-8
Balota DA Cortese MJ Sergent-Marshall S Spieler DH Yap MJ Visual word recognition of single-syllable words Journal of Experimental Psychology: General 2004 133 283 316 10.1037/0096-3445.133.2.283 15149254
Balota, D. A., Yap, M. J., Cortese, M. J., Hutchison, K. A., Kessler, B., Loftis, B., ... & Treiman, R. (2007). The English Lexicon Project. Behavior Research Methods, 39, 445–459.
Balota DA Yap MJ Hutchison KA Cortese MJ Adelman JS Megastudies: What do millions (or so) of trials tell us about lexical processing? Visual word recognition 2013 Psychology Press 90 115
Barca L Burani C Arduino LS Word naming times and psycholinguistic norms for Italian nouns Behavior Research Methods, Instruments, & Computers 2002 34 424 434 10.3758/BF03195471
Brysbaert M LEXTALE-FR: A fast, free, and efficient test to measure language proficiency in French Psychologica Belgica 2013 53 23 37 10.5334/pb-53-1-23
Cai Q Brysbaert M SUBTLEX-CH: Chinese word and character frequencies based on film subtitles PLoS One 2010 5 e10729 10.1371/journal.pone.0010729 20532192
Chang Y-N Hsu C-H Tsai J-L Chen C-L Lee C-Y A psycholinguistic database for traditional Chinese character naming Behavior Research Methods 2016 48 112 122 10.3758/s13428-014-0559-7 25630311
Chang Y-N Lee C-Y Semantic ambiguity effects on traditional Chinese character naming: A corpus-based approach Behavior Research Methods 2018 50 2292 2304 10.3758/s13428-017-0993-4 29124717
Chang Y-N Lee C-Y Age of acquisition effects on traditional Chinese character naming and lexical decision Psychonomic Bulletin & Review 2020 27 1317 1324 10.3758/s13423-020-01787-8 32789580
Chang Y-N Monaghan P Welbourne S A computational model of reading across development: Effects of literacy onset on language processing Journal of Memory and Language 2019 108 104025 10.1016/j.jml.2019.05.003
Chen H-C Vaid J Wu J-T Homophone density and phonological frequency in Chinese word recognition Language and Cognitive Processes 2009 24 967 982 10.1080/01690960902804515
Coltheart M Rastle K Perry C Langdon R Ziegler J DRC: A Dual Route Cascaded model of visual word recognition and reading aloud Psychological Review 2001 108 204 256 10.1037/0033-295X.108.1.204 11212628
Cortese MJ Khanna MM Age of acquisition predicts naming and lexical-decision performance above and beyond 22 other predictor variables: an analysis of 2,342 words Quarterly Journal of Experimental Psychology 2007 60 1072 1082 10.1080/17470210701315467
Cortese MJ Simpson GB Woolsey S Effects of association and imageability on phonological mapping Psychonomic Bulletin & Review 1997 4 226 231 10.3758/BF03209397 21331829
Faust ME Balota DA Spieler DH Ferraro FR Individual differences in information processing rate and amount: Implications for group differences in response latency Psychological Bulletin 1999 125 777 799 10.1037/0033-2909.125.6.777 10589302
Ferrand, L., New, B., Brysbaert, M., Keuleers, E., Bonin, P., Méot, A., ... & Pallier, C. (2010). The French Lexicon Project: Lexical decision data for 38,840 French words and 38,840 pseudowords. Behavior Research Methods, 42, 488–496.
Ferrand L Méot A Spinelli E New B Pallier C Bonin P Grainger J MEGALEX: A megastudy of visual and auditory word recognition Behavior Research Methods 2018 50 1285 1307 10.3758/s13428-017-0943-1 28791657
Gao X-Y Li M-F Chou T-L Wu J-T Comparing the frequency effect between the lexical decision and naming tasks in Chinese Journal of Visual Experiment 2016 110 e53815
Goh WD Yap MJ Chee QW The Auditory English Lexicon Project: A multi talker, multi-region psycholinguistic database of 10,170 spoken words and nonwords Behavior Research Methods 2020 52 2202 2231 10.3758/s13428-020-01352-0 32291734
Hendrix P Bolger P Baayen RH Distinct ERP signatures of word frequency, phrase frequency and prototypicality in speech production Journal of Experimental Psychology: Language, Memory, and Cognition 2017 3 128 149
Hoosain R Chen HC Tzeng OJL Psychological reality of the word in Chinese Language processing in Chinese 1992 Elsevier Science 111 130
Huang HW Lee CY Tsai JL Lee CL Hung DL Tzeng OJL Orthographic neighborhood effects in reading Chinese two-character words Neuroreport 2006 17 1061 1065 10.1097/01.wnr.0000224761.77206.1d 16791104
Hue C-W Chen H-C Tzeng OJ-L Recognition processes in character naming Advances in Psychology 1992 90 93 107 10.1016/S0166-4115(08)61888-9
Institute of Language Teaching and Research A frequency dictionary of Modern Chinese 1986 Beijing Language Institute Press
Jaccard J Turrisi R Interaction effects in multiple regression 2003 Sage
Kang, S. H. K., Yap, M. J., Tse, C.-S., & Kurby, C. A. (2011). Semantic size does not matter: “Bigger” words are not recognised faster. Quarterly Journal of Experimental Psychology, 64, 1041–1047.
Keuleers, E., Diependaele, K., & Brysbaert, M. (2010). Practice effects in large-scale visual word recognition studies: A lexical decision study on 14,000 Dutch mono- and disyllabic words and nonwords. Frontiers in Psychology, 1, 174. 10.3389/fpsyg.2010.00174
Keuleers E Lacey P Rastle K Brysbaert M The British Lexicon Project: Lexical decision data for 28,730 monosyllabic and disyllabic English words Behavior Research Methods 2012 44 287 304 10.3758/s13428-011-0118-4 21720920
Kim SY Yap MJ Goh WD The role of semantic transparency in visual word recognition of compound words: A megastudy approach Behavior Research Methods 2019 51 2722 2732 10.3758/s13428-018-1143-3 30291593
Leong C-K Cheng P-W Consistency effects on lexical decision and naming of two-character Chinese words Reading and Writing 2003 16 455 474 10.1023/A:1024243507278
Leong C-K Cheng P-W Mulcahy R Automatic processing of morphemic orthography by mature readers Language and Speech 1987 30 181 196 10.1177/002383098703000207 3505318
Li M-F Gao X-Y Chou T-L Wu J-T Neighborhood frequency effect in Chinese word recognition: Evidence from naming and lexical decision Journal of Psycholinguistic Research 2017 46 227 245 10.1007/s10936-016-9431-5 27119658
Li X Zang C Liversedge SP Pollatsek A Pollatsek A Treiman R The role of words in Chinese reading The Oxford handbook of reading 2015 Oxford University Press 232 244
Libben G Jarema G The representation and processing of compound words 2007 Oxford University Press
Lim Y-H Yap MJ Tse C-S Individual differences in Cantonese Chinese word recognition: Insights from the Chinese Lexicon Project Quarterly Journal of Experimental Psychology 2020 73 504 518 10.1177/1747021820906566
Liu Y Shu H Li P Word naming and psycholinguistic norms: Chinese Behavior Research Methods 2007 39 192 198 10.3758/BF03193147 17695344
Myers J Libben G Jarema G Processing Chinese compounds: A survey of the literature The representation and processing of compound words 2006 Oxford University Press 169 196
Myers J Libben G Derwing B The nature of transparency effects in Chinese compound processing Poster presented at the Fourth International Conference on the Mental Lexicon 2004
Packard JL Lexical access in Chinese speech comprehension and production Brain and Language 1999 68 89 94 10.1006/brln.1999.2102 10433744
Packard JL The morphology of Chinese: a linguistic and cognitive approach 2000 Cambridge University Press
Peng D Deng Y Chen B The polysemy effect in Chinese one-character word identification Acta Psychologica Sinica 2003 35 569 575
Perfetti CA Tan LH The time course of graphic, phonological, and semantic activation in Chinese character identification Journal of Experimental Psychology: Learning, Memory, and Cognition 1998 24 101 118
Perfetti CA Tan LH Inhoff A Chen H-C Wang J The constituency model of Chinese word identification Reading Chinese script: A cognitive analysis 1999 Erlbaum 115 134
Perry C Ziegler JC Zorzi M Nested incremental modeling in the development of computational theories: the CDP+ model of reading aloud Psychological Review 2007 114 273 315 10.1037/0033-295X.114.2.273 17500628
Que DL Longman Chinese Dictionary 2008 3 Longman
Roux F Armstrong B Carreiras M Chronset: An automated tool for detecting speech onset Behavior Research Methods 2016 49 1864 1881 10.3758/s13428-016-0830-1
Schmidtke D Kuperman V Gagne C Spalding T Competition between conceptual relations affects compound recognition: The role of entropy Psychonomic Bulletin & Review 2016 23 556 570 10.3758/s13423-015-0926-0 26340846
Schneider W Eschman A Zuccolotto A E-prime user’s guide 2001 Psychology Software Tools
Shen LM Zhu X Investigating stroke count and character frequency effects in the visual recognition of characters Psychological Science (in Chinese) 1994 17 245 247
Shuai L Malins JG Encoding lexical tones in jTRACE: A simulation of monosyllabic spoken word recognition in Mandarin Chinese Behavior Research Methods 2017 49 230 241 10.3758/s13428-015-0690-0 26850055
Spieler DH Balota DA Bringing computational models of word naming down to the item level Psychological Science 1997 8 411 416 10.1111/j.1467-9280.1997.tb00453.x
Sun, C. C., Hendrix, P., Ma, J., & Baayen, R. H. (2018). Chinese lexical database (CLD): A large-scale lexical database for simplified Mandarin Chinese. Behavior Research Methods, 50, 2606–2629.
Sze WP Rickard Liow SJ Yap MJ The Chinese Lexicon Project: A repository of lexical decision behavioral responses for 2,500 Chinese characters Behavior Research Methods 2014 46 263 273 10.3758/s13428-013-0355-9 23813237
Sze WP Yap MJ Rickard Liow SJ The role of lexical variables in the visual recognition of Chinese characters: A megastudy analysis Quarterly Journal of Experimental Psychology 2015 68 1541 1570 10.1080/17470218.2014.985234
Tan LH Perfetti CA Phonological activation in visual identification of Chinese two-character words Journal of Experimental Psychology: Learning, Memory, and Cognition 1999 25 382 393
Tsai J-L Lee C-Y Lin Y-C Tzeng OJ-L Hung DL Neighborhood size effects of Chinese words in lexical decision and reading Language and Linguistics 2006 7 659 675
Tsang Y-K Chen H-C Activation of morphemic meanings in processing opaque words Psychonomic Bulletin and Review 2014 21 1281 1286 10.3758/s13423-014-0589-2 24510473
Tsang Y-K Huang J Lui M Xue M Chan Y-WF Wang S Chen H-C MELD-SCH. A megastudy of lexical decision in simplified Chinese Behavior Research Methods 2018 50 1763 1777 10.3758/s13428-017-0944-0 28779457
Tse C-S Yap MJ Chan Y-L Sze WP Shaoul C Lin D The Chinese Lexicon Project: A megastudy of lexical decision performance for 25,000+ traditional Chinese two-character compound words Behavior Research Methods 2017 49 1503 1519 10.3758/s13428-016-0810-5 27734329
Tse C-S Yap MJ The role of lexical variables in the visual recognition of two-character Chinese compound words: A megastudy analysis Quarterly Journal of Experimental Psychology 2018 71 2022 2038 10.1177/1747021817738965
Xing H Shu H Li P The acquisition of Chinese characters: Corpus analyses and connectionist simulations Journal of Cognitive Science 2004 5 1 49
Yang JF McCandliss BD Shu H Zevin JD Simulating language-specific and language-general effects in a statistical learning model of Chinese reading Journal of Memory & Language 2009 61 238 257 10.1016/j.jml.2009.05.001 20161189
Yap MJ Balota DA Visual word recognition of multisyllabic words Journal of Memory & Language 2009 60 502 529 10.1016/j.jml.2009.02.001
Yap MJ Balota DA Tse C-S Besner D On the additive effects of stimulus quality and word frequency in lexical decision: Evidence for opposing interactive influences revealed by RT distributional analyses Journal of Experimental Psychology: Learning, Memory and Cognition 2008 34 495 513 18444751
Yap MJ Balota DA Sibley DE Ratcliff R Individual differences in visual word recognition: Insights from the English Lexicon Project Journal of Experimental Psychology: Human Perception and Performance 2012 38 53 79 21728459
Yap MJ Rickard Liow SJ Cook V Ryan D Processing the written word Routledge handbook of English writing system 2016 Routledge 453 469
Yap MJ Rickard Liow SJ Jalil SB Faizal SSB The Malay Lexicon Project: A database of lexical statistics for 9,592 words Behavior Research Methods 2010 42 992 1003 10.3758/BRM.42.4.992 21139166
Yarkoni T Balota DA Yap MJ Beyond Coltheart’s N: A new measure of orthographic similarity Psychonomic Bulletin & Review 2008 15 971 979 10.3758/PBR.15.5.971 18926991
Yu B Cao H A new exploration on the effect of stroke number in the identification of Chinese characters Acta Psychologica Sinica 1992 24 2 10 16
Ziegler JC Tan LH Perry C Montant M Phonology matters: the phonological frequency effect in written Chinese Psychological Science 2000 11 234 238 10.1111/1467-9280.00247 11273409
| 36443581 | PMC9707223 | NO-CC CODE | 2022-12-01 23:20:21 | no | Behav Res Methods. 2022 Nov 28;:1-21 | utf-8 | Behav Res Methods | 2,022 | 10.3758/s13428-022-02022-z | oa_other |
==== Front
Spat. Inf. Res.
Spatial Information Research
2366-3286
2366-3294
Springer Nature Singapore Singapore
498
10.1007/s41324-022-00498-7
Article
Spatial variations in COVID-19 risk perception and coping mechanism in Pakistan
http://orcid.org/0000-0002-3157-1186
Rana Irfan Ahmad [email protected]
[email protected]
1
Bhatti Saad Saleem [email protected]
2
Ahmad Junaid [email protected]
3
Aslam Atif Bilal [email protected]
4
Jamshed Ali [email protected]
5
1 grid.412117.0 0000 0001 2234 2376 Department of Urban and Regional Planning, School of Civil and Environmental Engineering (SCEE), National University of Sciences and Technology (NUST), H-12 Sector, 44000 Islamabad, Pakistan
2 grid.12641.30 0000000105519715 School of Geography and Environmental Sciences, Ulster University, Coleraine, UK
3 Prime Institute of Public Health, Peshawar Medical College, Peshawar, Pakistan
4 grid.444938.6 0000 0004 0609 0078 Department of City and Regional Planning, University of Engineering and Technology, Lahore, Pakistan
5 grid.5719.a 0000 0004 1936 9713 Institute of Spatial and Regional Planning (IREUS), University of Stuttgart, Stuttgart, Germany
29 11 2022
113
17 6 2022
5 11 2022
7 11 2022
© The Author(s), under exclusive licence to Korea Spatial Information Society 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The outbreak of novel coronavirus disease (COVID-19) was declared a pandemic by the World Health Organization, which instigated governments to impose lockdowns across their countries. Amidst the lockdown in Pakistan, this study comprised measures of the COVID-19 risk perception, coping mechanism, and spatial variations. The data from 40 selected indicators was collected using an online questionnaire and grouped into domains (4 risk perception and 3 coping mechanisms domains). The results revealed the spatial variations and the levels of risk perception and coping mechanisms within the study area. Relative to each other, overall risk perception was highest in Northern Areas (Gilgit-Baltistan and Azad Jammu and Kashmir) and Islamabad, and lowest in Balochistan province. Very little spatial variation was observed in terms of coping mechanisms. Age, gender, and marital status influenced the risk perception associated with COVID-19. The findings suggest spatial variation in risk perception, implying the need for localized and modified COVID-19 risk communication and risk reduction strategies.
Supplementary Information
The online version contains supplementary material available at 10.1007/s41324-022-00498-7.
Keywords
Community response
Coronavirus
Pandemic
Risk communication
Risk management
==== Body
pmcIntroduction
The recent worldwide outbreak of the coronavirus disease (COVID-19) has put forth discussions among the public, policymakers, scientists, and country heads about pandemics and their direct and indirect impacts on human lives. The novel COVID-19 was declared a pandemic by the World Health Organization (WHO) on 11 March 2020. According to the WHO, COVID-19 is an infectious disease that causes many cases of mild to moderate respiratory illness. At the same time, those with underlying health conditions or older age can develop severe conditions [1]. Since it mainly spreads through respiratory droplets and contact routes [2, 3], the governments advised people to maintain social distancing (stay approximately 2 m apart) when outside – some countries also imposed complete or partial/smart lockdowns bringing everyday life to almost a complete halt.
The first COVID-19 case was officially reported by China [4]. As of August 2022, 600 million confirmed cases and 6,484,700 associated deaths had been reported worldwide due to COVID-19. Over 12 billion vaccine doses have been administered [5]. To counter such a rapid spread of infection worldwide, a greater emphasis has been placed on mass screening, identification of cases in the community, isolating them, and quarantining the cases to prevent further spread [6, 7]. The strategy to contain COVID-19 with the social distancing and lockdown of cities and countries has impacted people, organizations, and governments in unprecedented ways [8–10]. Businesses have been shut down, people lost their jobs, and the whole supply chain has been disrupted [11]. But the impact has not been uniform for all people, as different people have different physical and psychosocial capacities [12]. However, the direct and indirect impacts of lockdowns are not entirely clear and are still unfolding.
COVID-19 has infiltrated almost every country, and Pakistan has not been an exception. The country is spatially located adjacent to the infection’s epicenters, i.e., The Peoples’ Republic of China and the Islamic Republic of Iran. Pakistan reported its first confirmed COVID-19 cases, imported from Iran, on 26 February, 2020 [13, 14]. Both cases had a travel history to the Islamic Republic of Iran. To date (as of 27 August 2022), 1,568,183 confirmed cases and 30,571 COVID-19 deaths had been reported, while 291,878,652 vaccine doses have been administered [13]. As a lockdown strategy was used to “flatten the curve”, provincial governments under the federal government’s ambit initiated the safety protocols. As the number of cases jumped, the provincial government of Sindh announced the lockdown on 23 March 2020, initially for 14 days [15]. To contain the further spread, the provincial governments of Punjab, Khyber Pakhtunkhwa, and Balochistan also went into lockdown on 24th and 25th March 2020 [16]. However, it was soon realized that community cooperation is vital for a successful lockdown.
Risk perception is widely used in disaster risk and climate change adaptation literature to examine individuals, communities, and nations regarding predisposition to accept government policies for a potential hazard [17]. It is also used for predicting the willingness of communities to undertake precautionary measures [18]. In the case of COVID-19, it has been asserted to assess risk perception to support a more effective response [19]. Similarly, for safeguarding mental health, coping mechanisms need to be developed. Despite well-proven knowledge about preventive measures, people still perceive the risk of infectious diseases differently. The level of risk perception varies significantly amongst different individuals. Hence, it becomes imperative to study the risk perception and coping mechanisms to mitigate the negative impacts of COVID-19, and design effective risk communication and reduction strategies. There is also a need to understand how risk perception may vary spatially. Examining the risk perception of the public in the pre-vaccination phase, this study aims to measure COVID-19 risk perception and coping mechanisms, and their spatial variations in Pakistan.
Risk perception and pandemics
The scientific community classified diseases as endemic, epidemic, and pandemic [20]. Endemic diseases are the one that continuously stays in a region all over the year, like tuberculosis and hepatitis in Pakistan, or human immunodeficiency virus (HIV) in African countries. Epidemics are diseases that rapidly increase in a specific region at a particular time, and if the epidemic spreads on a global scale, it is called a pandemic [21]. This classification is primarily based on the occurrence and spreading over a geographical area [20]. The word ‘pandemic’ comes from the Greek words; “pan” which means “all” and “demos” mean “people” [22]. However, pandemics are not something new to humanity. In the past, major pandemics happened in different parts of the world, resulting in an unbelievable number of people losing their lives. A few prominent pandemics are the black death (1331 to 1353), the third plague pandemic (1855), and the Spanish flu (1918 to 1920) [23]. However, the risk of pandemics was never as high in the past as it is in the current age of time. In today’s world, humans can travel thousands of miles within hours with multiple layovers. Traveling provides a decent medium for propagating viral infections from one part of the world to another in no time. COVID-19 is a recent example, as this took just four months from the first confirmed case to four million cases.
Risk perception is an important concept that helps gauge the community or individual judgment, reaction, or acceptance regarding an event [24]. It is defined as “…beliefs about potential harm or the possibility of a loss. It is a subjective judgment that people make about the characteristics and severity of a risk” [25]. Risk perception has been massively studied in the fields of psychology, social science, disaster risk, and climate change and has gained much importance in pandemic-related dissertations. In the case of a pandemic, risk perception can be imperative in assessing how severely it could impact a society [26]. The study further highlighted that those individual perceptions and behaviors in case of an infectious disease outbreak, such as staying home, limiting social interaction, taking medicine, etc., would reduce the infection risk at an individual and community levels. People with high-risk perceptions are likely to comply with relevant guidelines in case of an infection outbreak, which therefore lowers their probability of getting affected by a pandemic [27, 28]. Moreover, knowing how risk is perceived is crucial for formulating an effective plan for communicating risk and controlling the outbreak of pandemics [29, 30].
Several studies have investigated the risk perception of communities toward pandemics. A study examined the factors associated with risk perception of pandemic influenzas in Australia, and found location (urban or rural), language, age, and income being crucial in shaping risk perception [30]. Another study explored the dynamics of risk perceptions and precautionary behavior in response to the 2009 (H1N1) pandemic influenza in the United States [26]. They found that household size, gender, and geographical aspects were important in perceiving risk and participating in precautionary activities. While studying the risk perception of the adult population to Avian Influenza in Italy, a study found higher risk perception for less educated people with lower socioeconomic status[28]. Similar results were highlighted in the case of severe acute respiratory syndrome (SARS) risk perception in the Netherlands [31]. The study mentioned that higher risk perceptions were associated with more worry and self-reported precautionary actions. Another study observed the effect of risk perception on the dynamics of the H1N1 pandemic. The study found that mass media campaigns affect risk perception and behavior changes during a pandemic.
In the case of COVID-19, a recent trend in risk perception has started emerging. A study examined the role of socioeconomic factors and social media use on risk perception [32]. This study revealed that social media might contribute to unwarranted fear or overly pessimistic risk perception. Zeballos Rivas et al.(2021) indicated that high social media exposure leads to higher risk perception in Bolivia [33]. Using longitudinal analysis in the UK, Schneider et al. (2021) suggested that socioeconomic factors, direct experience, trust in government, science, and medical professionals, as well as personal and collective efficacy, influence the risk perception of COVID-19 [34]. Gerhold (2020) studied COVID-19 risk perception and coping strategies of people in Germany [35]. The results show that younger people perceive risk more than older people, while men are less concerned about COVID-19 than women. Zeballos Rivas et al.(2021) found the same in Bolivia [33]. A recent study in sub-Saharan Africa shows young people have a lower risk perception of COVID-19 [36]. Similarly, a study in the Netherlands shows that young adults perceive low risk where affective response for their well-being is also low, but is higher for other vulnerable people leading them to adhere to most preventive guidelines frequently [37]. Researchers studied the risk perception of COVID-19 among pharmacists and suggested that the role of media, gender, living in a city, and having children were all associated with an increased perception of COVID-19 risk [38]. Another study examined the risk perception among the public in Finland by analyzing multiple factors and providing recommendations for meaningful risk perception [39]. A study in Pakistan also highlighted the gender differences in COVID-19 risk perception [40]. Another important factor is vaccination which could influence risk perception and vice versa. Recent studies suggested that higher risk perception leads to acceptance and willingness to get vaccinated against COVID-19 in South Carolina [41], Italy [42], and the Netherlands [43]. Despite significant research on COVID-19, studies on the spatial aspect of COVID-19 risk perception and coping mechanisms are limited and require investigation.
Methods
This study relies on the primary data collected through an online questionnaire survey – the survey was designed to capture several aspects of COVID-19 risk perception, where one of the aspects, gender perspective, has been detailed in an earlier study [40]. The current study, touching upon the topic from a different dimension, presents the spatial interpretation of COVID-19 risk perception and coping mechanisms in Pakistan using the same data/indicators. The respondents participated voluntarily in the online survey. The purpose of using the collected data for academic and research purposes was clearly communicated, and consent was sought. The data collected was completely anonymous, and no personally identifiable information was obtained. It is also important to mention that the data was collected before any vaccination was developed.
This study assessed two components: (1) perception of risk related to COVID-19; and (2) behavioral approaches to cope with this pandemic in Pakistan. Apart from examining the relationship between different socioeconomic and risk perception indicators, this study also investigates the spatial variability in risk perception by grouping and analyzing the data from various administrative regions. The official figures on the number of COVID-19-related cases and deaths were also acquired and used to explain any relationship between risk perception (and its spatial variation) and the on-ground pandemic situation.
This study was conducted in Pakistan, which covers an area of around 796,096 sq. km. The country is administratively divided into four provinces, two autonomous territories, and one capital territory. In terms of area, the largest province is Balochistan, and the smallest is Islamabad Capital Territory (ICT). The study area map is presented in Fig. 1.
Fig. 1 Map of the study area
Questionnaire design and indicators
A detailed questionnaire was developed to collect the data from individual respondents within four dimensions: (1) socioeconomic characteristics, (2) public health emergency experience, (3) COVID-19-related risk perception, and (4) behavioral approaches to cope with the adverse impacts of this pandemic. The geographical location of the respondents was also attached to each response through a question asking the city name of the respondent. Table 1 gives an overview of the selected dimensions and domains. To describe the socioeconomic characteristics of the respondents, ten indicators - age, gender, education, marital status, household size, type of family, monthly income, profession, number of children/teenagers in the house (< 18 years old), and number of old people in the house (> 60 years old) were selected. The experience of respondents with public health emergencies was assessed through four indicators, namely isolation days (lockdown days), sources of information regarding COVID-19, storing food and other essential items, and past experience of dealing with any public health emergency.
Table 1 The various dimensions and indicators of risk perception established for this study
Dread, Fear, and Worry
F-1 How much are you afraid for your life from a COVID-19 infection?
F-2 How likely do you think to get a COVID-19 infection?
F-3 How much are you afraid that COVID-19 will persist/spread in the future?
F-4 How much are you worried that this pandemic will disturb your daily lifestyle?
F-5 How much are you worried that this pandemic is dangerous for your family?
F-6 How much do you think this pandemic is a danger for your community?
F-7 How much afraid are you about this pandemic based on current knowledge?
F-8 What are the chances of supply interruption during this pandemic?
Behavior and Attitude
B-1 How much do you think can deal with the consequences of this pandemic? *
B-2 How much can you adapt to lifestyles because of this pandemic?
B-3 How much do you think this pandemic can change your relationship with your neighbors and relatives?
Awareness and Knowledge
A-1 How much are you familiar with precautionary measures against this pandemic?
A-2 How much is your community protected from this pandemic? *
A-3 How much do you agree that COVID-19 is completely new for all of us?
Trust and Confidence
T-1 How much do you trust the information provided by the government about this pandemic?
T-2 How much do you rely on/trust the information about the pandemic obtained from different sources? *
T-3 How much do you trust disaster management agencies to deal with this pandemic?
T-4 How much do you trust emergency management policies to deal with this pandemic?
T-5 How much do you trust the information provided by your government about this pandemic?
T-6 How much do you trust the response provided by your government?
T-7 How much do you trust that science and experts will develop the vaccine within one year?
Scale (1-very low to 5-very high); * Reversed in analysis.
The data to assess risk perception was collected in four domains; (1) dread, fear, and worry (8 indicators); (2) behavior and attitude (3 indicators); (3) awareness and knowledge (3 indicators); and (4) trust and confidence (7 indicators). The behavioral attitude to cope with the COVID-19 pandemic was assessed in terms of three different mechanisms (Table 2), namely problem-oriented (6 indicators), emotion-oriented (7 indicators), and action-oriented (6 indicators).
Table 2 The various dimensions and indicators of coping mechanisms established for this study
Problem-oriented
P-1 I feel safe in my own home
P-2 I listen to the experts and follow their advice
P-3 I actively seek out new information about the current situation.
P-4 I am doing something completely new that I would never have done in other circumstances
P-5 I talk to someone who knows about it
P-6 I am seeking financial support from the government
Emotion-oriented
E-1 I turn to my work or other activities to distract myself
E-2 I actively seek meditation to calm myself
E-3 It will emerge over time; there is nothing more to do but wait
E-4 I hope for a miracle
E-5 I try to make myself feel better by eating, smoking, or taking medication
E-6 I refrain from things that can trigger bad moods
E-7 I refuse to believe what is happening
Action-oriented
BE-1 I wash my hands more than usual
BE-2 I avoid going out unnecessarily
BE-3 I avoid public spaces and transport
BE-4 I have bought disinfectants (soap, sanitizers, etc.) more than usual
BE-5 I have bought staple foods (flour, rice, lentils, meat, etc.) more than usual
BE-6 I have bought protective equipment (masks, gloves, etc.) more than usual
Scale: (1-strongly disagree to 5-strongly agree)
Sampling and data collection
The data was collected through a voluntary response technique, whereas a non-probability approach was used for sampling. The online survey was conducted between 7 and 16 March 2020 to collect the data for the aforementioned dimensions/domains/indicators. The majority of the responses were received from urban areas of the country. In addition, data on COVID-19-related cases and deaths reported by official sources were also collected to understand and explain the risk perception-related findings of this study.
Data analysis
The data analysis mainly comprised descriptive and statistical tests. Firstly, the socioeconomic characteristics and public health experience of sampled respondents were analyzed using descriptive statistics. Secondly, the indices of risk perception and coping mechanisms were developed using the aforementioned indicators. The index was calculated using the mean average method for all domains and the overall value. Each average value was tabulated and visualized for region-wise comparisons, and the analysis of variance (ANOVA) test (F-test) was performed to analyze significant differences among them. Lastly, a Pearson’s correlation was performed on the overall index to suggest a relationship between socioeconomic characteristics and public health emergency experience with risk perception and coping mechanisms. Table 3 summarizes the dimensions and domains used for assessment and statistical tests.
Table 3 Checklist of assessment and analyses of responses in different dimensions and domains
Data Assessment and General Analysis Correlation Analysis
Dimension Domain Descriptive statistics of data Index development and computation Spatial variability assessment (region-wise)* Correlation with risk perception Correlation with coping mechanisms
Socioeconomic characteristics - ✓ - ✓ ✓ ✓
Public health emergency experience - ✓ ✓ ✓ ✓
Risk perception - ✓ ✓ N/A -
Dread, fear, and worry - ✓ ✓ N/A -
Behavior and attitude - ✓ ✓ N/A -
Awareness and knowledge - ✓ ✓ N/A -
Trust and confidence - ✓ ✓ N/A -
Coping mechanisms ✓ - ✓ - N/A
Problem-oriented ✓ - ✓ - N/A
Emotion-oriented ✓ - ✓ - N/A
Action-oriented ✓ - ✓ - N/A
* Data grouped into six administrative regions: Punjab, Sindh, Khyber Pakhtunkhwa (KPK), Balochistan, Northern Areas (AJK + GB), and Islamabad. N/A: Not applicable
The index was constructed using the average weighted index method (Eq. 1). The selected indicators of risk perception and coping mechanisms were merged to form an index (Eqs. 2 and 3).\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$CI=({W}_{1}+{W}_{2}+{W}_{3}+ ...{W}_{n})/n$$\end{document} (Eq. 1)
=\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sum _{i=1}^{n}Wi/n$$\end{document}
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Overall risk perception=\frac{Fear+ Behavior+Awareness+Trust}{4}$$\end{document} (Eq. 2)
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Overall coping mechanism =\frac{Problem+ Emotion+Action}{3}$$\end{document} (Eq. 3)
Results and discussion
A total of 379 individual responses were collected from all over Pakistan. Given the vast variability in social norms and cultural traditions across the country, the data and findings must be interpreted cautiously. Table 4 shows descriptive statistics of the sampled population. Since the data was collected through an online survey, all of the respondents were mostly young (not many people in the older age group use the internet in the country) and educated (most of the people agreeing to complete the questionnaire had a good understanding of this study and possessed high education level). Descriptive statistics revealed that the respondents were 27.62 years old on average – most were young adults (aged 19–28). Around 63% of the total respondents were males and the rest females, whereas, in terms of marital status, around 68% were single, 30% married, and the rest were either divorced/widowed or did not prefer to disclose. The average household size was around 7. The majority of respondents earned more than PKR 25,001 a month (around 23% earned PKR 25,001-50,000, 29% earned PKR 50,001-100,000 and 27% earned more than 100,000 per month). The most common family type was nuclear (63.9%), and most respondents had at least 1–3 children (51.7%) in their households. However, most of the respondents (59.6%) did not have any older person (60 years old or above) living with them. In the case of past public health emergency experiences, about 8% of the respondents have faced it. These public health emergencies included prevalent diseases like tuberculosis, hepatitis, and poliovirus. In the current crisis, most respondents (82.8%) had stored food for less than one-month duration. Similarly, respondents were from all over the country, with various dates for the imposition of lockdowns in their respective areas. A different response was received on the number of lockdown days observed. At the time of the survey, the majority of the respondents (54.2) were under lockdown for a period of 14–21 days.
Table 4 Descriptive statistics of samples
Descriptive Statistics Groups Frequency Percentage
Socioeconomic Indicators
Age < 19
19–28
29–37
> 37
16
242
73
48
4.2
63.9
19.3
12.7
Mean
Std. Dev
27.62
9.62
Gender Male
Female
238
141
62.8
37.2
Education Up to Class 10
College
University
3
7
369
0.8
1.8
97.4
Marital Status Single
Married
Divorced/Widowed
I prefer not to say
259
114
3
3
68.3
30.1
0.8
0.8
Household size < 5
5–7
8–10
> 10
84
174
78
43
22.2
45.9
20.6
11.3
Mean
Std Dev
7.09
4.25
Type of family Living alone
Nuclear
Joint
I prefer not to say
40
242
88
9
10.6
63.9
23.2
2.4
Monthly Income
(in PKR)
< 10,000
10,000–25,000
25001-50,000
50,001-100,000
> 100,000
39
40
88
111
101
10.3
10.6
23.2
29.3
26.6
Number of children in the house (< 18 years old) 0
1–3
4–6
> 6
121
126
49
13
31.9
51.7
12.9
3.4
Mean
Std Dev
1.96
2.58
Number of old people in the house (> 60 years old) 0
1
2
> 2
226
100
43
10
59.6
26.4
11.3
2.6
Mean
Std Dev
0.79
3.2
Public health emergency experience
Past experience with a public health emergency Yes
No
30
349
7.9
92.1
Food storage Less than one month
One month to two months
Two months to three months
More than three months
314
56
6
3
82.8
14.8
1.6
8
Isolation days (lockdown days) Less than a week
1- Less than two weeks
2- less than three weeks
3- less than four weeks
More than four weeks
6
46
206
81
40
1.6
12.1
54.4
21.4
10.6
Mean
Std Dev
20.44
7.67
The geographical split of responses helped ascertain the spatial variations in risk perception and coping mechanisms within the study area (Pakistan). Using the provincial and regional administrative boundaries as a baseline to group the data, it was found that the lowest number of responses were received from Gilgit-Baltistan and Azad Jammu and Kashmir (AJK) areas. Since both the areas share a border (Fig. 1 study area map), these were dealt with as a single entity and named ‘Northern Areas’ for this study – the data from these regions was processed accordingly.
Punjab, the most populous province of the country, had the highest representation in the data; 186 responses which is around 49% of the data. A total of 51 (13.5%) and 81 (21.4%) responses were obtained from Sindh and Khyber Pakhtunkhwa provinces, respectively. Balochistan, the least populated province of Pakistan, constituted around 6% (24 responses) of the data, whereas Northern Areas (Gilgit-Baltistan + Jammu & Kashmir) represented around 3% (10 responses) of the total data. Around 7% (27 responses) of data came from the Islamabad Capital Territory.
The respondents were also asked about potential sources of information they pursued, gaining knowledge and awareness about potential COVID-19 risks. Most respondents relied on multiple sources of information, such as social media, television, local government, and others. The most widely used source of information, as per the data, was social media (86.5%), comprising apps/websites, namely Facebook, WhatsApp, Twitter, and YouTube, followed by television (73.5%). News channels and talk shows were the predominant sources of information on television. Around 31.1% of the respondents also used newspapers as a source of information, while 21.6% received announcements from local governments through community leaders and mosques. A relatively small portion of respondents relied on relatives, friends, and work colleagues, the Centers for Disease Control and Prevention (CDC)/WHO website, and government websites. This shows that social media and television can be used as effective mediums for risk communication in any future event in Pakistan.
COVID-19 risk perception
The risk perception regarding COVID-19 was measured using indicators and their domains. The mean and standard deviations were calculated to ascertain the level of risk perception against each indicator concerning regions and the whole of Pakistan.
Dread, fear, and worry
Fear is considered a vital domain and is sometimes used alternatively to perceive risk [44, 45]. Perceived fear for life was relatively highest in Sindh and Islamabad regions, probably because the initial cases of COVID-19 reported in the country were from these areas (Fig. 2a). The perceived likelihood of getting infected by this virus was highest in Balochistan. Interestingly, respondents from all regions reported (as perceived) below-average chances of getting infected, which shows a serious lack of risk perception among the citizens. Most of the respondents from all the regions showed a higher chance of increased occurrences of such events in the future. When higher perception values of future increases in occurrences (F-3) are seen in comparison with the chances of getting the infection (F-2), the fatalistic attitude of respondents becomes evident. This implies that although there are chances of increased occurrences, respondents believe that others will be in danger and not themselves. Since respondents were observing lockdown at the time of the survey, all the regions showed a serious concern regarding disruption to daily lifestyles. Similarly, most respondents worried that this pandemic would affect their family members or communities. Due to its novelty and limited scientific understanding to neutralize the infection, there was above-average fear based on current knowledge in all regions. Regarding the fear of supply chain interruption in lockdown extensions, all the regions showed above-average fear. Spatially, the fear domain showed almost similar values among all the regions of Pakistan, indicating that it was more or less similar throughout the country.
Fig. 2 Indicator-wise variation in (a) dread, fear, and worry; (b) behavior and attitude; (c) awareness and knowledge; (d) trust and confidence; and (e) overall risk perception on a relative scale from ‘low’ to ‘high’
Behavior and attitude
Behavior and attitude regarding a particular hazard/crisis may predict the actions of individuals, groups, or communities [17, 46]. The perceived capacity to deal with a particular hazard would decrease the risk perception of individuals. This indicator suggests that all regions of the country perceived a lack of capacity to deal with this COVID-19 situation (Fig. 2b). Moreover, most respondents perceived above-average chances of adapting to a new future lifestyle. Similarly, due to social distancing, there was a high chance of changing relationships with relatives, friends, and work colleagues. The overall behavior domain of risk perception showed a varying picture in different regions. Relatively, the lowest behavior and attitude values were observed for Sindh province.
Awareness and knowledge
Knowledge, awareness, and familiarity with a particular hazard affect risk perception [47]. A noticeably high value was observed regarding the perceived extent of familiarity with precautionary measures to reduce COVID-19 risk, implying that the respondents were familiar with precautionary measures against this pandemic (Fig. 2c). In terms of how much a community was not protected against this virus, most respondents indicated an above-average concern. This shows that respondents perceived risk to their communities or surroundings. Due to the novelty and originality of the COVID-19 crisis, almost all the respondents showed a high or very high level of danger from an unknown risk – which can significantly influence the risk perception of the respondents. Overall, all regions showed higher than average perceived risk levels in the awareness domain of risk perception. Islamabad showed the highest level of perceived risk with respect to awareness and knowledge regarding COVID-19.
Trust and confidence
Trust and confidence are vital in determining community and government cooperation in reducing risk [48]. Trust in government, policies, and institutions can significantly influence risk perception. Moreover, the reliability of information sources can act as a trust element between the community and the government [45]. Generally, the reliability of government-based sources of information was above average among the respondents (Fig. 2d). The respondents from Sindh, relatively, showed lesser reliability on government sources of information. In the age of “infodemic,” most respondents did not rely on the information sought from social media (most of the respondents were highly educated). In Pakistan, disaster management authorities and health departments lead the fight against COVID-19. Trust in disaster management and health authorities plays a significant role in shaping up the community response towards following the safety protocols or orders from concerned authorities. There was above average trust level in all regions. Relatively, Islamabad and Sindh regions showed a higher level of trust in their government departments. Clear, consistent and transparent government policies can significantly influence trust and risk perception. Except, Sindh and Balochistan, all regions demonstrated above average trust in governmental policies. Similar situation was observed with respect to trust in local administrations as well. Regarding responses about provincial or federal governments’ role in this pandemic, slightly above average trust was reported. However, the respondents from Balochistan province showed the least trust level onthe government response. With respect to trust in science and relevant experts, above-average values were observed. Overall, the trust domain of risk perception shows spatial variations in the study area, which can be attributed to the different approaches adopted by the local authorities, provincial governments, and the federal government to tackle this pandemic.
Overall risk perception
The risk perception index was constructed using all the domains and their indicators. With respect to fear, the same picture was observed in six regions (Fig. 2e). Fear in KPK and Sindh provinces was relatively highest. In terms of behavior and attitude domain, a significant difference was observed among all regions (F-value = 2.095, p-value = 0.065). In terms of awareness and knowledge, no significant difference was observed. Among all the different domains, awareness and knowledge had the highest value. In terms of the trust domain, a significant difference was observed (F-value = 3.568, p-value = 0.004), indicating that trust varied significantly among the six regions of Pakistan. Overall, the COVID-19 risk perception in the country was above average, albeit still low.
Coping mechanisms against COVID-19
The coping mechanisms against COVID-19 were measured using indicators and three domains, i.e., problem-oriented, emotion-oriented, and action-oriented. The mean and the standard deviations were calculated to ascertain the level of coping mechanism against each indicator with respect to different regions.
Problem-oriented
The problem-solving domain explains how various individuals use coping mechanisms against the COVID-19 crisis. According to the data, most respondents actively seek safety protocols to be adhered to in their homes to cope with/reduce the risk of infection exposure (Table 5). This strategy has been used worldwide to minimize infection rates and “flatten the curve”. As the country started systematically locking down, it was promising to find that educated respondents felt safe in their homes. Similarly, most of the respondents were listening to and following the advice of experts. However, some trusted more reliable sources of advice, such as WHO, while others relied on the expert opinions of government representatives and media outlets.
Because of the rapidly evolving situation, most respondents actively sought out new information about COVID-19 in Pakistan to stay updated about the number of cases/deaths, lockdown extensions, and vaccine development. Since a lockdown situation could induce mental stress, respondents were predisposed to try something new to relieve their minds. This was apparent in all six regions. The respondents also talked with others who may help them understand the current crisis. All the respondents were employed, and at the time of the survey, most were not seeking financial support from the government. However, the respondents from Northern Areas were partially inclined to seek financial support from their respective provincial governments. Overall, the problem-oriented coping mechanism was similar in all regions of Pakistan (Fig. 3a).
Fig. 3 Indicator-wise variation in: (a) problem-oriented; (b) emotion-oriented; (c) action-oriented; and (d) overall coping mechanism on a relative scale from ‘disagree’ to ‘agree’
Emotion-oriented
Mental health can get significantly affected in a pandemic crisis and prolonged lockdowns. Various statements were asked in the questionnaire to understand the psychological resilience against the COVID-19 pandemic. The above-average value was observed in all regions regarding respondents’ focus on work to distract themselves from the current crisis (Fig. 3b). Most of the respondents were not seeking any meditation to calm themselves – however, relatively speaking, respondents from Balochistan province had an above-average value in this regard. Most of the respondents from all the regions agreed that they could not do much other than wait for this crisis to get over. Moreover, the respondents agreed that lockdown was the only solution to combat the COVID-19 crisis, as evident in all the regions. The respondents were hopeful for a successful cure or drug to be developed soon to neutralize the virus. Except for the Northern Areas, all regions were hoping for a miracle to happen. Most respondents did not (over) eat, smoke, or take medication as coping mechanisms. Another indicator of psychological resilience showed that people actively avoided bad moods to reduce mental stress. All the regions showed above-average values for this statement. The statement regarding believing (or denying) this pandemic crisis shows good risk perceptions, which can help them agree or follow experts’ advice. Except for KPK, all regions had above-average emotional/psychological resilience values against COVID-19.
Action-oriented
Action-oriented indicators express respondents’ physical actions to cope with the COVID-19 pandemic. The foremost action suggested by WHO focused on washing hands for twenty seconds, which can significantly reduce infection risk. It was encouraging to observe that most respondents highly agreed with washing their hands more than usual (Fig. 3c). Relatively, the highest value was observed in Islamabad, implying respondents were acting upon the advice of government and health experts. Similarly, a high level of agreement was also observed regarding avoiding public spaces and public transport. Again, all regions exhibited high levels of agreement, with Islamabad leading the rest. However, regarding buying large quantities of disinfectants/staple foods, the data exhibited a varying value in all regions – generally, the respondents refrained from stockpiling in this crisis. They did not even buy additional protective equipment (gloves or masks). This shows a positive insight into stockpiling essential resources. The government, specifically the National Disaster Management Authority and the Ministry of National Health Services, had started advertising against stockpiling early on ratified ordinances and penalties for restricting food hoarding by traders capitalizing on the COVID-19 pandemic [49]. A slight variation in action-oriented mechanisms was observed among the regions (Fig. 3c).
Overall coping mechanism
The coping mechanism index was constructed using all the domains and indicators, i.e., problem-oriented, emotion-oriented, and action-oriented. With respect to the problem-oriented, a similar pattern was observed in all six regions - relatively. The problem-oriented mechanism was highest in Northern Areas (Fig. 3d). Comparatively, the emotion-oriented coping mechanism was the lowest among the three domains. KPK had the lowest scoring in the psychological domain in the country. Similarly, action-oriented coping mechanisms showed more or less same values in the country. Overall, coping mechanisms and their domains did not significantly differ among the regions. This implies that coping mechanisms in the country do not vary much spatially in Pakistan.
Relationship between socioeconomic factors and risk perception
Within socioeconomic domain indicators, only age, gender, and marital status were significantly correlated with the perception of risks associated with the COVID-19 pandemic, though the effects of these correlations are small (Table 5). These were women, younger and unmarried people who perceived COVID-19 risks more than men, elderly, and married people. Analyzing the relationship of the identified indicators with the risk perception domains reveals that not all of them are significantly correlated with the socioeconomic indicators, and there were notable variations. Fear was found to have a significant correlation with the indicator of gender only, as it implies that women have more fear of the COVID-19 pandemic than men, which confirms the findings of some of the studies [26, 38]. The role of gender has been explicitly examined in another study based on the same dataset [40]. Interestingly, it was also found that the behavior domain of risk perception did not have a significant relationship with any of the socioeconomic indicators, which is a deviation from some of the literature findings [50]. This might be because of the distinct variations in socioeconomic behavior and practices across various parts of Pakistan, making it difficult to draw any consistent relationship. The indicators of age and marital status were also found to have significant relations with awareness, implying that elderly and married people were more aware of the precautionary measures to be taken during the COVID-19 pandemic as compared to the younger and unmarried people. Some researchers also identified age as a significant predictor of risk perception during pandemic events. [30, 35]. Men, elderly and married people tended to have less trust in government institutions responsible for dealing with the COVID-19 crisis in Pakistan compared to younger women and younger and unmarried people. Within the public health emergency domain, only the indicator of isolation days was found to have a relationship with the risk perception domain of awareness, and that too is with a small effect size. It also shows that as the number of isolation days increases, people become impatient with respect to practicing precautionary measures consistently, and their awareness of the associated pandemic risks gradually starts declining. Only the gender variable was significant [40].
Table 5 Correlation between respondents’ risk perception, socioeconomic conditions, and previous public health emergency experience
Risk Perception Fear Behavior Awareness Trust Risk Perception
Socioeconomic
Age -0.63 0.12 0.47** -0.179*** -0.106*
Gender 0.119* 0.048 -0.030 0.172** 0.176**
Marital Status -0.081 0.028 0.120* -0.147** -0.103*
Public health emergency
Isolation Days -0.053 -0.048 -0.177** -0.012 -0.088
significant at *p < .05, **p < .01, ***p < .001
Note
Only indicators with a significant relationship are shown for brevity purposes.
Only the indicator of gender has been found to have a significant relationship with the coping mechanisms for reducing the risks of COVID-19, though with a small effect (Table 6; more detailed analysis and interpretations on the gender aspect are presented in an earlier study [40]). Women were more likely to adopt emotions and action-oriented mechanisms to cope with pandemic risks than men. These findings confirm the results of some of the past studies. A study found that people opt for action-oriented precautionary measures to reduce the epidemic risks while studying in five European and three Asian regions [51]. However, this study has not found a relationship between gender and problem-oriented coping mechanisms. This finding deviates from some studies that found problem-oriented mechanisms significant in coping with pandemic risks [52].
Table 6 Correlation between respondents’ coping mechanisms, socioeconomic conditions, and previous public health emergency experience
Problem Emotion Action Coping Mechanism
Gender 0.068 0.146** 0.106* 0.151**
significant at *p < .05, **p < .01, ***p < .001
Note
Only indicators with a significant relationship are shown for brevity purposes.
Conclusion
The COVID-19 crisis in Pakistan and other places worldwide is still unfolding. The direct and indirect impacts of the infection itself, and the opted government measures such as lockdown are developing gradually. This study has explored current risk perception concerning the COVID-19 crisis throughout Pakistan. The research has found that risk perception has varying levels in different regions of the country, depicting the spatial variation of COVID-19 risk perception and coping mechanisms. The main difference has been observed in the behavior and trust domains of risk perception, implying people may behave differently in dealing with the risks associated with the infection and complying with the instructions of the respective provincial governments. An alarming trend has been found in the community’s response to public health emergencies. As the isolation days progress, the awareness level of the COVID-19 risks tends to decline, showing that the community was not continuously engaged in pandemic risk reduction. Under such circumstances, it is difficult to reap the true benefit of the lockdown measures to contain an epidemic.
This research can be useful for concerned authorities to design risk communication and reduction measures. Findings from risk perception domains and coping mechanism perspectives also provide useful insights for the policymakers to establish synergies between health, industry, commerce, and information sectors for devising efficient communication mechanisms and resource flow structures according to the area-specific needs to deal with the COVID-19 situation.
This study has its limitations too. Firstly, non-probability sampling has been employed, which cannot be used to generalize its findings to the overall population. Secondly, as the data was collected through an online survey, most respondents were young and highly educated. They had access to the internet which led to an issue of under-representation in the sample. The other limitations of the online survey include a lack of sampling design due to the self-selection of the respondents and a low response mainly because of its self-administrative design and technical issues, especially in the case of a global pandemic. Moreover, measuring risk perception is difficult as many unforeseen factors are at play. Similarly, the coping mechanism is quantified using pre-determined indicators, and individuals or households may have more coping strategies that could not be incorporated in the current study. The findings of this study might not hold true for other developing or developed countries. More samples are needed to ascertain statistical relationships among indicators, domains, and regions. It is suggested that future studies investigate the role of vaccinations on COVID-19 risk perception.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Supplementary Material 1
Data availability
The data will be provided upon a reasonable request.
Declaration
Disclosure statement
The authors reported no potential conflict of interest.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. WHO (2020). Coronavirus disease (COVID-19) Situation reports. Geneva. Retrieved from https://www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports
2. Liu, J., Liao, X., Qian, S., Yuan, J., Wang, F., Liu, Y., & Zhang, Z. (2020). Community transmission of severe Acute Respiratory Syndrome Coronavirus 2, Shenzhen, China, 2020. Emerging Infectious Diseases, 26(6), 10.3201/eid2606.200239.
3. Chan JFW Yuan S Kok KH To KKW Chu H Yang J Yuen KY A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster The Lancet 2020 395 10223 514 523 10.1016/S0140-6736(20)30154-9
4. World Health Organization (2020). WHO Timeline - COVID-19. Retrieved May 11, 2020, from https://www.who.int/news-room/detail/27-04-2020-who-timeline---covid-19
5. Johns Hopkins University (2022). COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. Retrieved May 13, 2020, from https://github.com/CSSEGISandData/COVID-19
6. World Health Organization (2020). Country & Technical Guidance - Coronavirus disease (COVID-19). Retrieved May 13, 2020, from https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance
7. Sahu KK Kumar R Preventive and treatment strategies of COVID-19: from community to clinical trials Journal of Family Medicine and Primary Care 2020 9 5 2149 10.4103/jfmpc.jfmpc_728_20
8. Lee J Mental health effects of school closures during COVID-19 The Lancet Child & Adolescent Health 2020 10.1016/S2352-4642(20)30109-7
9. Journal of Criminal Justice, 68, 101692. https://doi.org/10.1016/j.jcrimjus.2020.101692
10. Campbell AM An increasing risk of family violence during the Covid-19 pandemic: strengthening community collaborations to save lives Forensic Science International: Reports 2020 2 100089 10.1016/j.fsir.2020.100089
11. Ivanov D Dolgui A Viability of intertwined supply networks: extending the supply chain resilience angles towards survivability. A position paper motivated by COVID-19 outbreak International Journal of Production Research 2020 58 10 2904 2915 10.1080/00207543.2020.1750727
12. Armantier, O., Koşar, G., Rachel, P., Skandalis, D., Smith, K., Topa, G., & Van der Klaauw, W. (2020). How Widespread Is the Impact of the COVID-19 Outbreak on Consumer Expectations? (No. 20200416b). New York. Retrieved from https://ideas.repec.org/p/fip/fednls/87811.html
13. Government of Pakistan (2022). COVID-19 Dashboard. Retrieved August 28, 2022, from http://covid.gov.pk/stats/pakistan
14. Sarwar S Waheed R Sarwar S Khan A COVID-19 challenges to Pakistan: is GIS analysis useful to draw solutions? Science of The Total Environment 2020 730 139089 10.1016/j.scitotenv.2020.139089 32387823
15. Ahmed, T., Tunio, H., & Rao, V. (2020, March 21). Sindh decides to go into COVID-19 lock down. The Express Tribune. Karachi. Retrieved from https://tribune.com.pk/story/2181138/1-sindh-govt-mulls-total-lockdown-provincial-covid-19-tally-crosses-350/
16. Naheed, N. (2020). Coronavirus spread timeline in Pakistan. Technology Times. Retrieved May 13, 2020, from https://www.technologytimes.pk/2020/04/08/coronavirus-spread-timeline-pakistan/
17. Rana IA Jamshed A Younas ZI Bhatti SS Characterizing flood risk perception in urban communities of Pakistan International Journal of Disaster Risk Reduction 2020 46 March 101624 10.1016/j.ijdrr.2020.101624
18. Slovic P Kunreuther H White G Slovic P Decision processes, Rationality and Adjustment to Natural Hazards The perception of risk 2000 London Routledge 39 69
19. Nature Human Behaviour. https://doi.org/10.1038/s41562-020-0884-z
20. Madhav N Oppenheim B Gallivan M Mulembakani P Rubin E Wolfe N Jamison DT Gelband H Horton S Jha P Laxminarayan R Mock CN Nugent R Pandemics: risks, impacts, and Mitigation Disease Control Priorities: improving Health and reducing poverty 2017 3 Washington The World Bank 315 346
21. Green MS Swartz T Mayshar E Lev B Leventhal A Slater PE Shemer J When is an epidemic an epidemic? The Israel Medical Association Journal 2002 4 1 3 6 11802306
22. Porta, M. (2014). A Dictionary of Epidemiology (6th ed.). Oxford: Oxford University Press. Retrieved from https://global.oup.com/academic/product/a-dictionary-of-epidemiology-9780199976737
23. Hays JN Epidemics and pandemics: their impacts on human history 2005 1 Santa Barbara Abc-clio
24. Renn O Risk governance: coping with uncertainty in a Complex World 2008 London Earthscan
25. Darker C Gellman MD Turner JR Risk perception Encyclopedia of behavioral medicine 2013 New York, NY Springer 1689 1691
26. Ibuka Y Chapman GB Meyers LA Li M Galvani AP The dynamics of risk perceptions and precautionary behavior in response to 2009 (H1N1) pandemic influenza BMC Infectious Diseases 2010 10 1 296 10.1186/1471-2334-10-296 20946662
27. Barr M Raphael B Taylor M Stevens G Jorm L Giffin M Lujic S Pandemic influenza in Australia: using telephone surveys to measure perceptions of threat and willingness to comply BMC Infectious Diseases 2008 8 1 117 10.1186/1471-2334-8-117 18793441
28. Di Giuseppe G Abbate R Albano L Marinelli P Angelillo IF A survey of knowledge, attitudes and practices towards avian influenza in an adult population of Italy BMC Infectious Diseases 2008 8 1 36 10.1186/1471-2334-8-36 18366644
29. Leung GM The impact of community psychological responses on outbreak control for severe acute respiratory syndrome in Hong Kong Journal of Epidemiology & Community Health 2003 57 11 857 863 10.1136/jech.57.11.857 14600110
30. Jacobs, J., Taylor, M., Agho, K., Stevens, G., Barr, M., & Raphael, B. (2010). Factors Associated with Increased Risk Perception of Pandemic Influenza in Australia. Influenza Research and Treatment, 2010, 1–7. 10.1155/2010/947906
31. Brug J Aro AR Oenema A de Zwart O Richardus JH Bishop GD SARS Risk Perception, Knowledge, Precautions, and information sources, the Netherlands Emerging Infectious Diseases 2004 10 8 1486 1489 10.3201/eid1008.040283 15496256
32. Huynh, T. L. D. (2020). The COVID-19 risk perception: A survey on socioeconomics and media attention. Economics Bulletin, 40(1), 758–764. Retrieved from http://www.accessecon.com/Pubs/EB/2020/Volume40/EB-20-V40-I1-P64.pdf
33. Zeballos Rivas DR Lopez Jaldin ML Canaviri N Portugal Escalante B Fernández LFA Ticona A Social media exposure, risk perception, preventive behaviors and attitudes during the COVID-19 epidemic in La Paz, Bolivia: a cross sectional study PLOS ONE 2021 16 1 e0245859 10.1371/journal.pone.0245859 33481945
34. Schneider CR Dryhurst S Kerr J Freeman ALJ Recchia G Spiegelhalter D van der Linden S COVID-19 risk perception: a longitudinal analysis of its predictors and associations with health protective behaviours in the United Kingdom Journal of Risk Research 2021 24 3–4 294 313 10.1080/13669877.2021.1890637
35. Gerhold, L. (2020). COVID-19: Risk perception and coping strategies. Results from Germany. Berlin.
36. BMC Public Health, 21(1), 1562. https://doi.org/10.1186/s12889-021-11600-3
37. Kollmann J Kocken PL Syurina EV Hilverda F The role of risk perception and affective response in the COVID-19 preventive behaviours of young adults: a mixed methods study of university students in the Netherlands British Medical Journal Open 2022 12 1 e056288 10.1136/bmjopen-2021-056288
38. Karasneh R Al-Azzam S Muflih S Soudah O Hawamdeh S Khader Y Media’s effect on shaping knowledge, awareness risk perceptions and communication practices of pandemic COVID-19 among pharmacists Research in Social and Administrative Pharmacy 2020 10.1016/j.sapharm.2020.04.027
39. Lohiniva, A. L., Sane, J., Sibenberg, K., Puumalainen, T., & Salminen, M. (2020). Understanding coronavirus disease (COVID-19) risk perceptions among the public to enhance risk communication efforts: a practical approach for outbreaks, Finland, February 2020. Eurosurveillance, 25(13), 10.2807/1560-7917.ES.2020.25.13.2000317.
40. Rana IA Bhatti SS Aslam AB Jamshed A Ahmad J Shah AA COVID-19 risk perception and coping mechanisms: does gender make a difference? International Journal of Disaster Risk Reduction 2021 55 102096 10.1016/j.ijdrr.2021.102096 33777688
41. Qiao S Tam CC Li X Risk exposures, risk perceptions, negative attitudes toward General Vaccination, and COVID-19 Vaccine Acceptance among College Students in south Carolina American Journal of Health Promotion 2022 36 1 175 179 10.1177/08901171211028407 34164998
42. Caserotti M Girardi P Rubaltelli E Tasso A Lotto L Gavaruzzi T Associations of COVID-19 risk perception with vaccine hesitancy over time for italian residents Social Science & Medicine 2021 272 113688 10.1016/j.socscimed.2021.113688 33485215
43. Hilverda F Vollmann M The role of risk perception in students’ COVID-19 vaccine uptake: a longitudinal study Vaccines 2021 10 1 22 10.3390/vaccines10010022 35062683
44. Sjoberg L Worry and risk perception Risk Analysis 1998 18 1 85 93 10.1111/j.1539-6924.1998.tb00918.x 9523446
45. Khan, A. A., Rana, I. A., & Nawaz, A. (2020). Gender-based approach for assessing risk perception in a multi-hazard environment: A study of high schools of Gilgit, Pakistan. International Journal of Disaster Risk Reduction, 44(November 2019), 101427. 10.1016/j.ijdrr.2019.101427
46. Burns WJ Slovic P Risk perception and behaviors: anticipating and responding to crises Risk Analysis 2012 32 4 579 582 10.1111/j.1539-6924.2012.01791.x 22500649
47. Wachinger G Renn O Begg C Kuhlicke C The risk perception paradox-implications for governance and communication of natural hazards Risk Analysis 2013 33 6 1049 1065 10.1111/j.1539-6924.2012.01942.x 23278120
48. Siegrist M Gutscher H Earle TC Perception of risk: the influence of general trust, and general confidence Journal of Risk Research 2005 8 2 145 156 10.1080/1366987032000105315
49. The News Staff (2020, April 17). Govt ratifies ordinance against hoarding; guilty to face three years in jail. The News. Islamabad. Retrieved from https://www.thenews.com.pk/latest/645742-govt-ratifies-ordinance-against-hoarding-guilty-to-face-three-years-in-jail
50. Poletti P Ajelli M Merler S The Effect of Risk Perception on the 2009 H1N1 pandemic Influenza Dynamics Plos One 2011 6 2 e16460 10.1371/journal.pone.0016460 21326878
51. Sadique MZ Edmunds WJ Smith RD Meerding WJ de Zwart O Brug J Beutels P Precautionary behavior in response to perceived threat of Pandemic Influenza Emerging Infectious Diseases 2007 13 9 1307 1313 10.3201/eid1309.070372 18252100
52. Balicer RD Omer SB Barnett DJ Everly GS Local public health workers’ perceptions toward responding to an influenza pandemic Bmc Public Health 2006 6 1 99 10.1186/1471-2458-6-99 16620372
| 0 | PMC9707224 | NO-CC CODE | 2022-12-01 23:20:21 | no | Spat. Inf. Res.. 2022 Nov 29;:1-13 | utf-8 | null | null | null | oa_other |
==== Front
Artif Intell Rev
Artif Intell Rev
Artificial Intelligence Review
0269-2821
1573-7462
Springer Netherlands Dordrecht
10332
10.1007/s10462-022-10332-z
Article
Deep learning on multi-view sequential data: a survey
Xie Zhuyang [email protected]
12
Yang Yan [email protected]
12
Zhang Yiling [email protected]
12
Wang Jie [email protected]
12
Du Shengdong [email protected]
12
1 grid.263901.f 0000 0004 1791 7667 School of Computing and Artificial Intelligence, Southwest Jiaotong University, Chengdu, 611756 China
2 grid.263901.f 0000 0004 1791 7667 Manufacturing Industry Chains Collaboration and Information Support Technology Key Laboratory, Southwest Jiaotong University, Chengdu, 611756 China
29 11 2022
144
© The Author(s), under exclusive licence to Springer Nature B.V. 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
With the progress of human daily interaction activities and the development of industrial society, a large amount of media data and sensor data become accessible. Humans collect these multi-source data in chronological order, called multi-view sequential data (MvSD). MvSD has numerous potential application domains, including intelligent transportation, climate science, health care, public safety and multimedia, etc. However, as the volume and scale of MvSD increases, the traditional machine learning methods become difficult to withstand such large-scale data, and it is no longer appropriate to use hand-craft features to represent these complex data. In addition, there is no general framework in the process of mining multi-view relationships and integrating multi-view information. In this paper, We first introduce four common data types that constitute MvSD, including point data, sequence data, graph data, and raster data. Then, we summarize the technical challenges of MvSD. Subsequently, we review the recent progress in deep learning technology applied to MvSD. Meanwhile, we discuss how the network represents and learns features of MvSD. Finally, we summarize the applications of MvSD in different domains and give potential research directions.
Keywords
Deep neural networks
Multi-view
Sequential data
Spatio-temporal
http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China Nos.61976247 Yang Yan Sichuan Science and Technology ProgramNO.2021YFG0312 Du Shengdong
==== Body
pmcIntroduction
With the development of social media, more and more human activities have become public and accessible. In addition, the application of a large number of acquisition equipments and sensors has made it easier for us to obtain information about the surrounding world. Humans collect these multi-source data in chronological order and obtain multi-view sequential data (MvSD). MvSD has broad researches and applications in various domains, including smart transportation, climate science, social media, health care, crime analysis, etc. However, as the volume and scale of MvSD increases, classical data mining methods are no longer applicable. On the one hand, the construction of hand-craft features is restricted by limited human knowledge, thus conventional methods are difficult to represent such complex data. On the other hand, MvSD changes dynamically over time and presents self-correlated, the traditional machine learning methods can not fully mine the knowledge mechanism in sequential data and it is difficult to effectively analyze the hidden attributes. Simultaneously, MvSD collected from various domains or obtained from diverse sensors leads to heterogeneity among views. Thus, how to make full use of the diversity among different views and fuse the latent knowledge in MvSD has attracted extensive research.
In recent years, deep learning has swept many fields and achieved remarkable achievements, such as object detection (Girshick 2015; Ren et al. 2015; He et al. 2016; Redmon et al. 2016; Liu et al. 2016a), image segmentation (Long et al. 2015; Ronneberger et al. 2015; Lin et al. 2017; He et al. 2017; Zhao et al. 2017c; Badrinarayanan et al. 2017), natural language processing (Kiros et al. 2014; Bahdanau et al. 2014; Cheng et al. 2016; Vaswani et al. 2017), etc. Deep learning has brought the possibility to solve the above problems with its general data understanding and parallel computing capabilities. First of all, the superiority of deep learning is based on its feature extraction ability, which breaks the performance of human-engineered features through end-to-end learning. Among them, convolutional neural network (CNN) achieves excellent performance on regular raster data, while recurrent neural network (RNN) is adapted to sequence data and model the correlations. Second, classical methods based on small datasets are untenable in MvSD. In contrast, the performance of deep models will be further improved with massive data samples. Therefore, deep models perform better feature representation and learning on larger datasets. Third, conventional machine learning methods generally exploits some linear functions to fit latent data structure and is not able to express complex models. As we all know, deep networks have nonlinear approximation capabilities and learn rules by optimizing the loss function as much as possible.
It is worth mentioning that MvSD is collected from multiple domains. For these multi-source sequential data, the analysis method based on multi-view learning has more sufficient feature representation capability than single-view learning. Multi-view deep learning can not only be used to analyze the implicit feature correlation and internal dynamic changes in sequential data, but also help solve the incompleteness and uncertainty in sequential data analysis. Taking video sentiment analysis as an example. It usually consists of three kinds of data: text sentences, images and audio clips. It is not comprehensive to analyze the expression only in text streams, because language is ambiguous in different situations. By combining facial features and speakers pronunciation, the speakers attitude is more accurately inferred. As for the traffic flow forecasting task, some external factors will also affect the prediction performance, such as weather, holidays and events. Injecting these factors into the network will further assist in prediction.
In the past few decades, a large number of machine learning techniques have been used for multi-view data, resulting in multi-view representation, multi-view clustering, multi-view fusion, etc. With the large amount of multimedia data available in recent years, multi-view learning has become a promising research. Benefiting from a large number of researches and sufficient theories of multi-view learning (Khan et al. 2022a; Yin and Sun 2019), some deep learning-based multi-view algorithms are gradually being studied (Yin and Sun 2019; Sun and Zong 2020; Mao and Sun 2020), forming deep multi-view learning (Yao et al. 2018; Wang et al. 2015; Kan et al. 2016; Sun et al. 2020d) and deep multi-view clustering (Li et al. 2019; Khan et al. 2022b; Xia et al. 2022). For example, Deep CCA (Andrew et al. 2013), which is extended from canonical correlation analysis (CCA) (Hotelling 1992), learns non-linear mappings between different views through stacked multi-layer neural networks. Deep matrix factorization (Zhao et al. 2017b; Huang et al. 2020a), which applies non-negative matrix factorization (NMF) from traditional clustering to the deep framework. In addition, deep subspace clustering (Ji et al. 2017; Wang et al. 2020b) is also extended on the basis of traditional subspace clustering, which further expands the application (Abavisani et al. 2020; Cai et al. 2021; Wang et al. 2020c). Therefore, adopting some ideas based on multi-view clustering can facilitate our research in the process of researching MvSD.
In the past few years, some works have investigated multi-modal data, demonstrating the effectiveness of deep learning for multi-modal data fusion. Several surveys reviewed the progress of multi-view deep learning (Wang 2021; Baltrušaitis et al. 2018; Chen et al. 2020c; Summaira et al. 2021; Rahate et al. 2021; Ramachandram and Taylor 2017; Zhao et al. 2017a). Wang (2021) discussed some recent researches on deep multi-modal models from two aspects of clustering and classification, focusing on the application of generative adversarial network (GAN) in clustering and cross-modal learning. Baltrušaitis et al. (2018) investigated the latest developments in multi-modal machine learning, and it stated five challenges: representation, translation, alignment, fusion and co-learning. Chen et al. (2020c) analyzed the prevailing multi-modal network structure and existing problems, including multi-modal feature extraction and latent feature learning. Summaira et al. (2021) discussed the latest advancements and trends in multi-modal deep learning, and adopted a new fine-grained taxonomy to classify existing multi-modal networks. Rahate et al. (2021) reviewed the relevant literature in multi-modal deep learning and categorized multi-modal co-learning from multiple perspectives. These aforementioned surveys are instructive for our MvSD investigation, at the same time, some spatio temporal data (STD) analysis works provide us with application fields and current progress for reference. Wang et al. (2020d) reviewed the recent development of deep learning technology in STD and classified existing literature according to the types of STD, data mining tasks, and deep learning models. It classified the spatio temporal data into five types: event, trajectory, point reference, raster, and video. Alam et al. (2021) classified the STD analytics systems into three categories and provided definitions and related applications for STD. Moreover, it conducted investigations and discussions on existing programming languages, development tools, and data platforms. Atluri et al. (2018) summarized traditional machine learning methods in STD, and discussed related data mining problems in analyzing different STD. Mazimpaka and Timpf (2016) summarized the application of deep learning in trajectory data and traffic prediction.
In this paper, we conduct research on MvSD. Existing multi-view surveys mostly focus on the applications, neural networks and fusion methods, do not specifically consider combining multi-view and sequential data for research. In addition, the aforementioned STD researches do not consider investigating from multi-view perspective. Our contributions are as follows:This paper conducts research from the perspective of multi-view sequence and discusses the challenges in MvSD data.
This survey reviews recent deep learning techniques for MvSD and categorizes MvSD into four data types. Then we organize different deep learning models for specific types of view data for representation and learning.
This survey summarizes some application domains and emerging tasks of MvSD, and points out some potential future research directions.
The rest of this paper is organized as follows. In Sect. 2, we divide the source data that constitutes MvSD into four categories and discuss the characteristics and challenges in each of these categories. In Sect. 3, we illustrate the existing deep representation methods for various types of view data. In Sect. 4, we investigate the deep learning models for MvSD. In Sect. 5, we summarize the applications of MvSD and related tasks. Finally, we discuss the future trends and conclude the survey. The taxonomy diagram of MvSD is presented in Fig. 1.Fig. 1 Taxonomy diagram of deep learning on MvSD
Multi-view sequential data
As illustrated in Fig. 2, we give the paradigm of MvSD, which is composed of m views from different sources. These view data consists of various types, such as continuous multiple images (video clips), character description (text sentences), spatial position changes (trajectory), and these view data lengths or time steps may not be aligned with each other. Among them, each view has order constraints and is arranged according to certain rules. The sequential data of different views usually comes from various domains, which has different statistical characteristics. Therefore, it is difficult for a single view model to handle heterogeneous data. We need to formulate corresponding representation methods according to different views and select appropriate models for feature representation, extraction and fusion.Fig. 2 Illustration of MvSD. MvSD can be composed of m views, we enumerate three types view data, including video clips, text sentences, and mobile data. Each view is arranged in a certain order (for example, chronological order, grammatical rules, etc)
Data types
There are many types of sequential data, such as meteorological data, time-series data, gene sequences, sensor data, audio clips, etc, all of which are the research objects of MvSD. In order to facilitate subsequent work, we first introduce four data types: point, sequence, graph and raster. Each data type can be directly or indirectly converted to sequential data. Different from the classification method in Refs. Atluri et al. (2018), we generalize point data into two categories, that is, individual instances are regarded as points (e.g., event data), and the instances themselves are points (e.g., LiDAR data). In addition, we categorize trajectory data and text data into sequence data.
Point
Point data describes discrete points in space with specific location coordinates (e.g., geographic latitude and longitude), indicating the existence in space and attaching some additional information. A point is usually represented by a tuple (pi, ei, ti), where pi represents the position of the point, ti represents the time when the point occurrs and ei represents additional features (such as temperature, humidity, color, etc). Event-type data regards individual instance (e.g., a traffic incident) as point, which usually means that it occurs in a certain location and is accompanied by information such as time and event category. Figure 3a shows an example of the event data. In addition, there are some instance data themselves that are point sets, which are scanned by sensors. Point cloud data is usually represented by three-dimensional coordinates, with additional information such as reflectivity, intensity, and color, etc. Figure 3b shows an illustration of the 3D point cloud data (Hackel et al. 2017). Point data has applications in many fields, such as transportation (e.g., traffic accidents), criminology (e.g., crime incidents), social media (e.g., social event), autonomous (e.g., point cloud data), etc.Fig. 3 An illustration of event and laser data
Sequence
Time series is a typical type of sequence data, which is a sequence obtained at consecutive and evenly spaced time points. For example, in mechanical fault diagnosis, the frequency of equipments is sampled at equal intervals. Figure 4a shows an example of audio signal. As shown in Fig. 4b, video data is viewed as a series of images arranged in chronological order. The trajectory data is also treated as a time series, which periodically records the moving position of the target. Figure 4c shows an example of the trajectory data. Time series is not the only case of sequence data, there are other cases such as text data, which need to consider the logic of language. We group trajectory data, audio, video, time series, and text into sequence data.Fig. 4 Illustration of sequence data
Graph
Graph data is a collection of vertices connected by a series of edges, each of which is assigned a weight. Graph data is used in many fields, including traffic networks, social networks, and recommendation systems. In a social network, each person is a vertex, and people who have a relationship with each other are connected by edges. Each edge has a direction to form a directed graph. In traffic forecasting, the traffic road networks are naturally modeled as graphs. Taking the road network as an example, where road segments are represented as edges, and nodes embedded in the spatial map represent intersections of these road segments.
Raster
Raster data is presented in a grid of pixels, and each pixel has a value, which represents information at a specific location (color or other statistics). Figure 5a shows an example of an image data, the position of each pixel is regarded as a fixed point, and each pixel is an observation value. In neuroscience, as a new neuroimaging method, functional magnetic resonance imaging (fMRI) is based on measuring changes in hemodynamics caused by neuronal activity. The scanned signals form raster data used to analyze brain activity. In urban big data, various fixed-position sensors collect data to form spatial map, air quality, and weather data. Figure 5b shows an example of raster traffic data (Zhou et al. 2020).Fig. 5 Illustration of raster data
Converting data format
The data formats mentioned above often need to be converted into appropriate formats according to specific tasks and models. These formats are often convertible to each other. Point data is naturally converted to raster data by quantifying in each grid cell. For example, the events (e.g., traffic accidents, crimes, etc) that occur in each grid are converted into event raster data, which in turn can be converted to point data. In autonomous driving, point cloud is converted into 3d voxel grid or 2d bird’s eye view (BEV) through quantization operation. Further, point data are treated as nodes in graph data. In spatial map, traffic sensors are viewed as nodes of the graph, and the distances between the sensors are used to construct an adjacency matrix. In some cases, sequence data is viewed as a series of observations in continuous time (e.g., sensor data), and the sequence data is converted to point data by sampling at equal intervals. In addition, some types of sequence data (e.g., trajectory data) are converted into raster data, and the positions of different time instants correspond to the coordinates of the grid in raster data. For some raster data (such as meteorological data), sequence data is obtained by performing continuous time statistics on the observations at each site.
Challenges
In this section, we discuss the challenges of MvSD and summarize the existing literature. We state the following five problems in MvSD: temporal dynamic, heterogeneity, cross-view dynamics, data missing, misalignment of asynchronous multi-view.
Temporal dynamics
For each sequential data, consequent changes are recorded in chronological order, showing the dynamics at different time slots. The data points at different times in sequential data depend on each other. If the dynamic information of the temporal granularity is ignored, the regularity of sequence change becomes difficult to model and the accuracy will decrease. For example, in the sentiment analysis task, the expressed opinions such as “I think it's...but...”, the semantics will be inconsistent or further enhanced, which is also known as intra-modality dynamic. Information at certain moments will drive sentiment recognition. In traffic forecasting, the detected flow in the same road section is affected by human travel and often shows closeness, period and trend. In crime prediction, the factors that cause crimes may change over time. For example, there are different crime patterns on weekdays and weekends. In air quality forecasting, air quality monitoring stations record changes in the next few hours or a day. These time series show dynamic changes, and even some unexpected factors attacks may lead to sudden changes.
Early sequence modeling methods were proposed under specific tasks. For example, Prophet (Taylor and Letham 2018) was proposed by Facebook in 2017 for the company’s internal business time series. And the early air quality prediction tasks were modeled by random forest (Fawagreh et al. 2014) and inverse distance weighting (Lu and Wong 2008). Autoregression (AR) models, used to describe certain time-varying processes, such as stock forecasts (Ferenstein and Gasowski 2004), climate changes (Janjua et al. 2014). In addition, AR models and their variants are used in prognostication and health monitoring (PHM) (Barraza-Barraza et al. 2017), some variants such as autoregressive moving average (ARMA) (Pham and Yang 2010), autoregressive integrated moving average (ARIMA) (Ordóñez et al. 2019). Further, some methods based on Gaussian process (Zhao and Sun 2016a, b), Markov chain model (Sun et al. 2015), ARIMA (Chen et al. 2011), etc., have been proposed for traffic prediction.
The current schemes for modeling the temporal dynamics of multi-view sequences is to use networks based on RNNs and their derivatives. In sentiment analysis, Refs. (Zadeh et al. 2017; Verma et al. 2020; Wang et al. 2019c) employed independent LSTM to model intra-modality dynamic separately for each view sequence. To model the context of the sequence, Refs. (Hazarika et al. 2020; Xu et al. 2019) introduced bi-directional LSTM to obtain feature representations for each view. In order to obtain the temporal dependence of each weather sequence, DAQFF (Du et al. 2019) utilized bi-directional LSTM to learn long-term temporal characteristics from multivariate time series. DeepAir (Yi et al. 2018) followed the method of DeepSD (Wang et al. 2017), using RNN to embed sequence data to find similarities in different time slots.
In addition, there are some literatures that combine attention-based structures to tackle temporal dynamics. Pham et al. (2018) learnt common representations for different modalities via sequence-to-sequence (Seq2Seq) and introduced an attention mechanism to handle long-term dependencies. To address temporal duplication content in the identical view, Tian et al. (2020) adaptively aggregated useful information through self-attention. DCRNN (Li et al. 2017) captured temporal dependencies among time series through gated recurrent unit (GRU). ST-MetaNet (Pan et al. 2019) proposed a meta RNN, which uses meta-knowledge to generate GRU weights from node embeddings to model diverse temporal dependencies. Forecaster Li and Moura (2019) applied graph transformer to model long-term temporal dependencie. In order to solve the cumulative error amplification in sequence prediction, GMAN (Zheng et al. 2020) directly encoded historical inputs and generated future time steps by transform attention, thereby mitigating the error propagation problem.
Heterogeneity
MvSD consists of a sequence of views from multiple domains, and these views are often heterogeneous. As shown in Fig. 6, the various data mentioned in Sect. 2.1 have their own distributions. For example, image and text data are presented in different forms. Images are usually composed of raster pixels, and the content is intuitive to humans. Whereas, textual data, which usually consists of words and symbols, follows linguistic logic and is therefore more complex than images.Fig. 6 Heterogeneity of MvSD. Different view data has different distribution
In order to solve the heterogeneity of different views, models trained from specific domains are usually used to extract feature representations on corresponding view. For example, Refs. (Zadeh et al. 2017; Verma et al. 2020) extracted language, audio and visual features through three independent modality-specific LSTMs, and then explored the relationships between these modalities in the feature space. ADAIN (Cheng et al. 2018) combined feedforward neural network (FNN) and RNN, where FNN extracted static features and RNN learnt time series features. The obtained features from different views are combined for subsequent networks. stMTMV (Liu et al. 2016b) introducesd linear functions to deal with spatial and temporal features separately, and then aligned spatio-temporal views on nodes. DeepCrime (Huang et al. 2018) proposed a category-dependent encoder that encoded regions and crimes separately and finally mapped them in a common latent space.
An encoder-decoder structure is used to implement transitions between modalities to address view heterogeneity. Pham et al. (2018) translated two modalities into another joint representation via Seq2Seq model. MCTN (Pham et al. 2019) converted one modality to another via circular translation. Furthermore, for the three modalities, the representation learned between the two modalities was further transformed into the other modal, thereby forming the final joint representation. Forecaster (Li and Moura 2019) adopted the encoder-decoder architecture, taking spatial information and auxiliary information as the encoder input, and the decoder predicted the future spatial information.
The heterogeneity gap between different views is minimized in the common feature space. ARGF (Mai et al. 2020) introduced an adversarial approach to transform the distribution of the source modality into the distribution of the target modality. Inspired by domain adaptation, MISA (Hazarika et al. 2020) mapped multiple modalities into a shared subspace according to a weight-sharing encoder, and aligned these features by introducing metric distances.
Cross-view dynamics
MvSD has temporal dynamics within a single view sequence, while there are dynamic interactions between different view sequences. We consider cross-view dynamics into two categories, spatio-temporal correlations and semantic interactions.
Spatio-temporal correlations MvSD changes continuously in time and manifests differently in space, with spatio-temporal dynamics within a single view sequence or across views. For example, each image in a video can be viewed as a continuous change in space over a time period. For another example, in traffic forecasting, the observations of each traffic sensor are closely related to the observations of the surrounding space, and each observation value is also related to its own historical observation. There are many studies exploring spatio-temporal correlations. A conventional way is to model the local space first and then mine the temporal dynamics with recurrent networks, such as combining local convolution with recurrent networks (Yao et al. 2018; Zhou et al. 2020; Bai et al. 2019; Yu et al. 2017; Song et al. 2020; Wang et al. 2020e; Yuan et al. 2018; Chen et al. 2019; Zhang et al. 2017). Bai et al. (2019) combined graph convolutional network (GCN) with LSTM, where the local spatial correlations captured by the graph convolutional network were fed into a multi-layer LSTM to model the temporal relationships. In addition, combining attention mechanism and encoder-decoder structure is also used for spatio-temporal dynamic modeling (Li and Moura 2019; Shi et al. 2020; Yin et al. 2021a; Wu et al. 2020). APTN (Shi et al. 2020) used an attention-based encoder to model spatial, temporal, and periodical. The decoder introduced temporal attention to explore the dependence of the time steps. Forecaster (Li and Moura 2019) integrated the dependency graph into Transformer for forecasting spatially and temporally related data.
Semantic interactions Semantic interactions are often manifested in interactions between multiple views. For specific tasks, these views contain supplemental information that enhances specific views. Taking video sentiment analysis as an example, it usually treats language as the primary modality, as for images and audio as auxiliary modality. To model the semantic dynamics across views, memory-based methods are usually employed (Tian et al. 2020; Zadeh et al. 2018b, c; Ismail et al. 2020; He et al. 2020b). Furthermore, encoder-based methods transform multiple views to a specific view to learn a common representation (Pham et al. 2018; Xu et al. 2019; Mai et al. 2020; Hazarika et al. 2020). In addition, some literatures employ contrastive learning to achieve feature-level and semantic-level interactions (Mai et al. 2021; Liu et al. 2021; Kim et al. 2021). Mai et al. (2021) performed intra-modal/inter-modal contrastive learning and semi-contrastive learning simultaneously to ensure that the intra-modal/inter-modal dynamics are fully learned.
Data missing
MvSD collects data from different sources, some human factors, communication delays, and sensor failure usually cause partial or fully missing of temporal data, thus data missing is a very common phenomenon. In other words, ideally complete MvSD is rare. As shown in Fig. 7, we illustrate different types of missing data. Figure 7a is the complete multi-view sequence, each view is intact during training and testing, and different views are paired with each other. Figure 7b and Figure 7c represent missing data during training and testing phases, respectively. Figure 7d indicates that there are missing data in both training and testing phases. In this section, we introduce recent deep learning methods about data missing in MvSD.Fig. 7 Different missing types of MvSD. Taking text and video sequences as examples, (a) is the complete multi-view sequence. We summarize the missing types as missing in the training phase (b), missing in the testing phase (c), and missing both in the training and testing phases (d)
Reconstructing missing data using autoencoders is one solution. The purpose of the autoencoder is to encode source data into latent features, and then use a decoder to decode the latent features into target domain data. DCC-CAE (Dumpala et al. 2019) combined deep canonical correlation analysis (DCCA) with cross-modal autoencoders. DCC-CAE assumes that audio and visual modalities are available during training, but only one modality is available during testing. DCC-CAE is composed of two decoders, which input available modalities and reconstruct the corresponding missing modality representations. CPM-Nets (Zhang et al. 2020) reconstructed the complete view by constructing latent representations through structural constraints. In the unsupervised situation, CPM-Nets proposed adversarial strategies to further improve the complete representation. MCTN (Pham et al. 2019) introduced cyclic consistency loss in the process of modality translation to make the learned joint representation contain as much of all modality information as possible. MFM (Tsai et al. 2018) decomposed the multimodal representation into two factors: multimodal discriminative and modality-specific generative factors. Among them, the discriminant factors contain the shared joint features used to discriminate the task. The information contained in generative factors is unique to generate specific modalities.
Meta-learning is a learning-to-learn algorithm that learns multiple tasks on training data and processes new tasks during testing. Meta-learning enables knowledge transfer for task-agnostic few-shot learning. SMIL (Ma et al. 2017) is the first work to study the lack of data in both training and testing phases. SMIL jittered the latent feature space through Bayesian meta-learning, making single modality embeddings approximate to full modality embeddings. Meta-learning based spatio-temporal network (Yao et al. 2019a) is suitable for solving the problem of unbalanced spatial distribution of collected data, and transferring knowledge from multiple cities to target cities.
Misalignment of asynchronous multi-view
Sequential data from different views are usually not strictly aligned. One is that the lengths of the view sequences are not equal. For example, affected by the sampling frequency, the number of images and the number of words in the video are not equal. The other is semantic misalignment. There is no complete correspondence between images and text in video data. Each image does not correspond to each word, and one character can be associated with multiple images. Figure 8a shows the ideal condition, that is, the sequence length and semantics of multiple views are completely aligned. However, in reality, more scenarios are shown in Figure 8b and c, where Fig. 8b shows length aligned but semantics unaligned, and Fig. 8c shows both length and semantics misaligned.Fig. 8 Misalignment of asynchronous multi-view sequence
Most of the existing work, such as (Zadeh et al. 2017; Wang et al. 2019c; Liang et al. 2018a), are based on the multi-view sequence alignment. Some recent work focuses on multi-view sequence misalignment (Tsai et al. 2019; Yang et al. 2020; Aytar et al. 2017; Le et al. 2018). Recently, some literatures adopt attention-based structure to achieve multi-view sequence alignment (Tsai et al. 2019; Yang et al. 2020; Le et al. 2018). Multimodal Transformer (Tsai et al. 2019) focused on the interaction between multi-view sequences of different time steps through cross-modal attention to transform one modality to another without explicitly aligning the data. Furthermore, multi-view sequence alignment is also achieved by using pretrained network. pre-trained networks. Aytar et al. (2017) trained a deep convolutional network, which used a large amount of alignment data (including language, visual and acoustic) for aligned cross-modal representation. There are other literatures that use multi-instance learning without explicit data alignment. Tian et al. (2020) formulated weakly supervised video parsing as multi-modal multi-instance learning (MMIL), and proposed MMIL pooling to aggregate multi-modal information.
Multi-view sequential representation
In Sect. 2.1, we introduce various data forms. In order to feed these data into the network for feature learning, we need to choose appropriate methods to represent these data. In this paper, we mainly consider three representations for network input: rasterized representation, sequential embedding and graph embedding.
Rasterized representation
As shown in Fig. 9, the rasterized data quantifies the points in each grid (such as events, track points, traffic flow, meteorological data, etc.). Each cell in the raster grid is regarded as the statistics of a region. For example, in autonomous driving, points in the scene are divided into 3D rectangular grids with a given resolution (Zhou and Tuzel 2018; Yang et al. 2018; Laddha et al. 2021; Fadadu et al. 2022) to obtain voxel grids or BEVs, and then 3D CNN can be naturally applied. In traffic flow forecasting, the whole city is split to n×n grids according to latitude and longitude, and each region is indexed by rows and columns, and each grid is aggregated according to time intervals (Zhou et al. 2020; Yuan et al. 2018; Wu et al. 2020; Liao et al. 2018; Wang et al. 2020f; Zhang et al. 2019, 2021a). Through rasterization, CNN extracts the spatial features of different regions. Using recurrent networks to model raster data across multiple time slices enables analysis of dynamic relationships between different regions.Fig. 9 Illustration of rasterized representation
Sequential embedding
For sequence data, such as text, trajectory data, and time series data, some feature transformation and feature embedding methods need to be used to process these data. For time series data, multi layer perceptron (MLP) is usually used to map it into latent vectors. Yi et al. (2018) transformed the raw features of each domain data into a low-dimensional space through embedding methods. Cheng et al. (2018) applied fully connected layers (FC) to extract features of points of interests (POIs) and meteorological features (such as weather, temperature, humidity, etc.). DAQFF (Du et al. 2019) set 1×1 convolution to transform multiple time series data. In the language model, pre-trained models are usually used to perform feature transformation on the language sequence. For example, Word2Vec (Mikolov et al. 2013), GloVe (Pennington et al. 2014), and BERT (Devlin et al. 2018) are commonly used in natural language processing. Some literatures use pretrained 300-dimensional Glove word embeddings to encode a sequence of transcribed words into a sequence of word vectors (Zadeh et al. 2017; Wang et al. 2019c; Liu et al. 2018; Zadeh et al. 2019, 2018b). Still others use the pretrained Transformer model BERT to extract utterance level textual features (Ismail et al. 2020; Rahman et al. 2020; Yu et al. 2021b; Sun et al. 2020e).
Graph representation
Graph models a set of objects and their correlations in sentences, images, and spatial. Embedding the graph into vector representation and then seamlessly connect with GCN in subsequent processing. The graph representation reflects the association of different areas of the spatial map (Yao et al. 2018; Li and Moura 2019; Bai et al. 2019; Huang et al. 2020b; Yu et al. 2017; Wu et al. 2020). Forecaster (Li and Moura 2019) learnt the weights of the non-zero entries of the adjacency matrix by introducing a sparse linear layer, taking into account that different locations may have different dependency strengths. In the spatial map, DMVST-Net (Yao et al. 2018) employed CNN to extract local feature representation of each local region and its surrounding neighbors, and embeded the local representation into a low-dimensional representation through FC to supplement the information of the graph nodes. In spatial event prediction, Wu et al. (2020) designed an embedding component to generate an embedding vector for each event time step in each time slot. The graph representation reflects the relationship between sentence and image (Wang et al. 2020a). Additionally, the graph representation also reflects the activity on social media (Islam and Goldwasser 2021).
Deep neural networks
In this Section, we mainly review deep network techniques commonly used to extract features of multi-view sequences. In the network, different data representations need to choose suitable network models.
CNN-based networks
Figure 10 shows the CNN structure, which usually consists of convolutional layers, pooling layer, and fully connected layer. BatchNorm (Ioffe and Szegedy 2015) can be appended after convolutional layers. CNN has excellent performance in processing regular grid data. By stacking multiple convolutional layers, the learning from the bottom-level information to the high-level semantic features is realized in bottom-up. It moves fixed-size filters (e.g., 3×3, 5×5) on the input grid from left to right and from top to bottom. The filters performs inner product operation on corresponding position and generates high-dimensional features.Fig. 10 Structure of CNN
Some works try to convert traffic networks at different time intervals into raster images, and use CNN to extract spatial correlations (Zhang et al. 2017, 2019). ST-ResNet (Zhang et al. 2017) designed a CNN to capture the spatial dependency of closeness, period, and trend. MDL (Zhang et al. 2019) converted the directed graph at each time slot into a tensor representation, which was used by CNN to extract the spatial relationship, and then fully convolutional network (FCN) (Long et al. 2015) was introduced to obtain the temporal dependencies.
RNN-based networks
RNN and its variants are models for processing sequence data, which use the previous output state of the sequence to predict the next state, as is shown in Fig. 11a. LSTM is an extension of RNN. It has a specially customized memory unit to remember longer input history information, as shown in the Fig. 11b. They are widely used in speech recognition, natural language processing and time series data analysis.Fig. 11 Structures of RNN, LSTM and Seq2Seq
A series of RNN-based methods are used in traffic forecasting (Bai et al. 2019; Yuan et al. 2018; Wang et al. 2020f). In passenger demand prediction, three LSTMs are used to model spatiotemporal maps, external meteorological data, and temporal metadata, respectively (Bai et al. 2019). Hetero-ConvLSTM (Yuan et al. 2018) extracted spatial features through ConvLSTM (Xingjian et al. 2015), and then fed the obtained features to LSTM to model temporal dynamics. MT-ASTN (Wang et al. 2020f) modeled the temporal features of dynamic graph sequence with different scales to capture crowd flow at different scales. In the sentiment analysis task, Refs. (Zadeh et al. 2017; Verma et al. 2020; Ismail et al. 2020) utilized three independent LSTMs to construct modality embedding subnetworks for language, visual, and acoustic respectively to model intra-modality.
Figure 11c shows the structure of Seq2Seq (Sutskever et al. 2014). The Seq2Seq model is designed for sequence data and is an encoder-decoder structure, where the encoder encodes the input sequence to obtain hidden state, and the decoder generates variable-length output according to the hidden state. Pham et al. (2018) performed unsupervised learning of joint multimodal representations via Seq2Seq. Liao et al. (2018) proposed a hybrid Seq2Seq model, which integrated auxiliary information in the encoder-decoder sequence learning framework.
Graph-based networks
GCNs are often used to model non-Euclidean structural data, and GCNs are usually divided into two categories, namely spectral-based graph networks and spatial-based graph networks. The spectral-based method defines convolution operation in the Fourier domain (Kipf and Welling 2016). The spatial-based method directly applies convolution on the graph to aggregate information from neighbors. We categorize recent literatures according to the above two methods.
Spectral-based GCN introduces filters from a signal processing perspective to define graph convolution. Song et al. (2020) proposed a spatio-temporal synchronization graph convolutional network, which could effectively capture complex local spatio-temporal correlations through spatio-temporal synchronization modeling mechanism. In order to meet the requirements of medium and long-term prediction tasks, Yu et al. (2017) introduced a spatio-temporal graph convolution network, which modeled the traffic network as a graph, and used spectral convolution to extract spatial features. Geng et al. (2019) encoded the correlations of different regions to obtain multiple graphs, which were then used for correlation modeling based on ChebNet-based multi-graph convolution.
Spatial-based GCN simulates the convolution operation of traditional CNN, and graph convolution is based on the spatial relationship of nodes. Wang et al. (2020f) converted the flow Origin-Destination (OD) matrix into a semantic graph, and then performed convolution operation on the semantic graph. Li et al. (2017) treated traffic flow as a directed graph and captured spatial dependencies according to the diffusion process on the graph through a diffusion convolutional recurrent neural network. Graph attention network (GAT) (Veličković et al. 2017) is another graph neural network, which calculates the weights of neighbor nodes through an attention mechanism, without knowing the structure of the graph. Pan et al. (2019) proposed a meta-graph attention network, which used attention mechanism to capture the dynamic spatial correlation between nodes, and the attention weights were generated from meta-knowledge.
Attention-based networks
Attention is originally proposed for natural language processing, and is now widely used in sequence-based tasks, where it models the relevant parts of the information and amplifies the most important parts. Attention can be used not only to focus on spatial dependencies, but also on spatio-temporal correlations.
Shi et al. (2020) proposed an attention mechanism to model spatial, short-term and long-term cyclic dependencies. Huang et al. (2020b) developed a graph attention network integrated with GCN into a spatial gated block to capture spatio-temporal features. GMAN (Zheng et al. 2020) introduced a transform attention layer between encoder and decoder to model the relationship between historical and future time steps. Refs. (Tian et al. 2020; Wu et al. 2021) developed hybrid attention network to jointly model temporal recurrence, co-occurrence, and asynchrony. Specifically, temporal recurrence was solved by self-attention, while co-occurrence and asynchrony were addressed by cross-modal attention. Zadeh et al. (2018b) studied a delta-memory attention network, which focused on cross-view interactions, and aggregated interactions over time with multi-view gated memory.
Transformer based on attention has been successfully applied (Tsai et al. 2019; Zadeh et al. 2019; Hasan et al. 2021; Wang et al. 2020g). The structure of the Transformer is shown in Fig 12. Tsai et al. (2019) developed a multi-modal transformer to model unaligned multi-modal language sequences, and integrated multi-modal time series from multiple pairs of cross-modal transformers. Zadeh et al. (2019) designed a multimodal transformer layer to capture the decomposition dynamics of multi-modal data and aligned temporally asynchronous information intra-modality and inter-modality. Hasan et al. (2021) encoded multi-modal sequences separately via Transformer, and learnt to represent punchline according to the background context.Fig. 12 Structure of Transformer (Vaswani et al. 2017)
Hybrid networks
Some works are dedicated to combining multiple modules, such as combining CNN and LSTM (Xingjian et al. 2015; Zhao et al. 2019; Yao et al. 2019b). Chen et al. (2019) combined GCN with GRU, GCN was used for spatial feature extraction, and GRU was used for capturing temporal dynamics. GC-LSTM (Chen et al. 2018) introduced LSTM to model the characteristics of the dynamic graph sequence and captured the temporal characteristics of the graph sequence. Poria et al. (2017) proposed a contextual attention-based LSTM network that focuses on contextual relations, and the attention-based fusion mechanism amplified higher quality and informative modalities. Huang et al. (2020b) adopted GCN to extract spatial features and graph attention network to extract road similarity, and finally integrated these two modules through a gate structure. Pan et al. (2019) designed a combination of meta-graph attention network and meta-recurrent network, where the meta-graph attention network captured spatial correlation, and the meta-recurrent network modeled temporal correlation.
Zadeh et al. (2018c) proposed a multi-attention recurrent network, composed of long-short term hybrid memory and multi-attention block, to discover the interactions between modalities and store them in hybrid memory. Wang et al. (2019c) combined LSTM and gating mechanism, where LSTM was used to model different view sequences, and the gated modality-mixing network was used for infering nonverbal shift vector. Xu et al. (2019) adopted a combination of bi-LSTM and attention mechanism, where bi-LSTM extracted text sequences and attention mechanism learnt the alignment weights between speech and text.
Discussion
In this section, we analyze the above-mentioned deep models, as shown in Table 1. We demonstrate popular deep models, and describe them in terms of model architecture, data form, model characteristic, and application areas.Table 1 Comparison of deep models
Models Approach Application Details
Architecture data characteristic
TFN (Zadeh et al. 2017) LSTM Text, audio, image RNN-based Sentiment analysis Using three independent LSTMs to model the temporal information of each modality separately
ST-ResNet (Zhang et al. 2017) CNN Rasterized map, time series CNN-based Traffic forecasting This paper constructs 3 views: closeness, period and trend, and uses CNN to model spatio-temporal correlations
GSTNet (Fang et al. 2019) GCN Time series, graph Graph-based Traffic forecasting This paper constructs a global spatio-temporal network, which uses temporal convolution to capture temporal features and a graph model to capture spatial dependencies
MLRF (Liang et al. 2018b) LSTM Image, audio RNN-based Emotion recognition Multimodal local ranking is used to obtain relative sentiment strength, then a Bayesian ranking algorithm infers global ranking, and finally combines multimodal behavior and relative sentiment ranking for inference
ADAIN (Cheng et al. 2018) LSTM, attention Time series Hybrid Climate science This paper models different local temporal information through LSTM, and these latent local features are finally fused through attention
DeepCrime (Huang et al. 2018) GRU, attention Time series Hybrid Crime analysis This paper models crime dynamics through hierarchical GRU and uses attention to capture the spatial and temporal associations of different crime patterns
Pham et al. (2018) Seq2Seq Text, audio, image RNN-based Sentiment analysis Using Seq2Seq for unsupervised joint multimodal representation learning
MISA (Hazarika et al. 2020) LSTM Text, audio, image RNN-based Sentiment analysis Using 3 independent LSTMs to extract the features of text, audio and image respectively. Specifically, to model intra-modal and inter-modal relationships, MISA performs distance metrics in two subspaces
HybridAtt (Yuan et al. 2019) CNN, Attention Time series Hybrid Sleep staging 1D CNN is used to process multivariate polysomnography records, and the designed hybrid attention mechanism further fuses multi-view features
MetaST (Yao et al. 2019a) CNN, LSTM Rasterized map Hybrid Water quality prediction A meta-learning paradigm spatiotemporal network is designed to solve the data imbalance problem through transfer learning
Yuan et al. (2018) ConvLSTM Rasterized map Hybrid Accident prediction A Hetero-ConvLSTM is proposed to handle both spatial heterogeneity and temporal autocorrelation
GraphSleepNet (Jia et al. 2020) GCN, attention Rasterized map Hybrid Sleep staging Based on spatiotemporal graph convolution network, graph convolution extracts spatial features and temporal convolution is used to capture transition rules between sleep stages
Hasan et al. (2021) Transformer Text, audio, image Attention-based Humor detection Language-, acoustic-, visual-, and humor-centric features are extracted separately using Transformers, followed by cross-attention layers for information interaction
Tian et al. (2020) Attention Image, audio Attention-based Video parsing A hybrid attention network is proposed to simultaneously explore unimodal and cross-modal temporal context. A multimodal multiple instance learning (MMIL) pooling method is developed to adaptively explore useful audio and video content
It can be seen that for sequence data processing, whether it is text, traffic flow or biological signals, RNN-based networks are general frameworks. The reason is that RNN-based networks can model the dependencies of these temporal information. For rasterized data, such as images, spatial maps, etc., CNN-based models are usually used for spatial modeling. In traffic prediction, specific nodes of some grid maps are turned into graph networks, which are suitable for GCN, spatial attention networks, etc. For multimedia data, such as text streams, audio, etc., in terms of feature representation, candidate extractors such as CNN, Transformer are available. Furthermore, it can be found that combining CNNs and recurrent networks becomes a paradigm for spatio-temporal modeling. For multi-view sequence modeling, the paradigm uses view-private recurrent networks for feature extraction, followed by cross-view interaction.
Applications
In this section, we summarize related work in different domains, including intelligent transportation, crime analysis, sentiment analysis, climate science and health care. We discuss these application areas separately and provide an overview of recent related techniques. In Table 2, we enumerate the application areas along with the aforementioned models.Table 2 Deep models for different applications of MvSD
Transportation Health care Crime analysis Sentiment analysis Climate science
CNN-based Zhou et al. (2020), Zhang et al. (2017, 2019), Lv et al. (2019), Chen et al. (2020), Guo et al. (2019a), Sun et al. (2020a) Akman et al. (2021), Jia et al. (2021b), Nessiem et al. (2021), Coppock et al. (2021), Piriyajitakonkij et al. (2020) Wang et al. (2019a), Okawa et al. (2019), Salama et al. (2021) Yi et al. (2018)
RNN-based Yuan et al. (2018), Yao et al. (2019a), Wang et al. (2020f), Huang et al. (2014), Yang et al. (2016) Le et al. (2018), Phan et al. (2019, 2021), Olesen et al. (2021) Huang et al. (2018) Zadeh et al. (2017), Hazarika et al. (2020), Pham et al. (2019), Liu et al. (2018), Sun et al. 2020e; Mai et al. 2019) Du et al. (2019, 2021), Cheng et al. (2018), Veiga et al. (2021)
Graph-based Pan et al. (2019), Song et al. (2020), Geng et al. (2019), Fang et al. (2019), Sun et al. (2020b), Bai et al. (2021), Chen et al. (2020), Guo et al. (2020) Jia et al. (2020, 2021a) Zhang et al. (2021b), Wang et al. (2021), Xia et al. (2021) Mai et al. (2020), Wang et al. 2022) Liu et al. (2016b), Chen et al. (2021c)
Attention-based Zheng et al. (2020), Shi et al. (2020), Huang et al. (2020b), Guo et al. (2019b) Phan et al. (2022) Rayhan and Hashem (2020) Tian et al. (2020), Zadeh et al. (2018b), Wu et al. (2021), Hasan et al. (2021), Brousmiche et al. (2021), Gu et al. (2018), Akhtar et al. (2019)
Hybrid Yao et al. (2018, 2019b), Li et al. (2017), Bai et al. (2019), Yu et al. (2017), Zadeh et al. (2019), Zhao et al. (2019), Guo et al. (2019b) Yuan et al. (2019), Supratak et al. (2017) Stec and Klabjan (2018), Ertugrul et al. (2019) Xu et al. (2019), Zadeh et al. (2018c), Tsai et al. (2019), Zadeh et al. (2019), Bie and Yang (2021), Chen et al. (2017), Han et al. (2021a) Cheng et al. (2018), Yao et al. (2019a), Zhong et al. (2020), Han et al. (2021c), Sasaki et al. (2021), Ouyang et al. (2021), Civitarese et al. (2021), Lin et al. (2020), Lu and Li (2020)
Intelligent transportation
With the development of mobile communication and human daily travel, a large amount of traffic data is generated, which usually includes traffic volume, speed, accidents, trajectories, spatial maps, and road networks, etc. Traffic data mining and analysis has become an urgent problem to be solved. Traffic forecasting plays an important role in smart cities, providing constructive guidance for urban planning, intelligent management and public safety, thereby promoting urban construction and avoiding waste of resources. Table 3 summarizes the performance of spatio-temporal models on several publicly available benchmark datasets. In TaxiBJ1 and NYC Bike2, we report root mean square error (RMSE) in terms of both flow and demand. For the two types of graph structure data, PeMSD43 and MeTR-LA4, we mainly summarize RMSE and mean absolute error (MAE).Table 3 Various traffic prediction models performance on TaxiBJ, NYC Bike, PeMSD4, and MeTR-LA
Models TaxiBJ NYC Bike Models PeMSD4 MeTR-LA
Flow (RMSE) Demand (RMSE) Flow (RMSE) Demand (RMSE) RMSE MAE RMSE MAE
HA 57.69 40.439 – 8.541 HA 54.14 36.76 7.8 4.16
ARIMA 22.78 – 10.07 – ARIMA 58.05 35.19 10.45 5.15
VAR 22.88 – 9.92 – VAR 51.62 33.63 9.13 5.41
DCRNN (Li et al. 2017) – 20.569 – 5.215 DCRNN (Li et al. 2017) 37.12 24.55 6.383 3.125
STGCN (Yu et al. 2017) – 19.101 – 4.759 STGCN (Yu et al. 2017) 38.16 27.02 7.24 3.47
DeepST (Zhang et al. 2016) 18.18 – 7.43 – TGCN (Zhao et al. 2019) – – 8.152 4.319
ST-ResNet (Zhang et al. 2017) 16.69 17.649 6.33 6.159 AGCRN (Bai et al. 2020) 32.26 19.83 6.535 3.209
ConvLSTM (Xingjian et al. 2015) – 18.788 – 4.745 DGCN (Guo et al. 2020) 31.61 19.6 – –
DMVST-Net (Yao et al. 2018) – 18.206 – 4.766 ASTGCN (Guo et al. 2019b) 32.82 21.8 6.78 3.841
As mentioned in Sect. 2.1, traffic data is typically represented as trajectories, events, and raster data. Among them, data such as trajectories and events are usually converted to raster data for processing, and then this type of data is consumed by networks such as CNN (Lv et al. 2019; Chen et al. 2020; Guo et al. 2019a; Sun et al. 2020a). The traffic forecasting of a single road segment can be regarded as sequence data, which is fed to RNN or LSTM (Huang et al. 2014; Yang et al. 2016). The road network is represented using a graph, and these road network associations are modeled by GCN (Yu et al. 2017; Geng et al. 2019; Fang et al. 2019; Sun et al. 2020b; Bai et al. 2021). In addition to the diversity of data forms, complex spatial-temporal correlations hinder traffic prediction. Among them, the spatial correlation is mainly reflected in different regions and different road segments. Traffic in adjacent areas is spatially causal, i.e. flows from one area to another. The temporal correlation is reflected in the fact that the flow of the area is affected by different time intervals, such as short-term or long-term changes. Therefore, spatial-temporal based models are used to deal with this problem (Bai et al. 2019; Zhao et al. 2019; Guo et al. 2019b).
Further, traffic data is affected by external factors (such as weather, accidents), and the data comes in various forms. Some literatures (Zhou et al. 2020; Wang et al. 2020f; Zhang et al. 2019) model the temporal dynamics into three views: closeness, period, and trend. In traffic flow forecasting, Guo et al. (2019b) combined spatio-temporal attention and spatio-temporal convolution to construct three basic components to model the recent, daily-periodic and weekly-periodic of traffic flow, respectively. In demand forecasting, Bai et al. (2019) took the structured city as one view, external meteorological data and time meta as the other two views. For the spatial view, the spatial features of each time slot were extracted by GCN, and then the spatio-temporal correlations were captured by LSTM. For the other two views, these factors were used to model passenger demand. Yao et al. (2018) built three views: spatial view, temporal view, and semantic view. Geng et al. (2019) studied multiple graphs to model the relationship between different regions, including neighborhood, functional similarity, and connectivity.
Example 1
Zhou et al. (2020) proposed a deep flexible structured spatial–temporal model (DFSSTM) for taxi capacity prediction. The structure of DFSSTM is shown in Fig. 13. First, the traffic data was rasterized into a n×n grid, and the flow relationships of vehicles were characterized by inflows and outflows. Due to the short-term and long-term dependencies of traffic data, DFSSTM divided temporal dependencies into three views: period, trend and closeness. Subsequently, DFSSTM tailored a siamese spatio-temporal network (SSTN), which took both inflows and outflows as inputs, to model spatio-temporal dependencies. Three SSTNs were used to model three temporal dynamics (period, trend and closeness). Finally, DFSSTM designed a fusion layer that automatically adjusted the weights to integrate different views.
Fig. 13 Structure of DFFSTM (Zhou et al. 2020)
Health care
In the medical system, clinicians diagnose patients through comprehensive consideration of multiple factors (for example, previous medical history, various biological indicators, patient physique, etc). This process is complicated and time-consuming. In order to better assist clinicians in diagnosis, many works apply deep learning technology in the field of medical intelligence (Yuan et al. 2019; Jia et al. 2020; Akman et al. 2021; Phan et al. 2022; Olesen et al. 2021; Piriyajitakonkij et al. 2020; Feng et al. 2021; Torres et al. 2016). The intelligent system discovers disease patterns by learning historical data and various clinical indicators, supplemented by experts knowledge. Table 4 summarizes the performance of the automatic sleep staging models on the SleepEDF-205, MASS6, and SHHS7 datasets. We report overall accuracy (Acc), Cohen’s kappa (κ), and macro F1-score (MF1).
Clinical data usually comes from multiple views and is heterogeneous (for example, verbal description, medical imaging, polysomnography, etc.). To handle asynchronous view sequences, Le et al. (2018) investigated memory technology to establish cross-view interactions and dependencies. It encoded the input view separately through two encoders and saved to two external memories. In sleep monitoring, Phan et al. (2021) took the original signal and time-frequency information as input. In order to solve the over-fitting rate between different views, it dynamically adjusted the learning steps between different modalities, and derived weights based on specific views for fusion different views. Jia et al. (2021b) proposed a temporal fully convolutional network based on U2-Net (Qin et al. 2020) for multimodal salient waves detection, which converted the time series classification problem into a salient detection problem. At the same time, the multi-scale features of the sleep stage are captured by the multi-scale extraction module. In recent years, the coronavirus (COVID-19) has spread worldwide, causing a large number of human casualties and economic losses. Some works use coughing and breathing audio to determine whether COVID is positive or negative (Akman et al. 2021; Nessiem et al. 2021; Coppock et al. 2021).Table 4 Performance of sleep staging models on SleepEDF-20, MASS, and SHHS
Models SleepEDF-20 MASS SHHS
Acc κ MF1 Acc κ MF1 Acc κ MF1
DeepSleepNet (Supratak et al. 2017) 82.0 0.760 76.9 86.4 0.805 82.2 – – –
FCNN+RNN (Phan et al. 2021) 83.5 0.775 77.7 86.4 0.806 82.1 88.1 0.832 80.9
TinySleepNet (Supratak and Guo 2020) 85.4 0.800 80.5 83.1 0.77 78.1 – – –
RobustSleepNet (Guillot and Thorey 2021) – – 81.7 – – 82.5 – – 80.0
SleepTransformer (Phan et al. 2022) – – – – – – 87.7 0.828 80.1
SeqSleepNet (Phan et al. 2019) 86.0 0.809 79.7 87.0 0.815 0.833 88.4 0.838 80.1
XsleepNet (Phan et al. 2021) 86.4 0.813 80.9 87.6 0.823 83.8 89.1 0.847 82.3
SalientSleepNet (Jia et al. 2021b) 87.5 – 83.0 – – – – – –
Crime analysis
Crime prediction plays a vital role in crime prevention. Recent works (Huang et al. 2018; Okawa et al. 2019; Stec and Klabjan 2018; Vomfell et al. 2018) can dig out the spatial-temporal patterns and trends of crimes through historical data combined with information such as crime incidents, time and location. This contributes to the early warning, allowing police to conduct inspections in high-risk regions, reducing the impact on society. However, unlike traffic data, the distribution of crime incidents in spatial and temporal is sparse, and there are fewer spatial-temporal associations between different crime incidents.
In order to solve the high heterogeneity of crime spatial distribution, the city is usually converted to raster data to extract spatial attributes, and then a recurrent network is used to capture the temporal dynamics. Wang et al. (2019a) selected crime incidents in a specific area, and divided the area into 16×16 raster images. Motivated by ST-ResNet (Zhang et al. 2017); Wang et al. 2019a) constructed three views to extract nearby, periodic, and trend features separately. To model the association between crime and different regions, DeepCrime (Huang et al. 2018) proposed a category-interactive encoder, which embeds information such as spatial and event categories into latent vectors for representation. In addition, DeepCrime (Huang et al. 2018) adopted three GRUs to separately encode crime sequences, abnormal sequences, and interdependent sequences to model the temporal dynamics. In order to make the model interpretable, Rayhan and Hashem (2020) considered an attention-based spatio-temporal network, which captured the dynamic spatio-temporal correlation of crimes based on past criminal events, external features and recurring trends. Specifically, two GAT variants were used to embed spatial hierarchical information and specific category features respectively. CASTNet (Ertugrul et al. 2019) designed a community-attentive spatio-temporal model to capture the spatio-temporal pattern of criminal events, which was used to predict opioid overdose. Specifically, CASTNet (Ertugrul et al. 2019) extracted opioid overdose at different locations through a multi-head attention network, and introduced hierarchical attention to allow interpretation of the contribution of features from different communities to local incident prediction. Table 5 summarizes the performance of the deep crime models on the New York City8 and Chicago9 datasets. Note that most of these models are derived from spatio-temporal models.Table 5 Summary of deep crime models performance on New York City, and Chicago
Models New York City Chicago
Categories Overall Categories Overall
Burglary Larceny Robbery Assault MAE MAPE Theft Battery Assault Damage MAE MAPE
SVM (Chang and Lin 2011) 0.449 0.565 0.553 0.426 1.086 0.683 0.654 0.643 0.507 0.631 0.990 0.586
ARIMA (Pham et al. 2019) 0.405 0.488 0.532 0.411 0.879 0.551 0.405 0.488 0.532 0.411 0.860 0.506
ST-ResNet (Zhang et al. 2017) 0.487 0.602 0.564 0.532 0.996 0.580 0.551 0.676 0.594 0.682 0.909 0.450
STGCN (Yu et al. 2017) 0.547 0.618 0.614 0.510 0.620 0.413 0.686 0.689 0.629 0.712 0.682 0.462
DeepCrime (Huang et al. 2018) 0.468 0.647 0.636 0.536 0.623 0.415 0.624 0.705 0.641 0.731 0.677 0.455
DCRNN (Li et al. 2017) 0.533 0.624 0.610 0.519 0.654 0.451 0.683 0.647 0.598 0.679 0.681 0.471
ST-MetaNet (Pan et al. 2019) 0.538 0.553 0.535 0.495 0.624 0.418 0.698 0.619 0.656 0.726 0.758 0.480
STDN (Yao et al. 2019b) 0.526 0.542 0.588 0.471 0.898 0.471 0.629 0.662 0.586 0.634 0.953 0.505
UrbanFM (Liang et al. 2019) 0.588 0.466 0.639 0.560 0.810 0.574 0.689 0.670 0.582 0.646 0.886 0.613
ST-SHN (Xia et al. 2021) 0.617 0.650 0.672 0.595 0.583 0.357 0.715 0.738 0.674 0.729 0.643 0.400
Sentiment analysis
The opinions expressed by humans in daily communication are usually complex and multi-modal, and it is of great significance for computer intelligence to understand these data. Sentiment analysis is also an important branch of future human-computer interaction.
In multi-modal sentiment analysis, these multi-view data are usually represented by vision, acoustic, and language. There are two challenges to be overcome in this task: intra-modality dynamics and inter-modality dynamics. Sun et al. (2020c) proposed a multi-view CRF model, which captured the correlation between features within a single view and considered the relationships between different views. Refs. (Zadeh et al. 2017; Liu et al. 2018) modeled the dynamics of specific modality sequences through three LSTMs and captured the interactions between modalities through a three-fold Cartesian product. Zadeh et al. (2018b) proposed a memory enhancement network, which modeled the interaction of multiple view sequences over time by introducing gated memory. Since different modalities are heterogeneous, MISA (Hazarika et al. 2020) mapped different modalities to a common subspace to learn shared feature representations. Mai et al. (2020) designed an adversarial encoder-decoder structure to embed different modalities into a common space and learn invariant representations of modalities. In order to deal with unaligned view sequences, Xu et al. (2019) introduced an attention mechanism to align speech and text, and realized the integration of speech and text at the word level. Table 6 summarizes the model performance on MOSI10, MOSEI11, and SIMS12, and the evaluation metrics follow previous work Rahman et al. (2020).Table 6 Performance of multimodal sentiment analysis models on MOSI, MOSEI, and SIMS
Models MOSI MOSEI SIMS
Acc-2 F1 MAE Corr Acc-2 F1 MAE Corr Acc-2 F1 MAE Corr
TFN (Zadeh et al. 2017) –/80.8 –/80.7 0.901 0.698 –/82.5 –/82.1 0.593 0.700 78.38 78.62 0.432 0.591
LMF (Liu et al. 2018) –/82.5 –/82.4 0.917 0.695 –/82.0 –/82.1 0.623 0.667 77.77 77.88 0.441 0.576
MFN (Zadeh et al. 2018b) 77.4/– 77.3/– 0.965 0.632 76.0/– 76.0/– – – 77.90 77.88 0.435 0.582
RAVEN (Wang et al. 2019c) 78.0/– 76.6/– 0.915 0.691 79.1/– 79.5/– 0.614 0.662 – – – –
MFM (Tsai et al. 2018) –/81.7 –/81.6 0.877 0.706 –/84.4 –/84.3 0.568 0.717 – – – –
Mult (Tsai et al. 2019) 81.5/84.1 80.6/83.9 0.861 0.711 –/82.5 –/82.3 0.580 0.703 78.56 79.66 0.453 0.564
MAG-BERT (Rahman et al. 2020) 84.2/86.1 84.1/86.0 0.712 0.796 84.7/– 84.5/– – – – – – –
MISA (Hazarika et al. 2020) 81.8/83.4 81.7/83.6 0.783 0.761 83.6/85.5 83.8/85.3 0.555 0.756 – – – –
Self-MM (Yu et al. 2021b) 84.0/85.98 84.42/85.95 0.713 0.798 82.81/85.17 82.53/85.3 0.530 0.765 80.04 80.44 0.425 0.595
MMIM (Han et al. 2021b) 84.14/86.06 84.00/85.98 0.700 0.800 82.24/85.97 82.66/85.94 0.526 0.772 – – – –
In audio-visual event recognition, Brousmiche et al. (2021) studied a multi-level attention fusion network, which could dynamically integrate visual and audio information for event recognition. Tian et al. (2020) presented a new task, named audio-visual video parsing (AVVP), which detected video events (labels them as audible, visible, or audible and visible) and located the duration via a weakly supervised approach. In this task, events may occur repeatedly in different views, so there are three challenges: unimodal-modal temporal recurrence, cross-modal co-occurrence, and cross-modal asynchronous. Tian et al. (2020) simultaneously employed a hybrid attention network to solve these problems, that is, using a self-attention mechanism to model unimodal-modal temporal recurrence, and using a cross-modal attention mechanism to simultaneously deal with cross-modal co-occurrence and cross-modal asynchronous.
Climate science
Weather data is usually collected by various sensor devices, including temperature, humidity, wind speed, pressure, air quality, etc. By studying the interrelationship of these meteorological data, it is helpful for human to further understand the earth’s environment and prevent natural disasters in advance.
Recently, some deep learning methods have been successfully applied to air quality prediction (Yi et al. 2018; Zhong et al. 2020; Sasaki et al. 2021; Ouyang et al. 2021; Lin et al. 2020). Refs. (Cheng et al. 2018; Du et al. 2021; Han et al. 2021c) adopted FC to extract local spatial features and LSTM to capture temporal dynamics. Yi et al. (2018) proposed a patial transformation component to aggregate spatial monitoring data of different scales. In addition, in order to model dynamic changes between cross-modal data, these features are fused in a distributed manner. Zhong et al. (2020) introduced reinforcement learning to predict air quality. The model mainly consisted of two components: site selector and air quality regressor. Among them, the site selector adaptively selected the relevant sites, and the quality regressor received the selected sites for air quality estimation. In water quality prediction, Liu et al. (2016b) studied a multi-task multi-view method to fuse the data from different domains to predict the water quality of a site. In extreme weather forecasting, Civitarese et al. (2021) proposes a temporal fusion transformer, which uses multiple variables (such as static, historical, and future) as input. In order to solve the imbalance in the spatial distribution of the collected data, Yao et al. (2019a) constructed a spatio-temporal network based on meta-learning, which transfered knowledge from multiple cities to help the target city make spatio-temporal predictions. Table 7 summarizes the performance of the deep air models on Beijing13 and Longdon14 dataset, we focus on the overall performance and the performance on PM2.5.Table 7 Air quality models performance on Beijing and LongDon
Datasets Models Overall PM2.5
RMSE MAE RMSE MAE
Beijing RF (Fawagreh et al. 2014) 26.68 15.59 – –
IDW (Lu and Wong 2008) 48.09 34.79 28.41 18.35
KNN 37.99 23.94 18.24 10.58
ADAIN (Cheng et al. 2018) 29.39 18.83 15.06 8.28
ANCL (Patel et al. 2022) 24.28 15.23 – –
MCAM (Han et al. 2021c) – – 12.57 6.83
LongDon RF (Fawagreh et al. 2014) 4.69 3.06 – –
IDW (Lu and Wong 2008) 8.01 5.50 – –
KNN 4.75 3.20 5.77 3.90
ADAIN (Cheng et al. 2018) 4.78 3.36 3.11 2.02
ANCL (Patel et al. 2022) 4.65 3.20 – –
MCAM (Han et al. 2021c) – – 2.81 1.78
Example 2
Du et al. (2019) proposed a deep air quality forecasting framework (DAQFF) for PM2.5 prediction. Figure 14 illustrates the network architecture of DAQFF. First, DAQFF customized multiple 1D convolutional neural networks for multivariate time series to model local features. Different from some spatio-temporal models, DAQFF concatenated the temporal features of multiple stations, which could simultaneously capture local features and spatial relationships across stations. Subsequently, to model long-term dependencies, DAQFF introduced bi-LSTM to extract long-term temporal dependencies. Finally, the obtained shared features were concatenated and fused for prediction.
Fig. 14 Structure of DAQFF (Du et al. 2019)
Future directions
In this Section, we enumerate some challenging deep learning techniques in MvSD in recent years, and point out potential future research directions.
Interpretable model research
Despite the impressive achievements of deep learning, the working principle of the model is in a black box, and the decision-making is difficult to establish a reasonable basis. Therefore, interpretability models have become a research hotspot, which builds trust with users to understand why decisions are made in different domains such as medical diagnosis, autonomous driving, or recommender systems.
Multi-view data tends to introduce bias because these views are heterogeneous and each view may have deviations, resulting in model error accumulation and amplification. Some works focus on feature-level interpretability (Rayhan and Hashem 2020; Ertugrul et al. 2019; Khanehzar et al. 2021), which achieves global interpretation by modeling local feature relationships. Meanwhile, attention is given to providing reliable explanations behind the predictions (Jia et al. 2021a; Zheng et al. 2021; Ma et al. 2018; Agyemang et al. 2020). Furthermore, it is also an approach to achieve interpretability by designing network architectures that conforms to human cognition (Choe et al. 2021). At present, there are no mature techniques and standards to estimate the performance of interpretability. Thus, it is impossible to compare the pros and cons of interpretable methods. In the future, interpretability will be further refined to guide the agent’s behavior or multi-view fusion decision-making.
Multi-modal architecture research
As the network structure becomes more and more complex, the cost and time of manually designing the network will be unbearable. Especially for MvSD, in the feature extraction stage, a feature extractor needs to be designed for a single view and there are many alternative network structures for different views. In the multi-view feature fusion stage, we need to consider aggregating multiple views strategies. The automated process of neural architecture search (NAS) can speed up research on MvSD.
Auto-MVCN (Li et al. 2020) tailored a multi-view architecture for 3D shape recognition, which explored correlations between view features by automatically searching for fused cells. In electronic health records, MUFASA (Xu et al. 2021) simultaneously searched modality-specific networks and feature fusion strategies. BM-NAS (Yin et al. 2021b) designed a bilevel search scheme, BM-NAS selected feature pairs from pre-trained unimodal and searched a feature fusion strategy. To make the model applicable to various multimodal tasks, MMnas (Yu et al. 2020a) defined a general network search framework to design task-specific heads for different tasks on a unified backbone. MFAS (Pérez-Rúa et al. 2019) found a reasonable network structure for multimodal fusion by constraining the search space and employing sequential model-based exploration methods. Although many recent works try to design fusion strategies, how to perform fusion of multi-view sequences is not well studied and needs more research in the future. Through multi-modal network architecture search, better models and fusion methods are obtained..
Data annotations
Deep learning benefits from massive amounts of data, however large-scale data annotation brings prohibitive costs, which becomes more severe when annotating MvSD. Therefore, some techniques based on unsupervised, semi-supervised, etc., are introduced to facilitate MvSD research.
Unsupervised learning uses unlabeled data. In multi-view representation learning, DUA-Nets (Geng et al. 2021) combined inverse networks through unsupervised learning to automatically evaluate the quality of different views. Through unsupervised training, contrastive learning has achieved great success in the computer vision domain (He et al. 2020a). In multi-modal sentiment analysis, Mai et al. (2021) performed intra-modal/inter-modal contrastive learning and semi-contrastive learning simultaneously to ensure that the intra-modal/inter-modal dynamics are fully learned. In semi-supervised learning, a small amount of labeled data is combined with a large amount of unlabeled data (Khanehzar et al. 2021; Chen et al. 2021a, b). ASM2TV (Chen et al. 2021a) designed a semi-supervised learning algorithm for fragmented time series that utilizes a large amount of unlabeled data to improve model performance. In weakly supervised learning, data labels are usually low quality. In the AVVP task, video-level labels are used for training, and precise labels are used at test time (Tian et al. 2020; Wu and Yang 2021; Yu et al. 2021a). Unsupervised, weakly supervised, etc., will continue to be researched in the future to solve the problem of manual multi-view data annotations.
Unaligned multi-view sequence learning
As mentioned in Sect. 2.2.5, multi-view sequence asynchronous (sequence length is not equal or semantic misalignment) is common in real applications. Therefore, ignoring the asynchrony between these sequences will hinder subsequent tasks.
Multi-view alignment is performed by using pre-trained models. Aytar et al. (2017) provided a model for downstream tasks, using a large amount of synchronized data to learn three modes (visual, sound, and language) aligned and modally robust deep representations. In addition, the attention mechanism provides a feasible solution for sequence alignment and cross-view alignment (Tian et al. 2020; Le et al. 2018). Le et al. (2018) designed a memory-augmented network to model the interaction between two unaligned sequences. Some works implements asynchronous sequence alignment by using Transformer (Tsai et al. 2019; Delbrouck et al. 2020). Delbrouck et al. (2020) investigated a Transformer-based joint encoding method to jointly encode one or more modalities, which established global dependencies between input and output through an attention mechanism. The study of unaligned multi-view sequences is no longer limited by prior manual data alignment, and will be further applied to practical scenarios.
Trusted multi-view learning
MvSD is collected from different data sources. Various sensors or environmental factors may affect the quality of these views and bring noise. Analyzing these low-quality data, especially when unreliable views are presented, will seriously hinder multi-view tasks. In addition, for a specific task, the value of information expressed by multiple views is different, so the weight of each view is not fixed. Therefore, uncertainty estimation of MvSD is helpful to improve the robustness of multi-view.
Han et al. (2021d) proposed a unified trusted multi-view classification framework that applied a Dirichlet distribution to model the probability of each class and parameterized evidence from different views to estimate the uncertainty of each view. Finally, the Dempster Shafer theory was used to integrate the multi-view opinions. Geng et al. (2021) designed an unsupervised multi-view learning method that estimated views quality online through uncertainty modeling and integrated inherent information from multiple views to obtain a noise-free representation, thereby reducing the impact of quality imbalances of different views. Wang et al. (2019b) studied a negative log-likelihood error loss, which achieved single-value prediction and uncertainty quantification simultaneously. It predicted the mean and variance of the parameterized Gaussian distribution at each time step. Through uncertainty estimation, the model utilizes valuable information as much as possible and reduces the impact of low-quality views.
Conclusions
In this paper, we review the latest deep learning techniques in MvSD. We introduce four common data types that make up MvSD, including point data, sequence data, graph data, and raster data. We also enumerate the technical challenges of MvSD: temporal dynamic, heterogeneity, cross-view dynamics, data missing and misalignment of asynchronous views. In addition, we summarize the representation methods of different data types in neural networks. Further, we reviewe the latest deep learning technology applied in MvSD. We also summarize some application areas of MvSD, and finally give several potential research directions in the future.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (No.6197624) and Sichuan Science and Technology Program (No.2021YFG0312).
1 TaxiBJ is the taxicab GPS data and meteorology data in Beijing from four time intervals: 1st Jul. 2013–30th Otc. 2013, 1st Mar. 2014–30th Jun. 2014, 1st Mar. 2015–30th Jun. 2015, 1st Nov. 2015–10th Apr. 2016.
2 The bike trajectories are collected from NYC CitiBike system. There are about 13000 bikes and 800 stations in total.
3 PeMSD4 describes the San Francisco Bay Area, and contains 3848 sensors on 29 roads dated from 1/1/2018 until 2/28/2018, 59 days in total.
4 MeTR-LA records four months of statistics on traffic speed, ranging from 3/1/2012 to 6/30/2012, including 207 sensors on the highways of Los Angeles County.
5 SleepEDF-20 is the Sleep Cassette subset of the Sleep-EDF Expanded dataset Kemp et al. (2000), consisting of 20 subjects (10 males and 10 females) aged 25–34.
6 MASS is pooled from different hospital-based sleep laboratories, consisting of whole-night recordings from 200 subjects (97 males and 103 females) aged 18–76.
7 The SHHS database Zhang et al. (2018) has two rounds of PSG records, namely Visit 1 and Visit 2. The former, consisting of 5791 subjects aged 39–90, was employed in this work.
8 https://data.cityofnewyork.us/Public-Safety/NYPD-Complaint-Data-Historic/qgea-i56i
9 https://www.kaggle.com/datasets/chicago/chicago-crime
10 CMU-MOSI Zadeh et al. (2016) dataset is one of the most popular benchmark datasets for MSA.
11 Compared to CMU-MOSI, CMU-MOSEI Zadeh et al. (2018a) dataset extends its data with more utterances, more samples, speakers, and topics.
12 SIMS Yu et al. (2020b) dataset is a Chinese MSA benchmark with fine-grained modal annotations.
13 This data collected in Beijing, China. The air quality data are recorded at 36 air quality stations every hour, from 2014/05/01 to 2015/04/30.
14 https://data.london.gov.uk/air-quality/
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Abavisani M Naghizadeh A Metaxas D Patel V Deep subspace clustering with data augmentation Adv Neural Inf Process Syst 2020 33 10360 10370
Agyemang B Wu W-P Kpiebaareh MY Lei Z Nanor E Chen L Multi-view self-attention for interpretable drug-target interaction prediction J Biomed Inform 2020 110 103547 10.1016/j.jbi.2020.103547 32860883
Akhtar MS, Chauhan DS, Ghosal D, Poria S, Ekbal A, Bhattacharyya P (2019) Multi-task learning for multi-modal emotion recognition and sentiment analysis. In: NAACL-HLT (1)
Akman A, Coppock H, Gaskell A, Tzirakis P, Jones L, Schuller BW (2021) Evaluating the covid-19 identification resnet (cider) on the interspeech covid-19 from audio challenges. https://arXiv.org/2107.14549
Alam MM, Torgo L, Bifet A (2021) A survey on spatio-temporal data analytics systems. https://arXiv.org/2103.09883
Andrew G, Arora R, Bilmes J, Livescu K (2013) Deep canonical correlation analysis. In: International Conference on Machine Learning. PMLR, pp 1247–1255
Atluri G Karpatne A Kumar V Spatio-temporal data mining: a survey of problems and methods ACM Comput Surv (CSUR) 2018 51 4 1 41 10.1145/3161602
Aytar Y, Vondrick C, Torralba A (2017) See, hear, and read: Deep aligned representations. https://arXiv.org/1706.00932
Badrinarayanan V Kendall A Cipolla R Segnet: a deep convolutional encoder-decoder architecture for image segmentation IEEE Trans Pattern Anal Mach Intell 2017 39 12 2481 2495 10.1109/TPAMI.2016.2644615 28060704
Bahdanau D, Cho K, Bengio Y (2014) Neural machine translation by jointly learning to align and translate. https://arXiv.org/1409.0473
Bai L, Yao L, Kanhere SS, Wang X, Liu W, Yang Z (2019) Spatio-temporal graph convolutional and recurrent networks for citywide passenger demand prediction. In: Proceedings of the 28th ACM International Conference on Information and Knowledge Management, pp 2293–2296
Bai L Yao L Li C Wang X Wang C Adaptive graph convolutional recurrent network for traffic forecasting Adv Neural Inf Process Syst 2020 33 17804 17815
Bai J Zhu J Song Y Zhao L Hou Z Du R Li H A3t-gcn: attention temporal graph convolutional network for traffic forecasting ISPRS Int J Geo-Inf 2021 10 7 485 10.3390/ijgi10070485
Baltrušaitis T Ahuja C Morency L-P Multimodal machine learning: a survey and taxonomy IEEE Trans Pattern Anal Mach Intell 2018 41 2 423 443 10.1109/TPAMI.2018.2798607 29994351
Barraza-Barraza D Tercero-Gómez VG Beruvides MG Limón-Robles J An adaptive arx model to estimate the rul of aluminum plates based on its crack growth Mech Syst Signal Process 2017 82 519 536 10.1016/j.ymssp.2016.05.041
Bie Y Yang Y A multitask multiview neural network for end-to-end aspect-based sentiment analysis Big Data Min Anal 2021 4 3 195 207 10.26599/BDMA.2021.9020003
Brousmiche M, Rouat J, Dupont S (2021) Multi-level attention fusion network for audio-visual event recognition. https://arXiv.org/2106.06736
Cai Y Zeng M Cai Z Liu X Zhang Z Graph regularized residual subspace clustering network for hyperspectral image clustering Inf Sci 2021 578 85 101 10.1016/j.ins.2021.07.003
Chang C-C Lin C-J Libsvm: a library for support vector machines ACM Trans Intell Syst Technol (TIST) 2011 2 3 1 27 10.1145/1961189.1961199
Chen C, Hu J, Meng Q, Zhang Y (2011) Short-time traffic flow prediction with arima-garch model. In: 2011 IEEE Intelligent Vehicles Symposium (IV). IEEE, pp 607–612
Chen M, Wang S, Liang PP, Baltrušaitis T, Zadeh A, Morency L-P (2017) Multimodal sentiment analysis with word-level fusion and reinforcement learning. In: Proceedings of the 19th ACM International Conference on Multimodal Interaction, pp 163–171
Chen J, Xu X, Wu Y, Zheng H (2018) Gc-lstm: Graph convolution embedded lstm for dynamic link prediction. https://arXiv.org/1812.04206
Chen C, Li K, Teo SG, Zou X, Wang K, Wang J, Zeng Z (2019) Gated residual recurrent graph neural networks for traffic prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp 485–492
Chen C Li K Teo SG Zou X Li K Zeng Z Citywide traffic flow prediction based on multiple gated spatio-temporal convolutional neural networks ACM Trans Knowl Discov from Data (TKDD) 2020 14 4 1 23 10.1145/3385414
Chen W, Chen L, Xie Y, Cao W, Gao Y, Feng X (2020b) Multi-range attentive bicomponent graph convolutional network for traffic forecasting. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp 3529–3536
Chen W, Wang W, Liu L, Lew MS (2020c) New ideas and trends in deep multimodal content understanding: a review. https://arXiv.org/2010.08189
Chen Z, Shi M, Zhang X, Ying H (2021a) Asm2tv: An adaptive semi-supervised multi-task multi-view learning framework. https://arXiv.org/2105.08643
Chen M, Du Y, Zhang Y, Qian S, Wang C (2021b) Semi-supervised learning with multi-head co-training. https://arXiv.org/2107.04795
Chen L, Xu J, Wu B, Qian Y, Du Z, Li Y, Zhang Y (2021c) Group-aware graph neural network for nationwide city air quality forecasting. https://arXiv.org/2108.12238
Cheng J, Dong L, Lapata M (2016) Long short-term memory-networks for machine reading. arXiv preprint https://arXiv.org/1601.06733
Cheng W, Shen Y, Zhu Y, Huang L (2018) A neural attention model for urban air quality inference: Learning the weights of monitoring stations. In: Thirty-second AAAI Conference on Artificial Intelligence
Choe J, Im S, Rameau F, Kang M, Kweon IS (2021) Volumefusion: Deep depth fusion for 3d scene reconstruction. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 16086–16095
Civitarese DS, Szwarcman D, Zadrozny B, Watson C (2021) Extreme precipitation seasonal forecast using a transformer neural network. https://arXiv.org/2107.06846
Coppock H, Gaskell A, Tzirakis P, Baird A, Jones L, Schuller BW (2021) End-2-end covid-19 detection from breath & cough audio. https://arXiv.org/2102.08359
Delbrouck J-B, Tits N, Brousmiche M, Dupont S (2020) A transformer-based joint-encoding for emotion recognition and sentiment analysis. https://arXiv.org/2006.15955
Devlin J, Chang M-W, Lee K, Toutanova K (2018) Bert: pre-training of deep bidirectional transformers for language understanding. https://arXiv.org/1810.04805
Du S Li T Yang Y Horng S-J Deep air quality forecasting using hybrid deep learning framework IEEE Trans Knowl Data Eng 2019 33 2412 10.1109/TKDE.2019.2954510
Du Y, Wang J, Feng W, Pan S, Qin T, Xu R, Wang C (2021) Adarnn: Adaptive learning and forecasting of time series. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management, pp 402–411
Dumpala SH, Sheikh I, Chakraborty R, Kopparapu SK (2019) Audio-visual fusion for sentiment classification using cross-modal autoencoder. In: 32nd Conference on Neural Information Processing Systems (NIPS 2018), pp 1–4
Ertugrul AM, Lin Y-R, Taskaya-Temizel T (2019) Castnet: Community-attentive spatio-temporal networks for opioid overdose forecasting. https://arXiv.org/1905.04714
Fadadu S, Pandey S, Hegde D, Shi Y, Chou F-C, Djuric N, Vallespi-Gonzalez C (2022) Multi-view fusion of sensor data for improved perception and prediction in autonomous driving. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp 2349–2357
Fang S, Zhang Q, Meng G, Xiang S, Pan C (2019) Gstnet: global spatial-temporal network for traffic flow prediction. In: IJCAI, pp 2286–2293
Fawagreh K Gaber MM Elyan E Random forests: from early developments to recent advancements Syst Sci Control Eng Open Access J 2014 2 1 602 609 10.1080/21642583.2014.956265
Feng C-M, Yan Y, Chen G, Fu H, Xu Y, Shao L (2021) Accelerated multi-modal mr imaging with transformers. https://arXiv.org/2106.14248
Ferenstein E Gasowski M Modelling stock returns with ar-garch processes SORT Stat Oper Res Trans 2004 28 55 68
Geng X, Li Y, Wang L, Zhang L, Yang Q, Ye J, Liu Y (2019) Spatiotemporal multi-graph convolution network for ride-hailing demand forecasting. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp 3656–3663
Geng Y, Han Z, Zhang C, Hu Q (2021) Uncertainty-aware multi-view representation learning. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, pp 7545–7553
Girshick R (2015) Fast r-cnn. In: Proceedings of the IEEE International Conference on Computer Vision, pp 1440–1448
Gu Y, Yang K, Fu S, Chen S, Li X, Marsic I (2018) Multimodal affective analysis using hierarchical attention strategy with word-level alignment. In: Proceedings of the Conference. Association for Computational Linguistics. Meeting, vol. 2018. NIH Public Access, p 2225
Guillot A Thorey V Robustsleepnet: transfer learning for automated sleep staging at scale IEEE Trans Neural Syst Rehabil Eng 2021 29 1441 1451 10.1109/TNSRE.2021.3098968 34288872
Guo S Lin Y Li S Chen Z Wan H Deep spatial-temporal 3d convolutional neural networks for traffic data forecasting IEEE Trans Intell Transp Syst 2019 20 10 3913 3926 10.1109/TITS.2019.2906365
Guo S, Lin Y, Feng N, Song C, Wan H (2019b) Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp 922–929
Guo K Hu Y Qian Z Sun Y Gao J Yin B Dynamic graph convolution network for traffic forecasting based on latent network of Laplace matrix estimation IEEE Trans Intell Transp Syst 2020 10.1109/TITS.2020.3019497
Hackel T, Savinov N, Ladicky L, Wegner JD, Schindler K, Pollefeys M (2017) SEMANTIC3D.NET: a new large-scale point cloud classification benchmark. In: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. IV-1-W1, pp 91–98
Han W, Chen H, Gelbukh A, Zadeh A, Morency L-P, Poria S (2021a) Bi-bimodal modality fusion for correlation-controlled multimodal sentiment analysis. In: Proceedings of the 2021 International Conference on Multimodal Interaction, pp 6–15
Han W, Chen H, Poria S (2021b) Improving multimodal fusion with hierarchical mutual information maximization for multimodal sentiment analysis. In: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp 9180–9192
Han Q, Lu D, Chen R (2021c) Fine-grained air quality inference via multi-channel attention model. In: IJCAI, pp 2512–2518
Han Z, Zhang C, Fu H, Zhou JT (2021d) Trusted multi-view classification. https://arXiv.org/2102.02051
Hasan MK, Lee S, Rahman W, Zadeh A, Mihalcea R, Morency L-P, Hoque E (2021) Humor knowledge enriched transformer for understanding multimodal humor. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, pp 12972–12980
Hazarika D, Zimmermann R, Poria S (2020) Misa: Modality-invariant and-specific representations for multimodal sentiment analysis. In: Proceedings of the 28th ACM International Conference on Multimedia, pp 1122–1131
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 770–778
He K, Gkioxari G, Dollár P, Girshick R (2017) Mask r-cnn. In: Proceedings of the IEEE International Conference on Computer Vision, pp 2961–2969
He K, Fan H, Wu Y, Xie S, Girshick R (2020a) Momentum contrast for unsupervised visual representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 9729–9738
He Y, Wang C, Li N, Zeng Z (2020b) Attention and memory-augmented networks for dual-view sequential learning. In: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 125–134
Hotelling H Kotz S Johnson NL Relations between two sets of variates Breakthroughs in statistics 1992 Berlin Springer 162 190
Huang W Song G Hong H Xie K Deep architecture for traffic flow prediction: deep belief networks with multitask learning IEEE Trans Intell Transp Syst 2014 15 5 2191 2201 10.1109/TITS.2014.2311123
Huang C, Zhang J, Zheng Y, Chawla NV (2018) Deepcrime: Attentive hierarchical recurrent networks for crime prediction. In: Proceedings of the 27th ACM International Conference on Information and Knowledge Management, pp 1423–1432
Huang S Kang Z Xu Z Auto-weighted multi-view clustering via deep matrix decomposition Pattern Recogn 2020 97 107015 10.1016/j.patcog.2019.107015
Huang R, Huang C, Liu Y, Dai G, Kong W (2020b) Lsgcn: long short-term traffic prediction with graph convolutional networks. In: IJCAI, pp 2355–2361
Ioffe S, Szegedy C (2015) Batch normalization: Accelerating deep network training by reducing internal covariate shift. In: International Conference on Machine Learning. PMLR, pp 448–456
Islam T, Goldwasser D (2021) Twitter user representation using weakly supervised graph embedding. https://arXiv.org/2108.08988
Ismail AA, Hasan M, Ishtiaq F (2020) Improving multimodal accuracy through modality pre-training and attention. https://arXiv.org/2011.06102
Janjua PZ Samad G Khan N Climate change and wheat production in Pakistan: an autoregressive distributed lag approach NJAS Wageningen J Life Sci 2014 68 13 19 10.1016/j.njas.2013.11.002
Ji P, Zhang T, Li H, Salzmann M, Reid I (2017) Deep subspace clustering networks. Adv Neural Inf Process Syst 30
Jia Z, Lin Y, Wang J, Zhou R, Ning X, He Y, Zhao Y (2020) Graphsleepnet: adaptive spatial-temporal graph convolutional networks for sleep stage classification. In: IJCAI, pp 1324–1330
Jia Z Lin Y Wang J Ning X He Y Zhou R Zhou Y Li-wei HL Multi-view spatial-temporal graph convolutional networks with domain generalization for sleep stage classification IEEE Trans Neural Syst Rehabil Eng 2021 29 1977 1986 10.1109/TNSRE.2021.3110665 34487495
Jia Z, Lin Y, Wang J, Wang X, Xie P, Zhang Y (2021b) Salientsleepnet: Multimodal salient wave detection network for sleep staging. https://arXiv.org/2105.13864
Kan M, Shan S, Chen X (2016) Multi-view deep network for cross-view classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 4847–4855
Kemp B Zwinderman AH Tuk B Kamphuisen HA Oberye JJ Analysis of a sleep-dependent neuronal feedback loop: the slow-wave microcontinuity of the eeg IEEE Trans Biomed Eng 2000 47 9 1185 1194 10.1109/10.867928 11008419
Khan GA Hu J Li T Diallo B Zhao Y Multi-view low rank sparse representation method for three-way clustering Int J Mach Learn Cybern 2022 13 1 233 253 10.1007/s13042-021-01394-6
Khan GA Hu J Li T Diallo B Wang H Multi-view data clustering via non-negative matrix factorization with manifold regularization Int J Mach Learn Cybern 2022 13 3 677 689 10.1007/s13042-021-01307-7
Khanehzar S, Cohn T, Mikolajczak G, Turpin A, Frermann L (2021) Framing unpacked: A semi-supervised interpretable multi-view model of media frames. https://arXiv.org/2104.11030
Kim D, Tsai Y-H, Zhuang B, Yu X, Sclaroff S, Saenko K, Chandraker M (2021) Learning cross-modal contrastive features for video domain adaptation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 13618–13627
Kipf TN, Welling M (2016) Semi-supervised classification with graph convolutional networks. https://arXiv.org/1609.02907
Kiros R, Salakhutdinov R, Zemel R (2014) Multimodal neural language models. In: International Conference on Machine Learning. PMLR, pp 595–603
Laddha A, Gautam S, Palombo S, Pandey S, Vallespi-Gonzalez C (2021) Mvfusenet: Improving end-to-end object detection and motion forecasting through multi-view fusion of lidar data. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 2865–2874
Le H, Tran T, Venkatesh S (2018) Dual memory neural computer for asynchronous two-view sequential learning. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 1637–1645
Liang PP, Liu Z, Zadeh A, Morency L-P (2018a) Multimodal language analysis with recurrent multistage fusion. https://arxiv.org/1808.03920
Liang PP, Zadeh A, Morency LP (2018b) Multimodal local-global ranking fusion for emotion recognition. In: the 2018
Liang Y, Ouyang K, Jing L, Ruan S, Liu Y, Zhang J, Rosenblum DS, Zheng Y (2019) Urbanfm: Inferring fine-grained urban flows. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 3132–3142
Liao B, Zhang J, Wu C, McIlwraith D, Chen T, Yang S, Guo Y, Wu F (2018) Deep sequence learning with auxiliary information for traffic prediction. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 537–546
Li Y, Moura JM (2019) Forecaster: a graph transformer for forecasting spatial and time-dependent data. https://arXiv.org/1909.04019
Li Y, Yu R, Shahabi C, Liu Y (2017) Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. https://arXiv.org/1707.01926
Li Z, Wang Q, Tao Z, Gao Q, Yang Z, et al (2019) Deep adversarial multi-view clustering network. In: IJCAI, pp 2952–2958
Li Z, Wang H, Li J (2020) Auto-mvcnn: neural architecture search for multi-view 3d shape recognition. https://arXiv.org/2012.05493
Lin T-Y, Dollár P, Girshick R, He K, Hariharan B, Belongie S (2017) Feature pyramid networks for object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 2117–2125
Lin Y, Chiang Y-Y, Franklin M, Eckel SP, Ambite JL (2020) Building autocorrelation-aware representations for fine-scale spatiotemporal prediction. In: 2020 IEEE International Conference on Data Mining (ICDM). IEEE, pp 352–361
Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C-Y, Berg AC (2016a) Ssd: single shot multibox detector. In: European Conference on Computer Vision. Springer, pp 21–37
Liu Y, Zheng Y, Liang Y, Liu S, Rosenblum DS (2016b) Urban water quality prediction based on multi-task multi-view learning
Liu Z, Shen Y, Lakshminarasimhan VB, Liang PP, Zadeh A, Morency L-P (2018) Efficient low-rank multimodal fusion with modality-specific factors. https://arXiv.org/1806.00064
Liu S, Fan H, Qian S, Chen Y, Ding W, Wang Z (2021) Hit: Hierarchical transformer with momentum contrast for video-text retrieval. https://arXiv.org/2103.15049
Long J, Shelhamer E, Darrell T (2015) Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 3431–3440
Lu GY Wong DW An adaptive inverse-distance weighting spatial interpolation technique Comput Geosci 2008 34 9 1044 1055 10.1016/j.cageo.2007.07.010
Lu Y-J, Li C-T (2020) Agstn: learning attention-adjusted graph spatio-temporal networks for short-term urban sensor value forecasting. In: 2020 IEEE International Conference on Data Mining (ICDM). IEEE, pp 1148–1153
Lv J Sun Q Li Q Moreira-Matias L Multi-scale and multi-scope convolutional neural networks for destination prediction of trajectories IEEE Trans Intell Transp Syst 2019 21 8 3184 3195 10.1109/TITS.2019.2924903
Ma T, Xiao C, Zhou J, Wang F (2018) Drug similarity integration through attentive multi-view graph auto-encoders. https://arXiv.org/1804.10850
Ma Mengmeng Ren Jian Zhao Long Tulyakov Sergey Wu Cathy Peng Xi SMIL: Multimodal Learning with Severely Missing Modality Proceedings of the AAAI Conference on Artificial Intelligence 2021 35 3 2302 2310 10.1609/aaai.v35i3.16330
Mai S Xing S Hu H Locally confined modality fusion network with a global perspective for multimodal human affective computing IEEE Trans Multimedia 2019 22 1 122 137 10.1109/TMM.2019.2925966
Mai S, Hu H, Xing S (2020) Modality to modality translation: An adversarial representation learning and graph fusion network for multimodal fusion. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp 164–172
Mai S, Zeng Y, Zheng S, Hu H (2021) Hybrid contrastive learning of tri-modal representation for multimodal sentiment analysis. https://arXiv.org/2109.01797
Mao L Sun S Multiview variational sparse gaussian processes IEEE Trans Neural Netw Learn Syst 2020 32 7 2875 2885 10.1109/TNNLS.2020.3008496
Mazimpaka JD Timpf S Trajectory data mining: a review of methods and applications J Spatial Inf Sci 2016 2016 13 61 99
Mikolov T, Chen K, Corrado G, Dean J (2013) Efficient estimation of word representations in vector space. https://arXiv.org/1301.3781
Nessiem MA, Mohamed MM, Coppock H, Gaskell A, Schuller BW (2021) Detecting covid-19 from breathing and coughing sounds using deep neural networks. In: 2021 IEEE 34th International Symposium on Computer-Based Medical Systems (CBMS). IEEE, pp 183–188
Okawa M, Iwata T, Kurashima T, Tanaka Y, Toda H, Ueda N (2019) Deep mixture point processes: spatio-temporal event prediction with rich contextual information. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 373–383
Olesen AN, Jennum P, Mignot E, Sorensen HB (2021) Msed: a multi-modal sleep event detection model for clinical sleep analysis. https://arXiv.org/2101.02530
Ordóñez C Lasheras FS Roca-Pardiñas J de Cos Juez FJ A hybrid arima-svm model for the study of the remaining useful life of aircraft engines J Comput Appl Math 2019 346 184 191 10.1016/j.cam.2018.07.008
Ouyang X, Yang Y, Zhang Y, Zhou W (2021) Spatial-temporal dynamic graph convolution neural network for air quality prediction. In: 2021 International Joint Conference on Neural Networks (IJCNN). IEEE, pp 1–8
Pan Z, Liang Y, Wang W, Yu Y, Zheng Y, Zhang J (2019) Urban traffic prediction from spatio-temporal data using deep meta learning. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 1720–1730
Patel ZB, Purohit P, Patel HM, Sahni S, Batra N (2022) Accurate and scalable gaussian processes for fine-grained air quality inference. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, pp 12080–12088
Pennington J, Socher R, Manning CD (2014) Glove: global vectors for word representation. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp 1532–1543
Pérez-Rúa J-M, Vielzeuf V, Pateux S, Baccouche M, Jurie F (2019) Mfas: multimodal fusion architecture search. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 6966–6975
Pham HT Yang B-S Estimation and forecasting of machine health condition using arma/garch model Mech Syst Signal Process 2010 24 2 546 558 10.1016/j.ymssp.2009.08.004
Pham H, Manzini T, Liang PP, Poczos B (2018) Seq2seq2sentiment: Multimodal sequence to sequence models for sentiment analysis. https://arXiv.org/1807.03915
Pham H, Liang PP, Manzini T, Morency L-P, Póczos B (2019) Found in translation: Learning robust joint representations by cyclic translations between modalities. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp 6892–6899
Phan H Andreotti F Cooray N Chén OY De Vos M Seqsleepnet: end-to-end hierarchical recurrent neural network for sequence-to-sequence automatic sleep staging IEEE Trans Neural Syst Rehabil Eng 2019 27 3 400 410 10.1109/TNSRE.2019.2896659 30716040
Phan H Chén OY Tran MC Koch P Mertins A De Vos M Xsleepnet: multi-view sequential model for automatic sleep staging IEEE Trans Pattern Anal Mach Intell 2021 10.1109/TPAMI.2021.3070057
Phan H Mikkelsen KB Chen O Koch P Mertins A De Vos M Sleeptransformer: automatic sleep staging with interpretability and uncertainty quantification IEEE Trans Biomed Eng 2022 69 2456 10.1109/TBME.2022.3147187 35100107
Piriyajitakonkij M Warin P Lakhan P Leelaarporn P Kumchaiseemak N Suwajanakorn S Pianpanit T Niparnan N Mukhopadhyay SC Wilaiprasitporn T Sleepposenet: multi-view learning for sleep postural transition recognition using uwb IEEE J Biomed Health Inform 2020 25 4 1305 1314 10.1109/JBHI.2020.3025900
Poria S, Cambria E, Hazarika D, Mazumder N, Zadeh A, Morency L-P (2017) Multi-level multiple attentions for contextual multimodal sentiment analysis. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE, pp 1033–1038
Qin X Zhang Z Huang C Dehghan M Zaiane OR Jagersand M U2-net: going deeper with nested u-structure for salient object detection Pattern Recogn 2020 106 107404 10.1016/j.patcog.2020.107404
Rahate A, Walambe R, Ramanna S, Kotecha K (2021) Multimodal co-learning: challenges, applications with datasets, recent advances and future directions. https://arXiv.org/2107.13782
Rahman W, Hasan MK, Lee S, Zadeh A, Mao C, Morency L-P, Hoque E (2020) Integrating multimodal information in large pretrained transformers. In: Proceedings of the Conference. Association for Computational Linguistics. Meeting, vol. 2020. NIH Public Access, p 2359
Ramachandram D Taylor GW Deep multimodal learning: a survey on recent advances and trends IEEE Signal Process Mag 2017 34 6 96 108 10.1109/MSP.2017.2738401
Rayhan Y, Hashem T (2020) Aist: An interpretable attention-based deep learning model for crime prediction. https://arxiv.org/arXiv:2012.08713
Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: unified, real-time object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 779–788
Ren S He K Girshick R Sun J Faster r-cnn: towards real-time object detection with region proposal networks Adv Neural Inf Process Syst 2015 28 91 99
Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. In: International Conference on Medical Image Computing and Computer-assisted Intervention. Springer, pp 234–241
Salama U, Chen X, Yao L, Paik H-Y, Wang X (2021) Deep multi-view spatio-temporal network for urban crime prediction. In: Australasian Database Conference. Springer, pp 50–61
Sasaki Y, Harada K, Yamasaki S, Onizuka M (2021) Airex: Neural network-based approach for air quality inference in unmonitored cities. https://arXiv.org/2108.07120
Shi X Qi H Shen Y Wu G Yin B A spatial-temporal attention approach for traffic prediction IEEE Trans Intell Transp Syst 2020 22 4909 10.1109/TITS.2020.2983651
Song C, Lin Y, Guo S, Wan H (2020) Spatial-temporal synchronous graph convolutional networks: A new framework for spatial-temporal network data forecasting. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp 914–921
Stec A, Klabjan D (2018) Forecasting crime with deep learning. https://arXiv.org/1806.01486
Summaira J, Li X, Shoib AM, Li S, Abdul J (2021) Recent advances and trends in multimodal deep learning: a review. https://arXiv.org/2105.11087
Sun S Zong D Lcbm: a multi-view probabilistic model for multi-label classification IEEE Trans Pattern Anal Mach Intell 2020 43 8 2682 2696 10.1109/TPAMI.2020.2974203
Sun S Zhao J Gao Q Modeling and recognizing human trajectories with beta process hidden Markov models Pattern Recogn 2015 48 8 2407 2417 10.1016/j.patcog.2015.02.028
Sun S Wu H Xiang L City-wide traffic flow forecasting using a deep convolutional neural network Sensors 2020 20 2 421 10.3390/s20020421 31940830
Sun J Zhang J Li Q Yi X Liang Y Zheng Y Predicting citywide crowd flows in irregular regions using multi-view graph convolutional networks IEEE Trans Knowl Data Eng 2020 10.1109/TKDE.2020.3008774
Sun S Dong Z Zhao J Conditional random fields for multiview sequential data modeling IEEE Trans Neural Netw Learn Syst 2020 10.1109/TNNLS.2020.3041591
Sun S Dong W Liu Q Multi-view representation learning with deep gaussian processes IEEE Trans Pattern Anal Mach Intell 2020 43 12 4453 4468 10.1109/TPAMI.2020.3001433
Sun Z, Sarma P, Sethares W, Liang Y (2020e) Learning relationships between text, audio, and video via deep canonical correlation for multimodal language analysis. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp 8992–8999
Supratak A, Guo Y (2020) Tinysleepnet: an efficient deep learning model for sleep stage scoring based on raw single-channel eeg. In: 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE, pp 641–644
Supratak A Dong H Wu C Guo Y Deepsleepnet: a model for automatic sleep stage scoring based on raw single-channel eeg IEEE Trans Neural Syst Rehabil Eng 2017 25 11 1998 2008 10.1109/TNSRE.2017.2721116 28678710
Sutskever I, Vinyals O, Le QV (2014) Sequence to sequence learning with neural networks. In: Advances in Neural Information Processing Systems, pp 3104–3112
Taylor SJ Letham B Forecasting at scale Am Stat 2018 72 1 37 45 10.1080/00031305.2017.1380080
Tian Y, Li D, Xu C (2020) Unified multisensory perception: Weakly-supervised audio-visual video parsing. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part III 16. Springer, pp 436–454
Torres C, Fragoso V, Hammond SD, Fried JC, Manjunath B (2016) Eye-cu: Sleep pose classification for healthcare using multimodal multiview data. In: 2016 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE, pp 1–9
Tran L, Liu X, Zhou J, Jin R (2017) Missing modalities imputation via cascaded residual autoencoder. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 1405–1414
Tsai Y-HH, Liang PP, Zadeh A, Morency L-P, Salakhutdinov R (2018) Learning factorized multimodal representations. https://arXiv.org/1806.06176
Tsai Y-HH, Bai S, Liang PP, Kolter JZ, Morency L-P, Salakhutdinov R (2019) Multimodal transformer for unaligned multimodal language sequences. In: Proceedings of the Conference. Association for Computational Linguistics. Meeting, vol. 2019, p 6558. NIH Public Access
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Ł, Polosukhin I (2017) Attention is all you need. In: Advances in Neural Information Processing Systems, pp 5998–6008
Veiga T, Ljunggren E, Bach K, Akselsen S (2021) Blind calibration of air quality wireless sensor networks using deep neural networks. In: 2021 IEEE International Conference on Omni-Layer Intelligent Systems (COINS). IEEE, pp 1–6
Veličković P, Cucurull G, Casanova A, Romero A, Lio P, Bengio Y (2017) Graph attention networks. https://arXiv.org/1710.10903
Verma S, Wang J, Ge Z, Shen R, Jin F, Wang Y, Chen F, Liu W (2020) Deep-hoseq: deep higher order sequence fusion for multimodal sentiment analysis. In: 2020 IEEE International Conference on Data Mining (ICDM). IEEE, pp 561–570
Vomfell L Härdle WK Lessmann S Improving crime count forecasts using twitter and taxi data Decis Support Syst 2018 113 73 85 10.1016/j.dss.2018.07.003
Wang Y Survey on deep multi-modal data analytics: collaboration, rivalry, and fusion ACM Trans Multimedia Comput Commun Appl (TOMM) 2021 17 1s 1 25
Wang W, Arora R, Livescu K, Bilmes J (2015) On deep multi-view representation learning. In: International Conference on Machine Learning. PMLR, pp 1083–1092
Wang D, Cao W, Li J, Ye J (2017) Deepsd: Supply-demand prediction for online car-hailing services using deep neural networks. In: 2017 IEEE 33rd International Conference on Data Engineering (ICDE), pp 243–254. IEEE
Wang B Yin P Bertozzi AL Brantingham PJ Osher SJ Xin J Deep learning for real-time crime forecasting and its ternarization Chin Ann Math Ser B 2019 40 6 949 966 10.1007/s11401-019-0168-y
Wang B, Lu J, Yan Z, Luo H, Li T, Zheng Y, Zhang G (2019b) Deep uncertainty quantification: A machine learning approach for weather forecasting. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 2087–2095
Wang Y, Shen Y, Liu Z, Liang PP, Zadeh A, Morency L-P (2019c) Words can shift: Dynamically adjusting word representations using nonverbal behaviors. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp 7216–7223
Wang J Wang W Wang L Wang Z Feng DD Tan T Learning visual relationship and context-aware attention for image captioning Pattern Recogn 2020 98 107075 10.1016/j.patcog.2019.107075
Wang Q Cheng J Gao Q Zhao G Jiao L Deep multi-view subspace clustering with unified and discriminative learning IEEE Trans Multimedia 2020 23 3483 3493 10.1109/TMM.2020.3025666
Wang Q Lian H Sun G Gao Q Jiao L Icmsc: incomplete cross-modal subspace clustering IEEE Trans Image Process 2020 30 305 317 10.1109/TIP.2020.3036717 33186106
Wang S Cao J Yu P Deep learning for spatio-temporal data mining: a survey IEEE Trans Knowl Data Eng 2020 10.1109/TKDE.2020.3025580
Wang X, Ma Y, Wang Y, Jin W, Yu J (2020e) Traffic flow prediction via spatial temporal graph neural network. In: WWW ’20: The Web Conference 2020
Wang S, Miao H, Chen H, Huang Z (2020f) Multi-task adversarial spatial-temporal networks for crowd flow prediction. In: Proceedings of the 29th ACM International Conference on Information & Knowledge Management, pp 1555–1564
Wang Z, Wan Z, Wan X (2020g) Transmodality: an end2end fusion method with transformer for multimodal sentiment analysis. In: Proceedings of The Web Conference 2020, pp 2514–2520
Wang C, Lin Z, Yang X, Sun J, Yue M, Shahabi C (2021) Hagen: Homophily-aware graph convolutional recurrent network for crime forecasting. https://arXiv.org/2109.12846
Wang J, Yang Y, Liu K, Xie P, Liu X (2022) Instance-guided multi-modal fake news detection with dynamic intra- and inter-modality fusion. In: PAKDD, pp 510–521
Wu Y, Yang Y (2021) Exploring heterogeneous clues for weakly-supervised audio-visual video parsing. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 1326–1335
Wu X, Huang C, Zhang C, Chawla NV (2020) Hierarchically structured transformer networks for fine-grained spatial event forecasting. In: Proceedings of The Web Conference 2020, pp 2320–2330
Wu J, Jiang Z, Wen S, Men A, Wang H (2021) Rethinking the constraints of multimodal fusion: case study in weakly-supervised audio-visual video parsing. https://arXiv.org/2105.14430
Xia L, Huang C, Xu Y, Dai P, Bo L, Zhang X, Chen T (2021) Spatial-temporal sequential hypergraph network for crime prediction with dynamic multiplex relation learning. In: IJCAI, pp 1631–1637
Xia W Wang S Yang M Gao Q Han J Gao X Multi-view graph embedding clustering network: joint self-supervision and block diagonal representation Neural Netw 2022 145 1 9 10.1016/j.neunet.2021.10.006 34710786
Xingjian S, Chen Z, Wang H, Yeung D-Y, Wong W-K, Woo W-c (2015) Convolutional lstm network: A machine learning approach for precipitation nowcasting. In: Advances in Neural Information Processing Systems, pp 802–810
Xu H, Zhang H, Han K, Wang Y, Peng Y, Li X (2019) Learning alignment for multimodal emotion recognition from speech. https://arXiv.org/1909.05645
Xu Z, So DR, Dai AM (2021) Mufasa: Multimodal fusion architecture search for electronic health records. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, pp 10532–10540
Yang H-F Dillon TS Chen Y-PP Optimized structure of the traffic flow forecasting model with a deep learning approach IEEE Trans Neural Netw Learn Syst 2016 28 10 2371 2381 10.1109/TNNLS.2016.2574840 27448371
Yang B, Luo W, Urtasun R (2018) Pixor: Real-time 3d object detection from point clouds. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 7652–7660
Yang J, Wang Y, Yi R, Zhu Y, Rehman A, Zadeh A, Poria S, Morency L-P (2020) Mtgat: multimodal temporal graph attention networks for unaligned human multimodal language sequences. https://arXiv.org/2010.11985
Yao H, Wu F, Ke J, Tang X, Jia Y, Lu S, Gong P, Ye J, Li Z (2018) Deep multi-view spatial-temporal network for taxi demand prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 32
Yao H, Liu Y, Wei Y, Tang X, Li Z (2019a) Learning from multiple cities: a meta-learning approach for spatial-temporal prediction. In: The World Wide Web Conference, pp 2181–2191
Yao H, Tang X, Wei H, Zheng G, Li Z (2019b) Revisiting spatial-temporal similarity: A deep learning framework for traffic prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp 5668–5675
Yi X, Zhang J, Wang Z, Li T, Zheng Y (2018) Deep distributed fusion network for air quality prediction. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 965–973
Yin J Sun S Multiview uncorrelated locality preserving projection IEEE Trans Neural Netw Learn Syst 2019 31 9 3442 3455 10.1109/TNNLS.2019.2944664 31670682
Yin X Wu G Wei J Shen Y Qi H Yin B Multi-stage attention spatial-temporal graph networks for traffic prediction Neurocomputing 2021 428 42 53 10.1016/j.neucom.2020.11.038
Yin Y, Huang S, Zhang X, Dou D (2021b) Bm-nas: Bilevel multimodal neural architecture search. https://arXiv.org/2104.09379
Yu B, Yin H, Zhu Z (2017) Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. https://arXiv.org/1709.04875
Yu Z, Cui Y, Yu J, Wang M, Tao D, Tian Q (2020a) Deep multimodal neural architecture search. In: Proceedings of the 28th ACM International Conference on Multimedia, pp 3743–3752
Yu W, Xu H, Meng F, Zhu Y, Ma Y, Wu J, Zou J, Yang K (2020b) Ch-sims: A chinese multimodal sentiment analysis dataset with fine-grained annotation of modality. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 3718–3727
Yu J, Cheng Y, Zhao R-W, Feng R, Zhang Y (2021a) Mm-pyramid: Multimodal pyramid attentional network for audio-visual event localization and video parsing. https://arXiv.org/2111.12374
Yu W, Xu H, Yuan Z, Wu J (2021b) Learning modality-specific representations with self-supervised multi-task learning for multimodal sentiment analysis. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, pp 10790–10797
Yuan Z, Zhou X, Yang T (2018) Hetero-convlstm: A deep learning approach to traffic accident prediction on heterogeneous spatio-temporal data. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 984–992
Yuan Y Jia K Ma F Xun G Wang Y Su L Zhang A A hybrid self-attention deep learning framework for multivariate sleep stage classification BMC Bioinform 2019 20 16 1 10
Zadeh A, Zellers R, Pincus E, Morency L-P (2016) Mosi: multimodal corpus of sentiment intensity and subjectivity analysis in online opinion videos. https://arXiv.org/1606.06259
Zadeh A, Chen M, Poria S, Cambria E, Morency L-P (2017) Tensor fusion network for multimodal sentiment analysis. https://arXiv.org/1707.07250
Zadeh AB, Liang PP, Poria S, Cambria E, Morency L-P (2018a) Multimodal language analysis in the wild: Cmu-mosei dataset and interpretable dynamic fusion graph. In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, vol. 1: Long Papers, pp 2236–2246
Zadeh A, Liang PP, Mazumder N, Poria S, Cambria E, Morency L-P (2018b) Memory fusion network for multi-view sequential learning. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 32
Zadeh A, Liang PP, Poria S, Vij P, Cambria E, Morency L-P (2018c) Multi-attention recurrent network for human communication comprehension. In: Thirty-Second AAAI Conference on Artificial Intelligence
Zadeh A, Mao C, Shi K, Zhang Y, Liang PP, Poria S, Morency L-P (2019) Factorized multimodal transformer for multimodal sequential learning. rint https://arXiv.org/1911.09826
Zhang J, Zheng Y, Qi D, Li R, Yi X (2016) Dnn-based prediction model for spatio-temporal data. In: Proceedings of the 24th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, pp 1–4
Zhang J, Zheng Y, Qi D (2017) Deep spatio-temporal residual networks for citywide crowd flows prediction. In: Thirty-first AAAI Conference on Artificial Intelligence
Zhang G-Q Cui L Mueller R Tao S Kim M Rueschman M Mariani S Mobley D Redline S The national sleep research resource: towards a sleep data commons J Am Med Inform Assoc 2018 25 10 1351 1358 10.1093/jamia/ocy064 29860441
Zhang J Zheng Y Sun J Qi D Flow prediction in spatio-temporal networks based on multitask deep learning IEEE Trans Knowl Data Eng 2019 32 3 468 478 10.1109/TKDE.2019.2891537
Zhang C Cui Y Han Z Zhou JT Fu H Hu Q Deep partial multi-view learning IEEE Trans Pattern Anal Mach Intell 2020 10.1109/TPAMI.2020.3037734
Zhang Y Yang Y Zhou W Wang H Ouyang X Multi-city traffic flow forecasting via multi-task learning Appl Intell 2021 51 6895 10.1007/s10489-020-02074-8
Zhang M, Li T, Li Y, Hui P (2021b) Multi-view joint graph representation learning for urban region embedding. In: Proceedings of the Twenty-Ninth International Conference on International Joint Conferences on Artificial Intelligence, pp 4431–4437
Zhao J Sun S High-order gaussian process dynamical models for traffic flow prediction IEEE Trans Intell Transp Syst 2016 17 7 2014 2019 10.1109/TITS.2016.2515105
Zhao J Sun S Variational dependent multi-output gaussian process dynamical systems J Mach Learn Res 2016 17 1 4134 4169
Zhao J Xie X Xu X Sun S Multi-view learning overview: recent progress and new challenges Inf Fusion 2017 38 43 54 10.1016/j.inffus.2017.02.007
Zhao H, Ding Z, Fu Y (2017b) Multi-view clustering via deep matrix factorization. In: Thirty-first AAAI Conference on Artificial Intelligence
Zhao H, Shi J, Qi X, Wang X, Jia J (2017c) Pyramid scene parsing network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 2881–2890
Zhao L Song Y Zhang C Liu Y Wang P Lin T Deng M Li H T-gcn: a temporal graph convolutional network for traffic prediction IEEE Trans Intell Transp Syst 2019 21 9 3848 3858 10.1109/TITS.2019.2935152
Zheng C, Fan X, Wang C, Qi J (2020) Gman: a graph multi-attention network for traffic prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp 1234–1241
Zheng L, Cheng Y, Yang H, Cao N, He J (2021) Deep co-attention network for multi-view subspace learning. In: Proceedings of the Web Conference 2021, pp 1528–1539
Zhong H, Yin C, Wu X, Luo J, He J (2020) Airrl: A reinforcement learning approach to urban air quality inference. https://arXiv.org/2003.12205
Zhou Y, Tuzel O (2018) Voxelnet: End-to-end learning for point cloud based 3d object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 4490–4499
Zhou W Yang Y Zhang Y Wang D Zhang X Deep flexible structured spatial-temporal model for taxi capacity prediction Knowl-Based Syst 2020 205 106286 10.1016/j.knosys.2020.106286
| 36466765 | PMC9707228 | NO-CC CODE | 2022-12-01 23:20:21 | no | Artif Intell Rev. 2022 Nov 29;:1-44 | utf-8 | Artif Intell Rev | 2,022 | 10.1007/s10462-022-10332-z | oa_other |
==== Front
Psychoanal Cult Soc
Journal for the Psychoanalysis of Culture & Society
1088-0763
1543-3390
Palgrave Macmillan UK London
332
10.1057/s41282-022-00332-x
Book Review
The Palliative Society: Pain Today
Byung-Chul Han, translated by Daniel Steuer, Polity Press, 2021, 76 pp, $59.95, hardback ISBN: 13-978-5095-4723-4
http://orcid.org/0000-0002-2622-6206
Kugelmann Robert [email protected]
grid.266229.b 0000 0001 2187 0206 University of Dallas, Irving, TX USA
28 11 2022
18
© The Author(s) under exclusive licence to Springer Nature Limited 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
==== Body
pmcThe Palliative Society is one of many books by the philosopher and cultural critic Byung-Chul Han, whose other titles include The Burnout Society (2010/2015) and Capitalism and the Death Drive (2019/2021). Eleven brief chapters constitute the book, examining pain from various perspectives, including its meaninglessness, its cunning, as a mode of truth, as a revelation of being, and our fear of it, our algophobia, throughout. The Palliative Society is the critique of modernity, signaled by the opening quotation from Ernst Jünger: “Tell me your relation to pain, and I will tell you who you are” (Jünger, 1934/2008, p. 32, quoted on p. 1). This slim volume, written aphoristically, explores the philosophical complexities of pain in societies marked by an extreme aversion to pain and discomfort. For Han, pain is not simply a sensation or feeling of displeasure; rather it constitutes important ways of what it means to be human. Indeed, for Han we are so much the worse for our penchant to flee it. Pain is an opening via negation to otherness, and without it we are lost “in the hell of the same” (p. 6, emphasis in original). What Han calls “the neoliberal dispositif of happiness” (p. 13) distorts happiness, which “is not at one’s disposal. Inherent in it is a certain negativity” (p. 13). “Dispositif,” a term of Foucault, designated “discursive and nondiscursive elements, … [that are] historical and culturally bound to a certain area or civilization, and … are answers to certain greater problems in a particular society” (Peltonen, 2004, p. 216). This neoliberal dispositif is palliative, seeking happiness by eliminating pain and discomfort without addressing the issues that, if attended to, might lead to radical social change: “Instead of revolution we thus get depression” (p. 12, emphasis in original). The neoliberal imperative, “be happy” (p. 11), draws upon our desire for “self-improvement and self-optimization” (p. 11), so that we discipline ourselves to conform to requirements for productivity, fully believing we are free in our self-subjugation.
Central to Han’s critique is that our palliative society has made health its supreme value. Health, however, is described in a specific way, as the functioning of the body defined exclusively in anatomical and physiological terms: “Life is reduced to a biological process that must be optimized. It loses any meta-physical dimension” (p. 16). Ours is a society of “survival,” “a society of the undead” (p. 17), as we have no concept of what a “good life” might mean. We are like Paul Valéry’s Monsieur Teste, who cannot narrate his pain, as it is meaningless, “outside the symbolic order” (p. 19). Life that is “reduced to a biological process” (p. 22) has no story, no “meaningful horizon” (p. 22).
Pain will have its way, nevertheless, with Jünger speaking of the “cunning of pain” (cited, p. 26). Where does pain insinuate itself in a palliative society? Han refers to the spread of chronic pain, of “cutting” (p. 28) and other forms of self-harm. We are beset with loneliness and isolation, and “narcissism and egotism are intensifying” (p. 28), echoing Christopher Lasch’s (1991) diagnosis. Cultural anesthesia leads to a need for “increasingly stronger stimuli … to provide people in an anesthetic society with any sense of being alive” (Illich, 1976, p. 152, quoted on p. 33).
Pain is a way of knowing, and it is essential to experience (German Erfahrung), as “a painful process of transformation that contains an element of suffering, of undergoing something” (p. 39). An experience in this sense, as Gadamer (1960/1989) wrote, “thwarts an expectation” (p. 356), and one undergoes a reversal, which is painful, a “learning through suffering” (p. 356), as in the catharsis of tragedy. Indeed, drawing on Heidegger’s notion of mood (Stimmung), Han writes that “pain is the fundamental mood of human finitude” (p. 45), thinking of “that area of being ‘in which pain and death and love belong together’” (Heidegger, 1950/2002, p. 205, quoted on p. 45). This mood attunes us to the “non-available” (p. 45), making pain Orphic: Orpheus loves Eurydice and descends into the underworld to rescue her from death with the enchantment of his music, only to lose her again by turning around to see her, to keep her present and to possess her—to keep her visible, thus losing her as other. Essential to pain as mood, then, is a desire that would overcome death, a love that cannot negate death, a longing for the face of the other. Han cites Heidegger again, who proposed that “the spirit [muot] which answers to pain, the spirit attuned by pain and to pain, is melancholy [Schwermut]” (Heidegger, 1950/2002, p. 153, quoted on p. 46). Melancholy, the disease of the philosopher according to an ancient text, attuning us to the saturnine, the flaws, cracks, and limits of human life, is a way of negation that gets at essences. Melancholy, which draws its significance from pain as a fundamental mood, recognizes that “a crucial part of taking care … is the experience of unavailability … otherness and strangeness” (p. 47). Like the melancholy angel in Dürer’s woodcut, Melencolia I of 1514, our eyes are on what is beyond us, on otherness, when we do not flee pain.
Han’s ultimate remedy is not heroic endurance of pain. It is, with Levinas, “a sensibility for the other [that] presupposes an ‘exposure’ that ‘offer[s] itself even in suffering” (Levinas, 1974/1991, p. 15, quoted on p. 52). This is primal pain: “pain toward the other,” “meta-physical pain” (p. 52), openness to the suffering of others, cracking the complacency and pursuit of the comfortable that defines our palliative society. Han closes on a somber note: because we cherish comfort more than freedom, we face a “transhuman” future, without pain and always happy, which is “not a human life” (p. 60). The undead will inherit the earth.
Much of the book is a dialogue with Ernst Jünger, whom Han quotes extensively. Jünger was a prominent and controversial German thinker of the twentieth century, a fierce critic of bourgeois society for its desire for comfort and security and its rejection of heroic virtues of endurance and courage. As an example of what Jünger meant by heroic virtue, to amplify Han’s account, consider the story of the Roman soldier, Gaius Mucius Cordus Scaevola, who held his arm over a flame without flinching to show his enemy his contempt for his own body and pain. Han does not follow Jünger in a celebration of militaristic derring-do, submission to authority, and a willingness to endure pain and self-sacrifice. He is, however, clear that our algophobia—morbid fear of pain—is leading us down a primrose path to a loss of individual freedom, autonomy, and authenticity (see Taylor, 1991), virtues that modernity at its best strives to cultivate. We seem to have a predilection, according to Han, to what I would call an Esau complex, a willingness to surrender our birthright for a mess of porridge.
Han writes that “every critique of society must … provide a hermeneutics of pain” (p. 1). Pain is too important to be left to medicine, where it primarily resides today. This medicalization of pain progressively destroys any meaning that pain might have, as it is something to avoid, eliminate, or conceal (the word “palliative” comes from the Latin palliare, “conceal”). Neither Jünger nor Han have been the first to make the charge that algophobia is one of modernity’s besetting flaws. For Jünger, “the bourgeois individual typically dwells in a ‘zone of sensitivity,’ where ‘security,’ ‘ease,’ and ‘comfort’—and ultimately ‘the body’ itself—become the essential core of life. Here, one seeks to avoid pain at all cost” (Durst, 2008, section 2). Jünger thus captured societal “algophobia.”
The charge that we moderns are more sensitive to pain than people in the premodern world has been made repeatedly (Kugelmann, 2017). In the nineteenth century, commentators tied increased sensitivity to pain to the upper classes and to the “civilized” races of Europe: “In the ideology of the [American] slave owners, it was a commonplace that slaves were relatively insensitive to pain” (Armstrong, 2012, p. 146). Weir Mitchell (1892), an important nineteenth-century neurologist, found an increasing sensitivity to pain taking place in the United States. An article in The Living Age (“The Meaning of Pain,” 1906) stated that “it is a well-established conclusion of science that the higher we rise in the scale of nervous organization the greater the possibilities of pain,” with “civilized races” feeling pain more exquisitely than “savages,” and men more than women (p. 699). Indeed, among the civilized and the men, “brain workers” feel pain more acutely than do manual laborers, which accounted for the more frequent occurrence of neurasthenia among these privileged groups, a view shared by the prominent psychologist and anthropologist, W. H. R. Rivers (1920). Such views, connecting intellect, level of civilization, and increased sensitivity to pain, were shared by Schopenhauer and Nietzsche (De Moulin, p. 541, n. 4).
Well before these accounts of algophobia, Alexis de Tocqueville (1835/1966), in Democracy in America, based on his journey to the United States in 1831–1832, noted that Americans continuously seek “improvement” in all spheres of life, especially material improvement, and that Americans have a great “taste for physical comfort” (p. 503), unlike European aristocrats and lower classes, the former because they take comforts for granted, the latter because they take their absence for granted. Only autonomous selves, it would appear, would eliminate all displeasure in the pursuit of happiness.
There is more to the story. In the middle of the nineteenth century came surgical anesthesia, word of which spread around the globe like wildfire and eventually made endurance of surgical pain absurd. It was with good reason that Weir Mitchell (1900), on the fiftieth anniversary in 1896 of the introduction of surgical anesthesia, could read his poem, “The Birth and Death of Pain,” in which we hear: “Whatever triumphs still shall hold the mind,/ Whatever gift shall yet enrich mankind,/ … No hour as sweet as when hope, doubt, and fears,/ ‘Mid deepening stillness, watched one eager brain,/ With Godlike will, decree the Death of Pain” (p. 18). I assume that even modernity’s fiercest critics avail themselves of anesthesia before going under the knife. Anesthesia altered existential possibilities for responding to pain.
Daniël de Moulin (1974) documents that René Leriche, acclaimed author of The Surgery of Pain (1939), found that “modern man is more sensitive to pain than even his immediate ancestors” (De Moulin, p. 542), this sensitivity being for Leriche a consequence of moderns having more methods available to eliminate pain, including anesthetics and analgesics, such as aspirin. The Dutch phenomenologist and physiologist F. J. J. Buytendijk (1943/1961), living in occupied Netherlands, wrote: “Modern man is irritated by things which older generations accepted with equanimity. He is irritated by old age, long illness, and even by death; above all he is irritated by pain. Pain must simply not occur. … The consequence is an immoderate state of algophobia … which is itself an evil and sets a seal of timidity on the whole of life” (pp. 15–16). His colleague, the phenomenological psychiatrist, J. H. van den Berg (1975), repeated the claim that modernity brings increased sensitivity to pain, explaining it through a loss of community and greater individual autonomy. For van den Berg, pain has been disembedded from social relationships, and pain is more painful when we face it alone. Ivan Illich, to whom Han refers on this topic (p. 19), charged that our medicalization of pain results in a cultural shift: “People unlearn the acceptance of suffering as an inevitable part of their conscious coping with reality and learn to interpret every ache as an indicator of their need for padding or pampering” (Illich, 1976, p. 133).
Han has extended “algophobia” to imply that “we live in a society of positivity that tries to extinguish any form of negativity” (p. 2). Power operates today not primarily by repression and overt violence—although that continues, especially in minority communities—but by “self-optimization” (p. 3). We discipline ourselves by striving “to be all that you can be,” as states an advertisement used to attract recruits to the US military (Singer, 2008). Our palliative society is also a “performance society” that, eschewing negation, finds opportunities for increased performance in any situation, such that we speak of “post-traumatic growth” and “resilience” (p. 2) come what may. Drawing on social media, Han also characterizes the palliative society as “the society of the like [Gefällt-mir], increasingly a society characterized by a mania for liking” (p. 3): Nothing should hurt. Han at this juncture introduces his sed contra by asserting that “what has been forgotten is that pain purifies. It has a cathartic effect” (p. 3), a claim echoing Jünger’s contempt for pain-averse middle class couch potatoes.
Han does not make a case for military discipline or for the contempt for life and comfort symbolized by the suicide bomber. No, for Han the palliative society is, to use the phrase of John McKnight (1996), a “careless society,” in the double sense of “not having a care in the world” and “I don’t have to care about you, because experts can handle whatever is ailing you.” What our palliative, performance society of the like faces is a loss of “nakedness of soul, exposure, the pain toward the other” (p. 54). Our algophobia is fundamentally fear of “pain towards the other” (p. 54), and not the quivering of the flesh in the face of discomforts and negativity.
With that Levinasian perspective, space opens between Han’s cultural critique and that of Jünger’s. For Jünger (1934/2008), the bourgeois individual lacks the heroic spirit: “The heroic … world presents an entirely different relation to pain than does the world of sensitivity. While in the latter … it is a matter of marginalizing pain and sheltering life from it, in the former the point is to integrate pain and organize life in such a way that one is always armed against it” (p. 16). For Han, it is not this heroic attitude that ultimately matters. It is pain as a fundamental mood of existence, exposing us to the pain of the other, that matters. Thus, by contrast, Han can write: “Pain is a gift” (p. 49). How do individuals and cultures receive this gift? Han, with Illich, sees the drift toward greater medicalization of pain as rendering people unable to cultivate what Illich (1976) called the “craft of suffering well” (p. 145), to which Han (p. 19) refers, emphasizing how with the atrophy of cultural ways to reckon with pain and suffering, pain becomes a “purely physical agony” (p. 19). In the anesthetic state inculcated by the palliative society, exposure to the pain and suffering of the other vanishes, along with the capacity to tolerate discomfort.
Nevertheless, to some extent, this craft of suffering is still very much with us. For example, professional and not-so-professional athletes learn to bear pain in the course of their training, their askesis. It can be a badge of honor to play through the pain of an injury. The heroic spirit thus endures, even in a palliative society. The historian Esther Cohen (1995) describes earlier forms of this craft, from antiquity to the late Middle Ages. Even though our primary attitude toward pain, she writes, is that of “rejection,” for which “physical suffering is not considered inevitable or unavoidable” (p. 51), Cohen does not see the development of modern pain-killers as causing this rejection of pain: “Many primitive societies are familiar with the analgesic qualities of various plants, yet they do not resort to them in situations in which modern Westerners would automatically demand relief. More important, in many societies the acceptance of pain is a cultural imperative” (p. 51). This observation supports the diagnosis of algophobia in contemporary society. Indeed, probably less common today than formerly, is an art of suffering Cohen calls “impassivity, … to endure without flinching” (p. 51). The Roman soldier who held his arm in the flame demonstrated impassivity. Cultural patterns of enduring pain “stoically” and “keeping a stiff upper lip” are still with us, although they may appear unenlightened with analgesics abounding. “Impassibility” is an attitude that seeks “the capacity of transcending pain completely” (p. 52), through trance and ritual and arduous training. Cohen notes that in the West, such freedom from pain “was a miraculous quality, a gift from heaven granted only occasionally to saints and martyrs” (p. 52). In the later Middle Ages, “philopassianism” developed, the deliberate evocation of pain, an attitude absent earlier in Christian Europe. Cohen explains philopassianism: “The idea of Imitatio Christi, fervently preached throughout the period to clerics and laymen alike, insisted that in order to follow Christ’s footsteps one must carry his cross and feel his pain” (p. 59). Practices such as self-flagellation were a means to feel this pain, ultimately of sin and repentance. Ariel Glucklich (2003) found that such practices break down the boundaries of the self and can open a person up to what is other; hence their use in religious practices throughout the world. Even a mild ascetical practice such as fasting can change one’s attunement in the everyday world, disrupting routine and exposing one to one’s lack. Thus, arts of suffering occur across cultures and history, and endure even among us. These arts keep their practitioners exposed, even potentially to the other. Nevertheless, a palliative society does make such arts more difficult to justify and practice.
The upshot is that Han uncovers the consequences of our pursuit of what we call health; algophobia-is-us. The book also addresses the Covid-19 pandemic, and here we see how the thesis that modernity puts us face to face with meaningless pain has its latest manifestation. The thrust of Han’s claims about the Covid pandemic center around “bare life,” our living defined in biological terms only. Let me extend Han’s critique to sources outside his text: Jeffrey Bishop on the “anticipatory corpse” and Illich on “life” as an idol. To put this into perspective, consider how Jeffrey Bishop (2011) distinguishes between zoē and bios in ancient Greek thought. Zoē is “bare life, the life we have by virtue of being alive” (p. 213). Bios is one’s “biography,” such as the “contemplative life,” the “life of pleasure,” and the “political life” (p. 213). Zoē “belongs to the realm of the oikos, or home, and not the realm of the polis, or city” (p. 214), whereas for us, with what Foucault called “biopolitics,” “the sphere of the polis reaches into the sphere of oikos” (Bishop, p. 214). We are thus confused, and do not know when this bare life begins or ends, and we tend to equate longevity, the continuance of zoē, as in itself a good. Illich (1992) goes further: In “The Institutional Construction of a New Fetish: Human Life,” Illich argued that “‘Human life’ is a recent construct, something which we now take so much for granted that we dare not seriously question it” (p. 219). Moreover, “thinking in terms of ‘a life’ and ‘human life’ vaguely connotes something of extreme importance and tends to abolish all limits that decency and common sense have so far imposed on the exercise of professional tutelage” (pp. 219–220). In his history of “a life,” Illich contended that our “a life” originated in a corruption of the Christian message that Christ was “Life,” such that today “life” is, in religious terms, an idol. While Han does not go that route, it is clear that he, along with Bishop and Illich, sees “bare life” as an impoverished rendering of human living. Bare life is visible in the x-ray and the lab results, in the anatomical text and health statistics. Bare life makes living available to increased biopolitical surveillance.
The pandemic accelerates the shift to “a biopolitical surveillance regime” (Han, p. 18) since the protection of zoē knows no limits. The privacy and autonomy of the modern individual succumbs to the demands for containing the spread of the virus. Thus, “the biopolitical regime of surveillance spells the end of liberalism” (p. 59). Han’s insight into bare life as an idol makes sense of what at first sounded to this reader as a rant against commonsense public health measures during the pandemic. For example: “Because of the pandemic, the society of survival has prohibited church services, even at Easter. Priests, too, practice ‘social distancing’ and wear protective masks. They sacrifice faith entirely to survival. … Virology deprives theology of its power” (p. 15). Finally, “faith degenerates into farce. It is replaced with intensive care units and respirators. The dead are counted daily” (p. 15). Han here engages in some sliding of the signifier as the Lacanians might say, with “distance” shifting from meaning steps taken to avoid infecting other people to meaning indifference and a lack of empathy. Indeed, Han asserts that “‘social distancing’ contributes to the loss of empathy” (p. 52), because we are apart and not near one another. That assertion is an empirical matter, with Pfattheicher et al. (2020) finding social distancing a sign of empathy. The encounter with an other is not measured in feet and inches. Han’s hyperbole in this matter—faith is replaced with intensive care units?—makes sense only if one sees that the idolatry of “bare life” is in play in public health measures, despite the goodwill that promotes them. The sacrifice of faith for survival is better understood in terms of Illich’s contention, surpassing Han’s on this point, that our efforts to preserve bare “life” perverts what it means to be an individual or a person. The palliative society’s valuation of bare life undermines the good that we would do in responding to the pandemic. After all, the road to hell is paved with good intentions.
Conflict of Interest
The author states that there is no conflict of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Armstrong T The logic of slavery: Debt, technology, and pain in American literature 2012 Cambridge University Press
Bishop JP The anticipatory corpse: Medicine, power, and the care of the dying 2011 University of Notre Dame Press
Buytendijk, F. J. J. (1961). Pain: Its modes and functions (E. O’ Shiel, Trans.). University of Chicago Press. (Originally published 1943)
Cohen E Towards a history of European physical sensibility: Pain in the later Middle Ages Science in Context 1995 8 1 47 74 10.1017/S0269889700001897 11639660
De Moulin D A historical-phenomenological study of bodily pain in Western man Bulletin of the History of Medicine 1974 48 4 540 570 4618152
Durst, D. C. (2008). Translator’s introduction. In E. Jünger, On pain. Telos Press.
Gadamer, H.-G. (1989). Truth and method (J. Weinsheimer & D. G. Marshall, Trans.) (2nd ed.). Continuum. (Original work published 1960)
Glucklich A Sacred pain: Hurting the body for the sake of the soul 2003 Oxford University Press
Han, B.-C. (2015). The burnout society (E. Butler, Trans). Stanford University Press. (Original work published 2010)
Han, B.-C. (2021). Capitalism and the death drive. Polity Press. (Original work published 2019)
Heidegger, M. (2002). Off the beaten track (J. Young & K. Haynes, Eds. & Trans.). Cambridge University Press. (Original work published 1950)
Illich I Medical nemesis: The expropriation of health 1976 Pantheon Books
Illich I In the mirror of the past: Lectures and addresses 1978–1990 1992 Marion Boyars
Jünger, E. (2008). On pain (D. C. Durst, Trans.). Telos Press. (Original work published 1934)
Kugelmann R Constructing pain: Historical, psychological, and critical perspectives 2017 Routledge
Lasch C The culture of narcissism: American life in an age of diminishing expectations 1991 Norton
Leriche, R. (1939). The surgery of pain (A. Young, Trans.). Williams & Wilkins.
Levinas, E. (1991). Otherwise than being or beyond essence (A. Lingis, Trans.). Kluwer Academic. (Original work published 1974)
McKnight J The careless society: Community and its counterfeits 1996 Basic Books
Mitchell SW Characteristics 1892 3 Century Co.
Mitchell SW The wager and other poems 1900 Century
Peltonen M From discourse to “dispositif”: Michel Foucault’s two histories Historical Reflections/réflexions Historiques 2004 30 2 205 219
Pfattheicher S Nockur L Böhm R Sassenrath C Petersen MB The emotional path to action: Empathy promotes physical distancing and wearing of face masks during the COVID-19 pandemic Psychological Science 2020 31 11 1363 1373 10.1177/0956797620964422 32993455
Rivers WHR Instinct and the unconscious: A contribution to a biological theory of the psycho-neuroses 1920 Cambridge University Press
Singer, P. W. (2008, May 2). How to be all that you can be: A look at the Pentagon’s five step plan for making Iron Man real. Brookings. https://www.brookings.edu/articles/how-to-be-all-that-you-can-be-a-look-at-the-pentagons-five-step-plan-for-making-iron-man-real/
Taylor C The ethics of authenticity 1991 Harvard University Press
The meaning of pain. (1906, March 17). The Living Age, 7th series, 30(3219), 699–701.
Tocqueville, A. de. (1966). Democracy in America (J. P. Mayer & M. Lerner, Eds.; G. Lawrence, Trans.). Harper & Row. (Original work published 1835)
Van den Berg JH Divided existence and complex society: An historical approach 1975 Duquesne University Press
| 0 | PMC9707242 | NO-CC CODE | 2022-12-01 23:20:22 | no | Psychoanal Cult Soc. 2022 Nov 28;:1-8 | utf-8 | null | null | null | oa_other |
==== Front
Educ Inf Technol (Dordr)
Educ Inf Technol (Dordr)
Education and Information Technologies
1360-2357
1573-7608
Springer US New York
11482
10.1007/s10639-022-11482-x
Article
A qualitative analysis of publicly available Standards and Guidance about Digital Literacies in U.S. States
http://orcid.org/0000-0002-8138-512X
Rice Mary F. [email protected]
Bailon Mark [email protected]
grid.266832.b 0000 0001 2188 8502 University of New Mexico, Albuquerque, NM USA
28 11 2022
120
21 3 2022
23 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Before the COVID-19 pandemic, schools had been adopting digital instruction in many parts of the world. The concept of digital literacies has also been evolving in complexity alongside the digital technologies that support it. However, little is known about what guidance available to support various levels of government in supporting digital literacies alongside digital instruction in local schools. The purpose of this study was to determine what guidance for digital literacies U.S. state departments of education had made available through their websites to local schools just prior to the onset of the pandemic. Using qualitative content analysis techniques, digital literacies guidance information was located on U.S. state departments of education websites and analyzed. Most states did not indicate that they used guidance from professional organizations about digital literacies. The 16 states that did have guidance used standards from the International Society for Technology in Education (ISTE), which have not been positioned by the organization as digital literacies standards, but instead reflect traditional understandings of Information Computing Technology (ICT). Implications of this study highlight potential strategies educational ministries might use to acknowledge and support digital literacies.
Keywords
Digital literacies
State educational policies
Professional educational organizations
Qualitative policy analysis
Digital learning
Digital education leadership
ISTE standards
University of New MexicoOFAC-2019
==== Body
pmcIntroduction
Digital technologies support meaning making and communication across modes (e.g., linguistic, visual, aural, and semiotic; Álvarez Valencia, 2016). These opportunities for making meaning with new technologies have been conceptualized as types of literacies because they rely on encoding/decoding processes to make meaning in social contexts (Hall, 1973/1991). As technologies have developed to display increasingly complex texts with video, music, animation, and more, digital literacies now encompass a wide and ever-growing variety of strategies, knowledges, and skills. While there is no unanimous agreement about what outcomes for young people are preferable for acquiring digital literacies, proposed goals include: advancing social democracy, maintaining a workforce, and supporting ongoing innovation and quality of life (Coldwell-Neilson, 2017). Each of these goals is a high-stakes one, not just for the learners, but for societies and communities. Therefore, it is important for learners to have opportunities to use digital literacies in their schooling.
Explicit attention to digital literacies as part of digital learning has become more important to formal educational experiences due to the recent COVID-19 pandemic. As school buildings in the United States and across the globe closed, teachers in many countries turned to using digitized instructional materials as part of government directives to provide education (Natanson & Strauss, 2020). While the uses of instructional materials that required digital literacies were expanding even before COVID-19 closed school buildings, the pandemic drew attention to the need to understand efforts to promote digital literacies (Digital Learning Collaborative, 2019). Understanding the digital literacies policies that U.S. states had when the pandemic began can provide important information to the states themselves and to other governing entities when decision makers are ready to revise policies that anticipate long-term school disruptions (Rice & Zancanella, 2021).
This local policy making is critical in the U.S. because the U.S. does not have a national educational policy (U.S. Department of Education, 2021). Instead, individual states develop educational policies in accordance with federal guidance that is usually tied to federal monies. Some policies at the state level are developed in state-level legislatures. Other policies are left to the discretion of state level educational officials hired to make such decisions. Therefore, it is to be expected that states would not be in alignment on any matter of educational policy. However, such differences provide the opportunity to see how different states have addressed important educational challenges so that states or countries with less developed policies can have models and examples in designing policies that work for them. Previous research suggests that many countries experience similar struggles to provide teachers with adequate policy guidance for using technologies generally and digital literacies specifically. For example, Dong and Newman (2018) found that teachers in China were using policy frameworks with inconsistent pedagogical recommendations for teaching digital literacies; Drajati et al., (2018) found that teachers in Indonesia had inconsistent access to common terms and language necessary for professional learning about technologies and information skills; and Luik et al. (2018) found that teachers in Estonia were integrating technologies into teaching without a pedagogical foundation. Considering findings from these studies, it seems useful to understand digital literacies policies and what organizations—with concomitant expertise and values—might be contributing to these.
To guide the leaders at the state level in the U.S., professional organizations with national and international reach have specialized in monitoring research and advocating for practice about digital literacies. Such organizations include the American Association of School Librarians (AASL), Computer Science Teachers of America (CSTA), International Society for Technology Education (ISTE), the National Council of Teachers of English (NCTE), and more. In addition, there are national consortiums that write standards and make recommendations about curriculum and instruction in the US, such as the National Governors Association Center for Best Practices and the Council of Chief State School Officers, which created the Common Core State Standards (2010). There was also a consortium of 26 states that formed the NextGen Science Standards (2013) with support from the National Science Teachers Association (NSTA), the American Association for the Advancement of Science (AAAS), and the National Research Council (NRC).
While it might seem to be a positive development that professional organizations and standards for digital literacies were abundant prior to the pandemic, there is no previous research about what guidance existed for digital literacies in different countries or in smaller government levels, such as a U.S. state. This gap extended to a lack of knowledge about how information, standards, and other materials from professional organizations might contribute to policy implementation in schools and classrooms. To address this gap, researchers investigated the following questions:
What professional organizations’ guidance were U.S. states drawing on about digital literacies leading up to the pandemic?
What were the key descriptive features of that guidance?
Defining Digital Literacies
Various professional organizations may use various types of related terms to describe digital literacies. These terms suggest various purposes and values. Also, Digital literacies as a term, is ever-evolving due to its reliance on technologies that are ever expanding (Lankshear & Knobel, 2006). Due to this changeability, policies for digital literacies seem challenging to develop. In research projects, key terms for digital literacies also need to be expansive. Many of the literacies of the digital originated in the pre-digital period but have gained importance because of their links to more sophisticated practices (Bawden, 2008; Martin & Grudziecki, 2006). This has happened in much the same way that early household computers have a relationship with more modern machines. The more modern computers have greater memories and connective capacities and a larger array of options for using and producing text (Bawden, 2001, 2008). While the terminology around using and producing text with digital literacies has undergone considerable development, many of the same concepts that emerged in the 1960s governing technological knowledge and skills are still used today (Dakers, 2006; van Joolingen, 2004, slide 5). For example, machines like typewriters and even older computers still have screens and keyboards. Many of the same keystrokes are required to do the same task in the 1880s through to the 1980s and now in 2021.
Early definitions and terms
There were many attempts to define digital literacies that compare them to information literacies (Gilster, 1997; Eshet, 2002). Specifically, the Prague Declaration (UNESCO, 2003) emphasised the global importance of information literacy in the context of the “Information Society.” The statement used words such as identify, locate, evaluate, organize, create, and use to convey its intentions. Specifically, the UNESCO document predicted a so-called new Information Society and indicated that Information Literacy was a basic human right.
Expansion of terms to Include Media Forms
Evaluating media has a critical bearing on individual health decisions, elections, and more (Potter, 2018). Media literacy studies have come to encompass understandings about how messages are constructed and interpreted, including understanding how characteristics of the author/sender and the receiver are crucial in understanding the meaning of the message and its content (Bawden, 2008; Martin & Grudziecki, 2006). Most recently, this includes social media and gaming (Rhodes & Robnolt, 2009). Part of understanding media requires understanding how images, text, and/or sound are used together to make meanings for various audiences (Blackall, 2005; Cameron, 2000; Mills, 2016). Multimodality also highlights the importance of symbolic systems and social semiotics (Kress & Leeuwen, 2006). Together, these elements of information, media, and mode shape what Lankshear & Knobel (2003) described as New Literacies and they include post typographic forms of texts, such as images, sounds, haptic manipulation, and more. These literacies help users access, interpret, criticise, and participate in culture and society (Kahn & Kellner, 2006; Kellner, 2002; Snyder, 2002; Tyner, 2014).
Recent definitions and expansions of Digital Literacies
More recently Çam & Kiyici (2017) merged elements from multiple time periods and frames, including ICT, information literacy, technological literacy, visual literacy, and media literacy. This definition summarised many of the repeating, overlapping ideas from previous definitions and provided the guidance for this study.
The term “digital literacy” today can be defined as the technical knowledge and skills necessary for the individuals who want to lead a productive life, to continue their personal development with lifelong learning activities, and to contribute positively to society. With this definition, the literacies incorporated in digital literacy can be listed as Information Literacy, Visual Literacy, Software Literacy, Technology Literacy, and Computer Literacy (p. 30).
For literacy researchers, there have been additional expansions on the idea of digital literacies to include more critical evaluation (Alvermann & Sanders, 2019; Digital Literacy High-Level Expert Group, 2008). In addition, Kellner (2002; 2004) expanded the notion of literacies even further to include aspects of social power as a site for production and consumption of digital texts. This layer of criticality requires teachers to focus more on issues of audience and purpose than skills like typing or even coding as aspects of digital literacies. Using Kellner (2003) as a frame for critical digital literacies might be precarious in the current moment as many U.S. states are banning curriculum and instruction that even uses the word critical (Sawchuk, 2021). It is under these shifting definitions and expanding understandings about literacies that states are working to determine the importance of these and make educational policies about them. It is also in this context that digital literacies guidance at the state level, might use multiple terms and take on multiple forms.
Professional Organizations and Advocacy
Professional agencies in education in the U.S. and even globally in some cases have been developing various standards (e.g., ISTE, CTSA, Next Generation Science, NETEA, AASL) for the specific purpose of informing the research and practice. Moreover, organizations rely on entities to take up their standards to demonstrate that their organization is viable, deserving of their members, and worthy of additional members. Besides the practical matter of demonstrating their professional prowess, these professional organizations and the guidance they give should enable conversations about equity when they are taken up by entities such as states.
Professional organizations all over the world engage in developing standards or guidance to set expectations for professional practices as well as provide some managerial oversight (Wadmann et al., 2019). Within the U.S., and for education specifically, these organizations have advanced new standards, guided education policies, and influenced legislative and regulatory practices (Dauenhauer et al. 2019; 2014; McGuinn, 2012).
Within such professional organizations, prominent members, boards, and/or task forces use research and perhaps federal guidance to craft standards and make recommendations for practice. Since the state leadership and the professional organizations might have the same access to members, research, and federal guidance, it stands to reason that most states would come to develop or adopt the similar standards from a narrow range of organizations, rather than each state crafting their own separate standards from multiple sources (DiMaggio & Powell, 1983). Moreover, states that are close together geographically may also be more likely to rely on the same professional organizations for standards. With these ideas about the relationship between professional organizations and leaders in-state entities, we undertook this policy scan study to explore what guidance for digital literacies U.S. state departments of education were providing to local schools.
Methodology
This policy scan drew on strategies from Ortiz et al. (2020) for obtaining information from state databases. The following policy scan activities were undertaken in 2018 through 2019: (1) open google search processes, (2) targeted searches on the state websites, (3) selection of information for inclusion, (4) content analysis of included items, and (5) verifying analysis for trustworthiness.
Open Google Searches
The first round of searching consisted of open Google searches. One researcher entered the terms “digital literacies [state name]”, “technological literacies [state name],” “information technologies [state name]” and “information computing technologies [state name]” to allow for the widest search possible. From there, the links that were provided were followed to see if the search result was linked to the state office of education or was sponsored by them in any way. Whatever was linked was captured either as a link or downloaded as a document and stored with a file in the state’s name.
Targeted state website searches
The second task was to go directly to the state website and search for guidance about digital/technological literacies. A second researcher engaged in this step. Where possible, searches were entered into search fields on the sites. Another strategy was to follow tabs and hyperlinks to literacy information and determine if there were any resources for digital/technological literacies. Resulting sites and documents were downloaded or captured as a link and then added to a file with the state’s name. When this was done, the two researchers compared and compiled documents.
Selection of information for inclusion
Once the information from the folders had been compiled, it was time to determine which documents constituted guidance. The researchers used the following criteria to determine whether this was the case. First, the document or site needed to be formally prepared for public viewing. This usually indicated formatting, a title, and an explanation of where the document came from. Second, the document or site needed to specify that it was for use in planning instruction. Third, the document or site is needed to name specific student outcomes. These could be formal objectives, or they could be more general aims. Finally, the document needed to contain the phrase “digital literacies” or related terms to separate it from general technological implementation or assistance with technologically supported patterns like blended learning. A table was made with fields for each state. Items were added as they related to the research question. Specifically, researchers recorded content areas, the organizations within the state office of education or under their direction who were sponsoring the development of the document, and what professional organizations’ information was cited in the document.
Content analysis of included items
The results underwent conventional content analysis to determine findings (Hsiu-Fang & Shannon, 2005). Content analysis leads to the “interpretation of content of text data through the systematic classification process of coding or identifying of themes or patterns” (Zhang & Wildemuth, 2009, p. 1278). Using a qualitative design for this content analysis emphasized “concepts rather than simply words” (Fraenkel & Wallen, 2011, p. 389) and also conveyed facts in a manner that was coherent and useful (Sandelowski, 2010). The content analysis for this project was conducted by having two researchers—one researcher from the initial identification and one entirely new researcher—move through the data in the first table to identify patterns. Several emerged. One was the prevalence of the ISTE standards as professional organizations whose work had been cited from the standards. The two researchers then made a new table of the ISTE states documenting whether the topic headings from ISTE were intact, the grade level articulations like those from ISTE had been used, the ISTE standards had been used as a primary framework or there were also other organizations cited, certain subjects or content areas were mentioned, and the definition of digital or technological literacy was included. The table was coded for patterns by the two researchers and then checked for accuracy by the researcher from the first round who participated in the identification of initial documents but did not participate in the initial content analysis.
Verifying analysis for trustworthiness
In qualitative research, trustworthiness reflects an effort to “control for potential biases that might be present throughout the design, implementation, and analysis of the study” (Bloomberg & Volpe, 2012, p176). The study was designed to allow other researchers to track “the processes and procedures used to collect and interpret the data” (p. 163). The account described above should enable other researchers to follow the same procedures and arrive at generally the same conclusions. During this process it was critical to ensure that major steps were taken and completed promptly so that websites would not change or that established issues of consensus among the researchers for inclusion would not be forgotten or shifted.
This process of working steadily also supported us in developing objectivity (Bloomberg & Volpe, 2018). The goal was to ensure that the findings come directly from the research rather than the biases and subjectivity of the researcher to the greatest extent possible. To ensure this, researchers used team processes where two researchers performed the identification task with a separate verification and then a team with one original researcher and a new researcher performed the analysis with a separate verification researcher. Further, one of the researchers was a digital literacies scholar and may have carried notions with them about what guidance should look like but was checked by another researcher who studied other aspects of technology in schools. A third researcher who did not research technology at all also contributed by providing an outsider view of what should constitute guidance about digital literacies (Hammersley, 2012). Thus, the findings presented should present a strong objective interpretation of the information that was discovered.
Findings
The purpose of this study was to learn what professional agencies U.S. states had been drawing on for guidance about digital literacies on their department of education websites. Each state department of education website was searched for guidance, but not every state had a plan. What follows is a description of the guidance that was found.
Overall description of findings
During the search process, the ISTE standards emerged as the most frequent source of professional guidance. Sixteen states used the ISTE standards as their primary source of guidance and support for digital literacies (See Fig. 1). Moreover, 11 states had adopted the standards as they were, with no modifications (Alabama, Alaska, Connecticut, Idaho, Mississippi, Nevada, Rhode Island, Vermont, Washington, West Virginia, Wisconsin). Nine states adopted the ISTE standards in concert with other professional organizations (Alabama, Minnesota, Mississippi, North Dakota, Oklahoma, Oregon, South Dakota, Washington, Wisconsin). These organizations will be discussed later. Finally, eight states even included the same grade level articulations as those that ISTE uses (Alabama, Alaska, Idaho, Nevada, South Dakota, Washington, West Virginia, Wisconsin).
Fig. 1 States that use ISTE standards for digital literacies
The states that used the ISTE standards placed them in context with a variety of subject matter orientations. Some states used the standards in more than one subject matter orientation: computer science (Alabama, Washington, West Virginia, Wisconsin); library science (Connecticut, Oregon, Wisconsin); general technology and information science (Idaho, Nevada, Washington, Wisconsin); Other sciences (Washington, Wisconsin); English language arts/general literacy (Mississippi, Washington, Wisconsin); mathematics (Mississippi, Washington, Wisconsin); and social science (Washington).
A few states that featured professional guidance about digital literacies used standards from groups other than ISTE. For example, the state of Washington had a document that incorporated guidance from NSCC, CTSA, and Next Generation Science Standards. Also, Washington referenced the largest number of subject matter specialties, including science, math, and technology, as well as library sciences, social studies, and English language arts. Table 1 contains sample descriptions from various states.
Table 1 Additional Sample Descriptions of State Plans
State Description from Website
Alaska The Alaska Digital Literacy Standards update the 2006 Alaska Technology Standards, bringing them into alignment with contemporary technology opportunities available to districts and students in Alaska. They focus on leveraging technology for learning, expression, digital citizenship, and communication. The standards are based on the International Society for Technology in Education (ISTE) student standards (2016 edition).
Idaho The standards align with the draft version of the K-12 Computer Science Framework (2016). The framework reflects the latest research in CS education, including learning progressions, trajectories, and computational thinking. The K-12 Computer Science Framework was steered by five organizations: Association for Computing Machinery, Code.org, Computer Science Teachers’ Association, Cyber Innovation Center, and National Math and Science Initiative; several states
(MD, CA, IN, IA, AR, UT, ID, NE, GA, WA, NC), large school districts (NYC, Chicago, San Francisco); technology companies (Microsoft, Google, Apple); and individuals (university faculty, researchers, K-12 teachers, and administrators).
North Dakota The North Dakota State content standards offer instructional guidance in core curriculum areas, while at the same time, they allow for, indeed encourage, a dynamic and living curriculum created at the local school district level. To ensure educational relevance, our state’s content standards are (1) based on academic standards developed nationally by various professional education associations, (2) validated by our state’s best educators based on classroom experience and local community expectations, and (3) widely supported by state and national education policymakers.
Oklahoma The Oklahoma Society for Technology in Education (OKSTE) provides instructional support, innovative leadership, and professional development as learning communities fuse the science of teaching and learning with EdTech. OKSTE is an ISTE affiliate and will use that relationship to create cadres of learning communities that are innovation leaders in the authentic integration of EdTech in Oklahoma classrooms.
Rhode Island The International Society of Technology in Education (ISTE) Standards act as a roadmap for educators and education leaders looking to re-engineer their schools and classrooms for digital age learning.
Vermont The Vermont State Board of Education adopted the International Standards for Technology Education (ISTE) Standards for student learning. The adoption of this framework is aligned with the state’s Education Quality Standards. These standards outline what Vermont students should know and be able to do with respect to information technology and will guide and inform the work of our schools as they prepare students for college and careers that have been dramatically transformed by information technology.
Washington Defines digital and media literacies and makes recommendations for legislative actions that support policies and practices.
West Virginia The West Virginia Board of Education and the West
Virginia Department of Education are pleased to present Policy 2520.14, 21st Century Learning Skills and Technology Tools
Content Standards and Objectives for West Virginia Schools. The West Virginia Standards for 21st Century Learning
includes 21st century content standards and objectives as well as 21st century standards and objectives for learning skills
and technology tools. This broadened scope of curriculum is built on the firm belief that quality engaging instruction must
be built on a curriculum that triangulates rigorous 21 st century content, 21 st century learning skills and the use of 21st
century technology tools.
Exploration of several States’ plans
States that incorporated ISTE standards to address digital literacies generally agreed on the importance of preparing students for digital citizenship as a critical aspect of digital literacies. However, states took different paths to the creation of their digital literacy guidance. Two examples come from Alabama, a politically conservative state in the southeastern U.S. and Connecticut, a politically progressive state in the northeastern part of the U.S. Alabama spends less money per pupil than most states and has a lower graduation rate than many other states. Connecticut spends more money per pupil than most states and has a higher graduation rate and students score higher on college entrance tests than in most other states. These differences provide an opportunity for comparison and contrast of their plans.
Alabama
The Alabama public school system (prekindergarten through grade 12) operates within smaller districts governed by locally elected school boards and superintendents (Ballotpedia, 2021a). The state has about 750,000 students enrolled in roughly 1,700 schools in 173 school districts (Ballotpedia, 2021a). The state has about 52,000 teachers in their schools (Ballotpedia, 2021a). On average Alabama spends close to $9,000 per pupil and ranked it 39th highest in the nation in spending (Ballotpedia, 2021a). The state’s graduation rate hovers around 80%.
In terms of digital access, about 88.6% of all Alabamians have access to high-speed internet, defined as internet with download speeds of 25 megabits per second or faster (Smith, 2020). There are about 475,000 people in Alabama without access to high-speed internet, including large swathes of rural Alabama (Smith, 2020). In fact, nine of Alabama’s 67 counties have less than 30% access to broadband, all of which are counties where many Black families live, some of which have poverty rates of greater than 30% (Smith, 2020). In addition, the US census bureau estimates only 73.3% of households subscribed to broadband internet between 2014 and 2018, suggesting that even when households have access to broadband, they may not be able to afford it (Smith, 2020). The governor of Alabama allocated 100 million dollars of federal disaster relief money to broadband access and schools partnered with local business to provide free Wi-Fi to schools for remote instruction. Given these circumstances, many K-12 students in the state of Alabama need access to the internet alongside access to digital literacies. Also, K-12 students in the state seem to need stronger educational support overall.
Alabama created a task force that was charged with conceptualizing and creating a Course of study for digital literacies and computer science (Richardson, n.d). The task force in Alabama that created their pre-pandemic digital literacies policy document consisted of stakeholders that represented multiple areas of the state of Alabama’s educational system. The task force included members from local Boards of Education, faculty members from local universities, school technology specialists, and one representative from industry. The industries represented were technological in nature, such as a company that sold email sorting software.
The Alabama document focused drew on the ISTE standards, saying:
The International Society for Technology in Education (ISTE) Standards for Students emphasize the skills and qualities we want to foster in students, enabling them to engage and thrive in a connected, digital world (Richardson, n.d., p.1).
This does not say anything explicit about literacy, but rather that the skills and the ethos of the standards are the reason for their inclusion in the development of the course of study. For defining digital literacies, the group members generated the following definition:
Digitally literate students can use technology responsibly and appropriately to create, collaborate, think critically, and apply algorithmic processes. They can access and evaluate information to gain lifelong knowledge and skills in all subject areas. Alabama includes computational literacy in its definition of digital literacy. Our goal is digital and computational literacy for all Alabama students (Richardson, n.d., p. 1).
Their definition includes many of the goals advanced for digital literacies, including advancing social democracy through the goal to “collaborate” and “use technology responsibly” (Mora, 2016), maintaining a workforce through the emphasis on “computational literacy” (Pothier & Condon, 2019), and supporting ongoing innovation and quality of life through “lifelong knowledge and skills” (Coldwell-Nelson, 2017). In this course of study, the team who created the document hybridized many ideas about digital literacies and digital learning to create a guidance document for their state.
Connecticut
The Connecticut public school system (prekindergarten through grade 12) was operating within districts governed by locally elected school boards and superintendents (Ballotpedia, 2021b). Connecticut has about 560,000 students enrolled in a total of 1,200 schools in 200 school districts (Ballotpedia, 2021b). There were about 44,000 teachers in the public schools (Ballotpedia, 2021b). On average Connecticut spends about $17,000 per pupil per year, which ranked it fifth highest in the nation in spending (Ballotpedia, 2021b). The state’s graduation rate is about 86% (Ballotpedia, 2021b).
During the pandemic, Connecticut spent 60 million dollars in federal monies and private donations to ensure that all students had an internet-ready device for school. Prior to the pandemic, about half of Connecticut’s student population lacked an internet connection and/or an internet-ready device in their home (Thomas & Watson, 2020). Many of these families without access were people of color, particularly Black and Hispanic/Latinx families (Thomas & Watson, 2020). Even so, thousands of school children were left without a broadband connection when remote schooling resumed in the fall (Thomas & Watson, 2020). These facts illustrate that even in states that spent more money on education and generally had better outcomes, there were large disparities when it comes to access to the internet, devices, and thus, digital literacies.
In Connecticut the department of education created a document called Charting New Frontiers in Student-Centered Learning (Casey et al., 2017). To create the document, there were multiple councils assembled into one large advisory committee. These councils included topics of data and privacy, digital learning, and infrastructure. Individuals who participated in these councils were representatives from public schools—mostly at the administrative level, higher education, library directors, law firms, data security officers from universities, and technology vendors. The town manager of Coventry, Connecticut also participated as a community member. Overall, there was far less participation from school representatives, but more from other groups.
In terms of digital literacies, the plan offered no direct definition, but the document expressed the intention for libraries to be sites of digital literacies:
Connecticut’s libraries offer digital and media literacy, cybersecurity, and other technology training to students of all ages, helping to bring about a more literate citizenry that can better understand and support the power of technology to deepen and personalize learning (Casey et al., 2017, p. 47).
The document includes references to media, data privacy, as well as ICT skills through “training” to students. The goals for democratic participation are also evident (Coldwell-Nelson, 2017). What is different about this document from Alabama, other than the greater emphasis on digital skills over digital literacies. In Alabama, the definition offered had a more community-centered approach and a stronger articulation of lifelong benefits to the students. Also, the Connecticut document does not mention ISTE at all, but instead includes several state-based technology commissions and the national organization Innovation Partners of America.
Together, these two examples demonstrated how states were offering very different documents—some with very narrow definitions in terms of content, and some with more open definitions going into the pandemic. Importantly, these extant policies were the baseline or starting point for what states had been providing to schools before massive remote learning set in. The comparison also highlights the various ways that digital literacies might be included in guidance documents to varying degrees while still expressing similar values for students to understand technologies and engage with them in ways that benefit them and possibility their communities.
Discussion
The purpose of this research was to investigate the integration of professional guidance about digital literacies into U.S. state education policies. Researchers wanted to know whose guidance states were relying on and how the states had incorporated that guidance within the subject matter and grade levels. Limitations to this study included the fact that states may post, update, move, or remove information at their will so not all information located during the search period may be located now, but available data can be provided to others upon request. With keyword searches, there is always a chance that additional keywording would have yielded additional information. Our purpose of learning about guidance for digital literacies focused on connecting information to literacy pages on state department of education websites, but it is possible that state technology education officials have taken up digital literacies without posting information online as evidence. Also, state officials change periodically or shift responsibilities, making it difficult to identify persons who might have information or who may be posting it.
While it is not a limitation, but rather an issue of scope, we remind readers that we did not gather data about how those policies might have changed during the pandemic. We might anticipate that not very much policy making directly about digital literacy occurred, however, since it would seem more likely that states were scrambling just to get remote online learning set up so that some instruction could be given. What is clear from an analysis of the data is that states took a variety of approaches that in future work can be examined by policy makers to see if there is need of revision.
Finally, it is important to remember that ISTE standards are used by many states, but that the purpose of this research was to find guidance for digital literacies. Not every state that used ISTE standards placed the standards alongside notions of digital literaciesand not every state that used ISTE standards as digital literacies has made strong connections between the standards and digital literacies.
What Professional Organisations’ Guidance do U.S. States draw on about Digital Literacies?
Although 16 states used the ISTE standards somewhere in their digital literacy guidance and a handful used standards from other places, over half of the states did not mention any organization’s guidance in connection with digital literacies. This could mean that states were not prioritizing digital literacies or any of its related concepts (information literacies, technological literacies). It could also mean that states did not know where to go for guidance. Afterall, Lankshear and Knobel (2006) pointed out that defining and understanding digital literacies outside of academic spaces is an imprecise endeavor. Indeed, in the two case study states, many families of color and families in rural areas were without internet access and devices, which might take priority for digital literacies away from officials in favor of other needs, including basic needs for connections and devices. However, when states encountered difficulties moving to digital learning, some of those challenges might come from an apparent lack of clarity about what are digital literacies and what constitutes digital knowledge and skills (Cameron, 2000; Mills, 2016). Reciprocally, one needs digital literacies to advocate for and leverage access to the internet and devices, but one also needs an internet connection and a device to use and develop digital literacies.
For states that had guidance, many chose the ISTE standards. This was an interesting choice because the ISTE standards do not explicitly claim to be literacy standards. In an explanation of the recent revision of the standards, Trust (2018), who was one a member of the revision team, noted that the standards were focused on helping teachers in using technology to “support student learning and creative thinking, design digital age activities and assessments, model digital work, promote and model digital citizenship, and engage in professional growth and leadership” (p. 1). Notice the similarity of these standards to many of van Joolingen’s (2004) definition of ICT literacies as:
the interest, attitude, and ability of individuals to appropriately use digital technology and communication tools to access, manage, integrate and evaluate information, construct new knowledge, and communicate with others to participate effectively in society (slide 5).
Reading the information from Trust (2018) and van Joolingen (2004) together, it is evident that ISTE standards touch on many of the same priorities as ICT literacies although ISTE has never claimed to be a literacy organization or claimed the ISTE standards are about literacies. Such circumstances again, raise the issue as to whether there is a meaningful distinction at this point in technological development between digital literacies and the knowledge, skills, and dispositions to engage in digital learning generally (Cameron, 2000; Mills, 2016). There may not be a meaningful distinction for state-level officials who design, approve, and promote this guidance. Other countries striving to support digital learning might also overlook digital literacies as a distinct need or embed it within general learning without reflection.
Further, while there was a wide variety of guidance taken up by states who were using guidance, there were some notable organizations that were not represented. One such example was the National Council of Teachers of English, which has a committee called Digital Literacies in Teacher Education (D-LITE). The D-LITE group has published several iterations of Beliefs about Integrating Technology into the English Language Arts Classroom (Lynch et al., 2018). Other countries might also have organizations that have guidance that has not been accessed by government leaders and could prove helpful.
Another challenge in the U.S. and abroad might be the participation of educational technology vendors in policy making without proper transparency. Such participation was more of an issue in Connecticut where there are many headquarters for educational technology companies versus Alabama, which prioritized wide participation from local school experts. Leaders in the U.S. and other countries might consider who is participating in these policy discussions, what their motives might be, and what can be done, not necessarily to discourage or eliminate vendor participation, but to declare it and understand its limitations the same as any other group. For example, many other countries struggle with non-governmental organizations (NGOs) that sometimes come into countries and promise support for underserved groups; however, people from those countries do not always trust that they are not just seeking government money from poor countries as a source of profit (Author, 2019).
In summary, the fact that the states that do have plans mostly drew their guidance from the same organization, even though that was not explicitly a literacy organization, supports the theory that many U.S. state leaders have converged on similar policies to ease the burden of trying to generate totally separate guidance within their states (DiMaggio & Powell, 1983). While there was only one state that had executive representation from ISTE on their board (Nevada), many of the planning board members may be members of ISTE and their familiarity with ISTE many have led them to think that adopting the standards was time and money saving. Again, the researchers of this study find nothing wrong with the ISTE standards—they merely question whether their wide use as literacy standards occurred because there was no real distinction between digital literacies and digital learning skills in policy makers’ minds, or these leaders were unable to reflect on the differences due to time, cost, or other pressures. This is also an important question in the U.S. and internationally when research studies use terms like ICT and e-learning as catch-all phrases for any learning that uses digital elements. As research, policy, and practice communities, we should be asking what the value is in having and using certain terms with precision, or if the terms should be left to local decision makers.
What are the key descriptive features of the Guidance documents?
As for question 2 about how professional guidance has been integrated, the answer is that in cases where the standards are used, they are used in their entirety. Moreover, the grade level distinctions recommended by ISTE were often included. Only a few states, such as Minnesota, Washington, and Alabama, had developed comprehensive digital literacies guidance that drew on multiple sources and had been customized to the needs of the states. Regardless of the unique needs that probably exist in every state, the common needs are the ones that are in the best position to be met because they enable complete adoption of standards that already exist and again, carry the perception of saved time and money (DiMaggio & Powell, 1983). While it is beyond the scope of this research to say how and whether local schools used the guidance, it is interesting to consider how participation in the creation of the plans might shape later use of them in the places that matter: classrooms. This also seems important in the international arena where there may be differing levels of federal oversight of education.
Implications
Now that the pandemic has changed education and offered insight into the best and worst of digital education experiences for school children and their teachers, there is a need to make stronger plans for digital learning during times of disruption and beyond (Rice & Zancanella 2021). Along these lines, the current study has several important implications for practice, research, and policy.
Practice
High quality digital literacy learning is not an accident and does not seem to be coercible in the moment of crisis (Rice & Zancanella, 2021). Therefore, educational agencies need to study guidance from professional organizations in their respective countries about digital literacies use to craft sound policies and practices. We also feel that these conversations can include all digital learning, but that some explicit conversation about digital literacies is important, given the interrelationship between digital literacies and digital learning. It is our contention that digital literacies is more than just ICT knowledge repackaged from the 1960 and 1970s (Martin & Grudziecki, 2006). In a true frame of digital literacies, practices that are oriented around analysis and evaluation of digital texts and tools, alongside awareness of audience and purpose for what is composed in digital spaces are most important (Alvermann & Sanders, 2019; Kellner, 2002; 2004; Mills, 2016).
Finally, U.S. state agencies and policy making agencies in other countries should consider how to make this guidance appropriate for the schools in their stewardship (U.S. Department of Education, 2021). There seems to be strong opportunities to make plans that draw on local strengths and expertise, such as was done in Alabama, since there has been so little federal guidance about digital literacies in the U.S. and so many national organizations with advice seem to stand at the ready. For countries without a wealth of professional organizations, leaders might consider looking to a geographic neighbor with shared history and values as a starting point. For example, Canada developed a nationwide overview of the digital literacy strengths and needs across their provinces and outlined steps that might be taken to develop policies at the level of provinces. (Hoechsmann & DeWaard 2015). Their work might be useful for countries like the U.S. that share some aspects of political organization, history, and geographic proximity. In turn, Canadian officials might consider what professional organizations in their country or internationally might have additional insight for policies that leverage the strengths for their various provinces.
Research
Future research projects should look at how states with various types of guidance from various agencies fared during the transition to emergency remote learning during the COVID-19 pandemic (Education Week, 2020). This research might be done in countries outside the U.S. with whatever guidance these political units were using. Besides achievement data, perception work might be useful in finding out whether educators within the state at various levels know about the guidance or use it. Moreover, some individual school districts and agencies might have their plans that draw on various sources of guidance. It seems worth investigating how these plans are formed and how professional organizations’ guidance makes its way (or does not) into the plans. In other countries, similar work might determine what educators in classrooms know about the policies for digital literacies (if there be any) in their jurisdictions. It might also be helpful to look at what aspects of guidance are most important to set a foundation for digital literacies development and digital learning in schools. For example, findings from a study in Ireland suggest that broadband access might be a critical foundational step for establishing engagement for digital learning generally (Mac Domhnaill et al., 2021).
Future research about professional learning about digital literacies provided by U.S. states or other agencies and local school entities internationally might shed additional light on how professional guidance makes it to teachers and into their practice. For example, it seems important to understand how teachers understand guidance about digital and technology education in cases where digital literacies and its accompanying ideologies about power construction and community practices are included and when they are not (Kellner, 2002; 2004). This is especially important considering that many U.S. states and many countries were scrambling to acquire internet connections and devices for students when the COVID-19 pandemic shut down many school buildings (Mac Domhnaill et al., 2021; Smith, 2020; Thomas & Watson, 2020). Who receives devices and access and why they have it are part of the critical digital literacies that students need (Kellner, 2002; 2004).
Policy
While ISTE redesigned its Standards for Educators regarding how to use technology to learn, collaborate, lead, and empower students based on input solicited from more than 2,000 educators and administrators, generally, the relationship between standards and academic research remains unclear (Trust, 2018). While practice to policy to research is certainly a possible pathway to understanding, there are alternative pathways that begin with collaborative research with practitioners and then move to policy before recommending procedures for practice.
While there were U.S. states that used ISTE standards that did not expressly connect the standards with digital literacies, several of them did. In terms of policies, ISTE might consider how many state-level agencies have adopted their standards into their digital literacies guidance. What should ISTE do with this information? They might be more explicit about how their standards promote digital literacies. They might also distance themselves from literacies and focus more on their work as a general technological organization. Other political units and entities might also consider reviewing the guidance they give to ensure that digital literacies are specifically addressed according to how it is conceptualized and applied in local settings. The questions remain: Is there a meaningful distinction between digital literacies and digital learning in the minds of practitioners and policy makers? Should there be?
Conclusion
The purpose of this study was to understand what professional agencies were providing the guidance that U.S. states were adding to their state department of education websites as part of their support for local schools. By locating and analyzing the various plans, the ISTE standards for technology learning were the most popular, but most states offer no guidance at all. However, the states that have adopted the ISTE standards, did seem to follow the isomorphic pattern of geographical clustering (Fig. 1). These findings suggest that leaders in various regions should consider including guidance that is focused on literacies directly, even when they have guidance about learning generally (Alvermann & Sanders, 2019; Kellner, 2002; 2004). With the COVID-19 pandemic’s disruption of schooling and the continuation of the trend toward the digital instructional materials, now is the time for states to make strategic decisions about how to include digital literacies in their formal guidance.
Data Availability
Some data may be available upon request.
Declarations
Conflicting Interests
The authors declare no conflicts of interest.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Álvarez Valencia JA Meaning making and communication in the multimodal age: ideas for language teachers Colombian Applied Linguistics Journal 2016 18 1 98 115 10.14483/calj.v18n1.8403
Alvermann, D. E., & Sanders, R. K. (2019). Adolescent literacy in a digital world. In R. Hobbs, & P. Mihailidis (Eds.), The International Encyclopedia of media literacy (pp. 1–6). John Wiley & Sons.
Ballotpedia (2021a). Public education in Alabama. Retrieved from https://ballotpedia.org/Public_education_in_Alabama
Ballotpedia (2021b). Public education in Connecticut. https://ballotpedia.org/Public_education_in_Connecticut
Bawden, D. (2008). Origins and concepts of digital literacy. In C. Lankshear, & M. Knobel (Eds.), Digital literacies: concepts, policies and practices (pp. 17–32). Peter Lang.
Blackall L Digital literacy: how it affects teaching practices and networked learning futures-a proposal for action research in International Journal of Instructional Technology and Distance Learning 2005 2 10 14 27
Çam, E., & Kiyici, M. (2017). Perceptions of prospective teachers on digital literacy.
Malaysian OnlineJournal of Educational Technology, 5(4),29–44.
Cameron, D. (2000). Book review: Multiliteracies: Literacy learning and the design of social futures. By Bill Cope, Mary Kalantzis (eds.) (2000). Routledge. Changing English, 7(2), 203–207.
Casey, D., Duty, L., & Kisiel, R. (2017). Chartering new frontiers in student-centered learning. Connecticut Association of School Superintendents.
Dakers, J. (Ed.). (2006). Defining technological literacy. Palgrave Macmillan.
Digital Learning Collaborative (2019). Snapshot 2019: A review of K-12 online, blended, and digital learning. Retrieved from https://www.evergreenedgroup.com/keeping-pace-reports
Digital Literacy High-Level Expert Group (2008). Digital literacy European commission working paper and recommendations from Digital literacy high-level expert group. Brussels, Belgium. Retrieved from http://www.ifap.ru/library/book386.pdf
DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields.American Sociological Review,147–160.
Dong C Newman L Enacting pedagogy in ICT-enabled classrooms: conversations with teachers in Shanghai Technology Pedagogy and Education 2018 27 4 499 511 10.1080/1475939X.2018.1517660
Drajati, N. A., Tan, L., Haryati, S., Rochsantiningsih, D., & Zainnuri, H. (2018). Investigating English language teachers in developing TPACK and multimodal literacy.Indonesian Journal of Applied Linguistics,575–582.
Education, & Week (2020). Map: Coronavirus and school closures https://www.edweek.org/ew/section/multimedia/map-coronavirus-and-school-closures.html
Eshet, Y. (2002). Digital literacy: A new terminology framework and its application to the design of meaningful technology-based learning environments. In P. Barker & S. Rebelsky (Eds.) Proceedings of EDMEDIA, 2002 World Conference on Educational Multimedia, Hypermedia, & Telecommunication (pp. 493–498). AACE.
Fraenkel, J. R., & Wallen, N. E. (2011). How to design and evaluate research in education. McGraw-Hill.
Gilster, P. (1997). Digital literacy. Wiley. Hall, S (1991 [1973]) Encoding, decoding. In S. During (ed.) The Cultural Studies Reader, (pp. 90–103). Routledge.
Hammersley, M. (2012). What is qualitative research?. A&C Black.
Hsiu-Fang, H., & Shannon, S. E. (2005). Qualitative health research. Three Appro.
Kahn, R., & Kellner, D. (2006). Reconstructing technoliteracy. Defining technological literacy. Palgrave Macmillan.
Kellner, D. (2002). Multiple literacies and critical pedagogies: new paradigms. In P. P. Trifonas (Ed.), Revolutionary pedagogies (pp. 218–244). Routledge.
Kellner D Technological transformation, multiple literacies, and the re-visioning of education E-learning and Digital Media 2004 1 1 9 37 10.2304/elea.2004.1.1.8
Kress, G. R., & Leeuwen, T. (2006). Reading images: the grammar of visual design (2nd ed.). Routledge.
Lankshear, C., & Knobel, M. (2003). New literacies: changing knowledge and classroom learning. Open University Press.
Lankshear C Knobel M Digital literacy and digital literacies: policy, pedagogy and research considerations for education Nordic Journal of Digital Literacy 2006 1 12 24 10.18261/ISSN1891-943X-2006-01-03
Luik P Taimalu M Suviste R Perceptions of technological, pedagogical and content knowledge (TPACK) among pre-service teachers in Estonia Education and Information Technologies 2018 23 2 741 755 10.1007/s10639-017-9633-y
Lynch, T., Hicks, T., Bartels, J., Beach, R., Connors, S., Damico, N., Doerr, Stevens, C., Hicks, T., LaBonte, K., Loomis, S., McGrail, E., Moran, C., Pasternak, D., Piotrowski, A., Rice, M., Rish, R., Rodesiler, L., Rybakova, K., Sullivan, S., Sulzer, M., Thompson, S., Young, C., & Zucker, L. (2018). Beliefs for integrating technology into the English Language Arts classroom Retrieved from http://www2.ncte.org/statement/beliefs-technology-preparation-english-teachers/
Martin A Grudziecki J DigEuLit: concepts and tools for digital literacy development Innovation in Teaching and Learning in Information and Computer Sciences 2006 5 4 249 267 10.11120/ital.2006.05040249
Mac Domhnaill C Mohan G McCoy S Home broadband and student engagement during COVID-19 emergency remote teaching Distance Education 2021 42 4 465 493 10.1080/01587919.2021.1986372
McGuinn P Fight club: are advocacy organizations changing the politics of education Education Next 2012 12 3 25 31
McLaughlin VL West JE Anderson JA Engaging effectively in the policy-making process Teacher Education and Special Education 2016 39 2 134 149 10.1177/0888406416637902
Mills, K. A. (2016). Literacy theories for the digital age: Social, critical, multimodal, spatial, material and sensory lenses. Multilingual Matters.
Mora RA Translating literacy as global policy and advocacy Journal of Adolescent & Adult Literacy 2016 59 6 647 651 10.1002/jaal.515
Natanson, H., & Strauss, V. (2020). America is about to start online learning, round 2: For millions of students, it won’t be any better. Washington Post (August 20, 2020). Retrieved from https://www.washingtonpost.com/local/education/america-is-about-to-start-online-learning-round-2-for-millions-of-students-it-wont-be-any-better/2020/08/05/20aaabea-d1ae-11ea-8c55-61e7fa5e82ab_story.html
National Governors Association Center for Best Practices, Council of Chief State School Officers. (2010). Common Core State Standards. National Governors Association Center for Best Practices, Council of Chief State School Officers.
Next Generation Science Standards (2013). Retrieved from https://www.nextgenscience.org/
Ortiz K Mellard D Rice MF Deschaine ME Lancaster S Providing special education services in fully online statewide virtual schools: a policy scan Journal of Special Education Leadership 2020 33 1 3 13
Pothier, W. G., & Condon, P. B. (2019). Towards data literacy competencies: Business students, workforce needs, and the role of the librarian.Journal of Business & Finance Librarianship,1–24.
Potter, W. J. (2018). Media literacy. Sage. Rhodes, J.A., & Robnolt, V.J. (2009). Digital literacies in the classroom. In L. Christenbury, R. Bomer, & P. Smagorinsky (Eds.), Handbook of adolescent literacy research (pp. 153–169). Guilford.
Rice M Zancanella D Framing policies and procedures to include digital literacies for online learning during and beyond crises Journal of Adolescent and Adult Literacy 2021 65 2 183 188 10.1002/JAAL.1194
Richardson, E. (n.d.). Alabama course of study: Digital literacy and computer science. Alabama Department of Education.
Sandelowski M What’s in a name? Qualitative description revisited Research in Nursing & Health 2010 33 1 77 84 10.1002/nur.20362 20014004
Sawchuk, S. (2021). What is critical race theory? and Why is it under attack? Education Week (May 18, 2021). https://www.edweek.org/leadership/what-is-critical-race-theory-and-why-is-it-under-attack/2021/05
Smith, M. (2020). No child left offline: Tackling the digital divide in Alabama. A + Education Partnership. Retrieved from https://aplusala.org/blog/2020/08/17/no-child-left-offline-tackling-the-digital-divide-in-alabama/
Snyder, I. (Ed.). (2002). Silicon literacies: communication, innovation and education in the electronic age. Psychology Press.
Thomas, J. R., & Watson, A. (2020). Almost all K-12 students in CT now have internet access. What’s the plan for seniors and others without school-aged families? CT Mirror. Retrieved from https://ctmirror.org/2020/10/27/almost-all-k-12-students-in-ct-now-have-internet-access-whats-the-plan-for-seniors-and-others-without-school-aged-families/
Trust T 2017 ISTE Standards for educators: from teaching with technology to using technology to empower learners Journal of Digital Learning in Teacher Education 2018 34 1 1 3 10.1080/21532974.2017.1398980
UNESCO (2003). Conference report of the information literacy meeting of experts. Prague, September 20–23, 2003. Retrieved from http://www.nclis.gov/libinter/infolitconf&meet/post-infolitconf&meet/FinalReportPrague.pdf
United States Department of Education (2021). The federal role in education https://www2.ed.gov/about/overview/fed/role.html
van Joolingen, W. (2004). The PISA framework for assessment of ICT literacy Retrieved from http://www.ericlondaits.com.ar/oei_ibertic/sites/default/files/biblioteca/7_pisa_framework.pdf
Wadmann S Holm-Petersen C Levay C ‘We don’t like the rules and still we keep seeking new ones’: the vicious circle of quality control in professional organizations Journal of Professions and Organization 2019 6 1 17 32 10.1093/jpo/joy017
Zhang, Y., & Wildemuth, B. M. (2009). Unstructured interviews. Applications of social research methods to questions in information and library science. University of Texas Press.
Dauenhauer B, Keating X, Stoepker P, Knipe R. State physical education policy changes from 2001 to 2016. J SchHealth. 2019; 89: 485-493. DOI: 10.1111/josh.12757
Bloomberg, L. D., & Volpe, M. (2018). Completing your qualitative dissertation: A road map from beginning to end. Sage.
Tyner, K. (2014). Literacy in a digital world: Teaching and learning in the age of information. Routledge.
Sandelowski Margarete What’s in a name? Qualitative description revisited Research in Nursing & Health 2010 33 1 77 84 10.1002/nur.20362 20014004
Coldwell-Neilson, J. (2017). Digital literacy–a driver for curriculum transformation. Research and development in higher education: Curriculum Transformation, 40, 84-94.
| 36465420 | PMC9707243 | NO-CC CODE | 2022-12-01 23:20:22 | no | Educ Inf Technol (Dordr). 2022 Nov 28;:1-20 | utf-8 | Educ Inf Technol (Dordr) | 2,022 | 10.1007/s10639-022-11482-x | oa_other |
==== Front
Proc.Indian Natl. Sci. Acad.
Proceedings of the Indian National Science Academy. Part A, Physical Sciences
0370-0046
2454-9983
Indian National Science Academy New Delhi
137
10.1007/s43538-022-00137-7
Research Paper
Circular economy based approach for green energy transitions and climate change benefits
Niwalkar Amol 1
Indorkar Tushar 1
Gupta Ankit 12
Anshul Avneesh 12
Bherwani Hemant [email protected]
12
Biniwale Rajesh 12
Kumar Rakesh 23
1 grid.419340.b 0000 0000 8848 8397 CSIR-National Environmental Engineering Research Institute (CSIR-NEERI), Nagpur, Maharashtra 440020 India
2 grid.469887.c 0000 0004 7744 2771 Academy of Scientific and Innovative Research (AcSIR), Gaziabad, Uttar Pradesh 201002 India
3 grid.418099.d Council of Scientific and Industrial Research (CSIR), Anusandhan Bhawan, 2 Rafi Ahmed Kidwai Marg, New Delhi, 110001 India
28 11 2022
114
8 4 2022
10 11 2022
© Indian National Science Academy 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Abstract
Concerning rapid resource depletion and the negative effects of climate change, the adaptation of Circular Economy (CE) strategies in the environmental sector is gaining global recognition and application. This study forecasts the energy demand and emissions scenarios for a circular economy-dependent green energy transition, based on learnings from the forced situation caused by the COVID-19 pandemic. This study focuses on the industrial, domestic, and transportation sectors of the National Capital Territory (NCT) of India i.e., Delhi. Implementation of circular economy strategies helps in limiting the impacts of rising energy demand by shifting towards green fuels and biofuels. The data related to energy demand, consumption, percentage share and growth, etc., was gathered and incorporated in Low Emission Analysis Platform model to generate three scenarios i.e., business as usual, circular economy (CE), and pandemic scenario to compare the outputs of energy demand and emissions in terms of CO2 and particulate matter. The results for the CE scenario, for the year 2020 to 2040 shows that there will be a reduction of 158.2 kt PM2.5 emissions (24.3%) and 540 Mt CO2 emissions (49%) as well as a reduction of 203 Mtoe total energy usage (49.3%) as compared to the business-as-usual scenario. The COVID-19 pandemic had a significant impact on societal and commercial activities in the concerned city. Activities came to a standstill during the COVID-19 pandemic, but the same had significantly improved the environment. This unusual forced situation of COVID-19 lockdown produced a unique scenario which shows that the total extra reduction in energy demand of 46%, CO2 and PM2.5 emissions of 45–60% could be achieved by 2040, as compared to CE scenario.
Graphical abstract
Supplementary Information
The online version contains supplementary material available at 10.1007/s43538-022-00137-7.
Keywords
Circular economy
COVID-19
Energy demand
CO2 emissions
PM2.5 emissions
Climate change
==== Body
pmcIntroduction
Activities in a circular economy (CE) based development, builds and rebuilds overall systems integrity. It is a structural shift that promotes long-term resilience, creates commercial and economic possibilities, and benefits the environment and society (Ellen MacArthur Foundation 2015). According to the World Economic Forum, the CE approach can conserve roughly half a trillion dollars of India’s GDP value while saving over one trillion dollars annually due to less carbon emissions, lower costs, and reduced supply chain risk (Fiksel et al. 2021; FICCI 2018). The extraction, manufacture, use and disposal processes as well as reduction, reuse, recycling, repair/refurbishment and recovery of energy and materials, are all based on the lineup process, which converts linear flows into circular flows and is the key to the system’s economic development (Abad-Segura et al. 2020). According to the current development scenario, India might produce an annual value of Rs. 14 trillion and Rs. 40 trillion in the CE sectors by 2030 and 2050, respectively. Currently, Indian government entities, industry associations, and nonprofit groups have lately connected their projects to the concepts of the CE (Bherwani et al. 2022; Ellen MacArthur Foundation 2016). This study developed the scenario of a CE using the CE strategies with the ReSOLVE framework to determine how much CE strategies in the city (Delhi, India) will assist in reducing climate change impacts. The CE strategies include reusing of scrap materials, improving efficiency and methods, using of solar powered or electric operated vehicles, using of green materials, switching to biofuels, sharing of vehicles, etc.
Establishing a CE necessitates multifaceted activities at the local, regional, national and global scales. Industries must use the ReSOLVE framework, which includes six actions i.e., regenerate, share, optimise, loop, virtualize, and exchange to transition from the current path to a circular one. All of these actions improve the use of tangible assets, increase their usable life, and shift away from conventional to renewable resources. Figure 1 depicts the ReSOLVE framework, which shows how the six actions for a CE strategy can be implemented to industries.
Fig. 1 The ReSOLVE framework for CE approach (adapted from Ellen MacArthur Foundation 2015)
In the forthcoming years, roughly two-thirds of the world’s population would be living in cities (Bocquier 2005). According to the World Cities Report (2020), more than 3/4th of the world’s energy is consumed by cities, and they emit more than 3/5th of GHG (Greenhouse gas) emissions. In India, the economic impact of air pollution is exceedingly large. India spends roughly 1.28% of its Gross domestic product (GDP) on health. Air pollution from fossil fuels in India is responsible for one million deaths annually and approximately 980,000 estimated premature births, amounting to an annual economic loss of 10.7 lakh crore (US$150 billion) or 5.4% of India’s GDP (Farrow et al. 2020; Greenpeace 2020). India’s coal supply has grown quickly, and coal remains the country’s primary source of energy and electricity generation (Energy Policy Review 2020). In India, sectors such as residential, commercial buildings, transportation, and industry consume the maximum share of energy in the form of electricity. Though there are number of factors that influence electricity use, one of the most significant is the climate. Frequent variations in climate induce thermal discomfort, and 10–20% of households utilize cooling and 30–40% utilize heating appliances within buildings to alleviate the discomfort in India (Akbari et al. 2001; MOEF 2018). Because of the hot climate, most Indian cities have adopted the practice of using cooling systems to improve thermal comfort (Kumari et al. 2021). With 97.5% of the urban population, Delhi is one of the most urbanized cities in India (MOHUA 2019–2020). In comparison to other Indian cities, Delhi has the highest rate of electricity usage and its per-capita power consumption has increased throughout the years from 522 kWh between 1984 and 1985 to 1548 kWh between 2018 and 2019 (Ramesh et al. 1988; PIB 2019).
According to the environmental performance index, Delhi is one of the world’s most polluted megacities (WHO 2016). In terms of PM10 pollution, Delhi stands at the top of the list, as per the World Health Organization’s (WHO) environmental monitoring database for the world’s biggest metropolises, which spans 100 nations and was published in April 2018 for the years 2011 to 2016 (State of Global Air 2018), PM2.5 concentrations in Delhi have been extremely high for many years, far exceeding the permissible limits set by the National Ambient Air Quality Standards (NAAQS) (Mohan and Kandya 2007; Kumar et al. 2017). In the year 2019, India’s average annual PM2.5 concentration was 58 µg/m3, while Delhi accounted for an average of 98.6 µg/m3 (Chatterji 2021). A high concentration of air pollution beyond the prescribed limit causes major health problems in society, including oxygen deficiency, chronic respiratory disorders, pneumonia, acute bronchitis, cancer and other serious public health issues (Gupta et al. 2021; Heal et al. 2012; Dholakia et al. 2013). Vehicle exhaust, large industries such as power generation, and small-scale enterprises are the main contributors to air pollution in Delhi (CPR 2017). Motor vehicle pollution is one of Delhi’s most critical and quickly developing environmental challenges. As the number of workplaces and residential regions grows, so does the length and number of trips per individual (Bose 1996; TRIPP 2018). Similarly, the industry contributes significantly to air pollution by consuming energy from a variety of fuel sources. Delhi contains India’s largest cluster of small-scale industries. According to the Central Pollution Control Board (CPCB), Delhi is surrounded by extreme pollutive industrial zones that surpass the limits of air, water, and soil emissions (Chatterji 2021). For the relief of the national capital, the Supreme Court in 2017 prohibited the use of other alternatives, such as petroleum coke and furnace oil (The Economics Times 2017). Most of the electricity used in Delhi is generated by thermal power plants, which use coal as input and contribute significantly to air pollution and climate change. The government must make significant efforts to reduce emissions and energy demand by enacting various policies and initiatives. Figure 2 illustrates how Delhi may demonstrate a new CE approach where materials are recycled or reused, renewable energy sources are utilised, and emissions can be decreased by implementing a CE as an initiative.
Fig. 2 Demonstrating the sectors with and without the CE of Delhi
The COVID-19 pandemic has thrown international and national initiatives into disarray. Although it is yet impossible to fully assess the pandemic’s effects, it has undoubtedly reduced worldwide economic production (Le Quéré et al. 2020). To help the economy recover, many governments have only partially eased the lockdown measures. According to the International Monetary Fund, the global economy declined by 3% in 2020 (Long and Ascent 2020). The electricity demand for commercial and industrial loads has reduced to 8–9% but has grown to 12% for residential loads in India, as industrial and commercial activities have been constrained, while people have been forced to stay at home and work from home (Elavarasan 2020). This could have a negative impact on network equipment such as substations, distribution transformers, and protection systems. To commemorate the people’s participation in the national lockdown, India held a 10-min nationwide light-off ceremony. If the utility does not properly handle this, it might have a significant impact on the national grid. The COVID-19 pandemic has had a wide-ranging impact on India’s public transportation system. Initially, during the strict lockdown, all public transportation, including buses, metros, taxis, auto-rickshaws, and other modes, came to a standstill which experienced a 90% drop in ridership (UITP 2020). The data for India was used to forecast the reduction and increase in energy demand of Delhi’s households, non-residential, and industrial sectors. The assumptions used to project the reduction in the transport sector in Delhi owing to the pandemic were based on population density. The concentration of PM2.5 decreased by about 16% in India during the COVID-19 pandemic’s lockdown (Bherwani et al. 2021b; Srivastava et al. 2020). During the lockdown, the city’s air quality temporary improved due to reduced regional traffic and reduced commercial and industrial activity (Bherwani et al. 2021a). Industrial pollution is responsible for 18.6% of the poor air quality in the Delhi-National Capital Region (NCR), which is home to approximately 3000 industries and this industrial zone emits 200 to 1000 tons of CO2 per year (Mint 2019). The federal government’s ineffective carbon tax policy exacerbates the situation. Because the carbon tax only applies to coal, large corporations have switched to less expensive fossil fuel-based alternatives in order to avoid the tax (India Briefing 2019).
The COVID-19 pandemic had a temporary positive effect on the environment, like air pollution and energy demand decreased drastically. In Delhi, the construction sector is one of the major sources of air pollution and during the pandemic, construction activities were halted, resulting reduction in material consumption and air pollution (Dasgupta and Srikanth 2020). However, this study focused on energy demand, and thus only the sectors with the highest footprints with energy demands were considered. The purpose of this study is to examine the energy demand and emissions generated by sectors such as industry, transportation, and built-in-area for Delhi in the upcoming years by implementing and comparing scenarios such as CE and pandemic. The study also aims to see how much CE-strategies in the city will help mitigate climate change and what effect they have on energy demand, carbon and particulate matter emissions. The findings of the study will not only assist decision-makers in comprehending the scope of the impact that CE strategies can have on a city, but also serve as a reference point for other cities across the nation to carry out the similar studies to understand their potential areas of priority for climate change mitigation. In order to comprehend the impacts of the circular economy on climate change and the scenario generated due to the pandemic, detailed and precise data on energy production and consumption for all sectors in the concentrated area of Delhi NCT are needed. Data points are required to understand the data collection procedure and to establish methods for the generation of factors to generate ratios and proportions for various parameters.
Materials and methods
Study area
The current study focuses on Delhi, which is India’s second economic capital. Delhi lies between the latitude 28°24′01′′N to 28°53′00′′N and longitude 76°50′24′′E to 77°20′37′′ shown in Fig. 3. It has a long history and covers an area of 1483 km2 with a population of about 28 million in 2019 (Economic Survey of Delhi 2018–2019). The maximum length covered by NCT is 51.90 kms and the width is 48.48 kms. The Indo-Gangetic alluvial plains surround the megacity in the north and east, with the Thar desert in the west and also the Aravalli hills in the south. The landscape of Delhi is mostly flat, with the exception of a low NNE-SSW trending ridge that is thought to be an expansion of Rajasthan’s Aravalli hills (DCMSME 2013; Sahay 2018; Yadav et al. 2017). A vast majority of the working population of NCT Delhi, nearly 90%, resides in urban areas, which is significantly higher than the national average of 31.16% (Statistical Abstract of Delhi 2014).
Fig. 3 Study area showing the region of NCT, Delhi
Data collection
In this study, the data for activity level and energy intensity of technology was gathered from India’s energy statistics, as well as the Niti Aayog, which determines the sector-wise energy intensity. The majority of the data was acquired through secondary sources such as reports and research articles. The electricity consumption data for the technology used in households was gathered through secondary research and further calculated based on per capita and household size in the Delhi NCT (MOEF 2018). The energy statistics in 2019 provided information on the total energy consumption by machines, equipment, and HVAC for the non-residential sector. The data was then converted to per capita data by dividing the population of India and then the data was considered for the calculations by multiplying it by Delhi’s population (Energy Statistics 2019). The energy statistics for the residential and non-residential sectors are given below in Table 1.
Table 1 Energy statistics for residential and non-residential sector
Branch Sub branch Technology Energy intensity
Residential (per year per household) Lighting Lighting 40.5 kWh
Entertainment Radio 35.4 kWh
CD and DVD player 104 kWh
TV 228 kWh
Computer 94 kWh
Kitchen appliances Refrigerator 497 kWh
Washing machine 62 kWh
Electric oven 131 kWh
Toaster 131 kWh
Microwave 76 kWh
Heating and cooling Electric water heater 645 kWh
Fans 106 kWh
Air cooler 403 kWh
AC 1207.5 kWh
Non-residential (per year) Industry Machines and HVAC 10,113 Gwh
Agriculture Equipment’s 4406.8 Gwh
Commercial Machines and HVAC 2073.85 Gwh
Traction and railways Machines 309.67 Gwh
Public transportation data, such as share percentage, passenger travelled per kilometer, fuel consumption per passenger travelled and fuel efficiency according to vehicle types was gathered from the research articles and further calculation was done for the energy intensity using the collected data (Bose et al. 2001; Errampalli et al. 2020; LEAP 2020). The energy intensity for each vehicles type was calculated by multiplying the fuel efficiency and load factor which is the ratio of passengers travelled per kilometer and distance travelled by each vehicles types per year. Similarly, for the fright sector the energy intensity was calculated considering the data such as the number of vehicles, distance travelled and weight carried per year by each vehicles type (Bose et al. 2001). The energy statistics for the transport sector are given below in Table 2.
Table 2 Energy statistics for the transport sector
Branch Sub branch Technology Energy intensity
Public transport Two-wheeler Gasoline 0.7 MJ/pass-km
Cars Gasoline 0.6 MJ/pass-km
Rail Electric 450 KJ
Autorickshaws CNG 0.9 MJ/pass-km
Buses CNG 0.42 MJ/pass-km
Freight Light commercial trucks Diesel 2.1 MJ/pass-km
Two axle trucks Diesel 1.3 MJ/pass-km
Muli axle trucks Diesel 0.84 MJ/pass-km
Goods auto CNG 4.2 MJ/pass-km
Goods van CNG 4.5 MJ/pass-km
The annual production of the steel and paper industries, as well as their energy consumption statistics such as coal requirement, electricity usage, etc., were gathered from research articles with the total number steel and paper industries and their production in India, and then further computed for Delhi NCT. The total production for the steel industry was taken from the literature and the required input consumption for the production such as coking coal, non-coking coal and natural gas was estimated based on the percentage distribution given in the research article. Similarly for the paper industry, the input required such as process heat and electricity were estimated (Production of Steel 2019; Tripathi 2014; Rao 2012; ASSOCHAM 2020). The consumption of the input products for the steel and paper industry is given below in Table 3.
Table 3 Consumption of the input product for steel and paper industries
Branch Sub branch Technology Consumption
Steel industry Thermal energy Coking coal 0.63 ton
Non-coking coal 0.23 ton
Natural gas 280 cubic meter
Electricity Electric machinery 440 Kwh/ton
Paper industry Process heat Heat 27.3 GJ/ton
Electricity Electricity 1400 Kwh/ton
Delhi’s climate is difficult to plan for since it is heterogeneous in nature: it has a dry-hot summer with temperatures as high as 45 °C, and cold winters with temperatures as low as 3 °C (Lall et al. 1991). Electric water heaters and air conditioners are extensively utilized in Delhi households for heating and cooling; around 43.3% of Delhi households use air conditioners to relieve discomfort (Khosla et al. 2021). About 20–25% of Indian households use kitchen appliances, 30–35% use entertainment media, 30–40% use heating and 10–20% use cooling systems, all of these appliances consume the majority of energy in the home (MOEF 2018). The non-residential attribute includes commercial, traction and railways, agriculture, and industry, where industry relates to all of Delhi’s industries and commercial refers to offices, hospitals, schools, and shopping malls, etc. The electricity demand in Delhi is high, but the local generation capacity is about 9920 MU. As a result, the government purchases electricity from other states to meet the demand (Economic Survey of Delhi 2018–2019). The Delhi transportation sector comprised of motorbikes, cars, autorickshaws, taxis, buses, and freight has risen by 3% annually, with 10.3 million vehicles entering and leaving the city on normal working days (Bose et al. 2001). On a normal working day, around 100,000 freight vehicles arrive and exit Delhi city, with approximately 21% of these freight vehicles travelling through the city. In terms of passing through freight traffic, Two Axle Trucks (HT) account for about half of the total, with Multi-Axle Trucks (MT) and Light Commercial Trucks (LT) accounting for around 18% each. Smaller goods vehicles, such as Goods Auto (GA) and Goods Van (GV), account for around 14% of passing through traffic. Diesel is used by the LT, HT, and MT, while CNG is used by the GA and GV. LT uses 45% of its diesel in the outer cordons and 75% of its diesel in the city (Errampalli et al. 2020). Vehicle activity in the transportation sector is increasing at an annual pace of 8%, so energy demand is increasing as well (Energizing India 2018; Aggarwal and Jain 2016). In the present study, steel and paper industries are considered for the industry sector and their production rate has yearly growth of 8% and 7%, while the energy intensity of steel and paper industries has grown by 12% (Pal 2013; Coal India 2015). The data used for each scenario is given in the supplementary data.
Leap model
In the current research, The Low Emissions Analysis Platform (LEAP) software application (Heaps 2022) is used to analyze the emission reductions of economies and industries in the energy system assessment. The LEAP is an energy environment modelling tool based on probable futuristic scenarios which are developed based on a detailed literature review and current energy production and use practices and used by thousands of organizations in almost 190 countries, including government agencies, academia, nonprofit organizations, consulting firms, and energy companies (Mohan and Kandya 2007). The scenarios are based on a thorough assessment of how energy is produced, consumed, and transformed in a given region under various assumptions such as population, economic growth, and technological advancement (LEAP 2005). LEAP’s modelling skills are divided into two stages. In stage one, LEAP’s built-in calculations take care of all the “non-controversial” emissions, energy and cost-benefit computations. Users enter spreadsheet-like equations at the second stage, which can be used to provide time-varying data or develop a range of sophisticated multi-variable models, allowing econometric and modeling approaches to be integrated into LEAP’s entire accounting standard. The Technology and Environmental Database (TED) in LEAP describes the technical qualities, prices, and environmental implications of a variety of energy technologies, encompassing present technologies, best practices, and next-generation technologies. TED has information on hundreds of techniques, including reports from hundreds of organizations such as the Intergovernmental Panel on Climate Change (IPCC), the United States Department of Energy, and the International Energy Agency (IEA). A user’s data can be added to or updated in TED’s main database of emission factors. Country-wise emission factors and other data from TED may be automatically integrated into LEAP analysis, making it simple to generate emission scenarios using LEAP’s energy scenarios and TED’s emission factors. The year 2019 was chosen as the base year for analysis and annual progress was mapped based on various scenarios. The outcomes of energy demand and environmental constraints can be reviewed at the end of the simulation level by providing inputs for energy demand. The results of the projections have been created up to the year 2040, in accordance with the IPCC (IPCC 2019).
Fig. 4 Schematic way of modelling framework in LEAP
Figure 4 depicts the three primary sectors of the modelling framework: built-in area, transportation, and industry. Furthermore, each of these sectors is subdivided into subbranches, each of which represents its own set of energy services based on technology and its energy consumption (final energy intensity). The energy demand for its energy services could be calculated by multiplying the activity level and its energy intensity. Carbon Dioxide (CO2) and Particulate Matter 2.5 (PM2.5) are the emissions employed to examine the environmental influences for each scenario which could be calculated using the general Eq. (1).1 Eft=∑ft(ED×NCV×EF)
where ‘E’ stands for total emission, ‘EF’ stands for emission factors in g/kg of fuel burned, ‘ED’ stands for the energy demand for fuel type ‘f’ and technology ‘t’ and ‘NCV’ is the net calorific value for fuel type ‘f’.
The current account in LEAP software depicts the condition for the year 2019 in this study. Business as usual (BAU), CE (CE) and Pandemic were the three scenarios implemented and examined for the energy projections and their emissions. BAU is a scenario in which the current account’s energy demand is depicted based on assumptions without any policies being implemented. The CE is implemented in such a way that Delhi’s energy demand is reduced by applying the ReSOLVE framework. The hotspots were subsequently identified and the ReSOLVE framework was used to apply CE strategies to those hotspots as shown in Table 4. The CE scenario is based on the business as usual (BAU), which is the baseline for calculating energy demand for the year 2019. The environmental consequences of the city’s energy production and consumption were investigated using the Input-Output Table (IOT). The Delhi Directorate of Economics and Statistics has created an “Input-Output Table” (IOT) for the city’s economy. This study approximates energy demand and its environmental impact by utilizing the whole upstream network of supply chains and the percentage reduction in the energy demand after applying the ReSOLVE framework and was calculated by the Leontief inverse matrix (Christis et al. 2019; Kitzes 2013).
Table 4 Execution of CE strategies using the ReSOLVE Framework in the industry and transport sectors for Delhi, India
Sectors CE strategies References
Minimizing the energy intensity Changing the expenditure
Industry Scrap materials can be reused, improvements in efficiency and methods (optimize, loop & regenerate)
–10% of all monetary inputs to the industry sector
No excessive consumption, matching supply & demand, use of green materials, enforcement of emission standards (optimize)
–10% of all expenditure on the industry sector
NITI Aayog (2018, 2021a, b), Sharma et al. (2016)
Transport Use of Solar powered or electric-operated vehicles, move away from large SUVs (regenerate)
–5% of all monetary inputs to the transportation sector
Switch to biofuels, sharing of vehicles, increase the use of public transport instead of own vehicles (share, exchange)
–5% of all expenditure on the transportation sector
Sharma et al. (2016)
The CE scenario then shows how Delhi’s energy shift is moving towards a more sustainable mode due to the use of CE strategies. The temporary effects of the COVID-19 Pandemic on Delhi’s energy track are represented in the Pandemic scenario, which depicts how the pandemic effect will alter until 2040. The impact of the CE and the COVID-19 pandemic is investigated by varying the activity level and technology’s energy intensity while analyzing the scenario.
Results and discussion
The results of the various scenarios are compared and discussed in this section. The study’s baseline scenario, known as “business as usual,“ estimates energy demands and emissions through 2040 based on population growth and rising demand with usual mitigation policies. The results are based on energy demand, consumption and emission factors; nevertheless, meteorological parameters, population density, and local climatic zones will have an influence at a local level, however, the current evaluation is based only on macro factors. Without introducing any new energy policies, the business-as-usual (BAU) scenario forecasts the current scenario energy demand through 2040. Figure 5a and b shows Delhi’s current energy demand and emissions by residential, non-residential, transport and industry sectors. Delhi’s energy is distributed among many industries, with a major portion going to freight and diesel fuel. Figure 5a shows that the freight sector consumes the most energy, up to 60% of the overall energy consumption of 20.7 Mtoe followed by the thermal plant, public transportation, and non-residential sectors. Less than 10% of total energy consumption is comprised of households, non-residential, and industrial sectors. Consequently, 61% of CO2 emissions come from the freight industry, while 23% and 15% of CO2 emissions come from thermal plants and public transportation. A switch to renewable energy fuels like green hydrogen, biofuels, and electrification in the transportation sector, as well as the use of agricultural residues as an input fuel in thermal power plants, is necessary for a green energy transition because Delhi has the highest rate of CO2 emissions from the transportation sector, followed by thermal power plants.
Kumar and Madlener (2016) investigated the implications of using renewable energy in electrical supply networks and calculated CO2 emissions using a least-cost approach to develop several scenarios and analyzed using the LEAP Model, and their findings indicate that under an ARET (accelerated renewable energy technology) scenario, renewables produce 23% of electricity and CO2 emissions can be reduced to 74% by 2050. Su and Urban (2021) compared various sustainable energy alternative scenarios for Meili, China, from 2020 to 2040 using the LEAP model. Their research shows that applying CE strategies can reduce emissions of CO2 and PM2.5 by 11% and 17%, respectively, and final energy consumption by an extra 8%.
Fig. 5 Sectoral energy use and emissions distribution in Delhi for 2019
As a substitute tactic for lowering pollution and energy consumption, the implementation of a CE in Delhi city can play a critical role (Bherwani et al. 2022). With reference to the current analysis Fig. 6a shows the CE scenario that can reduce the final energy demand by about 51% by 2040 when compared to BAU. However, if the CE is assessed to fall short in front of the pandemic scenario. It may be possible to achieve a 69% of reduction in the final energy demand is indicated by COVID-19 scenario. Similar trends can be seen in Fig. 6b, where the CE and pandemic scenarios reduce CO2 emissions by 54% and 86%, respectively, relative to BAU by 2040. The problem of climate change, which is primarily caused by anthropogenic activities grows gradually as industrialization and population grows. It can be seen from the results that the CE strategies have the potential to limit the growing climate change problems by reducing carbon footprints through improved quality, efficiency, and working conditions, while also providing a profitable model for industries. It takes time, resources and effort to implement new manufacturing techniques and adjust the design of products to achieve a CE in industry and these changes can be costly initially but will help the industry and combat climate change in the long run. It should also be noted that increasing the usage of renewable energy and changing the energy structure to one that is more sustainable are two ways whereby CE initiatives can reduce the effect of global climate change.
Fig. 6 Comparisons of BAU, CE, and pandemic scenarios for a final energy demand, b CO2 emissions, and c PM2.5 emissions for the year 2040
In Delhi, pollution is a serious issue that has a variety of negative effects on human health due to the city’s crowded transportation system (Bherwani et al. 2020a, b). Due to the pandemic, Delhi was spared from pollution because transportation and industries were suspended in the lockdown phase. The anticipated outcomes show that, when the pandemic scenario is taken into consideration, 86% of PM2.5 emissions might be decreased by 2040 as compared to BAU. The pandemic state might not persist for long, and pollution levels might start to increase once again. As a result, CE strategies may eventually become a vital tool for reducing emissions. Figure 6 c, illustrates that the CE scenario might reduce PM2.5 emissions in 2040 by 26% compared to BAU.
A CE strategy in the transportation sector can lower PM2.5 emissions by 42% when compared to BAU in 2040 as shown in Fig. 6. Improved air quality in Delhi can be achieved by implementing CE strategies, such as promoting vehicle electrification in the transportation sector or switching to renewable fuels such as green hydrogen and biofuels, which government intends to introduce as an alternate fuel source in the coming years. India seeks to achieve the target of 20% biofuel blending by 2025 (NITI Aayog 2021a, b) and in the forthcoming years, the blending percentage could be exceeded. As per BIS research, the Indian electric vehicle market is expected to rise at compounded annual growth rate (CAGR) of 43.13% between 2019 and 2030 (Inc 42 2021), and it may continue to grow until 2040. It is important to consider that the transportation industry must switch to electric vehicles in order to reach the climate change targets. For efficient resource use, CE practices like recycling, repairing, and refurbishing are crucial (Richter 2022). Batteries can be reused and converted into energy storage devices to increase their lifespan. Making ensuring that 85% of a vehicle’s weight is recyclable or reused is another CE method to mitigate climate change for the transportation sector (Paradowska 2017). Reusing and recycling are key components of a CE, which is what is needed to ensure efficient resource usage and successful climate change mitigation. CE strategies should be implemented in all industries in order to combat climate change since they ensure optimum resource utilisation and reduce carbon emissions. (Yang et al. 2022).
Fig. 7 Evolution trend in a final energy demand, b CO2 emissions, and c PM2.5 emissions for BAU, CE and Pandemic scenarios
As shown in Fig. 7, the long-term effects of the pandemic and the CE scenario could reduce Delhi’s energy demand and associated emissions. Energy demand, CO2 emissions, and PM2.5 emissions are increasing in Delhi as compared to pre-pandemic conditions and without implementing the CE approach, as indicated in the BAU scenario in Fig. 6a, b and c. In comparison to the BAU scenario, the CE scenario might save 203 Mtoe (49%) in final energy demand from 2020 to 2040. Similarly, CO2 and PM2.5 emissions might be saved by 500 and 150 tons, respectively. The mandate lockdown nearly halted all activity during the COVID-19 pandemic, and as a result, human interventions were limited, resulting in considerable reductions in pollution levels and other activities (Bherwani et al. 2020a, 2020b). Due to all these factors, the results of this study show that 23% of more final energy may be saved in when pandemic scenario compared to the CE scenarios as can be significantly seen from Fig. 7a and CO2 emissions follow a similar pattern which is observed from the Fig. 7b. If this is considered as a best-case scenario from environment point of view, the details of reduction in activities, pollution levels, energy and resource consumption can be computed, and the government and policymakers may use this information to make decisions about enacting laws and regulations that will be beneficial in the coming years. PM2.5 emissions are primarily caused by transportation and industry due to the combustion of gasoline, diesel, and coal. Its overall emissions are influenced by both the degree of transportation activity and the fuel efficiency of the vehicles. As shown in Fig. 7c, the PM2.5 emissions can be reduced by applying CE strategies and considering the pandemic situation will be around 157 tons (24%) and 450 tons (70%) respectively. Figure 8 shows how much each sector might contribute to Delhi’s green energy transition when comparing the CE scenario with BAU scenarios. Public transport and freight play a vital role in saving final energy, CO2 emissions, and PM2.5 emissions. The current outcomes may vary due to errors and uncertainties in the data as the results are based on secondary sources and assumptions, hence the results could be interpreted in conjunction with the data dependency of the modelled outputs and likely errors and uncertainties in data.
Fig. 8 Sectorial contribution to promoting Delhi toward green energy transition through CE initiatives
The Supreme Court ordered the CPCB to adopt a comprehensive action plan, the CPCB came up with the Graded Response Action Plan (GRAP), in Delhi, which contains several limitations to cope with air pollution levels, particularly PM intensity (GRAP 2018). The Delhi government’s electric vehicle strategy, termed the Delhi Electric Vehicle (EV) policy, was launched in 2020 and might help to reduce the transportation sector’s emissions intensity (RMI India 2020). One more policy introduced by the government in 2019 is the National Clean Air Program (NCAP), which is a five-year strategic plan which works to reduce air pollution and raise citizen awareness (Gupta 2020). Even with different policies and schemes in place, governments and policymakers continue to face obstacles in reducing emissions. As a result, specific strategies and solutions must be implemented that can directly strike at the root of the problem and alleviate energy demand and emissions.
The population and household size used in this study were gathered from secondary sources and are subject to vary due to annual growth projections. Also, because some data for the specific study area was not available, the electricity consumption by different appliances in households was obtained from the study India: Strategies for Low Carbon Growth (MOEF 2018). As a result, the amount of power consumed by a typical household may vary based on the number of appliances used and how long they were utilised, but the overall numbers may remain as stated in the literature. Based on the unique characteristics of each industry and the seasonal variations in their resource consumption patterns, the fuel usage, production and energy consumption data in the transportation and industrial sector may differ. Poor road conditions, the number of vehicles operated per day, and changes in the passenger and load concentration in vehicles are a few examples of circumstances that could cause spot variances, but overall data may still be valid based on the rationale followed in the research. Since all of the data was acquired from diverse sources, internal variations are possible and will lead to distinct outcomes. However, the most probable sources were considered for this study.
Conclusion
Energy demand and pollution have become important concerns in Delhi as a result of the city’s rising population, transportation, and industries. Only around 20% of the industries are authorized in the area, therefore industrial development has been erratic and unorganized. The remaining industries are scattered across the city’s residential and commercial areas. Road transport is the major mode of public transportation, and the vehicle population has grown exponentially. This study examines the sustainable energy transition alternatives for Delhi for a defined number of years, from 2020 to 2040. The scenarios were created considering the built-in area, transportation, and industry sectors, as well as changes in their consumption patterns throughout time. The findings in this study reveal that the CE strategies, which include the electrification of vehicles and the use of renewable fuels in the transportation sector, as well as the reuse of scrap material, the use of green materials, the enforcement of strict emission standards, and the restriction of excessive consumption in the industry sector, has a lot of potential for moving Delhi towards a green energy transition and lowering emission rates. In comparison to BAU, the improved savings in final energy consumption, CO2 and PM2.5 emissions could be 49%, 49% and 24% respectively by implementing a CE scenario for the years 2020 to 2040. The pandemic scenarios considered as the best-case scenario because it shows the temporary halt in activities and reduction in terms of environment pollution loads. The improved savings in final energy consumption, CO2 and PM2.5 emissions could be 72%, 72%, and 70%, respectively by implementing a pandemic scenario for the years 2020 to 2040. In terms of energy, CO2, and PM2.5 emissions, Delhi city might save 45%, 46%, and 61% by treating the pandemic as a baseline for the year 2040 as compared to the CE scenario. The COVID-19 pandemic casts doubt on the transition to a sustainable energy system since it was a forced situation that occurred globally and would not last for long. As a result, the CE scenario stands out as the best fit and has a significant potential to accelerate the clean energy transition by implementing CE strategies at several levels. The scarcity of data led to numerous obstacles like validations and uncertainty analysis of the data cannot be done because of the non-availability of actual data and also there can be variations in the results. Hence the future scope of the work may include validation and uncertainty analysis based on actual data and its variations. If CE strategies are to be implemented and further research is to be conducted, the government and authorities should make cities and state-wise data available for all sectors. However, this study might be expanded to include a wider range of sectors, with the results assisting policymakers and the government in implementing new policies to move toward renewable energy and green fuels in the coming years.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file 1 (PDF 239.3 kb)
Acknowledgements
Authors thankfully acknowledge Council of Scientific & Industrial Research (CSIR), India, and its constituent laboratory National Environmental Engineering Research Institute (NEERI) for supporting the research. The authors also thank Stockholm Environment Institute for providing access and student research license to LEAP tool for research purposes. The manuscript is checked for plagiarism using licensed version of iThenticate software.
Declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Abad-Segura E González-Zamar MD Belmonte-Ureña LJ Effects of circular economy policies on the environment and sustainable growth: worldwide research Sustainability 2020 12 14 5792 10.3390/su12145792
FICCI.: Accelerating India’s Circular economy shift. https://ficci.in/spdocument/22977/FICCI-Circular-Economy.pdf (2018). Accessed 28 July 2021
Aggarwal P Jain S Energy demand and CO2 emissions from urban on-road transport in Delhi: current and future projections under various policy measures J. Clean Prod. 2016 128 48 61 10.1016/j.jclepro.2014.12.012
Akbari H Pomerantz M Taha H Cool surfaces and shade trees to reduce energy use and improve air quality in urban areas Solar Energy 2001 70 3 295 310 10.1016/S0038-092X(00)00089-X
ASSOCHAM Indian paper industry likely to touch 25 MT by 2020: ASSOCHAM, 2020. www.business-standard.com/article/news-cm/indian-paper-industry-likely-to-touch-25-mt-by-2020-assocham-118022800150_1.html (2020). Accessed 20 July 2021
Bherwani H Gautam S Gupta A Qualitative and quantitative analyses of impact of COVID-19 on sustainable development goals (SDGs) in indian subcontinent with a focus on air quality Int. J. Environ. Sci. Tech. 2021 18 4 1019 1028 10.1007/s13762-020-03122-z
Bherwani H Gupta A Anjum S Anshul A Kumar R Exploring dependence of COVID-19 on environmental factors and spread prediction in India NPJ Climate Atmos. Sci. 2020 3 1 1 13 10.1038/s41612-020-00142-x
Bherwani H Kumar S Musugu K Nair M Gautam S Gupta A Ho CH Anshul A Kumar R Assessment and valuation of health impacts of fine particulate matter during COVID-19 lockdown: a comprehensive study of tropical and sub tropical countries Environ. Sci. Pollut. Res. Int. 2021 10.1007/s11356-021-13813-w
Bherwani H Nair M Musugu K Gautam S Gupta A Kapley A Kumar R Valuation of air pollution externalities: comparative assessment of economic damage and emission reduction under COVID-19 lockdown Air Qual. Atmos. Health 2020 13 6 683 694 10.1007/s11869-020-00845-3 32837611
Bherwani H Nair M Niwalkar A Balachandran D Kumar R Application of circular economy framework for reducing the impacts of climate change: a case study from india on the application of carbon and materials Footprint Nexus Energy Nexus 2022 10.1016/j.nexus.2022.100047
Bocquier P World urbanization prospects: an alternative to the UN model of projection compatible with the mobility transition theory Demogr. Res. 2005 12 197 236 10.4054/demres.2005.12.9
Bose RK Energy demand and environmental implications in urban transport—case of Delhi Atmos. Environ. 1996 30 3 403 412 10.1016/1352-2310(95)00111-5
Bose, R., Sperling, D., Delucchi, M.A., Nesamani, K.S., Redmond, L.S., Schipper, L., Tiwari, G., Transportation in developing countries:greenhouse gas scenarios for Delhi, India. (2001)
Centre for Policy Research (CPR): https://www.cprindia.org/news/6569 (2017). Accessed 24 July 2021
Chatterji, A. Air Pollution in Delhi: Filling the Policy Gaps. Massachusetts Undergraduate J. Eco., p 17. (2021)
Christis M Athanassiadis A Vercalsteren A Implementation at a city level of circular economy strategies and climate change mitigation—the case of Brussels J. Clean. Prod. 2019 218 511 520 10.1016/j.jclepro.2019.01.180
Coal requirement for power generation only. https://www.coalindia.in/media/documents/Normative_Coal_requirement_for_different_Industries_240220161.pdf (2015). Accessed 22 July 2021
DCMSME, Brief industrial Profile of NCT of Delhi.http://dcmsme.gov.in/old/dips/Brief%20industrial%20profile%20of%20Delhi.pdf (2013). Accessed 24 May 2022
Dasgupta P Srikanth K Reduced air pollution during COVID-19: learnings for sustainability from Indian cities Global Trans. 2020 2 271 282 10.1016/j.glt.2020.10.002
Delhi, E.V. policy’s twin boost: Faster EV adoption and improved air quality, RMI India. https://rmi-india.org/delhi-ev-policys-twin-boost-faster-ev-adoption-and-improved-air-quality/ (2020). Accessed 02 August 2021
Dholakia HH Purohit P Rao S Garg A Impact of current policies on future air quality and health outcomes in Delhi, India Atmos. Environ. 2013 75 241 248 10.1016/j.atmosenv.2013.04.052
Economic Survey of Delhi:, -19. http://delhiplanning.nic.in/sites/default/files/Final%20Economy%20survey%20English.pdf (2018). Accessed 30 July 2021
Elavarasan RM Shafiullah GM Raju K Mudgal V Arif MT Jamal T Subramanian S Balaguru VS Reddy KS Subramaniam U COVID-19: impact analysis and recommendations for power sector operation Appl. Energy 2020 279 115739 10.1016/j.apenergy.2020.115739 32904736
Ellen MacArthur, F., McKinsey center for business and environment, SUN. Growth within: a circular economy vision for a competitive Europe. (2015)
Ellen MacArthur, F.: 2016 circular economy in India: rethinking growth for long-term prosperity. https://ellenmacarthurfoundation.org/circular-economy-in-india (2021). Accessed 14 July 2021
Energy, S. http://mospi.nic.in/sites/default/files/publication_reports/Energy%20Statistics%202019-finall.pdf (2019). Accessed 15 July 2021
Energy Policy, R.: https://www.iea.org/reports/india-2020 (2020). Accessed 20 July 2021
Errampalli M Kayitha R Chalumuri RS Tavasszy LA Borst J Chandra S Assessment of urban freight travel characteristics—a case study of Delhi Transp. Res. Proc. 2020 48 467 485
Farrow A Miller KA Myllyvirta L Toxic air: the price of fossil fuels 2020 Seoul Greenpeace Southeast Asia 44
Fiksel J Sanjay P Raman K Steps toward a resilient circular economy in India Clean Technol. Environ. Policy 2021 23 1 203 218 10.1007/s10098-020-01982-0 33199980
Graded Response Action Plan (GRAP): The Solution to Delhi’s Air pollution (GRAP): https://www.aqi.in/blog/graded-response-action-plan-grap-the-solution-to-delhis-air-pollution/ (2018). Accessed 21 July 2021
Greenpeace https://www.greenpeace.org/india/en/press/4683/air-pollution-from-fossil-fuels-costs-india-us150-billion-annually-reveals-a-newreport/#:~:text=According%20to%20the%20report%2C%20India,US%24600%20billion%20and%20India (2020). Accessed 23 May 2022
Gupta A Bherwani H Gautam S Anjum S Musugu K Kumar N Anshul A Kumar R Air pollution aggravating COVID-19 lethality? Exploration in asian cities using statistical models Environ. Dev. Sustain. 2021 23 4 6408 6417 10.1007/s10668-020-00878-9 32837279
Gupta, A.: National clean air programme of India. Challenge and Strategies. (2020)
Heal MR Kumar P Harrison RM Particles, air quality, policy and health Chem. Soc. Rev. 2012 41 19 6606 6630 10.1039/c2cs35076a 22660420
Heaps, C.G., LEAP: the low emissions analysis platform. [Software version: 2020.1.80] Stockholm Environment Institute. Somerville, MA, USA. (2022). https://leap.sei.org
Inc 42, Will India Become the Largest EV Market in The World By 2030?: https://inc42.com/features/will-india-become-the-largest-ev-market-in-the-world-by-2030/ (2021). Accessed 03 August 2021
India briefing, Delhi’s Air Pollution and its Impact on Business: https://www.india-briefing.com/news/delhis-pollution-crisis-impact-business-15631.html/ (2019). Accessed 19 July 2021
IPCC Global warming of 1.5°C, An IPCC Special Report on the impacts of global warming of 1.5°C https://www.ipcc.ch/site/assets/uploads/sites/2/2019/06/SR15_Full_Report_High_Res.pdf (2019). Accessed 03 July 2021
Khosla R Agarwal A Sircar N Chatterjee D The what, why, and how of changing cooling energy consumption in India’s urban households Environ. Res. Lett. 2021 16 4 044035 10.1088/1748-9326/abecbc
Kitzes J An introduction to environmentally extended input-output analysis Resources 2013 2 4 489 503 10.3390/resources2040489
Kumar P Gulia S Harrison RM Khare M The influence of odd–even car trial on fine and coarse particles in Delhi Environ. Pollut. 2017 225 20 30 10.1016/j.envpol.2017.03.017 28343101
Kumar S Madlener R CO2 emission reduction potential assessment using renewable energy in India Energy 2016 97 273 282 10.1016/j.energy.2015.12.131
Kumari P Garg V Kumar R Kumar K Impact of urban heat island formation on energy consumption in Delhi Urban Climate 2021 36 100763 10.1016/j.uclim.2020.100763
Lall AB Pandit M Kulshreshta N Appasamy P Climate and housing form—a case study of New Delhi Energy Build 1991 16 3–4 837 849 10.1016/0378-7788(91)90079-I
Long, A., Ascent, D.: World Economic Outlook. International Monetary Fund (2020)
Low Emissions Analysis Platform Training Exercises (LEAP): https://leap.sei.org/documents/LEAPTrainingExerciseEnglish2020.pdf (2020). Accessed 20 July 2021
MOEF Residential Consumption of Electricity in India Documentation of Data and Methodology, MOEF, 2018 http://moef.gov.in/wp-content/uploads/2018/04/Residentialpowerconsumption.pdf (2018). Accessed 30 July 2021
Mint Why is Delhi’s air so toxic? https://www.livemint.com/news/india/why-is-delhi-s-air-so-toxic-11572782475997.html (2019). Accessed 15 July 2021
Mohan M Kandya A An analysis of the annual and seasonal trends of air quality index of Delhi Environ. Monit. Assess 2007 131 1 267 277 10.1007/s10661-006-9474-4 17242967
MOHUA, Ministry of Housing and Urban Affairs, Annual report 2019–2020 https://mohua.gov.in/upload/uploadfiles/files/Final%20Annual%20Report%20compressed-1-80.pdf (2019–2020). Accessed 23 July 2021
NITI Aayog Energizing India In A Joint Project Report of NITI Aayog and IEEJ, Government of India. https://smartnet.niua.org/content/570dbbef-ed57-44ea-9bbc-a7221095b126 (2018). Accessed 14 July 2021
NITI Aayog Ministry of Petroleum and Natural Gas, Roadmap for ethanol blending in India 2020-25, Report of the Expert Committee. https://www.niti.gov.in/sites/default/files/2021-06/EthanolBlendingInIndia_compressed.pdf (2021a). Accessed 03 August 2021
NITI Aayog, Strategy Paper on Resource Efficiency in Steel Sector Through Recycling of Scrap & Slag, https://www.niti.gov.in/writereaddata/files/RE_Steel_Scrap_Slag-FinalR4-28092018.pdf (2021b). Accessed 13 July 2021
Pal S Study on performance and prospect of indian steel industry from national perspective under globalization Int. J. Eco. Com. Res. (IJECR) 2013 3 3 2250 0006
Paradowska M May Grounds and challenges for implementing a circular economy in the european road transport sector. In TranSopot Conference 2017 Cham Springer 245 269
Press Information Bureau (PIB), Electrification of Villages: https://pib.gov.in/PressReleseDetailm.aspx?PRID=1592833 (2019). Accessed 27 July 2021
Production of Steel https://pib.gov.in/PressReleasePage.aspx?PRID=1576418 (2019). Accessed 22 July 2021
Le Quéré C Jackson RB Jones MW Smith AJ Abernethy S Andrew RM De-Gol AJ Willis DR Shan Y Canadell JG Friedlingstein P Temporary reduction in daily global CO2 emissions during the COVID-19 forced confinement Nat. Climate Change 2020 10 7 647 653 10.1038/s41558-020-0797-x
Ramesh S Natarajan B Bhagat G Peak load prediction using weather variables Energy 1988 13 8 671 679 10.1016/0360-5442(88)90097-7
Rao AL An industry analysis with special reference to Indian paper industry Int. J. Bus. Management Res. 2012 2 67 84
Richter JL A circular economy approach is needed for electric vehicles Nat. Electron 2022 5 1 5 7 10.1038/s41928-021-00711-9
Sahay S Urban adaptation to climate sensitive health effect: evaluation of coping strategies for dengue in Delhi, India Sustain. Citi. Soc. 2018 37 178 188 10.1016/j.scs.2017.11.017
Sharma SK Mandal TK Jain S Sharma A Saxena M Source apportionment of PM 2.5 in Delhi, India using PMF model Bull. Environ. Contam. Toxicol. 2016 97 2 286 293 10.1007/s00128-016-1836-1 27209541
Srivastava S Kumar A Bauddh K Gautam AS Kumar S 21-Day lockdown in India dramatically reduced air pollution indices in Lucknow and New Delhi, India Bull. Environ. Contam. Toxicol. 2020 105 9 17 10.1007/s00128-020-02895-w 32495123
State of Global Air: https://www.stateofglobalair.org/sites/default/files/soga-2018-report.pdf (2018). Accessed 28 July 2021
Statistical Abstract of Delhi: Directorate of Economics & Statistics. Government of NCT of Delhi, New Delhi. http://14.139.60.153/bitstream/123456789/8627/1/Statistical%20Abstract%20of%20Delhi%202014.pdf (2014). Accessed 30 July 2021
Su C Urban F Circular economy for clean energy transitions: a new opportunity under the COVID-19 pandemic Appl. Energy 2021 289 116666 10.1016/j.apenergy.2021.116666
TRIPP: Travel to Work in India, current patterns and future concerns. http://tripp.iitd.ac.in/assets/publication/WorkTravelReport.pdf (2018). Accessed 25 May 2022
The Economics times: https://economictimes.indiatimes.com/news/politics-and-nation/india-bans-use-of-dirtier-coal-alternative-in-new-delhi/articleshow/61205059.cms?from=mdr (2017). Accessed 18 July 2021
Tripathi JG Study of India’s paper industry-potential and growth in 21st century Indian J. Appl. res 2014 4 112 115 10.15373/2249555X/MAY2014/36
UITP: Covid-19 has hit Indian public transport from many directions, UITP, 2020 (2020). https://www.uitp.org/news/covid-19-has-hit-indian-public-transport-from-many-directions/ Accessed 28 July 2021
User Guide for LEAP: https://unfccc.int/resource/cd_roms/na1/mitigation/Module_5/Module_5_1/b_tools/LEAP/Manuals/Leap_Use_Guide_English.pdf (2005) Accessed 30 July 2021
WHO’s Urban Ambient Air Pollution database: http://www.who.int/airpollution/data/AAP_database_summary_results_2016_v02.pdf (2016). Accessed 27 July 2021
World cities report: The value of sustainable urbanization, https://unhabitat.org/sites/default/files/2020/10/wcr_2020_report.pdf (2020). Accessed 20 July 2021
Yadav N Sharma C Peshin SK Masiwal R Study of intra-city urban heat island intensity and its influence on atmospheric chemistry and energy consumption in Delhi Sustain. Citi. Soc. 2017 32 202 211 10.1016/j.scs.2017.04.003
Yang, M., Chen, L., Wang, J., Msigwa, G., Osman, A.I., Fawzy, S., Rooney, D.W., Yap, P.S.: Circular economy strategies for combating climate change and other environmental issues. Environ. Chem. Lett. 1–26 (2022). 10.1007/s10311-022-01499-6
| 0 | PMC9707248 | NO-CC CODE | 2022-12-01 23:20:22 | no | Proc.Indian Natl. Sci. Acad.. 2022 Nov 28;:1-14 | utf-8 | null | null | null | oa_other |
==== Front
Stud Russ Econ Dev
Stud Russ Econ Dev
Studies on Russian Economic Development
1075-7007
1531-8664
Pleiades Publishing Moscow
7350
10.1134/S1075700722060089
Economy and Enterprises
Russian Enterprises in the Spring of 2022: Adapting to the New Wave of Sanctions and Views on the ESG Agenda
Kuvalin D. B.
Zinchenko Yu. V. [email protected]
Lavrinenko P. A.
Ibragimov Sh. Sh.
grid.473260.4 0000 0004 0562 8681 Institute of Economic Forecasting, Russian Academy of Sciences, Moscow, Russia
28 11 2022
2022
33 6 697706
20 6 2022
28 6 2022
30 6 2022
© Pleiades Publishing, Ltd. 2022, ISSN 1075-7007, Studies on Russian Economic Development, 2022, Vol. 33, No. 6, pp. 697–706. © Pleiades Publishing, Ltd., 2022.Russian Text © The Author(s), 2022.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Abstract—
The article analyzes the results obtained by a regular survey of Russian enterprises in the real sector conducted by the Institute of Economic Forecasting, Russian Academy of Sciences. The consequences of economic sanctions against Russia for enterprises are assessed. The views of enterprises on the desirable measures of anticrisis policy aimed at countering sanctions are outlined. Methods for adaptation to sanctions that are applied by the enterprises themselves are described. The opinion expressed by enterprises about the role of the state in the economy is presented. The situation with the investment activity of enterprises is considered. Data on the purchases of Russian and foreign equipment are given, as well as comparative assessments of the quality of this equipment. The views of Russian enterprises on the relevance of the climate agenda are presented, as well as their attitude to the ESG principles. The share of Russian enterprises that publish financial statements on risks associated with climate change as well as nonfinancial statements in the field of sustainable development is estimated. The proportion of enterprises that set themselves the goal of reducing their carbon footprint is revealed.
Keywords:
business survey
economic sanctions against Russia
the role of the state in the economy
investment activity of enterprises
climate policy
ESG
carbon regulation
issue-copyright-statement© Pleiades Publishing, Ltd. 2022
==== Body
pmcA sharp change in the geopolitical situation in February 2022 led to a new wave of sanctions against Russia. The scale of the sanctions was unprecedented and affected almost all sectors of the Russian economy. On the one hand, Western countries completely banned or significantly restricted the import of various types of machinery and equipment as well as raw materials and components into Russia. On the other hand, many export deliveries from Russia to Western countries were completely or partially blocked. In addition, a wide range of financial sanctions were introduced, from the freezing of half of Russia’s gold and foreign exchange reserves, which is extremely controversial from the viewpoint of international law, to the disconnection of leading Russian banks from the SWIFT payment system. Secondary sanctions have also been stepped up, such as creating problems for banks and third-country companies wishing to work with Russian counterparties.
At the beginning of the geopolitical crisis, many observers assumed that due to Western sanctions, the fall in Russia’s GDP by the end of 2022 would be from 9 to 22% [1–3]. The real decline was not so dramatic but still quite substantial. As early as in April 2022, a negative trend was observed for a number of macroeconomic indicators. Thus, for example, according to the Institute of Economic Forecasting RAS (IEF RAS), Russia’s GDP growth in April 2022 was –1.7% compared to April 2021 and –4.5% compared to March 2021 (seasonally adjusted) [4]. At the same time, the Russian authorities managed to quickly stabilize the ruble exchange rate, returning it to precrisis levels, to stop the inflationary surge, and to avoid panic in the banking system and the consumer market.
A survey of Russian enterprises conducted by IEF RAS in April–May 20221 also confirmed the duality of the situation. On the one hand, sanctions have created serious economic problems. In particular, 2–3 months after the imposition of sanctions, the proportion of respondents affected by them was 59.20% (Table 1). On the other hand, this is less than 2–3 months after the start of the 2020 crisis associated with the COVID-19 pandemic, when the proportion of affected respondents was 73.60% [5]. In other words, the crisis associated with sanctions is still less overwhelming.
Table 1. Answers to the question: “Did your enterprise suffer from sanctions related to the situation in Ukraine?” (sum of answers = 100%)
Period Yes No No, but it might get hurt
in future
April–May 2022 59.20 18.50 22.30
At the same time, the majority of surveyed enterprises expected certain problems related to sanctions. Thirty-eight percent of respondents believed that the consequences of sanctions would be exclusively negative; another 32.60% believed that the consequences would be both negative and positive. At the same time, a small proportion of enterprises were more optimistic as 5.40% of respondents assumed that the sanctions would not have any special consequences for their activities, and another 2.30% hoped only for positive consequences (Table 2).
Table 2. Answers to the question: “What consequences do you expect for your enterprise as a result of the introduction of economic sanctions against Russia?” (sum of answers = 100%)
Period No special consequences are expected Negative Both negative
and positive Positive So far it’s hard
to tell
April–May 2022 5.40 38.00 32.60 2.30 21.70
Among the most acute problems caused by sanctions, 67.40% of enterprises most often named difficulties in obtaining imported raw materials and components; 62.00% named a rise in prices within the country; 44.20%, increased overall uncertainty in the economy; 40.30%, a rise in the cost of imports; and 38.80% cited a fall in effective demand within Russia (Table 3). It should also be noted that Russian enterprises were almost unconcerned about the difficulties in attracting foreign investors. Only 1.60% of respondents mentioned this problem. Apparently, this state of affairs means that the dependence of the Russian economy on foreign financing has become minimal. In the current situation, Russian enterprises are interested in foreign investors as in a source of high technology and advanced equipment but not as a source of money.
Table 3. Answers to the question: “What problems related to economic sanctions against Russia are the most acute for your company?” (sum of answers > 100%)
Period Difficulties in obtaining imported
raw materials and components,
the need to replace them A serious rise
in the cost of imports Reduced opportunities
for exporting our products Sagging solvent demand
within Russia Rising prices
within the country Rise in the cost of credit Difficulties in attracting
foreign investors Restrictions on technology
imports and a slowdown
in technological modernization Increasing overall
economic uncertainty Other
April–May 2022 67.4 40.3 17.1 38.8 62.0 32.6 1.6 22.5 44.2 3.1
Adapting to the crisis phenomena that were generated by the sanctions, enterprises most often launched a search for alternative suppliers that could replace the departing Western partners within Russia itself; this was 69.70% of the responses. At the same time, 36.10% of respondents began to look for new suppliers outside of Russia. Unfortunately, as in the course of previous crises, a significant part of the surveyed enterprises—36.90%—reduced their investment costs for the sake of anticrisis adaptation. Such a massive reduction in investment is always unfavorable from the viewpoint of long-term development. At the same time, 14.80% of enterprises reported that, on the contrary, they launched the processes of production modernization as an anticrisis measure. In addition, it should be noted that enterprises took other measures of active adaptation. In particular, 31.10% of respondents reported on the search for new sales markets, and 21.30% reported on the launch of new types of products (Table 4).
Table 4. Answers to the question: “What measures is your company taking to counter the ‘sanctions-related’ crisis?” (sum of answers > 100%)
Period The search for new suppliers
abroad was started The search for new suppliers
in Russia was started Investment costs
have been reduced Staff costs were reduced (layoffs,
cuts in wages, social package, etc.) The output of some products
was discontinued The output of new products
was launched The search for new markets
was started Actions were initiated in order
to restructure production
for the future (repairs,
modernization, etc.) Other
April–May 2022 36.10 69.70 36.90 11.50 14.80 21.30 31.10 14.80 3.30
The speed and effectiveness of enterprises’ adaptation to the problems caused by sanctions largely depends on the content of the state’s anticrisis policy. From the viewpoint of Russian enterprises, in the situation that developed in the spring of 2022, the most important actions of the authorities were as follows: 55.80% of responses cited limiting the growth in prices for fuel, energy, and transportation services; 52.70%, the reduction of the tax burden on manufacturers; 42.60%, financial support for manufacturers who have launched the production of import-substituting products; 39.50%, supporting demand in the economy through the growth of public procurement, payments of subsidies to the population, etc.; and 38.80%, a sharp expansion in the sphere of concessional lending (Table 5). In other words, according to enterprises, the anticrisis policy of the state should be both large-scale and very diverse.
Table 5. Answers to the question: “What measures to counter economic sanctions, in your opinion, should the Russian authorities take?” (sum of answers > 100%)
Period Reduce the tax burden
on manufacturers Reduce the tax burden
of the population Support demand in the economy
(due to the growth of public
procurement, payment of subsidies
to the population, etc.) Limit the rise in prices for fuel,
energy, transport services Reduce bureaucratic burden
(checks, reports, certificates, etc.) Strengthen the fight against
corruption Significantly expand the scope
of concessional lending Launch large infrastructure
and industrial projects
at the expense of the state budget Provide financial support for enterprises
that have launched the production
of import-substituting products Create alternative payment
systems to service import
and export transactions Other
April–May 2022 52.70 18.60 39.50 55.80 45.70 23.30 38.80 35.70 42.60 25.60 4.70
As is often the case during crises, there have been certain shifts in the views of enterprises on the role of the state in the economy [6]. In this regard, it should be noted that the current crisis boosted the popularity of two extreme viewpoints. For example, the share of those who support a complete rejection of state intervention in the economy increased from 9.20% in 2020 to 14.60% in 2022, reaching the highest level for the entire period of the surveys. At the same time, the share of those advocating stronger direct state participation in the economy also increased – from 9.20% in 2020 to 16.30% in 2022. However, the number of those supporting a more active role played by the state in the economy among Russian enterprises is still significantly higher than that of those who favor weakening of the state role. While the total proportion of respondents who, to one degree or another, were in favor of reducing the role of the state in the economy was 30.90% in the spring of 2022, the total share of supporters of increasing this role was 61.80% (Table 6). However, at the same time, most enterprises considered the state to be a less effective owner than private owners (Table 7). In this regard, it can be assumed that in the course of emerging from the crises, the share of proponents of active state participation among the Russian enterprises will gradually decrease.
Table 6. Answers to the question: “What role should the state play in the Russian economy in the next few years?” (sum of answers = 100%)
Period Completely refrain from interfering
in economic life
and only monitor observance of laws by all economic agents Maintain
a certain impact
on economic life,
but compared
with today its role should decrease Current degree
of state participation is close to optimal, so there is no need for any special change Step up the domestic economic policy by expanding the range of instruments and applying mainly indirect methods of economic regulation Step up the state’s direct participation in economic life and actively intervene
in economic practice
February–March 2005 6.94 19.08 2.89 57.22 13.87
February–March 2007 2.52 20.13 5.66 61.00 10.69
February–March 2010 6.02 14.46 6.63 53.01 19.88
April–May 2012 6.08 22.30 8.11 47.30 16.21
April–May 2014 6.25 14.38 7.50 48.75 23.12
April–May 2016 9.09 13.64 6.49 47.40 23.38
April–May 2018 6.40 19.20 8.80 46.40 19.20
April–May 2020 9.20 22.30 11.50 47.70 9.20
April–May 2022 14.60 16.30 7.30 45.50 16.30
Table 7. Answers to the question: “Could you please formulate your attitude towards the state as the owner of enterprises?” (sum of answers = 100%)
Period In Russia, under any circumstances, the state is undoubtedly less effective owner than private owners State is at present
a less efficient owner
than private owners, but
it due to low qualification of Russian officials and insufficient control over their activities The state is an ordinary owner neither better nor worse than the rest In modern conditions,
the state is a more efficient
owner than private owners;
at least the state more focused on solving problems
of enterprises and production growth
February–March 2005 18.54 53.94 16.85 10.67
February–March 2007 22.01 54.72 15.09 8.17
February–March 2010 18.18 45.45 20.00 16.37
April–May 2012 20.55 54.11 10.27 15.07
April–May 2014 13.84 48.42 20.13 17.61
April–May 2016 18.59 42.95 17.95 20.51
April–May 2018 13.49 36.51 26.98 23.02
April–May 2020 15.80 47.40 23.30 13.50
April–May 2022 30.60 27.30 20.60 21.50
As expected, the investment activity of Russian enterprises in the context of the sanctions crisis has generally declined. While in 2020, 59.40% of the surveyed enterprises were engaged in the implementation of investment projects at the time of the survey, and in 2021 it was 52.94%, in 2022 there were only 50.40% of those (Table 8). However, while in the case of current investment projects, the decline in activity turned out to be relatively small, in relation to investment plans for the future, one can speak of a dramatic failure.
Table 8. Answers to the question: “Does your enterprise currently implement any industrial investment projects?” (sum of answers = 100%)
Period Yes No
April–May 2012 54.11 45.89
April–May 2013 62.58 37.42
April–May 2014 58.75 41.25
April–May 2015 48.39 51.61
April–May 2016 44.87 55.13
April–May 2017 54.05 45.95
April–May 2018 49.57 50.43
April–May 2019 53.74 46.26
April–May 2020 59.40 40.60
April–May 2021 52.94 47.46
April–May 2022 50.40 49.60
While in 2020, 42.50% of respondents reported their intention to launch new investment projects within 1–2 years, and in 2021 it was 46.71%, in 2022 this share fell to 26.60% (Table 9). This is the lowest level of investment optimism since 2000. Apparently, this state of affairs is associated with a very high degree of uncertainty about the economic future. In this regard, it is no coincidence that the share of answers “difficult to say”—50.70%—also turned out to be the highest for the entire period of the surveys.
Table 9. Answers to the question: “Will your enterprise start implementing industrial investment projects in the next 1–2 years?” (sum of answers = 100%)
Period Yes No It’s hard to say
April–May 2012 47.59 18.62 33.79
April–May 2013 56.49 11.69 31.82
April–May 2014 52.20 14.47 33.33
April–May 2015 33.99 15.69 50.32
April–May 2016 35.06 16.88 45.06
April–May 2017 41.10 13.70 45.20
April–May 2018 46.96 19.13 33.91
April–May 2019 43.15 19.18 37.67
April–May 2020 42.50 14.90 42.50
April–May 2021 46.71 13.82 39.47
April–May 2022 26.60 22.70 50.70
Investment problems of Russian enterprises in the context of sanctions are largely associated with a high dependence on the purchase of imported equipment [7–10]. In 2022, new machinery and equipment manufactured abroad were purchased more often than Russian machinery and equipment (Table 10). At the same time, the vast majority of respondents note that there is either no Russian equipment that would not be inferior in quality to imported counterparts at all or it is very scarce (Table 11).
Table 10. Answers to the question: “What machines and equipment has your company purchased over the past 2–3 years?” (sum of answers > 100%)
Period Russian-made,
new Russian-made,
used Produced
in the CIS countries, new Produced
in the CIS countries, used Produced
in foreign countries, new Produced
in foreign countries, used
October–December 2011 64.19 17.57 10.81 1.35 68.92 11.49
April–May 2013 47.97 13.51 11.49 0.68 70.27 11.49
November–December 2014 56.72 17.16 13.43 2.99 64.18 10.45
April–May 2016 50.00 14.08 11.27 2.82 63.38 11.27
November–December 2017 67.83 15.38 11.89 2.10 53.85 9.79
April–May 2019 53.68 20.59 12.50 2.94 58.09 8.09
November–December 2020 53.96 15.11 9.35 0.72 51.80 9.35
April–May 2022 60.30 14.70 12.90 4.30 69.00 12.90
Table 11. Answers to the question: “Among the Russian-made machines and equipment that your enterprise needs, are there samples that are not inferior in quality to their counterparts from far abroad?” (sum of answers = 100%)
Period Yes, and quite a few Yes, but very few None
October–December 2011 6.34 54.93 38.73
April–May 2013 6.08 50.00 43.92
November–December 2014 3.85 53.08 43.07
April–May 2016 10.42 58.33 31.25
November–December 2017 12.75 50.34 36.91
April–May 2019 14.71 47.06 38.23
November–December 2020 11.81 50.39 37.80
April–May 2022 14.00 55.40 30.60
Nevertheless, one can note certain positive shifts in the enterprises’ assessing the quality of Russian equipment. Thus, for instance, when answering a question about the dynamics in the quality of Russian machinery and equipment, the share of those who believe that this quality has improved in most cases increased from 7.26% in 2020 to 20.50% in 2022.
The share of respondents who believe that the quality of Russian equipment has improved in some cases also rose, from 33.06% in 2020 to 38.50% in 2022 (Table 12). In addition, the proportion of those who believe that the gap in quality between Russian and foreign equipment is narrowing has increased. In 2022, this share reached 24.60% and for the second time in the survey period exceeded the share of those holding the opinion that the quality gap continues to widen (Table 13). At the same time, it is clear that, despite the gradual improvement in the situation with the quantity and quality of Russian engineering products, the problems caused by restrictions on the import of equipment from Western countries can be named among the most acute for most Russian enterprises.
Table 12. Answers to the question: “Has the quality of Russian machinery and equipment improved over the past 3–5 years?” (sum of answers = 100%)
Period Yes, in most cases the quality has improved For some types of equipment the quality has improved,
for others it has remained
at the same level Generally, the quality remained at the same level Generally,
the quality has deteriorated
October–December 2011 6.62 38.24 48.53 6.62
April–May 2013 7.59 31.03 46.21 15.17
November–December 2014 8.80 31.20 56.00 4.00
April–May 2016 7.91 35.25 53.24 3.60
November–December 2017 18.12 33.33 43.48 5.07
April–May 2019 10.69 33.59 48.85 6.87
November–December 2020 7.26 33.06 52.42 7.26
April–May 2022 20.50 38.50 37.60 3.40
Table 13. Answers to the question: “How do you assess the size of the gap between the quality of Russian and foreign-made machines and equipment in recent years?” (sum of answers = 100%)
Period In recent years, the gap
in quality has reduced The gap in quality remains
at about the same level The gap in quality continues to grow in favor of imported equipment
October–December 2011 12.95 50.36 36.69
April–May 2013 10.74 53.69 35.57
November–December 2014 16.80 60.00 23.20
April–May 2016 18.18 55.94 25.88
November–December 2017 28.47 52.78 18.75
April–May 2019 20.15 54.48 25.37
November–December 2020 15.20 56.80 28.00
April–May 2022 24.60 51.70 23.70
Arguably, another consequence of the current crisis may be some weakening of the overall shortage of labor resources in the Russian economy. While in November–December 2021 this deficit worsened so much that 49.00% of surveyed enterprises reported a shortage of labor resources, in the spring of 2022 the share of such answers fell to 33.60%. At the same time, the share of respondents who indicated an excess of labor resources increased from 0.70% at the end of 2021 to 5.50% in the spring of 2022 (Table 14).
Table 14. Answers to the question: “How is your enterprise currently provided with labor resources” (sum of answers = 100%)
Period There is an excess of labor
resources The enterprise
is optimally resourced with labor For some categories there
is an excess, for others there
is a lack of labor resources The enterprise
is experiencing
lack of labor resources
November–December 2014 3.50 32.17 37.76 26.57
April–May 2015 9.03 42.59 31.61 16.77
November–December 2015 5.65 39.55 40.68 14.12
April–May 2016 14.19 47.10 27.10 11.61
November–December 2016 5.00 48.13 31.87 15.00
November–December 2017 5.84 43.51 30.52 20.13
November–December 2018 4.93 38.73 33.80 22.54
April–May 2019 6.80 41.50 30.61 21.09
November–December 2019 3.01 36.75 31.93 28.31
April–May 2020 5.90 40.00 36.30 17.80
November–December 2020 3.60 35.97 25.90 34.53
April–May 2021 4.01 39.33 23.33 33.33
November–December 2021 0.70 29.70 20.60 49.00
April–May 2022 5.50 35.90 25.00 33.60
Nevertheless, in general, the shortage of labor in the Russian economy remains significant, which is why enterprises, adapting to the crisis, cut investment costs much more often than personnel costs (Table 14).
In the new Russian realities, the relevance of the sustainable development agenda has become problematic. Some experts have begun to voice the opinion that many aspects of sustainable development have lost relevance for Russia, in particular, those related to climate change and carbon regulation. Thus, the current attitude of real sector enterprises towards sustainable development issues, including the ESG agenda, has become another subject of study in the current survey.
Postpandemic sanctions, according to enterprises, will have a negative impact on their transition to sustainable development: 88% of respondents confirmed this to one degree or another (Table 15).
Table 15. Answers to the question: “Does your company think that tough sanctions in the postpandemic period will have a negative impact on the company’s transition to sustainable development?” (sum of answers = 100%)
Period Yes, the impact
will be significant Yes, the impact
will be medium Yes, the impact
will be marginal No, there
will be no impact
April–May 2022 30.40 41.60 16.00 12.00
During postpandemic sanctions, 78.90% of respondents did not publish nonfinancial reporting. Among those who published it, the majority were large-scale enterprises, most often representing the manufacturing industry. Enterprises publish reports on sustainability, corporate social responsibility, environmental performance, and greenhouse gas emissions. There is also a noticeable share of those planning to publish such reports in the future, almost 10% of the surveyed enterprises (Table 16).
Table 16. Answers to the question: “Does your company publish nonfinancial reports (in the field of sustainable development, social reports, environmental reports, integrated reports)?” (sum of answers = 100%)
Period Yes No We do not publish such reports, but we plan
to publish them in the future
April–May 2022 11.40 78.90 9.70
The climate agenda turned out to be of little relevance for Russian enterprises: 94.30% of respondents did not publish reports on financial risks associated with climate change (Table 17). Among those who planned to publish such reports were medium and large enterprises from manufacturing industries, as well as enterprises from the tourism sector (tourist centers and tour agencies).
Table 17. Answers to the question: “Does your business publish reports on financial risks associated with climate change?” (sum of answers=100%)
Period Yes No We do not publish such reports, but we plan
to publish them in the future
April–May 2022 0.80 94.30 4.90
Low concern about the climate agenda is also evident from the answers to the question about the goal to reduce the carbon footprint: 88% of the surveyed enterprises did not consider this goal relevant (Table 18).
Table 18. Answers to the question: “Does your business have a goal to reduce its carbon footprint?” (sum of answers = 100%)
Period Yes No, we do not consider this relevant
April–May 2022 12.00 88.00
Those who answered in the affirmative, however, most often include enterprises from the manufacturing industry, namely, foundries, cardboard and pulp manufacturers, manufacturers of machinery and equipment, manufacturers of petroleum products, as well as construction and installation companies. All these respondents have in common both the fact that they are large enterprises and the fact that they, due to the specifics of their activities, can make a significant contribution to reducing greenhouse gas emissions. Most often, these companies are leaders in the field of social and environmental policy in their industries and make significant investments in the field of environmental management.
However, many of those who do not set a goal to reduce their carbon footprint now admitted that the decarbonization agenda may affect them in the future: 52.90% of respondents found it difficult to unambiguously answer the question about the possible impact of carbon regulation on their business (Table 19).
Table 19. Answers to the question: “Can carbon regulation (imposing a cross-border carbon tax on products) affect your business in the future?” (sum of answers = 100%)
Period Yes No It’s hard to say
April–May 2022 13.20 33.90 52.90
In recent years, large corporations, banks and governments of developed countries, as well as international economic organizations of the United Nations, have been actively introducing ESG principles into business practice [11]. In this connection, the question of the attitude adopted by Russian enterprises towards this agenda, especially in modern conditions, has also become the subject of research. The results of the survey show that among the surveyed enterprises, the ESG agenda is perceived as more in demand than the climate one. In contrast to the situation with the climate agenda, in this case, the proportion of those who consider the ESG principles relevant turned out to be noticeably higher. Thirty-five point five percent of respondents adhered to the principles of ESG as inherent in the development of their enterprise. However, the proportion of those unfamiliar with these principles also turned out to be quite noticeable, 41.90% (Table 20).
Table 20. Answers to the question: “Do you follow the principles of ESG in the development strategy of your enterprise (ESG-Environmental, Social, Governance – these are the three parameters in accordance with which companies ensure sustainable development)?” (sum of answers = 100%)
Period Yes No Unaware of ESG principles
April–May 2022 35.50 22.60 41.90
Among the principles of ESG, the most popular are social and environmental ones. In 81.50% of cases, the respondents were engaged in improving corporate culture and social protection (labor protection, providing professional training, industrial safety, etc.), as well as participating in social policy in the area where the company operates (Table 21). In 67.70% of cases, enterprises took measures to protect the environment (greening of production, safe disposal of waste, increasing the energy efficiency of production, reducing material consumption, reforestation, etc.).
Table 21. Answers to the question: “To what ESG principles do you adhere in your enterprise development strategy?” (sum of answers >100%)
Period Environmental principle:
we take measures to protect
the environment (greening production, safe disposal
of waste, increasing the share
of recycling, increasing
the energy efficiency
of production, and reducing material consumption, reforestation, etc.) Social principle:
we improve corporate culture and social protection (labor protection, professional development, balance in the personnel structure, industrial safety, etc.), and also participate in social policy in the territory of the company’s presence Corporate Governance Principle: we follow the principles of business policy enshrined in the “Social Charter of Russian Business,” the company’s divisions comply with international standards (ISO 14001, etc.), we increase the number of independent directors in the board of directors, conduct an external audit, carry out various types of reporting, including nonfinancial reporting, etc. Other measures
April–May 2022 67.70 81.50 23.10 1.50
The principle of corporate governance is less popular. In only 23.10% of cases, was the adherence to the principles of business policy noted. These principles are enshrined in the “Social Charter of Russian Business” and require compliance with international standards (ISO 14001, etc.), external audit, various types of reporting, including nonfinancial. Enterprises that follow the principle of corporate governance include in their ESG strategies such sustainable development factors as the optimal distribution of powers within companies, an effective board of directors and management, transparent relations with shareholders and investors, etc.
Enterprises that take into account ESG principles in their strategy most often derive from this such benefits as an overall increase in operational efficiency, an improvement in the moral climate and improvement of corporate ethics, as well as maintaining the image and business reputation of the company and the opportunity to emphasize its leading position in the industry (Fig. 1). Such effects as an increase in the investment attractiveness of the company for foreign investors and an improvement in the company’s position in ESG indices and ratings were felt mainly by large and superlarge enterprises engaged in international trading operations.
Fig. 1. Answers to the question: “If your company takes into account ESG factors in its activities, what benefits does it receive?” (sum of answers>100%).
In the light of current events, the climate agenda and other issues of sustainable development in modern Russian conditions may undergo significant changes: development programs are being adjusted and the timing of their implementation is being shifted. It is expedient for Russia to focus on protecting the interests of the national economy by actively developing the legal and institutional infrastructure that should provide business with incentives to reduce carbon intensity and increase production efficiency [12].
The issues of climate challenges will remain in the focus of the world community, which means that Russia should not abandon them. The sustainable development agenda and the ESG concept will be truly relevant for our country, since the solution of related issues is necessary and important for the territories, the public, and business. However, at present, as the results of the survey show, sanctions have a negative impact on the processes associated with implementing the principles of sustainable development in enterprises.
MAIN CONCLUSIONS OF THE SURVEY.
1. The economic crisis in Russia caused by economic sanctions imposed by Western countries turned out to be massive but less significant than expected at the beginning of events and during the crisis caused by the COVID-19 pandemic.
2. Russian enterprises, adapting to the sanctions, began to act very quickly, using both passive (reducing investments and personnel costs) and active (search for new suppliers and new markets, launching new products and launching modernization measures) methods of adaptation.
3. Businesses believe that the Russian authorities should actively oppose sanctions by reducing the tax and bureaucratic burden on manufacturers, limiting prices for energy resources and transport, expanding the scope of concessional lending, supporting domestic demand, etc.
4. According to enterprises, there have been some positive changes in the quality of Russian machinery and equipment. However, in general, lagging in quality behind foreign analogs remains significant, and, therefore, the dependence of Russian manufacturers on advanced imported equipment remains high.
5. Climate change issues were of little concern to Russian enterprises. These issues are mainly addressed by those whose activities are directly related to the risks of climate change and a high carbon footprint.
6. In general, the practice of reporting on financial risks associated with climate change is not widespread among the surveyed enterprises. With nonfinancial sustainability reporting, the situation looks a little better, but still the share of those who publish it is small.
7. The majority of respondents did not perceive the concept of ESG as a tool for long-term competitive advantage. Due to this, the share of enterprises that follow the ESG principles and benefit from it is still small. Their priorities include the social and labor sphere and environmental regulation. So far, less attention has been paid to the quality of corporate governance as one of the components of the ESG agenda.
ACKNOWLEDGMENTS
The authors are grateful to G.P. Belyakova, L.I. Pisareva, N.I. Titova, R.A. Galetskaya, and O.Yu. Galetskaya. We would like to express our gratitude for their help in the organization of the survey to E.K. Vorobei (Sochi State University), S.V. Terebova (Vologda RC RAS), and N.N. Evchenko (Southern Federal University).
1 The survey covered 189 enterprises (power industry; ferrous and non-ferrous metallurgy; chemical industry; mechanical engineering; building materials industry; timber, woodworking and pulp and paper industries; light, food and medical industries; agriculture; construction; transport; communications; motor vehicle repair; healthcare; hotels; public catering) from 47 regions of Russia.
Translated by I. Pertsovskaya
==== Refs
REFERENCES
1 The World Bank predicts a reduction in GDP in Russia in 2022 by 11.2%. https://tass.ru/ekonomika/14338473. Cited June 19, 2022.
2 A. Kudrin, Russia will be in a very difficult situation for one and a half to two years, Kommersant, April 27, 2022. http://www.kommersant.ru/doc/5328933. Cited June 19, 2022.
3 Charles Riley, Investors beware: The West is going to hit Russia with more sanctions, CNN Business, March 23, 2022. https://edition.cnn.com/2022/03/23./investing/stocks-premarket-rading/index.html. Cited June 19, 2022.
4 A. A. Shirov et al., Short-term analysis of GDP dynamics, Institute of Economic Forecasting Russian Academy of Sciences, June 15, 2022. https://ecfor.ru/publication/ kratkosrochnyj-analiz-dinamiki-vvp-iyun-2022/. Cited June 19, 2022.
5 Kuvalin D. B. Zinchenko Yu. V. Lavrinenko P. A. Russian enterprises in the spring 2020: Covid-19 pandemic reactions and opinions on the role of the state in the economy Stud. Russ. Econ. Dev. 2021 32 111 122 10.1134/S1075700721010093 33642849
6 Blokhin A. A. Krivova E. S. The role of ‘soft’ state support in the development of big business Mir Nov. Ekon. 2021 15 29 40 10.26794/2220-6469-2021-15-4-29-40
7 Borisov V. N. Pochukaeva O. V. Developing import substitution as a consequence of the growth of competitiveness of investment equipment Razvitie Territorii 2021 2 10 18
8 V. N. Borisov and O. V. Pochukaeva, “Russian market of investment equipment: Stagnation or development,” O-vo Ekon., No. 12, 59–74 (2019).
9 S. V. Terebova, Mechanisms for Increasing the Innovative Activity of Industrial Enterprises: Problems of Development and Implementation (Vol. Nauch. Tsentr Ross. Akad. Nauk, Vologda, 2017) [in Russian]. https://elibrary.ru/item.asp?id=37115449. Cited June 19, 2022.
10 N. A. Ganichev and I. E. Frolov, “On the Prospects and Priorities of Import Substitution in the Aircraft Industry under International Sanctions,” in Strategic Planning and Development of Enterprises. Materials of the Seventeenth All-Russian Symposium, Ed. by G.B. Kleiner (Tsentr. Ekon.-Mat. Inst. Ross. Akad. Nauk, Moscow, 2016), pp. 45–47 [in Russian].
11 Silvestrov S. N. Lopukhin A. V. Plaksenkov E. A. Fintech as a factor in accelerating inclusive sustainable development Mir Nov. Ekon. 2021 16 28 44
12 B. N. Porfiriev, A. A. Shirov, A. Yu. Kolpakov, and E. A. Edinak, “Opportunities and risks of climate regulation policy in Russia,” Vopr. Ekon., No. 1, 72–89. (2022).
| 36466732 | PMC9707249 | NO-CC CODE | 2022-12-01 23:20:22 | no | Stud Russ Econ Dev. 2022 Nov 28; 33(6):697-706 | utf-8 | Stud Russ Econ Dev | 2,022 | 10.1134/S1075700722060089 | oa_other |
==== Front
Endocrine
Endocrine
Endocrine
1355-008X
1559-0100
Springer US New York
3264
10.1007/s12020-022-03264-9
Original Article
The prognostic utility of serum thyrotropin in hospitalized Covid-19 patients: statistical and machine learning approaches
http://orcid.org/0000-0002-2925-9679
Pappa E. [email protected]
1
Gourna P. 1
Galatas G. 1
Manti M. 1
Romiou A. 1
Panagiotou L. 1
Chatzikyriakou R. 2
Trakas N. 3
Feretzakis G. 45
Christopoulos C. 1
1 grid.413180.f 0000 0004 0638 7883 First Department of Internal Medicine, “Sismanoglio-A. Fleming” General Hospital, Athens, 15126 Greece
2 grid.413180.f 0000 0004 0638 7883 Department of Hematology, “Sismanoglio-A. Fleming” General Hospital, Athens, 15126 Greece
3 grid.413180.f 0000 0004 0638 7883 Department of Biochemistry, “Sismanoglio-A. Fleming” General Hospital, Athens, 15126 Greece
4 grid.55939.33 0000 0004 0622 2659 School of Science and Technology, Hellenic Open University, Patras, 26335 Greece
5 grid.413180.f 0000 0004 0638 7883 Department of Quality Control, Research, and Continuing Education, “Sismanoglio-A. Fleming” General Hospital, Athens, 15126 Greece
29 11 2022
17
25 7 2022
16 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose
To assess the prognostic value of serum TSH in Greek patients with COVID-19 and compare it with that of commonly used prognostic biomarkers.
Methods
Retrospective study of 128 COVID-19 in patients with no history of thyroid disease. Serum TSH, albumin, CRP, ferritin, and D-dimers were measured at admission. Outcomes were classified as “favorable” (discharge from hospital) and “adverse” (intubation or in-hospital death of any cause). The prognostic performance of TSH and other indices was assessed using binary logistic regression, machine learning classifiers, and ROC curve analysis.
Results
Patients with adverse outcomes had significantly lower TSH compared to those with favorable outcomes (0.61 versus 1.09 mIU/L, p < 0.001). Binary logistic regression with sex, age, TSH, albumin, CRP, ferritin, and D-dimers as covariates showed that only albumin (p < 0.001) and TSH (p = 0.006) were significantly predictive of the outcome. Serum TSH below the optimal cut-off value of 0.5 mIU/L was associated with an odds ratio of 4.13 (95% C.I.: 1.41–12.05) for adverse outcome. Artificial neural network analysis showed that the prognostic importance of TSH was second only to that of albumin. However, the prognostic accuracy of low TSH was limited, with an AUC of 69.5%, compared to albumin’s 86.9%. A Naïve Bayes classifier based on the combination of serum albumin and TSH levels achieved high prognostic accuracy (AUC 99.2%).
Conclusion
Low serum TSH is independently associated with adverse outcome in hospitalized Greek patients with COVID-19 but its prognostic utility is limited. The integration of serum TSH into machine learning classifiers in combination with other biomarkers enables outcome prediction with high accuracy.
Keywords
COVID-19
Thyroid stimulating hormone
Non-thyroidal illness syndrome
Machine learning
Artificial intelligence
Bayes classifier
==== Body
pmcIntroduction
Well before the outbreak of the COVID-19 pandemic, laboratory observations in patients with SARS-CoV-1 infection had revealed an association of coronavirus disease with distorted thyroid gland structure and function [1, 2]. The thyroid is rich in angiotensin-converting enzyme-2 (ACE-2), which some coronaviruses use as a vehicle for entering cells [3]. It, therefore, came as no surprise when Chen and colleagues reported that SARS-CoV-2 infection was also associated with deranged thyroid function tests [4]. This was subsequently confirmed by numerous investigators in Asian [5–9] and Western [10–14] populations. Although there have been reports of thyroiditis, often atypical, suggesting direct virus-induced or immunologically mediated damage to the gland [15, 16], in most cases the pattern of hormone abnormalities is compatible with the non-thyroidal illness syndrome (NTIS), also known as euthyroid sick syndrome (ESS), characterized by decreased levels of triiodothyronine (T3), normal or decreased levels of thyroxine (T4) and normal or decreased levels of thyrotropin (TSH) [5, 12, 13]. The changes are usually transient, and hormone levels return to normal after convalescence [12, 13]. The pathophysiological mechanisms underlying NTIS are not fully understood. It can occur in probably any serious illness, acute or chronic, and is associated with morbidity and mortality [17] In the setting of COVID-19, NTIS appears to reflect the effect of the “cytokine storm” on the hypothalamus-pituitary-thyroid axis at central (TSH suppression) and peripheral (deiodinase inhibition) levels [18]. The question of whether low TSH is causatively related to worse COVID-19 prognosis remains open [19]. A number of studies have shown that decreased levels of TSH and/or free T3 (FT3) or total T3 (TT3) are associated with unfavorable outcome in COVID-19 patients [6, 8, 9, 20]. However, the prognostic utility (i.e., the ability to discriminate between patients with good or bad prognosis) of thyroid function tests in COVID-19 has not been established. The objectives of the present study were: (a) to examine whether low serum TSH is a risk factor for adverse outcome in Greek patients with COVID-19, and (b) to evaluate the prognostic value of serum TSH and compare it with that of a number of biomarkers arbitrarily selected among those commonly used for assessment of disease severity.
Methods
This was a retrospective observational study involving statistical analysis of routinely collected patient data which were fully anonymized prior to analysis. The study was approved by our institution’s research and ethics committees (Decision No.: 6011/22-3-2022) and was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments.
Patients
We reviewed the records of all patients (n = 213) admitted to the COVID-19 unit of the Department of Internal Medicine of a tertiary public general hospital in Athens, Greece, from February to December 2021. In all cases, the diagnosis of SARS-CoV-2 infection was confirmed by polymerase chain reaction. The main admission criterion was hypoxic respiratory failure requiring supplemental oxygen. Serum levels of TSH were routinely measured at admission as screening for thyroid disease. Serum T4 and T3 measurement was at the discretion of the treating physician. Patients on thyroid hormone replacement treatment or with a known history of thyroid disease, pregnant patients, and patients taking amiodarone, somatostatin analogs, and dopamine agonists were excluded from the analysis. Also excluded were those with missing results of one or more of the laboratory indices under investigation (C-reactive protein [CRP], ferritin, D-dimers, albumin). Patients who had taken low-dose glucocorticoids (the equivalent of <30 mg prednisone daily) before admission were not excluded (see Online Resourse 1). The final study cohort consisted of 128 patients with a median age of 64 years (range 21–91), 68.5% of whom were male.
Laboratory assays
Serum TSH, albumin, CRP, ferritin, and D-dimers were measured as part of the routine laboratory work-up in all patients at the time of admission. TSH, free T4 (FT4), FT3, and T3 were measured on a Cobas e601 analyzer (Roche Diagnostics) using an electrochemiluminescence immunoassay. Albumin and CRP were measured by an immunoturbidimetric assay on a Cobas c501 analyzer (Roche Diagnostics). Ferritin was measured on an ADVIA Centaur analyzer (Siemens Healthcare Diagnostics) using a direct chemiluminescence 2-point sandwich immunoassay. D-dimers were measured with an immunoturbidimetric assay on a BCS XP Coagulation System (Siemens Healthcare Diagnostics).
Statistical analysis
IBM SPSS version 27.0 was used for data analysis, including correlations, comparisons of means and proportions with parametric and nonparametric methods as appropriate, multivariate logistic regression, and Receiver Operating Characteristics (ROC) curve analysis. Patient outcomes were classified as “favorable” (discharge without intubation during hospitalization) and “adverse” (intubation or in-hospital death of any cause). The optimal dichotomous cut-off values of continuous numeric variables (age, TSH, CRP, ferritin, D-dimers, albumin) being evaluated as prognostic factors were calculated from ROC curve analysis using the Juden’s index maximization rule.
Artificial Neural Network (ANN) analysis
The multilayer perceptron neural network module in IBM SPSS (version 27.0) was used to build, train, and test the relationship between the outcome and the independent covariates (sex, age, TSH, CRP, ferritin, D-dimers, albumin) constituting the input layer. The model used has a hidden layer with three units (neurons), excluding the bias unit. The activation function in the hidden layer was selected to be a hyperbolic tangent. The output layer of the network has one dependent variable with two units (adverse outcome, favorable outcome). Softmax was used as the activation function, while the cross-entropy error function was chosen in the output layer.
Machine learning classification algorithms
The Waikato Environment for Knowledge Analysis (WEKA) is an open-source data mining software developed at the University of Waikato, New Zealand [21]. It includes a number of popular classification algorithms which we used to build, train, and test models of COVID-19 outcome prediction based on combinations of the demographic (age, sex) and laboratory (TSH, CRP, ferritin, D-dimers, albumin) variables of our dataset. Besides overall accuracy (the percentage of correctly classified cases), specific quantitative indices considered for the selection of the best classifier were:True Positive Rate (TPR) for each class.
False Positive Rate (FPR) for each class.
Precision. A proportion calculated as the number of cases that are truly of a class, divided by the number of cases classified as that class. Equivalent to Positive Predictive Value.
Recall. The fraction of cases correctly predicted as belonging to a class. Equal to TPR.
F-measure. A combined measure reflecting the balance between precision and recall, calculated as: 2 * Precision * Recall/(Precision + Recall).
AUC. The area under the ROC (Receiver Operating Characteristic) curve.
Results
The basic demographic and clinical features of the patients are shown in Online Resource 1. Mean serum TSH in the study cohort was 0.93 mIU/L. TSH levels were significantly lower in male patients (0.84 mIU/L versus 1.13 mIU/L in females, P = 0.045). In the subsets of patients who had FT4 (N = 75), FT3 (N = 38), or T3 (N = 30) measurements, there was no correlation between TSH and thyroid hormone (FT4, FT3, T3) levels. For a detailed bivariate correlations matrix, see Online Resource 2. Patients of both sexes with adverse outcomes had significantly lower TSH levels compared to those with favorable outcomes (Fig. 1). Levels of CRP, ferritin, D-dimers, and albumin also differed between the two outcome groups, with varying degrees of statistical significance (Table 1). There was a weak positive correlation of TSH with serum albumin (Spearman’s Rho = 0.188, P = 0.034) but no significant correlation of TSH with the other laboratory indices of disease severity (Online Resource 2).Fig. 1 Boxplots of serum TSH levels of patients by sex and outcome. Differences between favorable and adverse outcome groups were significant in both sexes by the two-tailed Mann-Whitney U test. Created with SPSS v.27.0
Table 1 Comparison of variable levels by outcome (two-tailed Mann-Whitney U test)
Variable Outcome Median Interquartile range p
TSH (mIU/L) n.r.: 0.3–4.5 Favorable 0.77 1.51–0.51 0.00028
Adverse 0.45 0.86–0.27
CRP (mg/L) n.r.: <0.5 Favorable 50.3 83.9–25.1 0.00158
Adverse 73.6 154.5–46.7
Ferritin (μg/L) n.r.: 20–270 Favorable 432 742–250 0.01352
Adverse 790 1261–344
D-dimers (ng/mL) n.r.: <500 Favorable 556 982–335 0.00068
Adverse 1030 1310–669
Albumin (g/dL) n.r.: 3.5–5.1 Favorable 3.70 3.9–3.4 <0.00001
Adverse 2.90 3.2–2.6
n.r. normal range
ROC curve analysis
ROC curve analysis for adverse outcome prediction gave an AUC of 69.5% for TSH, compared to 86.9% for albumin (Fig. 2). The optimal cut-off value of serum TSH for discrimination between adverse and favorable outcomes was 0.5 mIU/L. Serum TSH below the optimal cut-off value was associated with an odds ratio of 4.13 (95% C.I.: 1.41–12.05) for adverse outcome. Detailed predictive performance characteristics of serum TSH and its comparison with other variables are shown in Table 2. TSH had the highest specificity in adverse outcome prediction (78.3%) but a low sensitivity (55.6%) and low positive predictive value (PPV = 58.1%). Serum albumin had a high sensitivity (93.3%) but low specificity (67.5%) and PPV (60.9%).Fig. 2 TSH and Albumin ROC curves for prediction of adverse outcome. The respective AUCs were 69.5% (95% C.I.: 60–79%) and 86.9% (95% C.I.: 80.4–93.4%). Created with SPSS v.27.0
Table 2 Analysis of ROC curves for predicting adverse outcome. Performance characteristics of individual variables at the optimal cut-off points
Variable AUC (%) Cut-off* Sens (%) Spec (%) PPV (%) NPV (%)
TSH 69.5 0.5 mIU/L 55.6 78.3 58.1 76.5
CRP 66.9 62 mg/L 66.7 65.1 50.8 78.3
Ferritin 63.2 752 μg/L 55.6 75.9 55.6 75.9
D-dimers 68.2 641 ng/mL 80.0 56.6 50.0 83.9
Albumin 86.9 3.5 g/dL 93.3 67.5 60.9 94.9
Age 68.8 60 years 82.2 51.8 48.1 84.3
*Refers to the optimal cut-off point, i.e., the variable value that maximizes Juden’s index (sensitivity + specificity-1)
AUC Area Under the ROC Curve
Sens sensitivity, Spec specificity, PPV Positive Predictive Value, NPV Negative Predictive Value
Binary logistic regression analysis
Multivariate binary logistic regression with sex, age, TSH, albumin, CRP, ferritin, and D-dimers as independent covariates and disease outcome (adverse—favorable) as dependent variable showed that only albumin (P < 0.001) and TSH (P = 0.006) were significantly associated with outcome (Table 3). The detailed SPSS logistic regression output can be found in the Online Resource 3a,b. The precision of this model was 82.9% for predicted adverse outcomes and 87.1% for predicted favorable outcomes. The probability equation of a simpler logistic regression model using only the statistically significant variables (albumin and TSH) was:P=11+e3.3*ALBUMIN+1.2*TSH−11.4
where P > 0.5 predicts an adverse outcome with a precision of 74.4%, while P < 0.5 predicts a favorable outcome with 84.7% precision.Table 3 Results of binary logistic regression analysis for adverse outcome prediction
Independent variables Coefficient Standard Error Significance (P) Odds Ratio 95% C.I. for O.R.
Lower Upper
SEX (male) 1.073 0.648 0.097 2.925 0.822 10.408
AGE −0.006 0.021 0.773 0.994 0.953 1.037
TSH −1.366 0.496 0.006 0.255 0.097 0.674
ALBUMIN −3.387 0.774 <0.001 0.034 0.007 0.154
CRP 0.002 0.004 0.659 1.002 0.994 1.010
FERRITIN 0.001 0.001 0.357 1.001 0.999 1.002
D-DIMERS 0.000 <0.001 0.447 1.000 1.000 1.000
Multilayer perceptron analysis
The multilayer perceptron model was trained with 69.5% of the cases (N = 89), while the remaining 30.5% (N = 39) were set aside for testing the performance of the model in outcome prediction. No data were excluded. The overall percentage of correct predictions was 83.1% (85.7% in the “favorable outcome” class, 78.8% in the “adverse outcome” class) in the training sample and 87.2% (92.6% in the “favorable outcome” class, 75.0% in the “adverse outcome” class) in the testing sample. The AUC of the ROC curve reflecting the overall predictive performance of the model was 90.7% for both outcomes (adverse-favorable). The absolute and normalized (%) predictive importance of the individual independent variables included in the model is shown in Fig. 3.Fig. 3 Multilayer Perceptron Neural Network analysis: Comparative predictive importance of independent variables. Created with SPSS v.27.0
WEKA classifier analysis
The best classification performance was obtained by a Naïve Bayes classifier using albumin and TSH as independent attributes. It was trained on 80% of the cohort cases and had an overall precision of 96.2%. Predictive accuracy was high in both the “favorable outcome” and “adverse outcome” classes. Detailed performance indices are shown in Table 4. The inclusion of additional attributes (sex, age, CRP, ferritin, D-dimers), did not improve model performance. Removal of the TSH attribute resulted in a 16.6% reduction in overall predictive accuracy with corresponding reductions in all the specific performance indices of the model.Table 4 Detailed accuracy of Naïve Bayes classifier by class
TP Rate FP Rate Precision Recall F-Measure AUC Class (Outcome)
1.000 0.143 0.950 1.000 0.974 0.992 Favorable
0.857 0.000 1.000 0.857 0.923 0.992 Adverse
Weighted Avg. 0.962 0.104 0.963 0.962 0.961 0.992
TP True Positive, FP False Positive, AUC Area Under the ROC Curve
Discussion
In the original report on deranged thyroid function in COVID-19, Chen, Zhou, and Xu found that 56% of 50 Chinese patients with moderate to critical severity COVID-19 had significantly lower TSH levels compared to healthy controls and non-COVID-19 pneumonia patients of similar severity [4]. In 34% of their patients, low TSH was the only abnormal thyroid function test. The degree of decrease in TSH levels was positively correlated with COVID-19 severity. In a retrospective study of 150 Chinese COVID-19 patients with NTIS (defined as decreased serum T3), Gong et al. found that low TSH was independently associated with a hazard ratio of 2.78 for 90-day mortality [6]. Similar results have been reported by Ahn et al., who analyzed retrospectively 119 Korean COVID-19 patients and found that the degree of decline in TSH was significantly associated with disease severity [5]. Studies in Western populations have yielded similar results. In the United Kingdom, Khoo et al. observed that patients admitted with COVID-19 had reduced serum TSH compared to their baseline levels, a phenomenon not seen in control patients admitted for non-COVID illnesses. Among patients with COVID-19, those who needed admission to the Intensive Therapy Unit (ITU) had significantly lower median TSH [13]. In a prospective Italian study of 506 patients admitted with COVID-19 of mild clinical severity, Sparano et al. found that admission levels of TSH in patients with unfavorable outcomes (death or transfer to ITU) were significantly lower than those of patients with favorable outcomes (0.56 versus 0.94 mIU/L) [14]. In another prospective Italian study, Campi et al. found that 39% of COVID-19 patients treated in ITU or sub-ITU had decreases in TSH levels inversely correlated with CRP and Interleukin-6 (IL-6). However, only FT3 levels were predictive of mortality [12]. In contrast to the above reports, a retrospective Danish study of 116 COVID-19 patients by Clausen et al. did not find an association between serum TSH and clinical outcome. They did detect weak negative associations of TSH levels with a number of cytokines (IL-8, IL-10, IL-15, IP-10, GM-CSF) but, interestingly, not with IL-6, a cardinal marker of COVID-19 severity. In their study, low serum TSH (<0.4 mIU/L) was not associated with increased mortality [22]. Similarly, Beltrão et al. did not find an association between serum TSH and mortality or disease severity in a prospective, observational study of 245 Brazilian patients with COVID-19 [11] The specific nature of the association of COVID-19 with low TSH has also been challenged: In a prospective study of critically and non-critically ill Greek patients, Vassiliadi et al. found that serum TSH levels of COVID-19 patients were not significantly different from those of control non-COVID patients, suggesting the absence of a specific link between low TSH and SARS-CoV-2 infection [23].
Our results lend further support to the argument that low serum TSH is associated with adverse outcome of COVID-19, independently of other laboratory indices of disease severity. The cause of low TSH in our cohort can only be speculated upon. Thyrotoxicosis due to thyroiditis is unlikely, as there was no inverse correlation between TSH and thyroid hormones. The discordance between TSH and thyroid hormone levels suggests the presence of a deranged hypothalamus-pituitary-thyroid axis as part of NTIS. However, a direct effect of the virus on the pituitary or hypothalamus cannot be excluded since SARS-CoV-2 receptors are known to exist in these locations [24, 25]. Although exogenous glucocorticoids can suppress TSH secretion, they are unlikely to be responsible for the low TSH levels seen in our patients since very few of them had received steroids prior to admission (see Online Resource 1).
Early identification of hospitalized COVID-19 patients who are likely to have an adverse outcome is important for optimization of treatment, rational use of resources, and advancing our understanding of disease pathophysiology. Be it a specific feature of SARS-CoV-2 infection or merely a manifestation of NTIS, low serum TSH in COVID-19 appears to be independently associated with adverse prognosis. However, the prognostic accuracy of low TSH levels in COVID-19 inpatients has not been studied in detail. In the present study, we compared the prognostic utility of serum TSH to that of a number of widely available biomarkers routinely used for disease monitoring and risk assessment in COVID-19. Among them, CRP and ferritin reflect the intensity of the inflammatory reaction, while D-dimers are a measure of the vascular occlusive complications characteristically associated with this disease. Serum albumin is a negative acute phase reactant and also a powerful predictor of outcome in hospitalized patients irrespectively of underlying illness [26]. Using ROC curve analysis, we showed that, in common with other prognostic markers, including CRP, ferritin, and D-dimers, TSH has limited prognostic accuracy, with an AUC of only 69.5%. Low serum albumin, which has been reported to be associated with increased disease severity and mortality in COVID-19 [27, 28] was the most accurate predictor of adverse outcome in our study cohort, with an AUC of 86.9%. However, its low specificity (67.5% versus 78.3 for TSH) resulted in a poor positive predictive value of just 60.9%. We then proceeded to test the predictive performance of combinations of prognostic variables using a number of supervised machine-learning algorithms included in the WEKA platform. A Naïve Bayes classifier obtained the best results (AUC of 99.2%) with serum albumin and TSH levels as independent attributes. This was in agreement with the Perceptron ANN analysis, which indicated that albumin and TSH were the most important predictors of outcome, and also with the multivariate binary logistic analysis from which albumin and TSH emerged as the only statistically significant, independent predictive variables.
To our knowledge, this is the first report on the prognostic utility of TSH, including a face-to-face comparison with commonly used prognostic laboratory markers in COVID-19. The main limitations of our study are its retrospective design and the small sample size which did not permit subgroup analysis to look for a possible modifying effect of clinical variables like comorbidities. It must be also emphasized that the prediction model described in this study has not yet been externally validated and the findings in our cohort of hospitalized Greek COVID-19 patients cannot be generalized to different populations, e.g., primary care patients or those of different ethnic origin.
Conclusion
This study presents additional evidence in favor of the association between low serum TSH levels and adverse outcome of hospitalized patients with COVID-19. However, it also shows that, as is the case with other biomarkers of disease severity, measurement of serum TSH alone has limited prognostic utility. Integration of TSH levels into multivariable machine learning classification algorithms shows promise for outcome prediction with high accuracy and should be the subject of further studies.
Supplementary Information
Supplementary material_revised_2
Supplementary information
The online version contains supplementary material available at 10.1007/s12020-022-03264-9.
Acknowledgements
We thank the anonymous reviewers for their constructive comments.
Author contributions
All authors contributed to the study conception and design. Material preparation and data collection were performed by E.P., P.G., G.G., M.M., A.R., L.P., R.C., and N.T. Data analysis was performed by C. Christopoulos and G.F. The first draft of the manuscript was written by C.C., E.P., and G.F., and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. C.C. and G.F. contributed equally to the supervision of the study.
Data availability
The datasets generated and/or analyzed during the current study are not publicly available but are available from the corresponding author upon reasonable request.
Compliance with ethical standards
Conflict of interest
The authors declare no competing interests.
Ethics approval
The study was approved by our institution’s research and ethics committees (Decision No. 6011/22-3-2022) and was in accordance with the principles of the Declaration of Helsinki.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Wang W Ye YX Yao H Evaluation and observation of serum thyroid hormone and parathyroid hormone in patients with severe acute respiratory syndrome J. Chin. Antituberc. Assoc. 2003 25 232 234
2. Wei L Sun S Xu C-H Pathology of the thyroid in severe acute respiratory syndrome Hum. Pathol. 2007 38 95 102 10.1016/j.humpath.2006.06.011 16996569
3. Li M-Y Li L Zhang Y Expression of the SARS-CoV-2 cell receptor gene ACE2 in a wide variety of human tissues Infect. Dis. Poverty 2020 9 45 10.1186/s40249-020-00662-x 32345362
4. Chen M Zhou W Xu W Thyroid function analysis in 50 patients with COVID-19: A retrospective study Thyroid 2021 31 8 11 10.1089/thy.2020.0363 32600165
5. Ahn J Lee MK Lee JH Thyroid hormone profile and its prognostic impact on the Coronavirus Disease 2019 in Korean patients Endocrinol. Metab. 2021 36 769 10.3803/EnM.2021.1109
6. J. Gong, D. Wang, H. Dong et al.: Prognostic significance of low TSH concentration in patients with COVID-19 presenting with non-thyroidal illness syndrome. BMC Endocr. Disord. 21 (2021). 10.1186/s12902-021-00766-x
7. S. Lang, Y. Liu, X. Qu et al.: Association between thyroid function and prognosis of COVID-19: A retrospective observational study. Endocrine research. (2021). 10.1080/07435800.2021.1924770
8. Zou R Wu C Zhang S Euthyroid Sick Syndrome in patients with COVID-19 Front Endocrinol. 2020 11 566439 10.3389/fendo.2020.566439
9. Zheng J Cui Z Shi N Suppression of the hypothalamic-pituitary-thyroid axis is associated with the severity of prognosis in hospitalized patients with COVID-19 BMC Endocr. Disord. 2021 21 228 10.1186/s12902-021-00896-2 34781943
10. C.D. Yazan, C. Ilgin, O. Elbasan et al. The Association of thyroid hormone changes with inflammatory status and prognosis in COVID-19. Int. J. Endocrinol. 2021 (2021). 10.1155/2021/2395212
11. F.E.L. Beltrão, D.C.A. Beltrão, G. Carvalhal et al. Thyroid hormone levels during hospital admission inform disease severity and mortality in COVID-19 patients. Thyroid: 31 (2021). 10.1089/thy.2021.0225
12. Campi I Bulgarelli I Dubini A The spectrum of thyroid function tests during hospitalization for SARS-COV-2 infection Eur. J. Endocrinol. 2021 184 699 709 10.1530/EJE-20-1391 33683214
13. B. Khoo, T. Tan, S.A. Clarke et al. Thyroid function before, during, and after COVID-19. J. Clin. Endocrinol. Metab. 106 (2021). 10.1210/clinem/dgaa830
14. C. Sparano, E. Zago, A. Morettini et al. Euthyroid sick syndrome as an early surrogate marker of poor outcome in mild SARS-CoV-2 disease. J. Endocrinol. Investig. (2021). 10.1007/s40618-021-01714-1
15. Muller I Cannavaro D Dazzi D SARS-CoV-2-related atypical thyroiditis Lancet Diabetes Endocrinol. 2020 8 739 741 10.1016/S2213-8587(20)30266-7 32738929
16. Lania A Sandri MT Cellini M Thyrotoxicosis in patients with COVID-19: the THYRCOV study Eur. J. Endocrinol. 2020 183 381 387 10.1530/EJE-20-0335 32698147
17. Fliers E Boelen A An update on non-thyroidal illness syndrome J. Endocrinol. Investig. 2021 44 1597 1607 10.1007/s40618-020-01482-4 33320308
18. Croce L Gangemi D Ancona G The cytokine storm and thyroid hormone changes in COVID-19 J. Endocrinol. Investig. 2021 44 891 904 10.1007/s40618-021-01506-7 33559848
19. W. Chen, J. Lei, Z. Li: Thyroid function changes and COVID-19 severity: egg or chicken? Endocrine. (2022). 10.1007/s12020-022-03129-1
20. Llamas M Garo ML Giovanella L Low free-T3 serum levels and prognosis of COVID-19: systematic review and meta-analysis Clin. Chem. Lab. Med. 2021 59 1906 1913 10.1515/cclm-2021-0805 34380183
21. E. Frank, M.A. Hall, Ian H. Witten. The WEKA Workbench: Online Appendix for “Data Mining: Practical Machine Learning Tools and Techniques” Morgan Kaufmann, Fourth Edition, 2016.
22. C.L. Clausen, Å.K. Rasmussen, T.H. Johannsen et al. Thyroid function in COVID-19 and the association to cytokine levels and mortality. Endocr. Connect. 10.1530/EC-21-0301
23. Vassiliadi DA Ilias I Pratikaki M Thyroid hormone alterations in critically and non-critically ill patients with SARS-CoV-2 infection Endocr. Connect. 2021 10 646 655 10.1530/EC-21-0029 34010152
24. S. Nampoothiri, F. Sauve, G. Ternier et al. The hypothalamus as a hub for SARS-CoV-2 brain infection and pathogenesis. bioRxiv (2020). 10.1101/2020.06.08.139329
25. Pal R Banerjee M COVID-19 and the endocrine system: exploring the unexplored J. Endocrinol. Invest. 2020 43 1027 1031 10.1007/s40618-020-01276-8 32361826
26. Akirov A Masri-Iraqi H Atamna A Shimon I Low albumin levels are associated with mortality risk in hospitalized patients Am. J. Med. 2017 130 1465.e11 1465.e19 10.1016/j.amjmed.2017.07.020 28803138
27. Bannaga AS Tabuso M Farrugia A C-reactive protein and albumin association with mortality of hospitalised SARS-CoV-2 patients: A tertiary hospital experience Clin. Med 2020 20 463 467 10.7861/clinmed.2020-0424
28. Bassoli C Oreni L Ballone E Role of serum albumin and proteinuria in patients with SARS-CoV-2 pneumonia Int J. Clin. Pr. 2021 75 e13946 10.1111/ijcp.13946
| 36445619 | PMC9707250 | NO-CC CODE | 2022-12-01 23:20:22 | no | Endocrine. 2022 Nov 29;:1-7 | utf-8 | Endocrine | 2,022 | 10.1007/s12020-022-03264-9 | oa_other |
==== Front
Ir J Med Sci
Ir J Med Sci
Irish Journal of Medical Science
0021-1265
1863-4362
Springer International Publishing Cham
3232
10.1007/s11845-022-03232-2
Original Article
The effect of full blood count and cardiac biomarkers on prognosis in carbon monoxide poisoning in children
http://orcid.org/0000-0002-1406-4895
İpek Sevcan [email protected]
1
http://orcid.org/0000-0002-5561-3598
Güllü Ufuk Utku [email protected]
2
http://orcid.org/0000-0002-0433-5970
Güngör Şükrü [email protected]
3
http://orcid.org/0000-0003-3596-2704
Demiray Şeyma [email protected]
4
1 grid.411741.6 0000 0004 0574 2441 Department of Pediatrics, Faculty of Medicine, Kahramanmaraş Sutcu Imam University, Kahramanmaraş, Turkey
2 grid.411741.6 0000 0004 0574 2441 Division of Pediatric Cardiology, Department of Pediatrics, Faculty of Medicine, Kahramanmaraş Sutcu Imam University, Kahramanmaraş, Turkey
3 grid.411741.6 0000 0004 0574 2441 Division of Pediatric Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, Faculty of Medicine, Kahramanmaraş Sutcu Imam University, Kahramanmaraş, Turkey
4 grid.411741.6 0000 0004 0574 2441 Department of Pediatrics, Kahramanmaraş Sutcu Imam University Medical Faculty, Kahramanmaraş, Turkey
29 11 2022
110
14 9 2022
21 11 2022
© The Author(s), under exclusive licence to Royal Academy of Medicine in Ireland 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Aims
In this study, cardiac biomarkers, blood parameters, electrocardiography (ECG), and echocardiography were investigated in children with carbon monoxide (CO) poisoning, and the diagnostic value of these parameters was investigated.
Methods
The demographical, clinical, and laboratory data of children aged 0–18 years who were admitted to the pediatric emergency department due to CO poisoning between January 2019 and January 2022 were retrospectively scanned from medical records. The patients were divided into two groups as troponin-I positive and troponin-I negative.
Results
There were 107 children aged 0–18 years (average age, 10.46 ± 5.77 years; 51% female) with CO poisoning. There were 13 patients with troponin-I positive myocardial injury. Troponin-I was positive in 3 patients whose carboxyhemoglobin (COHb) level was below 2% at the time of admission. In one patient, troponin-I, which was normal at admission, increased by the 24th hour of hospitalization. Hyperbaric oxygen therapy was given due to headache in one patient, although the COHb level of that patient was below 25%. An NT-proBNP level of ≥ 219.5 ng/L predicted the development of troponin-I positivity with a sensitivity of 70% and a specificity of 86.7% (AUC, 0.967 (0.58–0.994); p = 0.017). White blood cell (WBC), neutrophil, neutrophil-to-lymphocyte ratio (NLR), immature granulocyte (IG), and IG% levels were found to be significantly higher in the troponin-positive patient group.
Discussion and conclusion
NT-proBNP has been shown to be an early diagnostic marker for myocardial dysfunction. Additionally, when cardiac markers are not available, full blood parameters may assist clinicians for patient treatment and referral.
Keywords
Blood
Carbon monoxide poisoning
Carboxyhemoglobin
NT-proBNP
Pediatric emergency department
Troponin-I
==== Body
pmcIntroduction
Carbon monoxide (CO) is a tasteless, odorless, colorless, non-irritant, and lighter-than-air gas that is formed due to the incomplete combustion of organic materials and fossil fuels [1]. It is also known as a silent killer, as victims die without realizing that they have been poisoned [2]. CO poisoning is seen in places which are not adequately ventilated and mostly in cold climates and winter months [3, 4]. Sometimes, CO poisoning can be seen due to the blockage of discharge pipes by heavy snow clogs [5]. An increased rate of CO poisoning has also been reported after hurricanes [6]. In addition, CO poisoning can occur by inhaling solvents containing methylene chloride [7].
CO reduces the oxygen-carrying capacity of hemoglobin and impairs the release of oxygen to the tissues. In addition, it shows harmful effects by binding to other heme-containing substance, such as myoglobin, guanylate cyclase, certain cytochromes, and nitric oxide synthase [8, 9]. Vital organs are affected by CO poisoning, causing short-term and long-term morbidity, even death in some cases [10, 11].
BNP is released from the ventricles to the peripheral circulation mainly by mechanical stress, alpha adrenergic agonists, endothelin1, angiotensin 2, interleukin (IL)-1β, IL-18, tumor necrosis factor (TNF)-α, and transforming growth factor (TGF)-β. It antagonizes the renin–angiotensin–aldosterone system, causing a decrease in vascular tone, natriuresis and diuresis. In addition, its antifibrotic and antihypertrophic effects have also been demonstrated in the heart [12, 13].
Cardiovascular complications caused by CO poisoning have mostly been reported as left ventricular dysfunction, arrhythmia, myocardial stunning, arrhythmias, and pulmonary edema.
A literature search performed by our team has shown that although several studies have evaluated cardiac functions of adult patients with CO poisoning, similar studies in pediatric patients are scarce [14–16]. In this study, we investigated cardiac markers, blood parameters, electrocardiography (ECG), and echocardiography in children with CO poisoning. We aimed to investigate the diagnostic values of these parameters.
Methods
Study design and patients
In this single-center, retrospective study, we enrolled children aged 0–18 years, who were admitted to the pediatric emergency department of a tertiary university hospital for CO poisoning between January 2019 and January 2022. Annually, approximately thirty-four thousand children are admitted to the pediatric emergency department of that university hospital. The study was conducted in accordance with the Declaration of Helsinki. Ethical approval was obtained from the local ethics committee before the study.
Patients with congenital or acquired heart disease were excluded from the study.
The diagnosis of CO poisoning was made with a history of CO exposure, clinical findings and symptoms caused by CO poisoning, and blood carboxyhemoglobin (COHb) level [17]. Since COHb level may be low in patients with a long period of time between hospital admission and CO exposure, carbon monoxide poisoning was diagnosed using clinical symptoms, history, and carbon monoxide half-life in such patients [16].
Data analysis
Demographic, laboratory, and clinical data of the patients were obtained from the medical records. Full blood count, blood gas analysis, troponin-I, NT-proBNP, creatine kinase (CK), CK-MB, albumin, aspartate transaminase (AST), alanine transaminase (ALT), electrocardiography, and echocardiography were performed. While laboratory studies were performed in all patients at the time of admission, ECG and echocardiography were performed in patients with an elevated troponin-I level.
Myocardial damage was defined as a troponin-I value above 0.02 (μg/L). The patients diagnosed with CO poisoning were divided into 2 groups as those with troponin-I value above 0.02 (μg/L), which represented the troponin-positive group with myocardial damage, and those with a troponin-I value less than 0.02 (μg/L), which represented the troponin-negative group without myocardial damage.
Indications for hyperbaric oxygen therapy included a COHb level of 25% or greater, presence of neurological symptoms such as unconsciousness, confusion, headache, seizure, and evidence of myocardial ischemia [16].
The level of consciousness was assessed with the Glasgow coma score [18].
Troponin-I analysis
Serum troponin-I levels were measured via a one-step enzyme immunoassay using a Roche Cobas 8000 automatic analyzer (with an analytical range of 0.01–0.02 µg/L).
NT-proBNP analysis
Serum NT-proBNP levels were determined by using the AQT90 FLEX analyzer (Radiometer Medical, Copenhagen, Denmark); the cut-off value was set at 133 ng/L.
Electrocardiography and echocardiography
Electrocardiography (ECG) was performed at the time of admission, while echocardiography was performed in patients with troponin elevation within 1 h of the presentation at the latest. In addition, ECG and echocardiography were re-evaluated at discharge. ECGs and echocardiographic studies of the patients were evaluated by a pediatric cardiologist.
Statistical analysis
Statistical Package for Social Sciences for Windows 22 software was used for statistical analysis. Study variables were presented as number (n) and percentage (%) and mean ± standard deviation. The normality of the distribution of the study variables was tested using the Kolmogorov–Smirnov test. Normally distributed numerical parameters were evaluated with one-way analysis of variance or Student’s t test. Kruskal–Wallis or Mann–Whitney U test was used for numerical variables without a normal distribution. The relationship between the variables was analyzed by Pearson correlation analysis. Receiver operating characteristic (ROC) curve analysis was performed to find an estimated optimal cut-off point. A p value less than 0.05 was considered statistically significant.
Ethical approval
This study was approved by the ethics committee of the Kahramanmaraş Sutcu Imam University before its start (No.: 2022/09–04).
Results
There were 107 children aged 0–18 years with CO poisoning (average age 10.46 ± 5.77 years; 51% female). The source of CO poisoning was an improperly planned ventilation system of an indoor heating system or its accidental malfunction in all patients. There were 13 patients with troponin-positive myocardial injury and 94 troponin-negative patients. There was no significant difference between the patient groups in terms of gender distribution and mean age (Table 1). Glasgow coma scale of the patients was lower in the troponin-positive group (Table 1). Three troponin-I positive patients had a carboxyhemoglobin (COHb) level below 2% at the time of admission. A patient with a normal admission troponin-I level had an increased troponin-I level at the 24th hour of hospitalization. Eight patients were administered hyperbaric oxygen therapy. Hyperbaric oxygen therapy was given due to headache in one patient, although the COHb level of that patient was below 25%. It was applied to the other 7 patients because their COHb level was above 25%. None of our patients was intubated. None of our patients was connected to a mechanical ventilator, and no vasoactive drug was given to any patient. Sixteen patients were treated in hospital.Table 1 Comparison of demographic and laboratory data of children with carbon monoxide poisoning according to troponin-I
Troponin-I negative (n = 94) Troponin-I positive (n = 13) p
Age (year) 8.56 ± 4.56 7.48 ± 4.24 0.33
Sex (female/male) 49/45 6/7 0.77
GCS 13.9 ± 0.62 12.15 ± 1.99 < 0.001
pH 7.36 ± 0.04 7.28 ± 0.09 0.010
pCO2 (mmHg) 42.49 ± 8.07 43.13 ± 6.55 0.788
HCO3 (mmol/L) 22.76 ± 2.73 19.24 ± 4.44 0.015
COHb (%) 14.28 ± 9.96 17.05 ± 13.48 0.379
Lactate (mmol/L) 2.33 ± 1.07 4.70 ± 3.50 0.032
WBC (× 109/L) 10.26 ± 4.09 15.33 ± 4.41 < 0.001
Neutrophil (× 109/L) 5.60 ± 3.08 11.32 ± 4.78 < 0.001
Lymphocyte (× 109/L) 3.64 ± 2.30 3.02 ± 2.31 0.356
Monocyte (× 109/L) 0.67 ± 0.36 0.85 ± 0.48 0.102
NLR 2.89 ± 7.81 5.70 ± 3.84 0.041
LMR 5.68 ± 2.24 4.12 ± 2.79 0.064
IG% 0.259 ± 0.139 0.678 ± 0.382 0.001
IG (× 109/L) 28.44 ± 22.97 108.57 ± 75.12 0.002
Hb (g/dL) 12.78 ± 1.27 12.64 ± 1.25 0.697
PLT (× 109/L) 293.91 ± 79.19 321.64 ± 65.91 0.217
INR 1.59 ± 2.32 1.11 ± 0.09 0.652
Albumin (g/L) 4.54 ± 0.27 4.46 ± 0.35 0.401
AST (U/L) 26.28 ± 7.09 38.36 ± 24.61 0.091
ALT (U/L) 15.10 ± 5.85 15.14 ± 5.03 0.979
Statistics: Mann–Whitney U test; chi-square test
Abbreviations: pCO2, carbon dioxide partial pressure; HCO3, bicarbonate; COHb, carboxyhemoglobin; WBC, white blood cell count; NLR, neutrophil-to-lymphocyte ratio; LMR, lymphocyte-to-monocyte ratio; IG, immature granulocyte; Hb, hemoglobin; PLT, platelet; INR, international normalized ratio; AST, serum aspartate transaminase; ALT, alanine transaminase
Comparison of cardiac biomarkers by troponin-I status in children with CO poisoning was presented in Table 2. A comparison of the laboratory data of troponin-positive and troponin-negative patients revealed that the NT-proBNP values at admission, the 6th hour, and the 12th hour were significantly higher in the troponin-positive group (p = 0.005, p = 0.003, and p = 0.031, respectively). There was no significant difference between the two groups regarding the NT-proBNP value measured at the 24th hour (p = 0.092). Using a ROC curve analysis, we found that the best cut-off point of NT-proBNP to predict troponin positivity in patients with CO poisoning was ≥ 219.5 ng/L. Accordingly, we determined that a NT-proBNP level of ≥ 219.5 ng/L predicted troponin positivity with a sensitivity of 70% and a specificity of 86.7% (AUC, 0.967 (0.58–0.994); p = 0.017). An analysis of the CK values showed that the CK values were significantly higher in the troponin-positive group at admission and the 6th hour (p = 0.043, p = 0.030, respectively). There was no statistically significant difference between the groups with respect to the CK values at the 12th and 24th hours (p = 0.162, p = 0.419, respectively). In addition, although CK-MB values at admission and at the 12th hour were significantly higher in the troponin-positive patient group (p = 0.030, p = 0.047, respectively), there was no significant difference between the groups in terms of CK-MB values at the 6th and 24th hours (p = 0.085, p = 0.300, respectively).Table 2 Comparison of cardiac biomarkers of children with carbon monoxide poisoning according to troponin-I
Troponin-I negative (n = 94) Troponin-I positive (n = 13) p
NT-proBNP 0 h (ng/L) 113.63 ± 154.24 2317.01 ± 3583.79 0.005
NT-proBNP 6 h (ng/L) 177.40 ± 249.94 3750.60 ± 4117.46 0.003
NT-proBNP 12 h (ng/L) 118.67 ± 137.04 2983.36 ± 3773.64 0.031
NT-proBNP 24 h (ng/L) 120.16 ± 134.38 1942.60 ± 3053.46 0.092
CK 0 h (U/L) 115.73 ± 44.62 232.14 ± 193.13 0.043
CK 6 h (U/L) 103.42 ± 47.49 271.55 ± 190.78 0.030
CK 12 h (U/L) 91.30 ± 28.15 546.84 ± 1053.39 0.162
CK 24 h (U/L) 107.34 ± 42.11 455.30 ± 1007.90 0.419
CK-MB 0 h (µg/L) 3.11 ± 2.24 21.45 ± 28.18 0.030
CK-MB 6 h (µg/L) 3.82 ± 1.60 30.29 ± 45.83 0.085
CK-MB 12 h (µg/L) 1.75 ± 1.85 19.94 ± 29.60 0.047
CK-MB 24 h (µg/L) 1.88 ± 1.76 9.51 ± 17.05 0.300
Statistics: Mann–Whitney U test
h hour, NT-proBNP NT-pro–brain natriuretic peptide, CK creatine kinase
A comparison by the blood gas analysis performed at admission revealed that the pH and bicarbonate (HCO3) values were significantly lower in the troponin-positive patient group (p = 0.010, p = 0.015, respectively). There was no significant difference between the groups in terms of pCO2 and COHb levels (p = 0.788, p = 0.379, respectively) (Table 1). COHb levels of patients with CO poisoning with and without myocardial injury were presented in Figs. 1 and 2.Fig. 1 Graphical representation of COHb levels of in troponin-positive in children with CO poisoning
Fig. 2 Graphical representation of COHb levels of in troponin-negative in children with CO poisoning
Lactate, white blood cell (WBC), neutrophil, neutrophil-to-lymphocyte ratio (NLR), immature granulocyte (IG), and IG% levels were found to be significantly higher in the troponin-positive patient group (p = 0.032, p < 0.001, p < 0.001, p = 0.041, p = 0.002, p = 0.001, respectively) (Table 1). There was no statistically significant difference between the two groups in terms of lymphocyte, monocyte, lymphocyte-to-monocyte ratio (LMR), hemoglobin (Hb), platelet (PLT), international normalized ratio (INR), albumin, aspartate aminotransaminase (AST), and alanine aminotransaminase (ALT) values (p = 0.356, p = 0.102, p = 0.064, p = 0.697, p = 0.217, p = 0.652, p = 0.401, p = 0.091, p = 0.979, respectively) (Table 1).
Table 3 shows the data on the day on which the troponin-I, NT-proBNP, CK, and CK-MB values started to increase, for how many days they remained high, in how many days they reached the maximum level, and their maximum blood levels. In addition, the courses of troponin-I and NT-proBNP levels in troponin-positive patients were graphically presented in Figs. 3 and 4.Table 3 Evaluation of cardiac markers in children with carbon monoxide poisoning
Mean Median SD Range
The day on which the troponin-I level started to increase 1.1 1 0.316 1–2
How many days the troponin-I level remained high 3.2 3 2.044 1–7
The day on which the troponin-I level reached at its maximum value 1.11 1 0.557 0–2
Maximum value of troponin-I (μg/L) 0.212 0.065 0.329 0.03–1.1
The day on which the NT-proBNP level started to increase 1 1 0.00 1–1
How many days the NT-proBNP level remained high 6.33 6.5 3.724 2–12
The day on which the NT-proBNP level reached at its maximum value 1.33 1 0.516 1–2
Maximum value of NT-proBNP (ng/L) 6365 7615 4872.05 205–11,700
The day on which day the CK level started to increase 1 1 0.632 0–2
How many days the CK level remained high 2 2 1.871 0–5
The day on which the CK level reached at its maximum value 1 1 0.632 0–2
Maximum value of CK (U/L) 985.83 466.5 1412.92 154–3850
The day on which the CK-MB level started to increase 1 1 0.577 0–2
How many days the CK-MB level remained high 3.17 2.5 2.483 1–8
The day on which the CK-MB level reached its maximum value 1.17 1 0.408 1–2
Maximum value of CK-MB (µg/L) 49.33 16.5 56.246 9–130
Statistics: frequencies
Fig. 3 Graphical representation of the course of troponin-I in troponin-positive in children with CO poisoning
Fig. 4 Graphical representation of the course of NT-proBNP in troponin-positive in children with CO poisoning
There was no statistically significant difference between the ECG findings at the time of admission and after recovery (at discharge) in the troponin-positive patients (Table 3). When the patients were evaluated according to their echocardiography findings, we found statistically significant differences in left ventricular end-systolic diameter (LVSD) z-score, ejection fraction (EF), and fractioning shortening (FS) values at admission compared to those measured after recovery (p = 0.026, p = 0.047, p = 0.047, respectively). There was no statistically significant difference between the interventricular septum (IVS) z-score, the left ventricular end-diastolic dimension (LVDd) z-score, and the left ventricular posterior wall (LVPW) z-score values measured at admission and 1 week later (Table 4).Table 4 Comparison of the ECG and echocardiographic findings of children with CO poisoning at the time of admission and after recovering (control)
ECG findings
At the time of admission Control p
Heart rates (beats/min) 113.77 ± 26.71 95.67 ± 22.03 0.110
PR interval (sec) 0.127 ± 0.009 0.128 ± 0.010 0.984
QRS duration (sec) 0.069 ± 0.011 0.065 ± 0.008 0.440
QTc interval (sec) 0.404 ± 0.019 0.409 ± 0.017 0.535
Echocardiographic findings
At the time of admission Control p
IVS Zs 0.232 ± 0.333 0.656 ± 0.429 0.054
LV EDD Zs 0.288 ± 0.829 −0.018 ± 0.486 0.464
LV PW Zs 0.363 ± 0.512 0.658 ± 0.229 0.248
LV ESD Zs 0.154 ± 0.673 −0.760 ± 0.653 0.026
LV EF (%) 68.90 ± 4.15 74.00 ± 4.47 0.047
LV FS (%) 38.40 ± 3.56 42.60 ± 3.36 0.047
Statistics: independent student t test
IVS interventricular septum, LV left ventricle, EDD end-diastolic diameter, ESD end-systolic diameter, Zs z-score, PW posterior wall, FS fractioning shortening
An analysis of the correlation between cardiac markers and laboratory data showed a weak negative correlation between blood pH value and troponin-I and CK-MB and a moderately strong negative correlation between blood pH and CK (r, − 0.276, p = 0.015; r, − 0.378, p = 0.001; r, − 0.465, p < 0.001, respectively) (Table 4). There was no significant correlation between blood pH value and NT-proBNP (r, − 0.069, p = 0.663). There was no significant correlation between blood pCO2, COHb, NLR, and lymphocyte levels and cardiac markers (Table 4). We found a weak negative correlation between blood HCO3 value and troponin-I, CK-MB, and a moderately strong negative correlation between blood HCO3 and CK (r, − 0.265, p = 0.020; r, − 0.364, p = 0.001; r, − 0.432, p < 0.001, respectively). There was no significant correlation between blood HCO3 level and NT-proBNP (Table 5). There was a weak positive correlation between blood lactate level and troponin-I, CK-MB, and a moderately strong positive correlation between blood lactate and CK (r, 0.335, p = 0.004; r, 0.319, p = 0.007; r, 0.518, p < 0.001, respectively). We found a weak positive correlation between white blood cell count and NT-proBNP, CK, and CK-MB (r, 0.287, p = 0.037; r, 0.279, p = 0.012; r, 0.314, p, 0.003, respectively). There was no significant correlation between troponin-I and WBC. We found a moderately positive correlation between monocyte count and NT-proBNP and a weak positive correlation between monocyte count and CK and CK-MB (r, 0.422, p = 0.002; r, 0.331, p = 0.003; r, 0.308, p, 0.004, respectively). We found a statistically significant, albeit weak, negative correlation between lymphocyte-to-monocyte ratio and troponin-I, NT-proBNP, and CK and a negative correlation between lymphocyte-to-monocyte ratio and CK-MB (r, − 0.272, p = 0.010; r, − 0.276, p = 0.045; r, − 0.303, p, 0.006; r, 0.254, p = 0.017, respectively). We found a moderately strong positive correlation between blood IG, IG% values, and cardiac markers (Table 5). While there was a weak negative correlation between hemoglobin and NT-proBNP, there was no significant correlation between troponin, CK, and CK-MB (Table 4). We found a highly significant positive correlation between AST and troponin-I, NT-proBNP, and CK-MB. We found a moderately strong positive correlation between CK and AST. We also found a weak positive correlation between ALT level and NT-proBNP and CK-MB. There was no statistically significant correlation between troponin and CK levels and ALT (Table 5).Table 5 Evaluation of the correlation between cardiac markers and laboratory data of children with carbon monoxide poisoning
Troponin-I NT-proBNP CK CK-MB
pH r −0.276 −0.069 −0.465 −0.378
p 0.015 0.663 < 0.001 0.001
pCO2 r −0.051 −0.126 −0.052 −0.044
p 0.662 0.428 0.670 0.706
HCO3 r −0.265 −0.120 −0.432 −0.364
p 0.020 0.450 < 0.001 0.001
COHb r −0.160 −0.301 0.002 −0.160
p 0.164 0.053 0.986 0.165
Lactate R 0.335 0.224 0.518 0.319
p 0.004 0.166 < 0.001 0.007
WBC r 0.201 0.287 0.279 0.314
p 0.061 0.037 0.012 0.003
Neutrophil r 0.309 0.293 0.332 0.373
p 0.003 0.033 0.002 < 0.001
Lymphocyte r −0.131 −0.029 −0.059 −0.045
p 0.224 0.835 0.601 0.679
Monocyte r 0.190 0.422 0.331 0.308
p 0.077 0.002 0.003 0.004
NLR r 0.075 0.209 0.153 0.059
p 0.488 0.133 0.173 0.582
LMR r −0.272 −0.276 −0.303 0.254
p 0.010 0.045 0.006 0.017
IG% r 0.432 0.582 0.511 0.497
p < 0.001 < 0.001 < 0.001 < 0.001
IG r 0.409 0.539 0.562 0.516
p < 0.001 < 0.001 < 0.001 < 0.001
Hb r −0.051 −0.358 −0.070 −0.115
p 0.637 0.009 0.535 0.287
AST r 0.774 0.830 0.589 0.779
p < 0.001 < 0.001 < 0.001 < 0.001
ALT r 0.189 0.345 0.192 0.219
p 0.080 0.011 0.087 0.042
Statistics: Pearson correlation
NT-proBNP NT-pro-brain natriuretic peptide, CK creatine kinase, pCO2 carbon dioxide partial pressure, HCO3 bicarbonate, COHb carboxyhemoglobin, WBC white blood cell count, NLR neutrophil-to-lymphocyte ratio, LMR lymphocyte-to-monocyte ratio, IG immature granulocyte, Hb hemoglobin, PLT platelet, INR international normalized ratio, AST serum aspartate transaminase, ALT, alanine transaminase
Discussion
In our study, we investigated CO poisoning in children. We defined the troponin-positive patients as the group with myocardial damage. In a retrospective study on adults, conducted by Kim et al. the authors reported that cardiomyopathy developed in one of every eight patients with CO poisoning [19]. Similarly, in this pediatric study, we found that approximately one of every 8 patients had myocardial damage due to CO poisoning.
A comparison by blood gas analysis showed that pH (7.28 ± 0.09) and HCO3 (19.24 ± 4.44 mmol/L) values were significantly lower and lactate (4.70 ± 3.50 mmol/L) values significantly higher in the group with myocardial damage. In an experimental animal study of Simonsen et al. lactate was found higher and pH lower in the group with severe CO poisoning [20]. A literature search performed by our team revealed no information of the HCO3 values in patients with CO poisoning. Low pH, low HCO3, and high lactate levels all indicated metabolic acidosis in our patients. Metabolic acidosis, on the other hand, has been reported to negatively affect the prognosis of CO poisoning [10, 21, 22]. In addition, it was found that pH and HCO3 values were negatively correlated with troponin-I, CK, and CK-MB, whereas lactate value was positively correlated with those myocardial injury markers. On the other hand, no correlation was found between pCO2 and COHb levels and troponin-I, NT-proBNP, CK, and CK-MB values.
In this study, no significant difference was found between the study groups in terms of pCO2 and COHb levels. The half-life of COHb is between 250 and 320 min while breathing room air, about 90 min while breathing high-flow oxygen, and 30 min while breathing 100% hyperbaric oxygen [23]. Most of the time, oxygen therapy and hydration for patients with CO poisoning are started on the road to hospital or at the event scene, before blood samples are taken, and before patients reach the hospital. COHb can decrease to its normal blood range during ambulance transport because of its half-life, and the oxygen therapy started during transport. Therefore, a normal or less than 2% COHb value does not exclude severe CO poisoning [24, 25]. Moreover, clinical symptoms and COHb level are not necessarily correlated [9]. This was confirmed by the fact that three of our patients with normal COHb values had troponin-positive myocardial damage, and a patient with a COHb below than 25% received hyperbaric oxygen therapy for headache.
In previous studies, it has been suggested that troponin value may be normal at admission; thus, it should be measured serially in order to show myocardial damage [12, 26]. In the current study, the increase in troponin-I by 24th hour, which was normal at the beginning in one of our patients, supports this hypothesis. Therefore, we think that patients presenting to the emergency department with CO poisoning should be followed up for at least 24 h and cardiac markers should be re-checked before discharge.
In this study, NT-proBNP value was higher in the myocardial damage group at admission, and the elevation continued at the 6th and 12th hours. Moon et al. in a retrospective study conducted with adult patients with CO poisoning, showed that high NT-proBNP values at admission are a marker for poor prognosis and are important in identifying patients who may have a poor long-term neurological prognosis [27]. In another study conducted in adult patients with CO poisoning, NT-proBNP > 100 pg/mL at admission was found to be a significant marker in predicting left ventricular dysfunction [28]. In our previous studies, we reported the prognostic value of proBNP in various poisonings such as snake bites and scorpion bites [29, 30]. We also showed that it was predictive of cardiac involvement in children with COVID-19 at the time of admission [31]. In this study, we found that a NT-proBNP cut-off value of ≥ 219.5 ng/L predicted troponin positivity, i.e., myocardial injury, with sensitivity of 70% and a specificity of 86.7% in pediatric patients with CO poisoning. Similar to our findings, Turan et al. reported that NT-proBNP was predictive of myocardial injury in children with CO poisoning [32]. In addition, in a study by Abdel Aziz et al. it was stated that troponin-I did not differ in patients with mild, moderate, or severe CO poisoning and that NT-proBNP was more valuable than troponin-I in demonstrating myocardial damage [33].
In a retrospective study in adults, Lee et al. looked for a cut-off value for creatinine kinase for the prediction of delayed neurological sequelae. They suggested that a creatine kinase concentration of > 1603 U/L was an independent risk factor for delayed neurological sequelae [34]. Tekşam et al. found elevated CK-MB values at admission in 11 of 16 patients with myocardial injury after CO poisoning [35]. In our study, CK and CK-MB were also higher in the group with myocardial damage at admission.
CO causes tissue hypoxia and immunological, inflammatory damage at the cellular level. This causes neutrophil activation, lymphocyte proliferation, mitochondrial dysfunction, lipid peroxidation, oxidative stress, inflammation, and apoptosis [1, 36]. In a study, WBC, CK, and AST were found to be higher in patients with CO poisoning and a worse prognosis [22]. We also reached similar results in our study. Han et al. reported that NLR was predictive for myocardial injury in patients with CO poisoning [37]. In a study by Akcan Yildiz in pediatric patients with CO poisoning, it was reported that a high leukocyte count was useful in predicting prognosis and directing treatment [38]. In a study by Moon et al. with adults with CO poisoning, it was shown that both admission neutrophil count and NLR were predictive of troponin-I elevation when used together with clinical parameters in patients without troponin-I elevation or ischemic ECG changes at admission [26]. In our study, the cost-effective WBC, neutrophil, NLR, IG, and IG% levels were found to be higher in the troponin-positive group with myocardial damage; furthermore, neutrophil count, IG, and IG% were positively correlated with troponin-I, CK, CK-MB, and proBNP. Therefore, full blood count parameters, which can be measured with low-cost and almost everywhere, can guide a clinician about treatment and referral when measuring troponin-I is not an option.
There is scarce evidence on hepatic injury due to CO poisoning. In their retrospective study with 894 adult patients, Kim et al. reported that 1.6% of patients had hepatic injury and 14.3% had subclinical hepatitis-related injury [23]. In our study with 107 pediatric patients, hepatic damage was not detected in the group with and without myocardial damage. However, a weak positive correlation was detected between ALT and NT-proBNP. In various previous studies, a positive correlation was found between ALT and NT-proBNP; it was suggested that increased NT-proBNP and ALT showed deterioration in cardiac functions, and it was stated that ALT increased in relation to oxidative stress [39, 40]. In a study of Jung et al. with adult patients, AST was found to be higher in patients with poor neurological prognosis [22]. The high positive correlation between AST and cardiac markers in our study may be significant in terms of poor prognosis.
There was no significant difference between the ECG findings of our patients with myocardial injury at the time of admission and after recovery. When the patients were evaluated according to their echocardiographic findings, LVEF and LVFS values were significantly lower at the time of admission compared to the control values after recovery. In a study conducted by Teksam et al. in pediatric patients with CO poisoning, it was reported that normal ECG findings did not exclude myocardial damage [35]. Therefore, it seems more important to study the cardiac markers of the patients. In a retrospective study of Kim et al. that enrolled patients with CO poisoning, LV EF was lower in the cardiomyopathy group [19]. Park et al. suggested that left ventricular systolic dysfunction was found in the echocardiography of adult patients with CO poisoning and was associated with a poor prognosis [24]. Similarly, in our study, lower LVEF and LVFS values seen in the troponin-positive group at admission suggested that systolic dysfunction also developed.
The main limitations of our study are its retrospective, single-center design as well as the small number of patients. In addition, it did not include the clinical outcomes of patients, such as neurological deficits. There is a need for well-designed large-scale prospective studies on this subject.
Conclusion
In our study, we are of the opinion that NT-proBNP is a diagnostic marker recognizing myocardial dysfunction at an early stage, and the cut-off value we determined will help clinicians to decide on treatment in children with CO poisoning. In settings where cardiac markers are not available, we hope that full blood count parameters, which are low-cost and can be studied everywhere, will guide patient treatment and referral.
Data availability
Available on request.
Declarations
Conflict of interest
The authors declare no competing interests.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Rose JJ Wang L Xu Q McTiernan CF Shiva S Tejero J Gladwin MT Carbon monoxide poisoning: pathogenesis, management, and future directions of therapy Am J Respir Crit Care Med 2017 195 596 606 10.1164/rccm.201606-1275CI 27753502
2. Chatterjee PK Water-soluble carbon monoxide-releasing molecules: helping to elucidate the vascular activity of the 'silent killer' Br J Pharmacol 2004 142 391 393 10.1038/sj.bjp.0705826 15148242
3. Gozubuyuk AA, Dag H, Kacar A et al (2017) Epidemiology, pathophysiology, clinical evaluation, and treatment of carbon monoxide poisoning in child, infant, and fetus. Northern clinics of Istanbul 4:100–107. 10.14744/nci.2017.49368
4. Hullin T Aboab J Desseaux K Chevret S Annane D Correlation between clinical severity and different non-invasive measurements of carbon monoxide concentration: a population study PLoS ONE 2017 12 e0174672 10.1371/journal.pone.0174672 28350859
5. Habertürk (27.01.2022) Kar, kombinin baca borusunu tıkadı: 5 kişilik aile zehirlendi, baba öldü Kahramanmaraş’ta kar baca borusunu tıkayınca, Demir ailesi kombiden çıkan karbonmonoksit gazından zehirlendi. Haber Türk
6. Cukor J Restuccia M Carbon monoxide poisoning during natural disasters: the Hurricane Rita experience J Emerg Med 2007 33 261 264 10.1016/j.jemermed.2007.02.043 17976553
7. Amsel J Soden KJ Sielken RL Jr Valdez-Flora C Observed versus predicted carboxyhemoglobin levels in cellulose triacetate workers exposed to methylene chloride Am J Ind Med 2001 40 180 191 10.1002/ajim.1086 11494346
8. Kwon OY Chung SP Ha YR Yoo IS Kim SW Delayed postanoxic encephalopathy after carbon monoxide poisoning Emergency medicine journal : EMJ 2004 21 250 251 10.1136/emj.2002.002014 14988368
9. Jüttner B, Busch HJ, Callies A et al (2021) S2k guideline diagnosis and treatment of carbon monoxide poisoning. Ger Med Sci: GMS e-journal 19:Doc13. 10.3205/000300
10. Chang YC Lee HY Huang JL Chiu CH Chen CL Wu CT Risk factors and outcome analysis in children with carbon monoxide poisoning Pediatr Neonatol 2017 58 171 177 10.1016/j.pedneo.2016.03.007 27502424
11. Cho Y Kang H Oh J Lim TH Ryu J Ko BS Risk of venous thromboembolism after carbon monoxide poisoning: a nationwide population-based study Ann Emerg Med 2020 75 587 596 10.1016/j.annemergmed.2019.08.454 31759754
12. Cha YS Cha KC Kim OH Lee KH Hwang SO Kim H Features and predictors of myocardial injury in carbon monoxide poisoned patients Emergency medicine journal : EMJ 2014 31 210 215 10.1136/emermed-2012-202152 23349356
13. Nakagawa Y Nishikimi T Kuwahara K Atrial and brain natriuretic peptides: hormones secreted from the heart Peptides 2019 111 18 25 10.1016/j.peptides.2018.05.012 29859763
14. Alvarez Villela M, Wever-Pinzon O, Parikh M et al (2020) Patterns of cardiac dysfunction after carbon monoxide poisoning. Undersea Hyperb Med: Journal of the Undersea and Hyperbaric Medical Society Inc 47(3):477–485. 10.22462/03.07.2020.9
15. Cardiga R, Proença M, Carvalho C et al (2015) What do we know about carbon monoxide poisoning and cardiac compromise? Revista portuguesa de cardiologia : orgao oficial da Sociedade Portuguesa de Cardiologia = Portuguese journal of cardiology : an official journal of the Portuguese Society of Cardiology 34:557.e551–555. 10.1016/j.repc.2015.01.006
16. Park JH, Heo R, Kang H et al (2020) Diagnostic performance and optimal cut-off values of cardiac biomarkers for predicting cardiac injury in carbon monoxide poisoning. Clin Exp Emerg Med 7:183–189. 10.15441/ceem.19.072
17. Chi YJ Pan HY Cheng FJ Chang YI Chuang PC Experience of carbon monoxide poisoning and the outcome predicting score: a multicenter retrospective study Am J Emerg Med 2022 58 73 78 10.1016/j.ajem.2022.05.012 35636046
18. Hoffmann F Schmalhofer M Lehner M Zimatschek S Grote V Reiter K Comparison of the AVPU scale and the pediatric GCS in prehospital setting Prehosp Emerg Care 2016 20 493 498 10.3109/10903127.2016.1139216 26954262
19. Kim JS, Ko BS, Sohn CH et al (2020) High-sensitivity troponin I and creatinine kinase-myocardial band in screening for myocardial injury in patients with carbon monoxide poisoning. Diagnostics (Basel, Switzerland) 10. 10.3390/diagnostics10040242
20. Simonsen C Magnusdottir SO Andreasen JJ Wimmer R Rasmussen BS Kjaergaard B Maltesen RG Metabolic changes during carbon monoxide poisoning: an experimental study J Cell Mol Med 2021 25 5191 5201 10.1111/jcmm.16522 33949122
21. Hampson NB Hauff NM Risk factors for short-term mortality from carbon monoxide poisoning treated with hyperbaric oxygen Crit Care Med 2008 36 2523 2527 10.1097/CCM.0b013e31818419d8 18679118
22. Jung JW, Lee JH (2019) Serum lactate as a predictor of neurologic outcome in ED patients with acute carbon monoxide poisoning. Am J Emerg Med 37:823–827. 10.1016/j.ajem.2018.07.046
23. Kim SJ Oh HS Cha YS Kim MY Kim H Evaluation of hepatic injury in acute carbon monoxide-poisoned patients in emergency department Hum Exp Toxicol 2020 39 883 889 10.1177/0960327120909521 32129094
24. Park JS Seo KW Choi BJ Choi SY Yoon MH Hwang GS Tahk SJ Choi SC Min YG Shin JH Various echocardiographic patterns of left ventricular systolic dysfunction induced by carbon monoxide intoxication Cardiovasc Toxicol 2016 16 361 369 10.1007/s12012-015-9347-6 26498469
25. Moon JM Shin MH Chun BJ The value of initial lactate in patients with carbon monoxide intoxication: in the emergency department Hum Exp Toxicol 2011 30 836 843 10.1177/0960327110384527 20876159
26. Moon JM Chun BJ Cho YS Lee SM Diagnostic value of parameters related to white blood cell counts for troponin I elevation in CO poisoning Cardiovasc Toxicol 2019 19 334 343 10.1007/s12012-018-09501-w 30610672
27. Moon JM Chun BJ Shin MH Lee SD Serum N-terminal proBNP, not troponin I, at presentation predicts long-term neurologic outcome in acute charcoal-burning carbon monoxide intoxication Clin Toxicol (Phila) 2018 56 412 420 10.1080/15563650.2017.1394464 29067847
28. Lee JH Kim HS Park JH Kim MS Sun BJ Ryu S Kim SS Jin SA Kim JH Choi SW Jeong JO Kwon IS Seong IW Incidence and clinical course of left ventricular systolic dysfunction in patients with carbon monoxide poisoning Korean circulation journal 2016 46 665 671 10.4070/kcj.2016.46.5.665 27721858
29. Güllü UU İpek S Dalkıran T Dinçer S Yurttutan S Aynacı E The role of ProBNP on prognosis in scorpion stings Wilderness & environmental medicine 2021 32 137 142 10.1016/j.wem.2021.01.015 33994108
30. İpek S Gungor S Güllü UU Dalkıran T Mercan M Demiray Ş Gürbüz Y Snakebites in pediatric patients in Kahramanmaraş: is pro-brain natriuretic peptide a prognostic biomarker for snakebites? Cureus 2022 14 e21570 10.7759/cureus.21570 35228929
31. Güllü UU Güngör Ş İpek S Yurttutan S Dilber C Predictive value of cardiac markers in the prognosis of COVID-19 in children Am J Emerg Med 2021 48 307 311 10.1016/j.ajem.2021.06.075 34246919
32. Turan C Dogan E Yurtseven A Saz EU Usefulness of N-terminal pro-B-type natriuretic peptide (NT-ProBNP) as a marker for cardiotoxicity and comparison with echocardiography in paediatric carbon monoxide poisoning Cardiol Young 2020 30 1103 1108 10.1017/s1047951120001651 32571446
33. Abdel Aziz MH El Dine F Hussein H Abdelazeem AM Sanad IM Prediction of troponin I and N-terminal pro-brain natriuretic peptide levels in acute carbon monoxide poisoning using advanced electrocardiogram analysis, Alexandria Egypt Environmental science and pollution research international 2021 28 48754 48766 10.1007/s11356-021-14171-3 33929669
34. Lee H Kang H Ko BS Oh J Lim TH Cho Y Initial creatine kinase level as predictor for delayed neuropsychiatric sequelae associated with acute carbon monoxide poisoning Am J Emerg Med 2021 43 195 199 10.1016/j.ajem.2020.02.054 32165069
35. Teksam O Gumus P Bayrakci B Erdogan I Kale G Acute cardiac effects of carbon monoxide poisoning in children European journal of emergency medicine : official journal of the European Society for Emergency Medicine 2010 17 192 196 10.1097/MEJ.0b013e328320ad48 20195157
36. Roderique JD Josef CS Feldman MJ Spiess BD A modern literature review of carbon monoxide poisoning theories, therapies, and potential targets for therapy advancement Toxicology 2015 334 45 58 10.1016/j.tox.2015.05.004 25997893
37. Han YY, Wang Y, Zhao GQ et al (2018) Relationship between neutrophil-to-lymphocyte ratio and myocardial injury induced by acute carbon monoxide poisoning. Zhonghua lao dong wei sheng zhi ye bing za zhi = Zhonghua laodong weisheng zhiyebing zazhi. Chin J Ind Hyg Occup Dis 36:362–364. 10.3760/cma.j.issn.1001-9391.2018.05.010
38. Akcan Yildiz L Gultekingil A Kesici S Bayrakci B Teksam O Predictors of severe clinical course in children with carbon monoxide poisoning Pediatr Emerg Care 2021 37 308 311 10.1097/pec.0000000000001580 30106865
39. Sato YZ Molkara DP Daniels LB Tremoulet AH Shimizu C Kanegaye JT Best BM Snider JV Frazer JR Maisel A Burns JC Cardiovascular biomarkers in acute Kawasaki disease Int J Cardiol 2013 164 58 63 10.1016/j.ijcard.2011.06.065 21777987
40. Zhang L Hou J Ma FZ Li J Xue S Xu ZG The common risk factors for progression and mortality in COVID-19 patients: a meta-analysis Adv Virol 2021 166 2071 2087 10.1007/s00705-021-05012-2
| 36445626 | PMC9707252 | NO-CC CODE | 2022-12-01 23:20:22 | no | Ir J Med Sci. 2022 Nov 29;:1-10 | utf-8 | Ir J Med Sci | 2,022 | 10.1007/s11845-022-03232-2 | oa_other |
==== Front
Econ Change Restruct
Economic Change and Restructuring
1573-9414
1574-0277
Springer US New York
9465
10.1007/s10644-022-09465-6
Article
Assessment of environmental tax and green bonds impacts on energy efficiency in the European Union
Li Ying [email protected]
1
Zhou Mi [email protected]
2
Sun Huaping [email protected]
3
Liu Jia [email protected]
4
1 grid.464478.d 0000 0000 9729 0286 School of Economics, Tianjin University of Commerce, Tianjin, China
2 grid.216938.7 0000 0000 9878 7032 College of Economic and Social Development, Nankai University, Tianjin, China
3 grid.440785.a 0000 0001 0743 511X School of Finance and Economics, Jiangsu University, Zhenjiang, China
4 grid.464479.c 0000 0000 9459 2326 School of Accounting, Tianjin University of Finance and Economics, Tianjin, China
29 11 2022
119
31 7 2022
15 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Owing to the increased greenhouse gas emissions and threat of environmental pollution, scholars have drawn attention to the issue of energy efficiency in recent decades. One of the main practical policies related to energy efficiency is deploying green tax and green financing tools. The main purpose of this paper is to analyze the impacts of issued green bonds and environmental taxes on energy efficiency in the 27 European Union (EU) member states from 2010 to 2021. The major results confirm that green bonds accelerate the process of improvement of the level of energy efficiency in EU members. In the EU, environmental taxation is an efficient fiscal policy to encourage enterprises to conduct different policies and projects to improve their energy efficiency levels. In addition, a 1% increase in the gross domestic product leads to an increase in energy efficiency of the EU by nearly 0.39%. The recommended practical policies are promoting the green bond market through more transparency and marketing among private investors, paying attention to the concept of green economic recovery, and issuing digital green bonds (DGBs).
Keywords
Green bonds
Green fiscal policy
European Union
Energy efficiency
JEL Classification
N10
O14
O23
C14
National Statistical Science Research Project2022LY067 Li Ying
==== Body
pmcIntroduction
Energy efficiency has become an essential concept in the energy economics literature. The simple meaning of energy efficiency is using less energy input to produce goods and services. Petterson (1996), Ning et al. (2022), and Bianco and Marmori (2022) defined energy efficiency as a wide range of policies to increase energy efficiency in the process of commodities production. Popkova and Sergi (2021) argued that energy efficiency had become a multifaceted and complex concept in recent decades, and countries should understand the different aspects of this concept with deep research. During recent decades, the challenges of lack of capital to invest in the field of energy efficiency, inefficient policies of the government in the field of supporting energy efficiency projects, the spread of COVID-19, severe fluctuations in global oil prices, etc., have caused the slow movement of improving energy efficiency in different countries (Rasoulinezhad and Taghizadeh-Hesary 2022; Phung et al. 2022). Bergman and Foxon (2020) expressed that despite the desire of countries to improve energy efficiency indicators, this issue has become challenging and complex. According to the Energy Efficiency Report 2021 (IEA), the energy efficiency index has experienced a 2.3% improvement between 2011 and 2016. Since 2016, due to various global problems and challenges, the countries of the world have not been able to improve significantly in this field. Figure 1 illustrates the improvement rate of energy intensity (an appropriate proxy for energy efficiency) from 2011 to 2020:Fig. 1 Energy intensity enhancement rate, 2011–2020, %.
Source Authors based on the Energy Efficiency Report IEA 2020
One of the reasons why policymakers pay much attention to the issue of the effectiveness of energy usage is the importance of improving this index in protecting the environment and eliminating the global threats of global warming or climate change (Saboori et al. 2022). In other words, the more energy efficiency improves, the amount of energy consumption will decrease, and the efficiency of an energy unit in the production of goods will increase. This will cause less emission of greenhouse gases or, in other words, environmental pollution. Li et al. (2022) confirmed that a 1% improvement in energy efficiency reduces carbon dioxide emissions by nearly 0.92%. In another study, Mirza et al. (2022) depicted the significant contributions of energy efficiency to CO2 mitigation because any improvement in energy efficiency indicators can help a country use fewer fossil fuels and lower energy intensity, causing less carbon dioxide emission. It can be said that an increase in energy efficiency can lower the amount of energy needed to generate power and produce different commodities and residential consumption, which has a practical consequence of the reduction of greenhouse gas emissions.
Despite the importance and necessity of improving the energy efficiency index, the countries of the world have not been able to achieve the desired level of this index in recent years. One of the reasons for this is the use of inappropriate policies and tools in improving energy efficiency and developing projects related to this index. Many researchers like Tu et al. (2020) and Rasoulinezhad (2022) have indicated that to develop energy efficiency projects countries should move toward developing the market of green financing instruments such as green bonds. Another group of previous studies (e.g., Yoshino et al. 2021; Chien et al. 2021) insists that proper use of green financial policies, such as establishing a green (carbon) tax, can open the way to improve the energy efficiency index in the countries of the world. Which tool among green bonds or environmental tax impacts energy efficiency is vital for research. Because the policy makers for each of these two policies need to spend money and have formal planning, so knowing the sign and magnitude of the effect of these two policies on the effectiveness of energy can be a useful and practical finding for the policymakers and managers of the sustainable development of countries.
The necessity of this research is of great importance for EU countries. The member states of the EU are known as pioneers in the field of sustainable development and achieving a zero-carbon economy (according to the EU climate-neutral by 2050 plan) in the world. Therefore, studying the effects of green bonds and green taxes on the effectiveness of energy in these countries can provide strategic and practical results. On the other hand, due to the import of fossil fuels and dependence on foreign countries in this field, EU members are trying to implement strategies to reduce dependence on fossil fuels. One of the main strategies in this field is to improve energy efficiency, which can lead to less consumption of fossil fuels (Akram et al. 2020). Figure 2 shows the dependency of the EU on energy imports from abroad from 1990 to 2020.Fig. 2 Dependency of EU on energy imports, 1990–2020, %.
Source Authors based on Eurostat
According to the figure above, it is clear that the EU member states have become more than 50% dependent on energy imports from abroad in recent years, which shows the importance of prioritizing the improvement of energy efficiency in these countries.
Regarding the problem mentioned above and motivations to study the issue of energy efficiency and the impacts of green tax and green bonds on it for the case of the EU, this paper tries to contribute to earlier studies from the following aspects:First, the paper seeks to model energy efficiency by considering green bonds and green tax as explanatory variables. This type of energy efficiency modeling allows the researchers to explore the impacts of both variables of green tax and green bonds on energy efficiency.
Second, the panel causality test analyzes the causal relationships between energy efficiency-green bonds and energy efficiency-green tax. It provides the direction of causal linkages among these variables, bringing in-detail results for policymakers.
Third, an advanced econometric analysis based on the existence of cross-sectional dependency among series is employed to estimate the coefficients of independent variables. Following Feng et al. (2021), cross-sectional dependence is a common characteristic in the panel data approach that researchers should address.
The research structure is as follows: In the second Section, the authors try to clarify the literature gap the paper seeks to fill. Next Section debates data and estimation strategy. Section 4 provides empirical estimations results and a brief discussion. Section 5 presents robustness checks to ascertain the validity of empirical findings. The concluding remarks are provided in the final section, and some practical policies are recommended.
Literature review
A vast number of scholars have drawn the energy efficiency matter attention. For instance, in a study, Petterson (1996) highlighted the importance of energy efficiency to support of environmental protection. Lopes et al. (2012) argued that energy efficiency is a concept of an energy economy with social properties. Because households are an essential player in energy efficiency, consumption patterns can improve/weaken energy efficiency in a country. Backlund et al. (2012) revealed the need to invest in energy efficiency technologies that require appropriate policies and governance. Zakari et al. (2022) addressed the concept of energy efficiency as the main pillar of sustainable development. They argued that improvement in energy efficiency could bring countries closer to the sustainable development indicators defined by the United Nations in 2015. In another study, Hassan et al. (2022) investigated the relationship between energy efficiency and air pollution reduction in OECD countries. The results of this research show the importance of energy effectiveness in operationalizing the concept of energy transition (moving toward less consumption of fossil fuels) and reaching a zero-carbon economy. Baniassadi et al. (2022) mentioned the role of energy efficiency improvement in the living conditions of the next generations on the earth. Any increase in the level of energy efficiency can lower greenhouse emissions. In another new study, Bertoli (2022) expressed that in post-COVID era, countries should implement practical policies to enhance energy efficiency and sufficiency to reach the goals of sustainable development.
A group of studies has concentrated on the importance of energy efficiency in reducing CO2 emissions. For instance, Li and Lin (2015) highlighted the reasons for energy inefficiency in China and remarked on the role of fiscal policy and investment motivations in promoting energy efficiency. Iftikhar et al. (2016) analyzed data for 26 countries and concluded that improving energy efficiency in the industrial sector will reduce the consumption of non-renewable energy and, as a result, reduce the emission of greenhouse gases such as carbon dioxide. Moretto et al. (2018) argued that energy efficiency is a crucial concept to environmental protection. Mirza et al. (2022) studied 30 developing economies and found out the significant role of energy efficiency in reducing CO2 emissions. Regarding the case of China, Wang and Jia (2022) investigated the relationship between CO2 emissions and energy efficiency in a panel of Chinese provinces. The major results confirmed the big contribution of energy efficiency to carbon dioxide emissions and sustainable economic growth.
Another strand of literature focused on the impacts of green tax and green bonds on the process of enhancement of energy efficiency in different economies. Miguel and Manzano (2011) discussed the necessity of green tax reforms to enhance the energy efficiency, energy transition, green recovery, and environmental protection. Silva et al. (2020) investigated the consequences of carbon tax policy on green energy promotion and lowering fossil fuel consumption. They confirmed the significant coefficient of green tax to impact green energy consumption, which is a way to improve energy efficiency. He et al. (2021) and Wang and Yu (2021) tried to seek the influence of environmental tax on the enhancement of energy efficiency levels in the OECD (Organization for Economic Co-operation and Development) economies. They found out that the impact is not clear and depends on countries’ economic size. In other words, in a large economic size country, environmental tax has a higher potential to accelerate energy efficiency improvement. GarciaQuevedo and Jove-Liopis (2021) considered four different environmental policies in Spain and analyzed their impacts on energy efficiency investment. The estimation results depicted the low effect of green tax on energy efficiency investment in the industrial sector. Goenka et al. (2021) discussed the positive impacts of green taxes on decarbonization and green economic recovery progress. In another study, Bashir et al. (2021) aimed to seek the impacts of green tax on energy intensity for the case of OECD economies. The concluding remarks highlighted the positive role of an environmental tax on the improvement of energy efficiency and energy usage. Employing the game theory approach, Li et al. (2021) confirmed the scenarios of efficiency of carbon tax policies to the reduction of energy usage to produce different commodities.
Regarding green bonds, Devine and McCollum (2022) expressed that green bonds can provide sufficient capital for countries to develop green energy projects and promote technologies to improve energy efficiency. In another study, Quang and Thao (2022) focused on the ASEAN member countries and found that the issued green bonds can reduce energy intensity (energy usage/GDP). This finding has been confirmed by Mamun et al. (2022), who also studied the relationship between green finance and the decarbonization process in 46 different nations. They showed long-term and short-term positive impacts of green financing on energy efficiency. Tan et al. (2022) investigated the different impacts of green tax on economic recovery. They found that the success of environmental tax depends on the appropriate tax based on the tax preferences in developing economies.
Regarding the literature gaps
By summarizing the results of the above studies, it can be realized that energy efficiency is an essential matter for the countries of the world in order to achieve sustainable development and green economic growth recovery during the COVID-19 and post-COVID-19 period.
Second, governments’ green policies (green financing or environmental taxation) affect energy efficiency differently.
This article will try to analyze the effect of green bonds and green taxes on the energy efficiency of EU member states. Based on the knowledge of the authors, such a study has not been done so far, so this article will fill in this literature gap.
Data and estimation strategy
Energy efficiency enhancement can be achieved through the implementation of various policies and strategies by the government. One of the new and effective policies in this field is the issuance of green bonds as a green financing tool by the government. By targeting projects related to the environment, including energy efficiency, green bonds can appropriately solve the lack of sufficient capital to invest in these projects and create appropriate capital accumulation from private sector investors and foreign investors. Another effective tool of the government for developing energy efficiency projects is the imposition of a green tax (environmental tax) on activities that cause environmental pollution. The imposition of such a tax leads to economic enterprises developing energy efficiency to reduce fossil energy consumption.
In terms of economic theory, the utility function of an economic unit to invest in a project related to energy efficiency can be assumed as Eq. 1:1 Ut=Urt,σt,taxt=rt-βσt2+δtaxt+Zt
where the rate of return of investment in an energy efficiency project is showed by r, while σ and β represent the risk weight and total risk for the project is related to energy efficiency. Tax denotes environmental tax, whereas Z is representative of all the variables (such as inflation rate, exchange rate, governance, carbon dioxide emissions per capita, geopolitical risk, etc.) that can affect the economic unit's desirability to invest in energy efficiency projects.
It is important to note that other factors (Z) have many direct and indirect effects on decision-making and the desirability of the economic unit and risk, the investment return, and the green tax rate. For example, increasing carbon dioxide per capita will increase the green tax rate to pressure economic enterprises to improve their energy efficiency. Or, an increase in geopolitical risk in a country greatly increases the risk of investing in low-return or late-return projects. Therefore the desire to invest in energy efficiency projects, which are usually late-return or low-return projects, will decrease. In this situation, the tool of green bonds, which according to Markus and Adriana (2018)’s findings, significantly reduces the risk of green projects, and can increase the desirability of the economic unit to participate in energy efficiency projects.
If it is assumed that an economic unit invests in the project related to energy efficiency through bank facilities (with an interest rate of r), and buys issued green bonds (with an interest rate (rtGBOND), then the total risk this investor is Eq. 2 will be:2 rt=αt.rtD+1-αt·rtGBOND
With a simple mathematical operation, Eq. 2 can be inserted into the first equation, and the result is Eq. 3:3 Ut=Urt,σt,taxt=αt.rtD+1-αt.rtGBOND-βσt2+δtaxt+Zt
Based on Eq. 3 and previous studies in the energy efficiency field, the energy efficiency index is the dependent variable, and the variables of issued green bonds, environmental tax, carbon dioxide emissions, gross domestic product, green energy consumption, and inflation as explanatory variables were selected. The increase in the volume of issued green bonds is expected to lead to energy efficiency improvement due to the creation of favorable investment conditions. Environmental tax can also be a deterrent to the consumption of more fossil fuels by economic enterprises. Therefore, with the imposition and increase in green taxes, economic enterprises will be encouraged to develop more energy efficiency projects. Carbon dioxide emissions can also positively or negatively affect energy efficiency. With the increase in carbon dioxide emissions, governments will try to reprogram and pressure economic enterprises to develop green activities.
On the other hand, more carbon dioxide emissions can have a causal relationship with energy efficiency and increase carbon dioxide emissions. Carbon means the economic growth of most countries through the consumption of fossil fuels and not paying attention to sustainable development. Therefore, an increase in carbon dioxide emissions can mean a decrease in energy efficiency in a country. In the case of GDP, which can mean the size of the economy and the economic power of a country, it is expected to have a positive effect on the energy efficiency index of a country. Because of the increase in GDP in a country, the ability of the government and the private sector to invest in green projects such as energy efficiency has increased, and it is expected that the energy efficiency index will improve. Also, the consumption of clean energy can significantly affect energy efficiency. With the increase in clean energy consumption in a country, the energy efficiency index is expected to increase. And finally, the inflation rate is expected to harm the development of energy efficiency since any increase in the general level of prices can lead to a rise in the cost of projects, technology imports, and energy transition.
In this research, to find out how issued green bonds and environmental taxes can change energy efficiency index, the case of 27 EU member countries is selected, and the econometric analysis is done through the panel data framework of these economies from 2010 to 2021. Hence the total observations for the estimation are 324 (12*27). The econometric equation (Eq. 4) based on the selected variables, period, and case study can be organized as follows:4 ENEFi,t=β0+β1GBONDi,t+β2GTAXi,t+β3INFi,t+β4GDPi,t+β5COi,t+β6GECi,t+εit
In Eq. 4, ENEF stands for energy efficiency index. GBOND, GTAX, INF, GDP, CO, and GEC represent issued green bonds, environmental tax, inflation, gross domestic product, carbon dioxide emissions, and green energy consumption. Initial introduction of the selected variables is represented in Table 1.Table 1 Introduction of the variables.
Source Authors
Variable symbol Unit Sources
Energy efficiency index of industry (ODEX) ENEF – European environmental agency
Issued Green bond GBOND US$ Climate bond initiative
Inflation, consumer prices INF Annual % World bank
Green tax GTAX Million EURO EUROSTAT
CO2 emissions CO Metric tons per capita BP
GDP GDP Current US $ World bank
Green energy consumption GEC Exajoules BP
Table 2 reports the variables amount in the EU from 2010 to 2021. According to the data in Table 2, it is clear that the energy efficiency index has decreased compared to 2010 despite the improvement of the situation from 2015 to 2021. Issuance of green bonds has grown exponentially in the EU, from 12 billion US dollars in 2010 to more than 260 billion US dollars in 2021. Also, the income of EU members from environmental taxes increased from 2010 to 2021, from about 251 billion euros in 2010 to more than 315 billion euros in 2021. Also, renewable energy consumption in the EU has made significant progress and reached from 0.65 exajoules in 2010 to about 7.5 exajoules in 2021.Table 2 Variables amounts for the EU, 2010–2021.
Source Authors
Variable Unit 2010 2015 2020 2021
Energy efficiency index of industry (ODEX) Index 1990 = 100 74 70 70 72
Issued green bond Billion US$ 12 20 158.2 264.9
Inflation, consumer prices Annual % 1.5 −0.1 0.5 2.6
Green tax revenue Billion EURO 251.6 284.5 299.9 315.7
CO2 emissions per capita Metric tons per capita 7.3 6.5 3.9 4.6
GDP Current Trillion US $ 14.56 13.55 15.3 17.09
Green energy consumption Exajoules Input-equivalent 0.65 6.02 6.85 7.50
In order to know the initial dimensions of the variables and their relationship with each other, several pre-tests should be performed before econometric estimation. The first required test is to study the existence or non-existence of cross-sectional dependence among the variables of the model. In order to study this matter, two Breusch-Pagan LM test (Breusch and Pagan 1980) and Pesaran CD test are implemented. If there is an inter-departmental dependence between the variables, it can be realized that for the next test, which includes the study of the existence of the unit root, an appropriate method should be used. In this research, the Cross- sectionally augmented IPS test (CIPS) method, proposed by Pesaran (2007), is employed as a reliable and efficient method. This method has the following equation (Eq. 5):5 CIPSN,T=N-1∑i=1NtiN,T
If the panel data unit root test shows the existence of the mean of the model variables after differentiating once, it is possible to perform the co-integration relationship test between the model variables. For this purpose, Westerland’s (2007) method, which has four co-integration test statistics (Eqs. 6–9), is used.6 Pr=α^iSEα^i
7 Pα=Tα^
8 Gr=1N∑i=1Nα^iSEα^i)
9 Gα=1N∑i=1NTα^iα^i1)
In the above equations, α^i, SEα^i, α^i1 represent the estimation of the error correction parameter, standard error, and 1-∑j=1piα^ij, respectively.
In the next step, the empirical estimations of coefficients of explanatory variables are conducted through a continuous-updated fully modified (CUP-FM) estimator proposed by Bai and Kao (2006). Finally, the causality relationships between the variables are evaluated by the panel causality test method of Demitrescu and Hurlin (2012).
Empirical findings and discussion
In the first step, two cross-sectional dependence tests were performed, and the related findings, reported in Table 3, revealed that there is cross-sectional dependency among variables.Table 3 Result of test of cross-sectional dependency.
Source Authors
Breusch-Pagan LM Pesaran CD
401.592 (0.00) 34.103 (0.00)
Hence, the appropriate panel data unit root test is the CIPS that its findings are represented in Table 4 as follows:Table 4 Results of panel unit root test.
Source Authors
Variable statistic
ODEX
D (ODEX)
−0.664
−4.920
Green bonds
D(Green bonds)
−1.058
−4.782
Inflation
D(Inflation)
−2.119
−4.093
Green tax
D(Green tax)
−1.033
−4.804
CO2 emissions per capita
D(CO2 emissions per capita)
−1.004
−4.111
GDP
D(GDP)
−1.132
−5.860
Green energy consumption
D (Green energy consumption)
−0.948
−4.505
D shows the first difference
The CIPS panel unit root test results show that all the variables become stationary after the first difference, so there is a possibility of a long-term relationship between the variables. The Westerlund (2007) co-integration approach is employed to determine whether there is any long-term relationship between variables. The results of this test are expressed in Table 5 and confirm the existence of co-integration in the empirical model.Table 5 Results of panel co-integration test.
Source Authors
Tests Stat. (prob.)
Pr −9.834 (0.00)
Pα −13.138 (0.00)
Gr −4.014 (0.00)
Gα −17.804 (0.00)
According to the conducted tests and their findings, especially the presence of cross-sectoral dependence, the CUP-FM estimator was used to estimate the coefficients of the explanatory variables. The results of the implementation of this estimator are listed in Table 6:Table 6 Estimation of coefficients.
Source Authors
Explanatory variable coefficient p-value
Issued green bond 0.831 0.035
Inflation, consumer prices −0.044 0.000
Green tax revenue 0.052 0.061
CO2 emissions per capita −0.04 0.204
GDP 0.397 0.000
Green energy consumption 0.518 0.029
According to Table 6, the coefficient of issued green bonds positively affects the energy efficiency enhancement in the EU, expressing that by a 1% rise in the volume of issued green bonds in the EU, energy efficiency will be improved by nearly 0.83%. Therefore, for the EU member states, the green bond instrument is fruitful in improving the energy efficiency level, which is an essential factor for reaching a suitable level of environmental protection. It is consistent with Devine and McCollum (2022) and Ning et al. (2022), who confirmed that issued green bonds are a fruitful instrument for developing projects related to energy efficiency improvement. In addition, the positive impacts of green bonds on energy efficiency improvement have been confirmed by Rasoulinezhad and Taghizadeh-Hesary (2022) for the case of the top ten economies that support green finance by Zhang et al. (2022) for the case of E-7 economies. The inflation rate has a negative impact meaning that any increase in the general level of prices of commodities in the EU member countries is an obstacle to improving energy efficiency. The main reason is that inflation increases the cost of imports, the price of commodities in the local markets, affecting the expenditure required to run the economic projects and energy efficiency. Consequently, it reduces profit and return on investment. It is in line with Tang et al. (2022), who found out negative impacts of inflation on enterprise-environmental protection expenditure in the case of China. Carbon dioxide emissions have a statistically insignificant coefficient. It can be interpreted that in the EU, the issue of environmental protection through different policies like energy coefficient has become an accepted and integrated program and task in the EU, which does not affect by CO2 emissions. The environmental tax has a positive and statistically significant (at 1% level) coefficient, mentioning that the level of energy efficiency in the EU rises by nearly 0.05% by a 1% increase in green tax revenue. This result is consistent with Telatar and Birinci’s (2022) findings for the case of Turkey. The critical point here is that the magnitude of the coefficient of green bonds is larger than the coefficient of green tax, meaning that the impact of issued green bonds on energy efficiency improvement in the EU is more than the impact of green tax. The gross domestic product positively impacts energy efficiency, denoting that with a 1% rise in the economic size of the EU, the energy efficiency improves by approximately 0.39%. The increase in the production volume of EU economies means the availability of more capital to advance energy efficiency projects and increase the speed of the energy transition process. Finally, the estimation results reveal the positive and significant coefficient of green energy consumption, highlighting the positive impact of renewable energy consumption on energy efficiency improvement in the EU. The development of renewable energy consumption (in industry, electricity production, household consumption, and vehicle fuel) in the economies of the EU leads to more attention to increasing the effectiveness of energy and the development of investment in the technology progress and financial support for patents and innovative ideas in the field of energy efficiency in the EU.
To have more in-detail insights about the relationship between issued green bonds- energy efficiency and environmental tax energy efficiency, the Dumitrescu and Hurlin (2012) panel causality test, with the results shown in Table 7, is conducted.Table 7 Dumitrescu and Hurlin’s causality test.
Source Authors
Green bond–energy efficiency causality relationship Issued green bonds do not homogeneously cause energy efficiency 3.014 (0.02)
Energy efficiency does not homogeneously cause issued green bond 0.284 (0.48)
Green tax–energy efficiency causality relationship Green tax does not homogeneously cause energy efficiency 4.699 (0.04)
Energy efficiency does not homogeneously cause green tax 3.683 (0.00)
p values are shown in parenthesis
According to the results and the p-values, it can be concluded that a uni-directional causality relationship exists between the volume of issued green bonds and the level of improvement of energy efficiency in the EU. It can be highlighted that there is a bi-directional causal link between green tax and energy efficiency in the EU. The main reason for this finding is that, generally, tax is an instrument for applying fiscal policy and as a means of regulation; it has impacted and is affected by energy efficiency. The governments of EU member states can adopt the green tax rate based on the level of energy efficiency and play a role as a regulator in this sector by changing the green tax rate according to the enhancement/deterioration of the energy efficiency level in the EU.
Robustness check
In order to ensure the reliability of the obtained results, two robustness tests are performed.The first test is based on the strategy of changing the dependent variable. In this way, energy intensity, one of the essential proxies of energy effectiveness, is used instead of the Energy efficiency index of industry (ODEX). Energy intensity data were extracted from the World Bank website for EU countries, and coefficients were re-estimated. The results for only issued green bonds and environmental tax are reported in 8:
According to Table 8, it is clear that the sign of the coefficients of the two variables of published green bonds and environmental tax is positive and is in full agreement with the previous estimation results.2. The second robustness check is based on the strategy of changing the sample size. For this purpose, the coefficients of the explanatory variables were estimated for a subset of 20 EU member states. In re-estimation, all pre-tests were performed as before. Table 9 outlines the findings of the new estimation as follows:
Table 8 Robustness check (dependent variable: energy intensity).
Source Authors
Explanatory variable coefficient p-value
Issued Green bond 0.019 0.000
Green tax revenue 0.284 0.048
Table 9 Robustness check (changing the sample size).
Source Authors
Explanatory variable coefficient p-value
Issued green bond 0.116 0.009
Inflation, consumer prices −0.103 0.164
Green tax revenue 0.009 0.000
CO2 emissions per capita 0.031 0.113
GDP 0.109 0.001
Green energy consumption 0.483 0.003
According to the results shown in Table 9, despite some differences (lack of significance of coefficients of the inflation rate and per capita carbon dioxide emissions), the signs of the coefficients of issued green bonds and environmental taxes are positive and utterly consistent with the previous results. Therefore, based on the two robustness tests’ findings, it can be concluded that the results of the estimations in Sect. 4 have complete reliability and validity.
Conclusions and policy recommendations
Protecting the environment, which has been heavily polluted in recent decades due to economic growth and industrialization, has become a vital global matter. Most countries are trying to reduce and control the rate of environmental pollution by implementing policies over time to achieve a zero carbon economy target in a horizon of 30–50 years which would be a good signal for moving toward sustainable development. One of the anti-environmental pollution policies is to focus on the issue of energy efficiency to reduce fossil energy consumption and carbon dioxide emissions. This paper investigates the relationship between issued green bonds and environmental tax on the energy efficiency of the EU members. The methodology approach for econometric modeling is the panel data and annual data of variables from 2010 to 2021.
Conclusions
According to the major estimation findings, the concluding remarks can be highlighted as bellows:The energy efficiency level can be enhanced by issuing green bonds in the EU members. Since the efficiency of this green financing instrument has been confirmed in the energy efficiency of the EU, the member states of this union can rely more on the green bond market in the post-COVID era to provide sufficient financing for the development of energy efficiency projects. The development of the green bond market can, on the one hand, increase the number of green projects in EU countries and, on the other hand, significantly improve the flow of investment in green projects. The development of the green bond market means a more favorable outlook for sustainable development.
The second conclusion is that environmental taxation is an efficient fiscal policy that encourages enterprises to conduct different policies and projects to improve their energy efficiency levels. However, the power of the green tax on energy efficiency in EU members has been less than that of the green bonds issued in these countries. Therefore, to increase the positive effect of the green tax on the market, EU member states need to set more efficient green tax rates and adjust the regulations related to green taxation. The effectiveness of environmental tax requires the appropriate determination of green tax rates and the existence of transparent laws.
Third, inflation can be addressed as an obstacle to promoting energy efficiency in the EU. An increase in prices will lead to more expensive production of economic enterprises, more expensive economic projects, and a decrease in the expected profit of investors. Therefore, investors and economic enterprises (especially small and medium enterprises) will be less willing and able to pay attention and give priority to energy efficiency matter. It can be concluded that policy making and controlling the EU’s inflation rate to develop energy efficiency in the member countries of this union should be given more attention. Uncontrolled inflation can cause stagnation in the sector of green projects and the development of the green bond market.
Another concluding remark regards the gross domestic product as representing economic size and power. This research confirmed the positive impact of GDP on energy efficiency enhancement. In general, with the increased GDP of the member countries of the EU, the financial power of governments in defining and implementing green projects such as energy efficiency has increased (according to Phung et al. (2022)). GDP increases FDI volume accelerating green recovery in the post-COVID era, as well as the power of banks in granting green facilities and supporting energy efficiency projects.
Green energy consumption positively impacts energy efficiency in the EU. The increase in renewable energy consumption means a decrease in the share of fossil energy consumption in different sectors of the economy (industry, residential, transportation, and electricity generation) of the EU, which will cause more demand to improve energy efficiency in these countries.
There is a bi-directional causal link between green tax and energy efficiency in the EU. The governments of EU member states can adopt the green tax rate based on the level of energy efficiency and play a role as a regulator in this economic sector by changing the green tax rate according to the enhancement/deterioration of the energy efficiency level in the EU.
Practical policies and recommendations for further research
As practical policy implications, the EU member states can promote the green bond market through more transparency and marketing among private investors. Considering the development of the digital economy in the era of social distancing caused by the COVID-19 pandemic (Taghizadeh-Hesary et al. 2022), it is suggested to create an infrastructure for the issuance of digital green bonds (DGBs). This would reduce both the costs of issuing physical bonds and enables greater geographical scope for the participation of foreign investors in the energy efficiency projects of EU. Also, considering the efforts of the countries in the world, including the EU, in economic growth recovery in the post-COVID era, it is recommended to pay attention to the concept of green economic recovery. Following Phung et al. (2022), the green economic recovery is a new concept developed in the era of Covid-19. It refers to the return of the economic power of the countries to the level before the outbreak of the COVID-19 pandemic. In the shadow of green economic recovery, economic growth will be created by paying attention to environmental issues, one of which is the improvement of the energy efficiency index. Furthermore, developing projects related to the green generation of electricity (sustainable electricity proposed by Taghizadeh-Hesary et al. 2022) in the EU is another practical policy proposed by this research. Although the generation of green electricity in some countries of the EU has a history of several decades (For example, the contributions of green power generation in Germany have improved from 3.4% in 1990 to over 42% in 2019 to total power generation in the country), but considering the political tension between Russia and Ukraine since February 24, 2022, and the decrease in the security of the EU’s fossil energy supply from Russia, the need for faster and broader implementation of green electricity generation is essential.
In conducting this research, there were limitations, such as the lack of new data on environmental taxes and green financing tools, the lack of clarity about the effects of the war between Russia and Ukraine, and the spread of COVID-19 on the global economic mechanism. Despite the limitations in conducting this research, according to the authors of this paper, it has a good and worthy contribution to the previous literature related to energy efficiency, the green bond market, and environmental tax. Future researchers can obtain additional findings by considering the COVID-19 pandemic as a variable in the econometric model, using country-level analyzes, and evaluating the effectiveness of different green tax rates. Other statistical and econometric methods such as scenario building, futurology, and artificial neural networks are also suggested to predict the trends and mega-trends of the research variables.
Author contributions
We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all the authors.
Declarations
Conflict of interest
We declare that there is no conflict of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Akram R Chen F Khalid F Ye Z Majeed M Heterogeneous effects of energy efficiency and renewable energy on carbon emissions: evidence from developing countries J Clean Prod 2020 247 119122 10.1016/j.jclepro.2019.119122
Backlund S Thollander P Palm J Ottosson M Extending the energy efficiency gap Energy Policy 2012 51 392 396 10.1016/j.enpol.2012.08.042
Bai j, Kao C, Baltagi BH On the estimation and inference of a panel co-integration model with cross-sectional dependence Panel data econometrics: theoretical contributions and empirical applications 2006 Amsterdam Elsevier
Baniassadi A Heusinger J Gonzalez P Weber S Samuelson H Co-benefits of energy efficiency in residential buildings Energy 2022 238 121768 10.1016/j.energy.2021.121768
Bashir MF Benjiang MA Shahbaz M Shahzad U Vo XV Unveiling the heterogeneous impacts of environmental taxes on energy consumption and energy intensity: empirical evidence from OECD countries Energy 2021 226 120366 10.1016/j.energy.2021.120366
Bergman N Foxon T Reframing policy for the energy efficiency challenge: insights from housing retrofits in the United Kingdom Energy Res Soc Sci 2020 63 101386 10.1016/j.erss.2019.101386
Bertoli P Policies for energy conservation and sufficiency: review of existing policies and recommendations for new and effective policies in OECD countries Energy Build 2022 264 112075 10.1016/j.enbuild.2022.112075
Bianco V Marmori C Modelling the deployment of energy efficiency measures for the residential sector. The case of Italy Sustain Energy Technol Assess 2022 49 101777 10.1016/j.seta.2021.101777
Breusch T Pagan A The lagrange multiplier test and its applications to model specification in econometrics Rev Econ Stud, Oxf Univ Press 1980 47 1 239 253
Chien F Hsu C Zhang Y Tran T Assessing the impact of green fiscal policies and energy poverty on energy efficiency Environ Sci Pollut Res 2021 29 4363 4374 10.1007/s11356-021-15854-7
Devine A McCollum M Advancing energy efficiency through green bond policy: multifamily green mortgage backed securities issuance J Clean Prod 2022 345 131019 10.1016/j.jclepro.2022.131019
Dumitrescu EI Hurlin C Testing for granger non-causality in heterogeneous panels Econ Model 2012 29 4 1450 1460 10.1016/j.econmod.2012.02.014
Feng L Zhao P Ding Y Liu B Rank-based tests of cross-sectional dependence in panel data models Comput Stat Data Anal 2021 153 107070 10.1016/j.csda.2020.107070
Garcia-Queved J Jove-Liopis E Environmental policies and energy efficiency investments An Ind-Level Anal Energy Policy 2021 156 112461 10.1016/j.enpol.2021.112461
Goenka A Liu L Nguyen M COVID-19 and a green recovery? Econ Model 2021 104 105639 10.1016/j.econmod.2021.105639 34539023
Hassan T Song H Khan Y Kirikkaleli D Energy efficiency a source of low carbon energy sources? Evidence from 16 high-income OECD economies Energy 2022 243 123063 10.1016/j.energy.2021.123063
He P Sun Y Niu H Long C Li Sh The long and short-term effects of environmental tax on energy efficiency: perspective of OECD energy tax and vehicle traffic tax Econ Model 2021 97 307 325 10.1016/j.econmod.2020.04.003
Iftikhar Y He W Wang Z Energy and CO2 emissions efficiency of major economies: a non-parametric analysis J Clean Prod 2016 139 779 787 10.1016/j.jclepro.2016.08.072
Li K Lin B Metafroniter energy efficiency with CO2 emissions and its convergence analysis for China Energy Econ 2015 48 230 241 10.1016/j.eneco.2015.01.006
Li L Zhang S Cao X Zhang Y Assessing economic and environmental performance of multi-energy sharing communities considering different carbon emission responsibilities under carbon tax policy J Clean Prod 2021 328 129466 10.1016/j.jclepro.2021.129466
Li R Li L Wang Q The impact of energy efficiency on carbon emissions: evidence from the transportation sector in Chinese 30 provinces Sustain Cities Soc 2022 82 103880 10.1016/j.scs.2022.103880
Lopes M Antunes C Martins N Energy behaviours as promoters of energy efficiency: a 21st century review Renew Sustain Energy Rev 2012 16 6 4095 4104 10.1016/j.rser.2012.03.034
Mamun A Boubaker S Nguyen D Green finance and decarbonization: evidence from around the world Finance Res Lett 2022 46 102807 10.1016/j.frl.2022.102807
Markus D, Adriana N (2018) Are green bonds a viable way to finance environmental goals? An analysis of chances and risks of green bonds. IW-Report No. 28/2018. URL: https://www.econstor.eu/handle/10419/180209 [accessed 17 Apr 2021]
Miguel C Manzano B Green tax reforms and habits Resour Energy Econ 2011 33 1 231 246 10.1016/j.reseneeco.2010.05.001
Mirza F Sinha A Khan J Kalugina O Zafar M Impact of energy efficiency on CO2 Emissions: empirical evidence from developing countries Gondwana Res 2022 106 64 77 10.1016/j.gr.2021.11.017
Moretto D Branca T Colla V Energy efficiency and reduction of CO2 emissions from campsites management in a protected area J Environ Manage 2018 222 368 377 10.1016/j.jenvman.2018.05.084 29870965
Ning Y Cherian J Sial M Alvarez-Otero S Comite U Zia-Ud-Din M Green bond as a new determinant of sustainable green financing, energy efficiency investment, and economic growth: a global perspective Environ Sci Pollut Res 2022 10.1007/s11356-021-18454-7
Pesaran H A simple panel unit root test in the presence of cross-section dependence J Appl Economet 2007 22 2 265 312 10.1002/jae.951
Petterson M What is energy efficiency? Concepts, indicators and methodological issues Energy Policy 1996 24 5 377 390 10.1016/0301-4215(96)00017-1
Phung T Rasoulinezhad E Thu H How are FDI and green recovery related in Southeast Asian economies? Econ Chang Restruct 2022 10.1007/s10644-022-09398-0
Popkova E Sergi B Energy efficiency in leading emerging and developed countries Energy 2021 221 119730 10.1016/j.energy.2020.119730
Quang P Thao D Analyzing the green financing and energy efficiency relationship in ASEAN J Risk Finance 2022 10.1108/JRF-02-2022-0046
Rasoulinezhad E Identification of the success factors of the green bond market for sustainable development in the COVID-19 era Energy Res Lett 2022 3 3 1 4 10.46557/001c.29979
Rasoulinezhad E Taghizadeh-Hesary F Role of green finance in improving energy efficiency and renewable energy development Energy Effic 2022 10.1007/s12053-022-10021-4
Saboori B Gholipour H Rasoulinezhad E Ranjbar O Renewable energy sources and unemployment rate: evidence from the US states Energy Policy 2022 168 113 155 10.1016/j.enpol.2022.113155
Shidong L Chupradit S Maneengam A Suksatan W The C Ngoc Q The moderating role of human capital and renewable energy in promoting economic development in G10 economies: evidence from CUP-FM and CUP-BC methods Renew Energy 2022 189 180 187 10.1016/j.renene.2022.02.053
Silva S Soares I Pinho C Green tax reforms with promotion of renewable energy sources and carbon capture and sequestration: comparison of different alternatives Energy Rep 2020 6 1 620 625 10.1016/j.egyr.2019.09.036
Taghizadeh-Hesary F Phoumin H Rasoulinezhad E COVID-19 and regional solutions for mitigating the risk of SME finance in selected ASEAN member states Econ Anal Policy 2022 74 506 525 10.1016/j.eap.2022.03.012
Taghizadeh-Hesary F Rasoulinezhad E Phoumin H Ways to achieve universal access to sustainable electricity in Southeast Asia Econ Chang Restruct 2022 10.1007/s10644-022-09386-4
Tan Z Wu Y Gu Y Liu T Wang W Liu X An overview on implementation of environmental tax and related economic instruments in typical countries J Clean Prod 2022 330 129688 10.1016/j.jclepro.2021.129688
Tang Y Yue S Ma W Zhang L How do environmental protection expenditure and green technology innovation affect synergistically the financial performance of heavy polluting enterprises? Evid Ch, Environ Sci Pollut Res 2022 10.1007/s11356-022-21908-1c
Telatar O Birinci N The effects of environmental tax on ecological footprint and carbon dioxide emissions: a nonlinear co-integration analysis on Turkey Environ Sci Pollut Res 2022 29 44335 44347 10.1007/s11356-022-18740-y
Tu C Rasoulinezhad E Sarker T Investigating solutions for the development of a green bond market: evidence from analytic hierarchy process Financ Res Lett 2020 34 101457 10.1016/j.frl.2020.101457
Wang Z Jia X Analysis of energy consumption structure on CO2 emission and economic sustainable growth Energy Rep 2022 8 4 1667 1679 10.1016/j.egyr.2022.02.296
Wang Y Yu L Can the current environmental tax rate promote green technology innovation?: evidence from China’s resource-based industries J Clean Prod 2021 278 123443 10.1016/j.jclepro.2020.123443
Westerlund J Testing for error correction in panel data Oxford Bull Econ Stat 2007 69 9 709 748 10.1111/j.1468-0084.2007.00477.x
Yoshino N Rasoulinezhad E Taghizadeh-Hesary F Economic impacts of carbon tax in a general equilibrium framework: empirical study of Japan J Environ Assess Policy Manage 2021 23 2250014 10.1142/21464333222500144
Yunzhao L Modelling the role of eco innovation, renewable energy, and environmental taxes in carbon emissions reduction in E−7 economies: evidence from advance panel estimations Renew Energy 2022 190 309 318 10.1016/j.renene.2022.03.119
Zakari A Khan I Tan D Alvarado R Dagar V Energy efficiency and sustainable development goals (SDGs) Energy 2022 239 122365 10.1016/j.energy.2021.122365
Zhang L Huang F Lu L Ni X Iqbal S Energy financing for energy retrofit in COVID-19: recommendations for green bond financing Environ Sci Pollut Res 2022 29 23105 23116 10.1007/s11356-021-17440-3
| 0 | PMC9707254 | NO-CC CODE | 2022-12-01 23:20:22 | no | Econ Change Restruct. 2022 Nov 29;:1-19 | utf-8 | null | null | null | oa_other |
==== Front
J Bionic Eng
J Bionic Eng
Journal of Bionic Engineering
1672-6529
2543-2141
Springer Nature Singapore Singapore
297
10.1007/s42235-022-00297-8
Research Article
Boosting Whale Optimizer with Quasi-Oppositional Learning and Gaussian Barebone for Feature Selection and COVID-19 Image Segmentation
Xing Jie [email protected]
http://orcid.org/0000-0003-3044-9223
Zhao Hanli [email protected]
Chen Huiling [email protected]
Deng Ruoxi [email protected]
Xiao Lei [email protected]
grid.412899.f 0000 0000 9117 1462 Key Laboratory of Intelligent Informatics for Safety & Emergency of Zhejiang Province, Wenzhou University, Wenzhou, 325035 China
28 11 2022
122
11 8 2022
9 10 2022
19 10 2022
© Jilin University 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Whale optimization algorithm (WOA) tends to fall into the local optimum and fails to converge quickly in solving complex problems. To address the shortcomings, an improved WOA (QGBWOA) is proposed in this work. First, quasi-opposition-based learning is introduced to enhance the ability of WOA to search for optimal solutions. Second, a Gaussian barebone mechanism is embedded to promote diversity and expand the scope of the solution space in WOA. To verify the advantages of QGBWOA, comparison experiments between QGBWOA and its comparison peers were carried out on CEC 2014 with dimensions 10, 30, 50, and 100 and on CEC 2020 test with dimension 30. Furthermore, the performance results were tested using Wilcoxon signed-rank (WS), Friedman test, and post hoc statistical tests for statistical analysis. Convergence accuracy and speed are remarkably improved, as shown by experimental results. Finally, feature selection and multi-threshold image segmentation applications are demonstrated to validate the ability of QGBWOA to solve complex real-world problems. QGBWOA proves its superiority over compared algorithms in feature selection and multi-threshold image segmentation by performing several evaluation metrics.
Supplementary Information
The online version contains supplementary material available at 10.1007/s42235-022-00297-8.
Keywords
Whale optimization algorithm
Quasi-opposition-based learning
Gaussian barebone
Image segmentation
Feature selection
Bionic algorithm
http://dx.doi.org/10.13039/501100004731 Natural Science Foundation of Zhejiang Province LZ21F020001 Zhao Hanli Basic Scientific Research Program of WenzhouS20220018 Zhao Hanli
==== Body
pmcIntroduction
Due to the rapid development of various industries, people face more complex optimization problems in real life. Conventional optimization techniques have limitations in resolving complex and massive optimization problems, which cannot meet the requirement of convergence speed and calculation accuracy [1]. Compared with conventional optimization techniques, meta-heuristic algorithms (MAs) have the characteristics of flexibility, simplicity, derivation-free mechanism, and local optimal avoidance [2]. Therefore, MAs have been widely used to resolve complex optimization issues in recent years. The research inspiration for MAs mainly comes from biological behavior or natural physical phenomena. Furthermore, according to the different simulated natural behaviors, MAs can be divided into three main classes: evolution-based algorithms (EAs), physical-based algorithms (PAs), and swarm-based algorithms (SAs) [3]. EAs are inspired by Charles Darwin’s theory of natural selection, in which the best individuals always combine to produce better offspring [4]. The main representatives of the evolution-based algorithm are genetic algorithms (GA) and differential evolution (DE). PAs imitate the physical rules and chemical reactions of the universe, such as simulated annealing (SA), particle swarm optimization (PSO), and sine cosine algorithm (SCA) [5]. SAs is a kind of heuristic algorithm that simulates the swarm behavior to solve optimization problems, such as firefly algorithms (FA) [6], moth flame optimization (MFO) [7], flame optimization algorithm (FOA) [8], Harris Hawks optimizer (HHO) [9], and Slime Mould algorithm (SMA) [10]. SAs have become an important method for solving optimization problems because of their excellent self-organization, self-adaptation, and self-learning characteristics. It has been adopted in various domains [11, 12], such as image segmentation [13], wireless networks [14], unmanned aerial vehicles [15], target tracking [16], neural network [17], MRI classification [18], feature selection [19], and engineering problems [20], and vehicle design [21, 22].
Mirjalili and Lewis proposed a novel meta-heuristic algorithm in 2016 [23], named whale optimization algorithm (WOA), which was encouraged by the humpback whale’s foraging behavior. In recent years, WOA has attracted much attention and has been utilized to find optimal solutions in many fields. For example, based on chaotic and multi-swarm strategies, Wang et al. [24] developed an improved WOA and used it in two optimization scenarios. Chen et al. [25] introduced a chaotic local search strategy and Levy flight (LF) into WOA (BWOA), which was applied to solve three well-known problems in mathematical modeling studies. Abdel-Basset et al. [26] proposed two new variants of WOA, based on the ranking method and cyclic exploration–exploitation operator, and named RWOA and HWOA, respectively. They were applied to identify the parameters of the three-diode photovoltaic model. Peng et al. [27] proposed a hybrid WOA to improve the performance of cloud load forecasting. Chakraborty et al. [28] proposed a modified WOA variant and used it to solve problems in the engineering domain. Peng et al. [27] introduced a hybrid WOA based on the Levy and migration strategies into cloud load forecasting. Ye et al. [29] devised a novel modified WOA using the strategy of LF and pattern search and applied it to the field of energy optimization. Mostafa et al. [30] studied a WOA-based liver segmentation method from magnetic resonance images. Chao et al. [31] embedded orthogonal crossover into WOA to improve its exploration ability and applied it to estimate the surface duct. Hassib et al. [32] combined WOA with bidirectional recurrent neural network algorithms to train a deep learning approach. M.Hassi et al. [32] combined WOA with bidirectional recurrent neural network algorithms to train a deep learning approach. Darwish et al. [33] developed a novel whale optimization algorithm based on chaotic maps, which was used to select feature sets with high classification performance and a small number of features. Tripathi et al. [34] proposed a WOA variant for the recommendation over large-scale datasets in managing large-scale datasets.
In the original WOA, although the number of parameters to be adjusted is less and the convergence ability is strong enough, it still has the disadvantages of slow convergence speed and low convergence accuracy. Therefore, many optimization schemes were proposed to mitigate them to overcome these shortcomings. Hussien et al. [35] introduced a binary whale optimization algorithm based on two transfer functions. Chakraborty [36] et al. improved WOA in three aspects: the original algorithm’s parameters, the prey's search range, and the inertia weights. While reducing the complexity of the algorithm, it effectively improves the performance of the algorithm. Luo et al. [37] improved WOA based on the regularity of chaos and the mutational character of Gaussian mutation. Saha et al. [38] proposed a cosine-adapted modified whale optimization by incorporating cosine parameters into the selection of control parameters. In the study [1], LF and ranking-based mutation operators were embedded into WOA, which can prevent the algorithm from falling into local optimum and help the algorithm find the optimal solution quickly. Tu et al. [39] enhanced WOA with the strategies of communication mechanism and biogeography-based optimization algorithm, which overcome the shortcomings of WOA in slow convergence speed and easy trapping into the local optimum. Heidari et al. [40] integrated the learning mechanism and hill-climbing local search into WOA to enhance the exploitation process of the original WOA, which is called BMWOA. Chen et al. [41] combined two strategies, random replacement and double adaptive weight, with WOA to improve the performance of WOA. In the study [42], the exploration and exploitation capabilities of WOA were enhanced by embedding the modified mutualism phase of symbiotic organisms search and the DE mutation operator. Wang et al. [43] integrated elite strategy and spiral motion from MFO into WOA. Elhosseini et al. [44] introduced an inertia weight strategy to improve the parameters of WOA nonlinearly as a way to strengthen the ability of WOA to find optimal solutions. Chakraborty et al. [45] improved the exploration ability of WOA using the DE mutation strategy and balanced the exploration and exploitation capabilities by introducing a new parameter. Wang et al. [46] used opposition-based learning and a global grid ranking mechanism to enhance WOA's performance.
Although the above meta-heuristic algorithms, including WOA and improved WOA variants, have demonstrated their effectiveness in many optimization problems, there are still some shortcomings in the convergence speed and accuracy [11, 47]. The original WOA can be further improved when solving some complex practical problems, especially in the field of feature selection [48] and image segmentation [49]. In the late iterative stage, the exploitation capabilities of WOA do not enable much change in the location of search agents [44, 50]. When the algorithm is trapped in a local optimum, it does not solve this type of situation well. This may make the WOA-based multi-threshold image segmentation, and feature selection methods have insufficient power to jump out of the local optimum and thus fail to achieve satisfactory optimization performance. In addition, the diversity of the population decreases with the number of evaluations [29, 41, 51]. The global exploration capability of WOA is insufficient, which may result in some regions with better solutions not being found. As a result, when WOA is applied in the multi-threshold image segmentation problem and feature selection problem, it may miss the thresholds and feature subsets that have better effects.
To this end, we designed an improved WOA, called QGBWOA, by combining a quasi-opposition-based learning (QOBL) strategy and the Gaussian barebone (GB) mechanism. To address the shortcoming of WOA in jumping out of the local optimum in the later evaluation, the QOBL strategy is introduced. QOBL strategy is characterized by considering the opposite position of the current solution in the solution space. Therefore, if the current solution falls into a local optimum, its opposite solution may help it to break out of the local optimum and find a more optimal solution. The local exploitation capability of the algorithm is thus enhanced. The GB mechanism is introduced to address the problem of population diversity decreasing in the late evaluations. GB mechanism can generate the position of individuals in the search region by Gaussian distribution. Therefore, it enhances the population’s diversity and improves the algorithm’s global exploration ability, which gives the algorithm more possibility to move to regions with better solutions for searching.
In summary, the main contributions of this paper are as follows:A new enhanced WOA algorithm combining QOBL and GB, called QGBWOA, is proposed, with the experimental results showing that QGBWOA has higher accuracy and a faster convergence rate in obtaining global solutions.
A QGBWOA-based wrapper feature selection method is proposed for tackling feature selection tasks.
A QGBWOA-based image segmentation method of 2D histograms combined with Kapur’s entropy is proposed and applied to real COVID-19 pathology images.
QGBWOA achieves higher classification accuracy and a smaller number of features in the feature selection task and shows excellent performance in multi-threshold image segmentation problems in all three evaluation metrics, including Peak Signal to Noise Ratio (PSNR) [52], Structural Similarity (SSIM) [53], and Feature Similarity (FSIM) [54].
The rest of this article is organized as follows. The original WOA is presented in Sect. 2. Section 3 describes the proposed QGBWOA algorithm. Section 4 analyzes the experimental results of QGBWOA in the benchmark function test. Section 5 provides the application of QGBWOA to the feature selection and image segmentation problems. The conclusion and future work are summarized in Sect. 6.
Whale Optimization Algorithm
WOA [23] simulates the hunting actions of whales: encircling prey, bubble net attack, and searching for prey.
Encircling Prey Phase (Exploitation)
In the exploitation phase, whales use bubble nets to attack their prey, including two models of shrinking encircling and spiral updating. In the encircling prey stage, the best search agent position obtained so far is selected as the optimal position, and other individuals gradually approach the best agent. Its mathematical model is as Eq. (2):1 Ddist=C∙Xbest(t)-X(t),
2 Xt+1=Xbestt-A∙Ddist,
where X(t+1) is the position of the search agent in the next evaluation, X(t) is the position of the search agent in the current evaluation, and Xbest(t) is the best agent explored so far. Let FEs represent the counter of evaluation, A and C are two control parameters that can be presented as follows:3 A=2a∙r1-a,
4 a=2-2×FEsMaxFEs,
5 C=2r2,
where r1,r2 are two random numbers in [1], MaxFEs denotes the maximum number of evaluations of the algorithm, and a linearly decreases from 2 to 0 over evaluations.
Spiral Updating Phase (Exploitation)
The spiral updating phase is realized by Eq. (6). Odist=Xbestt-X(t) denotes the distance between the search agent in the current evaluation and the best agent obtained so far.6 Xt+1=Odist∙efl∙cos(2πl)+Xbestt,
where f is a constant that controls the logarithmic spiral’s shape and is set to 1 according to the original text. The parameter l is a random number in [− 1, 1].
The probability of being selected for the spiral moves and shrinking encircling phase is 50% each:7 Xt+1=Xbestt-A∙Ddistpro<0.5;Odist∙efl∙cos(2πl)+Xbesttpro≥0.5,#
where pro is a random number between 0 and 1.
Search for Prey Phase (Exploration)
When A is less than − 1 or more than 1, whales use a random walk mechanism to search for prey based on the locations of other individuals. The mathematical model for the exploration phase is as follows:8 Ddist=C∙Xrand(t)-X(t),
9 Xt+1=Xrand(t)-A∙Ddist,
where Xrandt represents a random search agent selected from the current population.
The flow chart of WOA is as shown in Fig. 1.Fig. 1 The flow chart of WOA
The proposed QGBWOA
In this section, the proposed QGBWOA will first be described by flowchart and pseudo-code. Then, the two strategies, QOBL and GB, will be described in detail. Finally, the time complexity of QGBWOA is analyzed.
Algorithm Overview
The flow chart of QGBWOA is shown in Fig. 2. We present the pseudo-code of the proposed QGBWOA in Algorithm 1.Fig. 2 Flow chart of QGBWOA
The pipeline of QGBWOA is described as follows. First, the population is randomly initialized. Then, find the optimal individual in the current evaluation based on the fitness value, and update the individual position. In the position updating phase, by incorporating the QOBL mechanism, the ability of the algorithm to find a superior solution is boosted to some extent. Thus, the convergence rate and the quality of the solution can be improved. After the individual position is updated, the GB strategy is used to update the population position again. The increased diversity of the population enhances the exploration ability; thereby, the frequency of the algorithm falling into the local optimum is significantly reduced. The details of QOBL and GB mechanisms are presented in the following subsections.
Quasi-Opposition-Based Learning
As shown in Eq. (3), the value of parameter A will be less than 1 at the late stage of evaluation in the original WOA, and the encircling prey phase is executed. The position of the new individual is only related to the position of the optimal individual and current individual, so the final new individual’s position will not change significantly in the late evaluation, which may cause it to fall into the local optimum. Therefore, QOBL [55] is taken to enhance the original WOA in local search ability in the late evaluation process, to reduce the frequency of WOA falling into the local optimum. QOBL is an improved version of opposition-based learning (OBL) [56], which considers the individual with the opposite position to the current individual may be closer than the current individual. In recent years, the QOBL strategy has been used in MAs [57–60] to improve convergence speed accuracy.
The mathematical model is depicted as follows:10 xjqo=randlbj+ubj2,xjo,#
where xjqo represents the quasi-opposite individual of the current search agent in j-th dimension, lbj+ubj2 represents the center of [lbj,ubj], randlbj+ubj2,xjo represents a uniformly distributed random number between in lbj+ubj2 and xjo, xjo denotes the opposite individual of the current search agent in j-th dimension:11 xjo=lbj+ubj-xj,
where lbj denotes the lower bound of the search space, ubj denotes the upper bound of the search space, and xj denotes the position of the current individual in j-th dimension.
Gaussian Barebone Mechanism
As we mentioned earlier, when the algorithm is evaluated in the later phase, the diversity of WOA will reduce, which can cause insufficient convergence speed and convergence accuracy. The GB mechanism can help individuals choose the most suitable direction and continuously approach the optimal solution to avoid prematurely falling into local optimality. Therefore, after the position of all the search agents has been updated, the characteristics of the randomness of GB are incorporated into WOA to enhance the population diversity. This balances the algorithm’s local exploitation and global search capabilities and further improves the convergence speed.
The GB [61] strategy is on the basis of bare bones PSO (BBPSO) [62], and the parameter CR is employed in the GB strategy to guide each individual. If the probability of random generation is less than CR, the Gaussian distribution is used to update the individual’s position in the next evaluation; otherwise, the idea of differential evolution is introduced to update the individual’s position. The GB strategy is as follows:12 Vi,j=GPLeader-Xi,j2,PLeader-Xi,jr3<CR;Xt1,j+r4×Xt2,j-Xt3,jotherwise,
where Vi,j denotes the position of the i-th individual in the j-th dimension, PLeader denotes the global optimal position in the population, Xi,j is the current individual in the j-th dimension, G represents the Gaussian distribution, r3andr4 are random numbers within [1], Xt1,j, Xt2,j, Xt3,j are three arbitrarily selected individuals that are diverse from the current individual.
Time Complexity of QGBWOA
The time complexity of QGBWOA is subject to the population size (N), the number of dimensions (Dim), and the maximum number of algorithm evaluations (MaxFEs). Then, the overall time complexity is as follows:The population size is N. The time complexity of initializing all individual whales is O (N).
The time complexity of population fitness and updating the position and the fitness of the current optimal solution is MaxFEs×O(2N).
The primitive WOA search mechanism also causes the position of each search agent to change during the search process of QGBWOA. The time complexity of updating the position of each search agent is MaxFEs×(ON×Dim+5×N).
Implementing the QOBL mechanism is MaxFEs×O(N×Dim).
Performing GB strategy is MaxFEs×O(N×(Dim+5)).
Therefore, the total time complexity is O (QGBWOA) = O (Initialization) + O (Calculation of initial whales and Selection) + O (WOA) + O (QOBL strategy) + O (GB mechanism) = O(N) + MaxFEs× (O(2N) +ON×Dim+5×N + O(N×Dim) + O(N×(Dim+5))).
Experimental Results and Discussion
In this section, the algorithm stability and strategy combination are analyzed, and experimental simulation results on the IEEE CEC 2014 and CEC 2020 benchmark functions are shown to verify the performance of QGBWOA comprehensively.
Benchmark Functions
CEC 2014 benchmark functions and CEC 2020 benchmark functions are used to verify the efficacy of QGBWOA. The details of the functions are shown in Appendix A.
For the fairness of the experiment results, all tested algorithms were performed in the same environment: the population size was 30, the maximum number of evaluations was set to 300,000, and the algorithms were independently estimated 30 times on each benchmark function. We used the Friedman and WS tests to evaluate the experiment results.
Balance and Diversity Analysis on QGBWOA and WOA
In this section, QGBWOA is qualitatively analyzed on CEC 2014 in five aspects: search history, search trajectory, average fitness, and diversity and population balance.
The search history, search trajectory, and average fitness results are reported in Appendix B. In Fig. B.1, The first, second, third, and fourth columns of the figure show the three-dimensional of corresponding functions, the historical search trajectories in 2-dimensional (2D), the trajectories of the search agents, and the average fitness of individuals, respectively.
The red dots shown in Fig. B.1 (b) represent the global optimal solution, while the black dots represent the position of the search agent. The figure clearly shows that with the increase of the number of evaluations, the black dots gradually approach the red dots to find the optimal solution. In Fig. B.1 (c), the individual trajectory fluctuation is small in F2, while in the early evaluation process of F4, F10, and F16, the individual trajectory fluctuation is relatively strong. This shows that QGBWOA can reach most of the search space. Figure B.1 (d) shows that the average fitness decreases faster in F2, while the average fitness has strong fluctuations during early evaluations in F4 and F10. This indicates that QGBWOA can quickly determine the approximate range of optimal solutions in the early evaluations and further explore the optimal solution in later evaluations to achieve accurate convergence.
Figure B.2 shows the result of QGBWOA and WOA balance analysis. The figure's red, blue, and green curves represent the exploration, exploitation, and incremental–decremental curves. When the exploration effect is weaker than the exploitation effect, the green curve decreases, and vice versa. This is because the algorithm usually performs the global exploration first in the solution space, determines the approximate solution location with superior quality, and then performs the solution's local exploitation to find a more optimal solution. As shown in Fig. B.2, in the beginning of the evaluation, the search curve always starts with a higher value and the algorithm is mainly based on a global search. Then, local exploitation soon dominated. From Fig. B.2, it can be observed that the search phase of QGBWOA ends earlier than WOA, indicating that the local exploitation time of QGBWOA in the target area is longer.
Figure B.3 shows the results of QGBWOA and WOA diversity analysis. The diversity of the population is high at the beginning because the algorithm initializes the population randomly. As the evaluation progresses, the search range of the algorithm continues to decrease, and the population diversity also continues to decrease. It can be seen from Fig. B.3 that the average diversity of QGBWOA falls faster than WOA, which indicates that QGBWOA converges more quickly than WOA.
In summary, QGBWOA demonstrates remarkable advantages in terms of convergence speed and global search capability compared with WOA.
Ablation Study on QGBWOA
To demonstrate the influence of QOBL and GB mechanisms, ablation experiments with QGBWOA were conducted. In Table 1, “Q” and “GB” mean the quasi-opposition-based learning mechanism and the Gaussian barebone mechanism, respectively. “1” implies that the corresponding mechanism is employed, and a “0” conversely indicates that it is not used. For example, QWOA indicates that WOA only uses the quasi-opposition-based learning mechanism and does not use the Gaussian barebone mechanism. Figure B.4 shows the convergence curves of QGBWOA with the other two mechanisms and the original WOA on the CEC 2014 benchmark functions, and the Dim is set to 30. It can be seen from the figure that QGBWOA is far superior to WOA in terms of convergence speed and convergence accuracy, which demonstrates the effectiveness of the GB mechanism in the global exploration capability of WOA. Experimental results show that QGBWOA is the best way to solve these different functions.Table 1 The ablation experiment of QGBWOA
Q GB
WOA 0 0
QWOA 1 0
GBWOA 0 1
QGBWOA 1 1
The WS test and Friedman test were used to compare the algorithms for statistical difference calculation. The results are shown in Appendix A. Table A. 3 records the average (Avg), standard deviation (Std), and average ranking (ARV) for each algorithm. "+", "−" and "=" in Table A. 3 indicate that QGBWOA is better than other algorithms, inferior to other algorithms, and equal to other algorithms, respectively. Table A. 3 shows that QGBWOA ranks first among its two variants of the mechanism and the original WOA algorithm. On the benchmark functions F23–F25 and F27–F30, the Std values of QGBWOA are 0, which indicates that QGBWOA has good robustness. This is because the addition of the QOBL strategy improves the local exploitation ability of QGBWOA, and the addition of the GB strategy comprehensively improves the global exploration ability of QGBWOA, improves the population diversity, and better helps the algorithm find the global optimal solution. The final average ranking of QGBWOA is the first, which indicates that the algorithm is optimized best when these two mechanisms work together.
Table A. 4 shows the p values results of the WS test. When the value in the WS test is less than 0.05, QGBWOA has remarkable performance over its peer. The values less than 0.05 in the table have been bolded. From Table A. 3 shows that QGBWOA outperforms QWOA, GBWOA, and WOA in 20, 23, and 27 of the 30 benchmark functions, respectively. It can be seen that with the integration of the proposed two mechanisms, the performance is gradually improved.
Comparison with Other Metaheuristic Algorithms on CEC 2014 Test
In this section, QGBWOA is compared with 7 of the MAs, including WOA, DE [63], FA [6], FOA [64], PSO [65], SCA [5], and MFO [7]. Table A. 5 shows the results of F1-F30 with Dim values of 10, 30, 50, and 100, respectively.
It can be found that QGBWOA performs better in F1 when Dim values are 10, 30, 50, and 100. DE performs better on F2 and F3. Furthermore, compared with FOA, SCA, MFO, and WOA, QGBWOA is superior to them on all hybrid and composition functions when Dim is 10, 30, 50, and 100, respectively. DE can obtain the optimum on F15 in all four dimensions. QGBWOA finally ranks first when the Dim is 30, 50, and 100. The average mean is 1.8, which is 26% higher than the second-ranked DE algorithm and 63% higher than the original WOA. The p values are shown in Table A. 6 when Dim is 30. In Table A. 6 shows that the values of FA, FOA, and SCA on all functions are less than 0.05, indicating that QGBWOA outperforms these algorithms.
Figure 3 reports convergence curves and boxplots for each algorithm on 10 functions. In Fig. 3, F1 and F2 are unimodal functions. F4, F5, and F11 are multi-model functions. F17 and F18 are hybrid functions. F23, F24, and F29 are composition functions. The first and third columns of the figure show the convergence curves, and the second and fourth columns show the corresponding box plots. In unimodal, multimodal, and partial hybrid functions, although QGBWOA did not discover the optimal solution in the beginning phase, it converged to the optimal solution in later evaluations, indicating that QGBWOA has a good ability to avoid falling into the local optimal solution. In composition functions, the convergence rate and convergence accuracy of QGBWOA are significantly better than other algorithms. In the box plot, the center marker of each box indicates the intermediate value. Each box has the 25th and 75th percentiles at the lower and upper margins. Red "+" marks are used to mark outliers. The box plot in Fig. 3 shows that QGBWOA has more stable optimization results and fewer outliers than the compared algorithms in most cases. These results confirm the greatly improved performance of QGBWOA compared to WOA and other peers. The proposed QGBWOA has better performance due to the QOBL strategy and GB mechanism. The impact of these two strategies is shown in the convergence plots of the benchmark functions. Observing the convergence curves of F1, F2, F11, and F17, it can be found that the proposed algorithm does not fall into the local optimum at the evaluation times of about 50,000, and continues to converge to the region with higher solution quality, which illustrates that the QOBL strategy can be a good way to improve the local exploitation ability and the accuracy of the solution. In the convergence plots of F18, F23, F24, and F29, QGBWOA improves the exploration ability of the population due to the GB strategy so that the global optimal solution can be fast located, enabling the improvement of convergence faster.Fig. 3 Convergence curves and boxplots of QGBWOA and other meta-heuristics (Dim = 30)
Comparison with The State-of-The-Art WOA Variants on CEC 2014 Test
To confirm the efficacy of QGBWOA, we compared QGBWOA with 6 WOA variants on CEC 2014 test. These variants include ACWOA [44], CCMWOA [37], OBWOA [46], RDWOA [41], BMWOA [40], and BWOA [25].
Table A.7 shows the statistical results in different dimensions. The table shows that QGBWOA obtains the smallest optimization result among the 30 test functions. QGBWOA ranks first when Dim values are 10, 30, 50, and 100, respectively. Compared with the second-ranked RDWOA, QGBWOA outperforms RDWOA in these four categories of dimensions with 8, 19, 19, and 21 functions, respectively. Moreover, QGBWOA performs better in all four dimensions on functions F1, F2, F4–F7, F15, F17, F19, F21–F23, and F27–F30.
Table A.8 reports the comparison results of the WS test between QGBWOA and the compared variants of WOA when Dim is 30. The p values on all test functions of BMWOA except F14 are less than 0.05, which indicates that QGBWOA’s performance is superior to that of BMWOA. The p values of RDWOA, CCMWOA, and BWOA on the F23, F25, F27, and F28 functions are 1, indicating that RDWOA, CCMWOA, and BWOA can obtain the same optimization results as QGBWOA.
In Fig. 4, the convergence rate of QGBWOA is faster than the compared state-of-the-art algorithms on most of the benchmark functions. In addition, it can be observed that its convergence accuracy is the best, whereas other variants of WOA are trapped in local optimal values to varying degrees. Figure 4 depicts the box plots of the fitness of the best individuals found in the final generation. These comparison results show that QGBWOA is better than these state-of-the-art algorithms for complex optimization problems. The results affirmed the ability of QGBWOA to solve benchmark problems in different dimensions.Fig. 4 Convergence curves and boxplots of QGBWOA and other variants algorithms
Comparison of CEC 2020 Benchmark Functions
In this subsection, QGBWOA is tested with 7 group intelligence algorithms, including WOA, HHO [9], SMA [10], HGS [66], SCA [5], RDWOA [41], ACWOA [44], on the CEC 2020 benchmark functions with Dim equals to 30. Table A.9 illustrates the statistical results of all optimizers on Avg and Std. It can be seen in the table that QGBWOA outperforms the compared optimizers for 9, 8, 4, 3, 10, 6, and 8 of the 10 CEC 2020 benchmark functions, respectively. QGBWOA ranks first in total with ARV equal to 1.4.
The p value results of QGBWOA against the compared optimizers are shown in Table A.10, where p values greater than 0.05 are shown in bold. The results show that QGBWOA significantly differs from the optimizer for most of the tested functions except F9 and F10.
Statistical Analysis of QGBWOA
Because the Freidman test can only give a conclusion on whether there is a variance in performance among algorithms, a post hoc test is needed to find the statistical difference in the performance of the algorithms. Commonly used follow-up tests include the Nemenyi and Bonferroni–Dunn tests [67]. The Nemenyi test is employed to compare the performance of algorithms with each other, while the Bonferroni–Dunn test is used to compare an algorithm with the rest algorithms. Bonferroni–Dunn post hoc statistical analysis is used in this article to verify the performance difference between QGBWOA and the compared algorithms. Assume the difference in the average rank between the two algorithms is better than the critical difference (CD). The CD is described as Eq. (13).13 CD=qαkk+16Num,
where α denotes the significant level, qα is the critical value, k is the number of algorithms, and Num represents the number of test functions.
In the experiment in Sect. 4.4, eight algorithms were chosen, so k is 8. Thirty benchmark functions were used, so Num=30. The significant levels of α were selected as 0.05 and 0.1. According to Eq. (13), it can be calculated that when α values are 0.05 and 0.1, respectively, the corresponding CD values are 1.7 and 1.55, respectively. The average rank of QGBWOA is ARVQGBWOA=1.70 when Dim is 30. When the average rank of the comparison algorithm is greater than CD+ARVQGBWOA=3.4/3.25(α=0.05/0.1), there is a significant difference between QGBWOA and this algorithm. In Fig. 5, the solid line represents the threshold when the significant level is 0.1, and the dotted line indicates the threshold when the significant level is 0.05. As shown in Fig. 5, QGBWOA outperforms FA, FOA, PSO, SCA, MFO, and WOA at these two significant levels. In the Bonferroni–Dunn test, QGBWOA and DE showed no remarkable difference in performance.Fig. 5 Bonferroni–Dunn test results of experiments in Sect. 4.4 (Dim = 30)
The post hoc test analysis is performed for the experiment in Sect. 4.5. The results are shown in Fig. 6. Because eight algorithms were tested in 30 functions, the k is 8 and Num is 30. Similarly, CD values are 1.70 and 1.55 when the corresponding α values are 0.05 and 0.1, respectively. From the figure, we can see that QGBWOA is significantly better than ACWOA, BMWOA, CCMWOA, BWOA, OBWOA, and WOA at both significant levels and exhibits no obvious difference from RDWOA in the Bonferroni-Dunn test.Fig. 6 Bonferroni–Dunn test results of experiments in Sect. 4.4.5 (Dim = 30)
Applications
In this section, we will show the applications of QGBWOA in feature selection and image segmentation.
Feature Selection Based on Proposed QGBWOA
Binary QGBWOA
Data mining technology has been widely used in medicine in recent years. Most medical data are high-dimensional, and extracting a sub-set of useful features from high-dimensional data is tricky. Therefore, the dimensionality of the dataset needs to be reduced before data mining. Feature selection is one of the dimensionality reduction methods. However, it is often difficult to exhaust the combination of feature subsets, especially when dealing with high-dimensional data, and using a heuristic algorithm can solve this problem well [68]. As a heuristic method, the SAs have the intelligent selection and random search advantages, which can search for an ideal solution set. In recent years, the combination of SAs and feature selection methods has received more and more attention from researchers [69–72]. In this subsection, a QGBWOA-based wrapper feature selection method is proposed. Twenty-four datasets from the UCI machine learning repository are utilized to measure the effectiveness of the proposed method.
Since the search space in the feature selection problem is represented in binary, the real value of each solution found by QGBWOA must be converted into a Boolean type. The function is defined as follows:14 Xi,jt+11,ifr5≥TXi,jt;0,otherwise,#
15 Tx=11+e-2x,#
where r5 is a random number between in 0 and 1. If Xi,j=1, it means the solution of Xi in the j-th dimension is regarded as a relevant feature; otherwise, if Xi,j=0, it means the solution of Xi in the j-th dimension is regarded as an irrelevant feature. Since feature selection aims to obtain better classification accuracy with fewer features, the classifier error rate and the number of selected features are utilized to form the fitness function. The fitness function is defined as follows:16 Fitness=θ·Acc+υ·ST,#
where Acc denotes the classification accuracy of the K-nearest neighbors (KNN) classifier, θ and υ represent the weight coefficients of the classifier error rate and the number of selected features, respectively, S represents the number of the selected feature sub-set, and T means the total number of features in the dataset.
The overall algorithm pipeline is shown in Fig. 7. First, the dataset is processed. Second, using ten-fold cross-validation to divide the dataset into ten parts, nine of which are used as training data, and the remaining one is used as test data. Before utilizing QGBWOA to search for the best feature combination in the dataset, the number of attributes of the dataset is set to the dimension of the population in QGBWOA. The KNN classifier is used to evaluate the accuracy of the selected features. The fitness of the population is calculated. Then, QGBWOA is applied to update the position of the population in the discrete search space. After reaching a specified number of evaluations, an optimal feature sub-set is obtained. The KNN classifier evaluates the classification accuracy of the obtained feature sub-set. Finally, the best sub-set of features is obtained.Fig. 7 The architecture of feature selection using the binary QGBWOA
Experiment on Feature Selection
Twelve low- and high-dimensional datasets are used to examine the efficacy of the proposed approaches. Both these two categories of datasets are chosen from the California Irvine (UCI) Machine Learning Repository [73]. Table 2 describes datasets in terms of the number of samples, features, and classes. It can be seen that datasets in low-dimension contained less than 350 features, and most of the high-dimensional datasets have more than 5000 features. And QGBWOA was tested together with bMFO [74], BSMA [75], BSO [76], bWOA [77], and BDE [78]. Table 3 reports their parameter settings. Moreover, the datasets are divided by ten-fold cross-validation [79]. The wrapped feature selection method is based on the KNN (K=1) classifier [80]. The maximum iteration is set to 50.Table 2 Datasets of the feature selection experiment
Category Datasets Samples Attributes Classes
Low-dimension Blood 997 69 2
Breast cancer 286 9 2
Congress 435 17 2
Ionosphere 351 35 2
Krvskp 208 61 2
M-of-n 100 14 2
Parkinson 195 23 2
Sonar 208 61 2
Spect 267 23 2
Heart 270 14 5
Lymphography 148 19 4
Penglung 73 326 7
High-dimension Brain_tumor1 90 5921 5
Brain_tumor2 50 10,368 4
CNS 60 7130 2
DLBCL 77 5470 2
Leukemia 72 7131 2
Leukemia1 72 5328 3
Leukemia2 72 11,226 3
Lung_cancer 203 12,601 3
Prostate_tumor 102 10,510 2
SRBCT 83 2309 4
Tumors_9 60 5727 9
Tumors_11 174 12,534 11
Table 3 Parameter setting of compared algorithms
Algorithm Other parameters Common parameters
BMFO b=1 Dim = number of features in the current data set; population size = 20
BSMA a=
BSO C1=0.5
C2=2
Threshold=0.25
bWOA b=1
r1=
r2=
BDE beta_min=0.2
beta_max=0.8
pCR=0.2
Appendix A reports the results of QGBWOA and other methods based on average classification error, the number of selected features, fitness values, and time cost. Table A.11–Table A.13 and Table A.15–Table A.17 show the fitness, number of selected features, and average classification error results on low-dimensional and high-dimensional datasets, respectively. It can be observed that QGBWOA significantly outperforms the compared algorithms in terms of fitness and can select a smaller number of features. QGBWOA ranked first in total compared to other methods. Table A.14 and Table A.18 show the timing results on each data set. The results in the table report that the time cost of the proposed QGBWOA is higher than the compared algorithms, such as bSMA and BSO. To a certain extent, the embeddedness of QOBL and GB strategies increases the time overhead of the proposed algorithm.
Figure B. 5 shows the convergence curves of QGBWOA and the other five algorithms on 12 low-dimensional datasets. The figure shows that QGBWOA performs well on all these 12 datasets. The convergence value of QGBWOA reaches the minimum, indicating that QGBWOA can obtain higher classification accuracy than the algorithms. Furthermore, it can be seen from Fig. B. 6 that the convergence value of the fitness of QGBWOA on all 12 high-dimensional datasets is smaller than that of the compared methods except Brain_tumor2 and Lung_cancer.
Image Segmentation Based on Proposed QGBWOA
Proposed Image Segmentation Method
Image segmentation is a fundamental technique in a variety of image processing applications. The characteristics of the image divide the image into multiple discrete regions, which are characterized by continuity or similarity in the same region and have obvious contrast between different regions [81]. The multi-threshold method is a commonly used image segmentation method, which uses multiple thresholds to mark target regions of interest in an image. The choice of threshold has an impact on the image segmentation effect. One of the often-considered methods is the histogram-based method. The histogram describes the frequency of the corresponding gray value in the image. The one-dimensional histogram only reflects the magnitude of the pixel’s gray level, and the 2D histogram can reflect the spatial correlation information between the pixel and its neighborhood. Abutaleb et al. [82] proposed an image segmentation method combining the local pixel gray average with the original gray histogram. Kapur’s entropy can be used to evaluate the optimal thresholds. It divides the image into different categories and determines whether the categories are consistent according to the entropy size. Kapur’s entropy finds the optimal threshold by maximum fitness value [83]. Kapur’s entropy is chosen in this article to determine the n best thresholds. Let [T1,T2,T3,⋯,Tn] be the threshold set values to divide the image into n+1 classes and the formula is described as:17 FT1,T2,T3,⋯,Tn=H0+H1+⋯+Hn,#
18 H0=-∑s=0T1-1psω0lnpsω0,ω0=∑s=0T1-1ps,#
19 H1=-∑s=T1T2-1psω1lnpsω1,ω1=∑s=T1T2-1ps,#
⋮
20 Hn=-∑s=TnL-1psωnlnpsωn,ωn=∑s=TnL-1ps,#
21 ps=h(s)P,s=0,1,⋯,L-1,#
where P is the total number of pixels in the image, L represents the total gray levels of the given image, s is the gray level, ps is the probability of the s-th gray level, H1,H2,⋯,Hn denote the Kapur's entropies of corresponding classes, and ω0, ω1,⋯,ωn denote the probabilities of corresponding classes.
Enumerating all combinations of thresholds and selecting the optimal one is quite difficult, and the time complexity will grow exponentially with an increasing number of thresholds [84]. The use of MAs to find the optimal threshold has attracted attention in recent years [85–88]. An improved butterfly optimization algorithm was proposed by Sharma et al. [89], which is utilized for image segmentation problems. Chakraborty et al. [90] introduced an enhanced version of WOA to tackle image segmentation problems. A new pulse couple neural network model based on grey wolf optimizer was proposed by Wang et al. [88] for medical image segmentation. Zhao et al. [91] improved the Salp swarm algorithm (SSA). This version is used to find optimal segmentation threshold for images.
In this subsection, we put forward a QGBWOA-based multi-threshold image segmentation method by integrating the 2D histogram with the entropy of Kapur. QGBWOA is utilized to find the optimal set of thresholds, where the entropy of Kapur is the objective function, and the image is segmented according to the threshold. The detailed flowchart of the method is given in Fig. 8.Fig. 8 Flow chart of multi-threshold image segmentation using QGBWOA
Simulation Experiment
We selected six COVID-19 patients’ images collected by Cohen et al. [92] as the segmentation images. These six images are named A, B, C, D, E, and F in this experiment. Computed tomography (CT) of the lungs of COVID-19 patients often shows high-gray diffuse ground-glass and pulmonary nodular shadows. This is because the COVID-19 virus enters the pulmonary bronchi and further invades the alveolar epithelial cells, where it replicates itself. The rapid replication of the virus leads to significant swelling of the epithelial cells, which will appear as high-gray shadows on CT images with significant gray differences from the normal lung parenchyma.
Figure 9 shows the original images and the 2D histograms of these six images. The population number of the algorithm in this experiment was set to 20, the iteration times were set as 100, and the size of the image was set as 512 × 400. The performance of the QGBWOA-based multi-threshold image segmentation method was evaluated at various thresholds, including low threshold levels (4 and 6) and high threshold levels (10 and 15). We compared our QGBWOA-based multi-threshold image segmentation with the multi-threshold segmentation methods based on other methods, including WOA, Cuckoo search (CS) algorithm, MFO, SSA, and biogeography-based learning PSO (BLPSO). In addition, the experiment results are evaluated by three indicators, including PSNR, SSIM, and FSIM.Fig. 9 The original images and their 2D histograms
The set of ideal segmentation thresholds found by the proposed algorithm for these images is shown in Fig. 10. The solid line in Fig. 10 represents the grayscale histogram, and the dotted line represents the ideal threshold set produced by QGBWOA. Because the threshold is 10, there are ten red lines in each picture.Fig. 10 The optimal threshold of images at threshold level 10
The Avg and Std values of PSNR, SSIM, and FSIM in each threshold are shown in Appendix A, where the optimal results have been bolded. The results in Tables A.19 to Table A.21 show that QGBWOA achieves optimal results at most thresholds. The mean value of the overall ranking of these evaluation metrics is shown in Tables 4, 5 and 6. The results show that the mean of the overall ranking of QGBWOA is the smallest, which confirms the strong competitiveness of the proposed method. The best results for Kapur’s entropy obtained by the proposed method are illustrated in Appendix A. In Table A.22, QGBWOA has a noticeable superiority in searching for the optimal value of Kapur’s entropy compared with other algorithms. Figure 11 illustrates the final segmentation results with a threshold of 10. From the result, we can see that depending on the threshold found by QGBWOA; the image can be segmented into blocks of pixels with different gray values with sharper borders, which is useful for evaluating suspected cases of COVID-19.Table 4 PSNR average results rankings
Threshold QGBWOA WOA CS MFO SSA BLPSO
4
ARV 3.500 4.000 3.833 3.000 3.000 3.667
Rank 3 6 5 1 1 4
6
ARV 2.833 3.833 3.333 3.833 4.333 2.833
Rank 1 4 3 4 6 1
10
ARV 3.167 3.500 4.167 2.833 3.333 4.000
Rank 2 4 6 1 3 5
15
ARV 1.833 4.167 4.833 4.000 3.667 2.500
Rank 1 5 6 4 3 2
Table 5 SSIM average results ranking
Threshold QGBWOA WOA CS MFO SSA BLPSO
4
ARV 2.833 4.667 3.833 2.833 3.167 3.667
Rank 1 6 5 1 3 4
6
ARV 2.500 2.500 3.833 3.833 4.500 3.833
Rank 1 1 3 3 6 3
10
ARV 3.333 3.833 3.333 3.667 3.833 3.000
Rank 2 5 2 4 5 1
15
ARV 2.667 4.167 3.667 4.333 2.833 3.333
Rank 1 5 4 6 2 3
Table 6 FSIM average results ranking
Threshold QGBWOA WOA CS MFO SSA BLPSO
4
ARV 2.833 4.167 4.167 3 3 3.834
Rank 1 5 5 2 2 4
6
ARV 2.667 3.667 4.000 3.500 4.167 3.000
Rank 1 4 5 3 6 2
10
ARV 2.500 3.167 4.000 3.667 3.667 4.000
Rank 1 2 5 3 3 5
15
ARV 2.500 4.333 4.167 4.500 2.500 3.000
Rank 1 5 4 6 1 3
Fig. 11 The segmentation results of all images obtained by QGBWOA at threshold level 10
Conclusion and Future Work
In this paper, a new WOA variant QGBWOA based on QOBL and GB strategy has been proposed to improve the inadequacies of the original WOA. The QOBL strategy is introduced to strengthen the local exploitation ability and to assist the proposed method to jump out of the local optimum. The GB strategy is employed to balance the algorithm’s exploitation and exploration capabilities and help the algorithm find regions with better solutions. QGBWOA was tested on CEC 2014 and CEC 2020 benchmark functions with different dimensions, in which the performance of QGBWOA is compared with the basic methods and the state-of-the-art WOA variants. The experimental results show that QGBWOA can provide optimum solutions and effectively avoid premature convergence. Finally, the ability of QGBWOA to solve real-world problems is validated by the feature selection and the multi-threshold image segmentation applications.
The increase in time complexity is the inherent result of mechanism embedding, and how to reduce the time complexity is also our next work. In future, parallel computing techniques [93] can be considered to reduce the time complexity while keeping the performance of QGBWOA. Additionally, we would like to apply QGBWOA to other fields, such as multi-objective optimization [94], fuzzy definition optimization [95], and dynamic optimization [96].
Data Availability Statement
The data involved in this study are all public data, which can be downloaded through public channels.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 14917 KB)
Acknowledgements
The authors would like to thank the editors and the anonymous reviewers for their valuable comments on improving this paper. Many thanks also to Ali Asghar Heidarid for his helpful proofreading. This work was supported by the Zhejiang Provincial Natural Science Foundation of China (no. LZ21F020001) and the Basic Scientific Research Program of Wenzhou (no. S20220018).
Declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Yan ZP Zhang JZ Zeng J Tang JL Nature-inspired approach: An enhanced whale optimization algorithm for global optimization Mathematics and Computers in Simulation 2021 185 17 46 10.1016/j.matcom.2020.12.008
2. Mirjalili S Mirjalili SM Lewis A Grey Wolf Optimizer Advances in Engineering Software 2014 69 46 61 10.1016/j.advengsoft.2013.12.007
3. Faris H Mirjalili S Aljarah I Mafarja M Nature-inspired optimizers: Theories, literature reviews and applications 2020 Springer International Publishing 185 199
4. Li CY Li J Chen HL Jin M Ren H Enhanced Harris hawks optimization with multi-strategy for global optimization tasks Expert Systems with Applications 2021 10.1016/j.eswa.2021.115499
5. Mirjalili S SCA: A sine cosine algorithm for solving optimization problems Knowledge-Based Systems 2016 96 120 133 10.1016/j.knosys.2015.12.022
6. Yang X. S. (2009). Firefly algorithms for multimodal optimization. In Proceedings of stochastic algorithms: Foundations and applications, Berlin, Heidelberg (pp. 169–178).
7. Mirjalili S Moth-flame optimization algorithm: A novel nature-inspired heuristic paradigm Knowledge-Based Systems 2015 89 228 249 10.1016/j.knosys.2015.07.006
8. Yang X Li W Su L Wang Y Yang A An improved evolution fruit fly optimization algorithm and its application Neural Computing and Applications 2020 32 9897 9914 10.1007/s00521-019-04512-2
9. Heidari AA Mirjalili S Faris H Aljarah I Mafarja M Chen H Harris hawks optimization: Algorithm and applications Future Generation Computer Systems-The International Journal of Escience 2019 97 849 872 10.1016/j.future.2019.02.028
10. Li SM Chen HL Wang MJ Heidari AA Mirjalili S Slime mould algorithm: A new method for stochastic optimization Future Generation Computer Systems-the International Journal of Escience 2020 111 300 323 10.1016/j.future.2020.03.055
11. Kang H Bei F Shen Y Sun X Chen Q A diversity model based on dimension entropy and its application to swarm intelligence algorithm Entropy 2021 10.3390/e23040397
12. Hu J Wu H Zhong B Xiao R Swarm intelligence-based optimisation algorithms: An overview and future research issues International Journal of Automation and Control 2020 14 656 693 10.1504/IJAAC.2020.110077
13. Anam S. & Fitriah Z. (2021). Early blight disease segmentation on tomato plant using K-means algorithm with swarm intelligence-based algorithm. International Journal of Mathematics and Computer Science, 16, 1217–1228. https://ijmcs.future-in-tech.net/16.4/R-Anam.pdf
14. Al-Mousawi AJ Wireless communication networks and swarm intelligence Wireless Networks 2021 27 1755 1782 10.1007/s11276-021-02545-x
15. Kumar G Anwar A Dikshit A Poddar A Soni U Song WK Obstacle avoidance for a swarm of unmanned aerial vehicles operating on particle swarm optimization: A swarm intelligence approach for search and rescue missions Journal of the Brazilian Society of Mechanical Sciences and Engineering 2022 44 56 10.1007/s40430-022-03362-9
16. Xiang T Multi-scale feature fusion based on swarm intelligence collaborative learning for full-stage anti-interference object tracking Journal of Ambient Intelligence and Humanized Computing 2020 10.1007/s12652-019-01671-x
17. Tunay M Pashaei E Pashaei E Hybrid hypercube optimization search algorithm and multilayer perceptron neural network for medical data classification Computational Intelligence and Neuroscience 2022 2022 16 10.1155/2022/1612468
18. Bharanidharan N Rajaguru H Performance enhancement of swarm intelligence techniques in dementia classification using dragonfly-based hybrid algorithms International Journal of Imaging Systems and Technology 2020 30 57 74 10.1002/ima.22365
19. Bach HN Xue B Zhang M A survey on swarm intelligence approaches to feature selection in data mining Swarm and Evolutionary Computation 2020 10.1016/j.swevo.2020.100663
20. Nautiyal B Prakash R Vimal V Liang G Chen H Improved Salp swarm algorithm with mutation schemes for solving global optimization and engineering problems Engineering with Computers 2021 10.1007/s00366-020-01252-z
21. Yıldız B Pholdee N Bureerat S Erdaş M Yildiz A Sait S Comparison of the political optimization algorithm, the Archimedes optimization algorithm and the Levy flight algorithm for design optimization in industry Materials Testing 2021 63 356 359 10.1515/mt-2020-0053
22. Yildiz BS Patel V Pholdee N Sait SM Bureerat S Yildiz AR Conceptual comparison of the ecogeography-based algorithm, equilibrium algorithm, marine predators algorithm and slime mold algorithm for optimal product design Materials Testing 2021 63 336 340 10.1515/mt-2020-0049
23. Mirjalili S Lewis A The whale optimization algorithm Advances in Engineering Software 2016 95 51 67 10.1016/j.advengsoft.2016.01.008
24. Wang M Chen H Chaotic multi-swarm whale optimizer boosted support vector machine for medical diagnosis Applied Soft Computing 2020 10.1016/j.asoc.2019.105946
25. Chen HL Xu YT Wang MJ Zhao XH A balanced whale optimization algorithm for constrained engineering design problems Applied Mathematical Modelling 2019 71 45 59 10.1016/j.apm.2019.02.004
26. Abdel-Basset M Mohamed R El-Fergany A Askar SS Abouhawwash M Efficient ranking-based whale optimizer for parameter extraction of three-diode photovoltaic model: Analysis and validations Energies 2021 14 1 21 10.3390/en14133729
27. Peng H Wen WS Tseng ML Li LL A cloud load forecasting model with nonlinear changes using whale optimization algorithm hybrid strategy Soft Computing 2021 25 10205 10220 10.1007/s00500-021-05961-5
28. Chakraborty S Kumar SA Sharma S Mirjalili S Chakraborty R A novel enhanced whale optimization algorithm for global optimization Computers and Industrial Engineering 2021 10.1016/j.cie.2020.107086
29. Ye XJ Liu W Li H Wang MJ Chi C Liang GX Chen HL Huang HL Modified whale optimization algorithm for solar cell and PV module parameter identification Complexity 2021 2021 1 23 10.1155/2021/8878686
30. Mostafa A Hassanien AE Houseni M Hefny H Liver segmentation in MRI images based on whale optimization algorithm Multimedia Tools and Applications 2017 76 24931 24954 10.1007/s11042-017-4638-5
31. Yang C Wang YZ Inversion of the surface duct from radar sea clutter using the improved whale optimization algorithm Electromagnetics 2019 39 611 627 10.1080/02726343.2019.1675443
32. Hassib EM El-Desouky AI Labib LM El-kenawy ESM WOA plus BRNN: An imbalanced big data classification framework using Whale optimization and deep neural network Soft Computing 2020 24 5573 5592 10.1007/s00500-019-03901-y
33. Sayed GI Darwish A Hassanien AE A new chaotic whale optimization algorithm for features selection Journal of Classification 2018 35 300 344 10.1007/s00357-018-9261-2
34. Tripathi AK Mittal H Saxena P Gupta S A new recommendation system using map-reduce-based tournament empowered Whale optimization algorithm Complex and Intelligent Systems 2021 7 297 309 10.1007/s40747-020-00200-0
35. Hussien AG Hassanien AE Houssein EH Amin M Azar AT New binary whale optimization algorithm for discrete optimization problems Engineering Optimization 2020 52 945 959 10.1080/0305215x.2019.1624740
36. Chakraborty S Saha AK Chakraborty R Saha M An enhanced whale optimization algorithm for large scale optimization problems Knowledge-Based Systems 2021 10.1016/j.knosys.2021.107543
37. Luo J Chen HL Heidari AA Xu YT Zhang Q Li CY Multi-strategy boosted mutative whale-inspired optimization approaches Applied Mathematical Modelling 2019 73 109 123 10.1016/j.apm.2019.03.046
38. Saha N Panda S Cosine adapted modified whale optimization algorithm for control of switched reluctance motor Computational Intelligence 2022 38 978 1017 10.1111/coin.12310
39. Tu J Chen H Liu J Heidari AA Zhang X Wang M Ruby R Pham Q-V Evolutionary biogeography-based whale optimization methods with communication structure: Towards measuring the balance Knowledge-Based Systems 2021 10.1016/j.knosys.2020.106642
40. Heidari AA Aljarah I Faris H Chen H Luo J Mirjalili S An enhanced associative learning-based exploratory whale optimizer for global optimization Neural Computing and Applications 2020 32 5185 5211 10.1007/s00521-019-04015-0
41. Chen HL Yang CJ Heidari AA Zhao XH An efficient double adaptive random spare reinforced whale optimization algorithm Expert Systems with Applications 2020 10.1016/j.eswa.2019.113018
42. Chakraborty S Saha AK Sharma S Chakraborty R Debnath S A hybrid whale optimization algorithm for global optimization Journal of Ambient Intelligence and Humanized Computing 2021 10.1007/s12652-021-03304-8
43. Zhao D Liu L Yu FH Heidari AA Wang MJ Oliva D Muhammad K Chen HL Ant colony optimization with horizontal and vertical crossover search: Fundamental visions for multi-threshold image segmentation Expert Systems with Applications 2021 10.1016/j.eswa.2020.114122
44. Elhosseini MA Haikal AY Badawy M Khashan N Biped robot stability based on an A-C parametric whale optimization algorithm Journal of Computational Science 2019 31 17 32 10.1016/j.jocs.2018.12.005
45. Chakraborty S Sharma S Saha AK Chakraborty S SHADE–WOA: A metaheuristic algorithm for global optimization Applied Soft Computing 2021 10.1016/j.asoc.2021.107866
46. Wang WL Li WK Wang Z Li L Opposition-based multi-objective whale optimization algorithm with global grid ranking Neurocomputing 2019 341 41 59 10.1016/j.neucom.2019.02.054
47. Yang L Chen HX Fault diagnosis of gearbox based on RBF-PF and particle swarm optimization wavelet neural network Neural Computing and Applications 2019 31 4463 4478 10.1007/s00521-018-3525-y
48. Got A Moussaoui A Zouache D Hybrid filter-wrapper feature selection using whale optimization algorithm: A multi-objective approach Expert Systems with Applications 2021 10.1016/j.eswa.2021.115312
49. Abdel-Basset M Mohamed R AbdelAziz NM Abouhawwash M HWOA: A hybrid whale optimization algorithm with a novel local minima avoidance method for multi-level thresholding color image segmentation Expert Systems with Applications 2022 10.1016/j.eswa.2021.116145
50. Qi A Zhao D Yu F Heidari AA Chen H Xiao L Directional mutation and crossover for immature performance of whale algorithm with application to engineering optimization Journal of Computational Design and Engineering 2022 9 519 563 10.1093/jcde/qwac014
51. Wang G Gui W Liang G Zhao X Wang M Mafarja M Turabieh H Xin J Chen H Ma X Sun Y Spiral motion enhanced elite whale optimizer for global tasks Complexity 2021 2021 1 33 10.1155/2021/8130378
52. Huynh-Thu Q Ghanbari M Scope of validity of PSNR in image/video quality assessment Institution of Engineering and Technology 2008 44 13 800 801
53. Zhou W Bovik AC Sheikh HR Simoncelli EP Image quality assessment: From error visibility to structural similarity IEEE Transactions on Image Processing 2004 13 600 612 10.1109/TIP.2003.819861 15376593
54. Zhang L Zhang L Mou X Zhang D FSIM: A feature similarity index for image quality assessment IEEE Transactions on Image Processing 2011 20 2378 2386 10.1109/TIP.2011.2109730 21292594
55. Rahnamayan S., Tizhoosh H. R. & Salama M. M. A. (2007). Quasi-oppositional differential evolution. In Proceedings of 2007 IEEE Congress on Evolutionary Computation (pp. 2229–2236).
56. Tizhoosh H. R. (2005). Opposition-based learning: A new scheme for machine intelligence. In Proceedings of International Conference on Computational Intelligence for Modeling Control and Automation, Vienna, Austria (pp. 695–701).
57. Wan J Chen H Li T Yuan Z Liu J Huang W Interactive and complementary feature selection via fuzzy multigranularity uncertainty measures IEEE Transactions on Cybernetics 2021 10.1109/TCYB.2021.3112203
58. Agarwal M Srivastava GMS Opposition-based learning inspired particle swarm optimization (OPSO) scheme for task scheduling problem in cloud computing Journal of Ambient Intelligence and Humanized Computing 2021 12 9855 9875 10.1007/s12652-020-02730-4
59. Deng W Shang SF Cai X Zhao HM Song YJ Xu JJ An improved differential evolution algorithm and its application in optimization problem Soft Computing 2021 25 5277 5298 10.1007/s00500-020-05527-x
60. Jiao S Chong G Huang C Hu H Wang M Heidari AA Chen H Zhao X Orthogonally adapted Harris hawks optimization for parameter estimation of photovoltaic models Energy 2020 10.1016/j.energy.2020.117804
61. Wang H Rahnamayan S Sun H Omran MGH Gaussian bare-bones differential evolution IEEE Transactions on Cybernetics 2013 43 634 647 10.1109/tsmcb.2012.2213808 23014758
62. Kennedy J. (2003). Bare bones particle swarms. In Proceedings of IEEE swarm intelligence symposium (pp. 80–87).
63. Storn R Price K Differential evolution – a simple and efficient heuristic for global optimization over continuous spaces Journal of Global Optimization 1997 11 341 359 10.1023/A:1008202821328
64. Pan W-T A new fruit fly optimization algorithm: Taking the financial distress model as an example Knowledge-Based Systems 2012 26 69 74 10.1016/j.knosys.2011.07.001
65. Kennedy J. & Eberhart R. (1995). Particle swarm optimization. In Proceedings of ICNN’95—international conference on neural networks (vol.1944, pp. 1942–1948).
66. Yang YT Chen HL Heidari AA Gandomi AH Hunger games search: Visions, conception, implementation, deep analysis, perspectives, and towards performance shifts Expert Systems with Applications 2021 10.1016/j.eswa.2021.114864
67. Demšar J Statistical comparisons of classifiers over multiple data sets Journal of Machine Learning Research 2006 7 1 30
68. Arora S Anand P Binary butterfly optimization approaches for feature selection Expert Systems with Applications 2019 116 147 160 10.1016/j.eswa.2018.08.051
69. Thaher T Chantar H Too J Mafarja M Turabieh H Houssein EH Boolean particle swarm optimization with various evolutionary population dynamics approaches for feature selection problems Expert Systems with Applications 2022 10.1016/j.eswa.2022.116550
70. Takieldeen AE El-kenawy E-SM Hadwan M Zaki RM Dipper throated optimization algorithm for unconstrained function and feature selection Cmc-Computers Materials and Continua 2022 72 1465 1481 10.32604/cmc.2022.026026
71. Qaraad M Amjad S Hussein NK Elhosseini MA Large scale salp-based grey wolf optimization for feature selection and global optimization Neural Computing and Applications 2022 34 8989 9018 10.1007/s00521-022-06921-2
72. Kitonyi PM Segera DR Hybrid gradient descent grey wolf optimizer for optimal feature selection Biomed Research International 2021 10.1155/2021/2555622
73. Dua, D. & Graff, C. (2019). UCI Machine Learning Repository. Irvine, CA: University of California, School of Information and Computer Science. http://archive.ics.uci.edu/ml
74. Nadimi-Shahraki MH Banaie-Dezfouli M Zamani H Taghian S Mirjalili S B-MFO: A binary moth-flame optimization for feature selection from medical datasets Computers 2021 10 136 10.3390/computers10110136
75. Abdel-Basset M Mohamed R Sallam KM Chakrabortty RK Ryan MJ BSMA: A novel metaheuristic algorithm for multi-dimensional knapsack problems: Method and comprehensive analysis Computers and Industrial Engineering 2021 10.1016/j.cie.2021.107469
76. Hashim FA Hussien AG Snake optimizer: A novel meta-heuristic optimization algorithm Knowledge-Based Systems 2022 242 108320 10.1016/j.knosys.2022.108320
77. Kumar V Kumar D Binary whale optimization algorithm and its application to unit commitment problem Neural Computing and Applications 2020 32 2095 2123 10.1007/s00521-018-3796-3
78. Chen Y Xie W Zou X A binary differential evolution algorithm learning from explored solutions Neurocomputing 2015 149 1038 1047 10.1016/j.neucom.2014.07.030
79. Arlot S Celisse A A survey of cross-validation procedures for model selection Statistics Surveys 2010 4 40 79 10.1214/09-ss054
80. Hu J Chen HL Heidari AA Wang MJ Zhang XQ Chen Y Pan ZF Orthogonal learning covariance matrix for defects of grey wolf optimizer: Insights, balance, diversity, and feature selection Knowledge-Based Systems 2021 10.1016/j.knosys.2020.106684
81. Chen Y Wang M Heidari AA Shi B Hu Z Zhang Q Chen H Mafarja M Turabieh H Multi-threshold image segmentation using a multi-strategy shuffled frog leaping algorithm Expert Systems with Applications 2022 10.1016/j.eswa.2022.116511
82. Abutaleb AS Automatic thresholding of gray-level pictures using two-dimensional entropy Computer Vision, Graphics, and Image Processing 1989 47 22 32 10.1016/0734-189X(89)90051-0
83. Ji W He X Kapur’s entropy for multilevel thresholding image segmentation based on moth-flame optimization Mathematical Biosciences and Engineering 2021 18 7110 7142 10.3934/mbe.2021353 34814242
84. Wang S Sun K Zhang W Jia H Multilevel thresholding using a modified ant lion optimizer with opposition-based learning for color image segmentation Mathematical Biosciences and Engineering 2021 18 3092 3143 10.3934/mbe.2021155 34198377
85. Zhao D Liu L Yu F Heidari AA Wang M Liang G Muhammad K Chen H Chaotic random spare ant colony optimization for multi-threshold image segmentation of 2D Kapur entropy Knowledge-Based Systems 2021 10.1016/j.knosys.2020.106510
86. Cai J Luo T Xu G Tang Y A Novel biologically inspired approach for clustering and multi-level image thresholding: Modified Harris hawks optimizer Cognitive Computation 2022 14 955 969 10.1007/s12559-022-09998-y
87. Chakraborty S Saha AK Nama S Debnath S COVID-19 X-ray image segmentation by modified whale optimization algorithm with population reduction Computers in Biology and Medicine 2021 139 104984 10.1016/j.compbiomed.2021.104984 34739972
88. Wang X Li Z Kang H Huang Y Gai D Medical image segmentation using PCNN based on multi-feature grey wolf optimizer bionic algorithm Journal of Bionic Engineering 2021 18 711 720 10.1007/s42235-021-0049-4
89. Sharma S Saha AK Majumder A Nama S MPBOA—a novel hybrid butterfly optimization algorithm with symbiosis organisms search for global optimization and image segmentation Multimedia Tools and Applications 2021 80 12035 12076 10.1007/s11042-020-10053-x
90. Chakraborty S Sharma S Saha AK Saha A A novel improved whale optimization algorithm to solve numerical optimization and real-world applications Artificial Intelligence Review 2022 55 4605 4716 10.1007/s10462-021-10114-z
91. Zhao S Wang P Heidari AA Chen H He W Xu S Performance optimization of salp swarm algorithm for multi-threshold image segmentation: Comprehensive study of breast cancer microscopy Computers in Biology and Medicine 2021 139 105015 10.1016/j.compbiomed.2021.105015 34800808
92. Cohen J. P., Morrison P., Dao L., Roth K., Duong T. Q. & Ghassemi M. 2020. COVID-19 image data collection: Prospective predictions are the future. 10.48550/arXiv.2006.11988
93. Ewees AA Al-qaness MAA Abd EM Enhanced salp swarm algorithm based on firefly algorithm for unrelated parallel machine scheduling with setup times Applied Mathematical Modelling 2021 94 285 305 10.1016/j.apm.2021.01.017
94. Wang BC Li HX Zhang QF Wang Y Decomposition-based multiobjective optimization for constrained evolutionary optimization IEEE Transactions on Systems, Man, and Cybernetics: Systems 2021 51 574 587 10.1109/tsmc.2018.2876335
95. Naderipour A Abdullah A Marzbali MH Arabi NS An improved corona-virus herd immunity optimizer algorithm for network reconfiguration based on fuzzy multi-criteria approach Expert Systems with Applications 2022 187 115914 10.1016/j.eswa.2021.115914 34566274
96. Li F Su Z Wang G An effective dynamic immune optimization control for the wastewater treatment process Environmental Science and Pollution Research 2022 10.1007/s11356-021-17505-3
| 36466725 | PMC9707266 | NO-CC CODE | 2022-12-01 23:20:22 | no | J Bionic Eng. 2022 Nov 28;:1-22 | utf-8 | J Bionic Eng | 2,022 | 10.1007/s42235-022-00297-8 | oa_other |
==== Front
Int Urol Nephrol
Int Urol Nephrol
International Urology and Nephrology
0301-1623
1573-2584
Springer Netherlands Dordrecht
36443608
3416
10.1007/s11255-022-03416-w
Urology - Original Paper
The danger of hyperoxia on the rat kidneys: is tadalafil a real shield?
Kilicarslan Nermin 1
http://orcid.org/0000-0002-3089-7870
Demir Aslan [email protected]
5
Yeni Sezgin 2
Cicek Mehmet Cagatay 2
Saricetin Aysun 3
Dirican Melahat 4
1 grid.488643.5 0000 0004 5894 3909 Anesthesia Department, Bursa Yüksek İhtisas Training and Research Hospital, University of Health Sciences, Bursa, Turkey
2 grid.34538.39 0000 0001 2182 4517 Urology Department, Medical Faculty, Bursa Uludag University, Bursa, Turkey
3 grid.34538.39 0000 0001 2182 4517 Pathology Department, Veterinary Faculty, Bursa Uludag University, Bursa, Turkey
4 grid.34538.39 0000 0001 2182 4517 Biochemistry Department, Medical Faculty, Bursa Uludag University, Bursa, Turkey
5 Urology Department, Medical Faculty, Dragos Hospital, Bezmialam Foundation University, Dragos Hastanesi, Sahil Yolu Sok. No: 16 Maltepe, 34844 Istanbul, Turkey
29 11 2022
17
25 9 2022
14 11 2022
© The Author(s), under exclusive licence to Springer Nature B.V. 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Purpose
Continuous oxygen therapy to compensate for decreased oxygen saturation in the blood is a life-saving treatment used in case lung involvement. Excess oxygen delivery was reported to be a common situation, in which about 50% of the patients showed hyperoxemia and 4% in severe hyperoxemia. In this work, we investigated the effects of hyperoxia on the rat kidneys and whether tadalafil has an effect to reduce this damage.
Materials and methods
Three groups of 8 male rats each weighing 300–350 g were formed. The groups were divided into the control group, hyperoxia group, and hyperoxia and tadalafil administered group for 10 days. At the end of the 10th day, blood and kidney samples were taken for biochemical analysis (SOD and NO levels) and histopathological examination.
Results
While our findings showed that SOD levels were significantly different among the control and experimental groups and within the experimental groups, no statistical difference was found in terms of NO levels among the groups (Table 1).
While the glomerular and tubular injury was higher in the Hyperoxia group and the Hyperoxia + Tadalafil group than in the control group (p < 0.001), as a result of the rate of severe glomerular and tubular injury in the hyperoxia group, was 62.5% and 43.8% and in the group given tadalafil was 43.8% and 31.3%, respectively (Table 2).
Conclusions
Exposure to hyperoxia condition causes renal glomerular and tubular damage, and tadalafil does not show a protective effect on this damage according to this study’s dose and exposure time.
Keywords
Hyperoxia
Tadalafil
Rat kidneys
Superoxide dismutase
Nitric oxide
==== Body
pmcIntroduction
Many patients have been exposed to oxygen therapy due to their decreased saturation depending on lung involvement because of some diseases, such as chronic obstructive pulmonary disease, cancer, some respiratory disorders in newborns, and some infectious diseases, especially, currently COVID-19 [1, 2]. Continuous oxygen therapy to compensate for decreased oxygen saturation in the blood is a life-saving treatment commonly used in critically ill patients [3, 4]. High-flow oxygen therapy was recommended for patients with acute respiratory failure and who do not respond to conventional therapy [5]. Excess oxygen delivery was reported to be a common situation, in which about 50% of the patients showed hyperoxemia and 4% in severe hyperoxemia [4].
Some animal studies in the literature show that the state of hyperoxia causes histopathological injuries in some organs, such as the lung, brain, cardiac and vascular system, pampiniform plexus, and testes, and may affect biological systems such as antioxidant enzymes and cytokine production through excessive production of reactive oxygen species (ROS) and generate free radical-mediated damages in various organs [1–3, 6–8] and thus may cause the affected organ failure [1–3, 6–8].
ROS are metabolites of oxygen and hence are produced under hyperoxic conditions. All living aerobic cells are normally exposed to some ROS that is toxic to living organisms [9, 10]. In fact, the body tries to eliminate the harmful effects of ROS with its antioxidant defense systems such as superoxide dismutase (SOD) and catalase that interact with them and protect the organism against these metabolites [1, 11, 12]. As a result of normal living conditions, the living organism manages to survive in a healthy way with a delicate balance between the positive and negative effects of ROS products to which it is constantly exposed [13]. The antioxidant defense mechanism achieves these processes endogenously in an enzymatic way (such as SOD, glutathione peroxidase (GPx), catalase (CAT) and exogenously in a non-enzymatic way [ascorbic acid (Vitamin C), -tocopherol (Vitamin E), glutathione (GSH), carotenoids, flavonoids, and other antioxidants] [1, 14].
Exogenous antioxidants are needed when the endogenous enzymatic system is insufficient to reduce ROS-related damage. In the literature, it is possible to find many studies using tadalafil as an exogenous antioxidant to reduce various organ damage due to ROS. However, these studies differ from each other in terms of their results [1, 15–18]. With this study model, we investigated the effects of hyperoxia on the rat kidneys and whether tadalafil has an effect to reduce this damage, to shed some light on this issue.
Materials and methods
All experiments were performed in accordance with the guidelines for animal research from the National Institutes of Health and were approved by our Animal Local Ethics Committee with the number of 2022–01/10. B.30.2.ULU.0.8Z.00.00/168).
Animal model
Twenty-four mature male Sprague–Dawley rats were 6 months and weighing 300–350 g kept in standard conditions with free access to food and water. The rats were randomly divided into three groups with an equal number of 8. The groups were formed as the control (C), the hyperoxia (H), and the hyperoxia + tadalafil (H + T) groups, respectively.
Groups
Control group (C) Eight adult male rats were kept in room conditions (FiO2 = 0.21), humidity rate of 60 ± 5%, and room temperature of 21 ± 1 °C. Surgery was performed under general anesthesia. Hyperoxia group (H) Eight male rats were exposed to hyperoxia (FiO2 = 0.96) in a hyperoxia cabinet for 8 h/day for 10 days. All rats were removed from the oxygen cabinet and placed in separate cages. Each rat was fed with an amount of 10–12 ml/100 g of water and sufficient food, daily. Hyperoxia + tadalafil (HT) group Eight adult male rats were exposed to hyperoxia in a hyperoxia cabinet for 8 h/day for 10 days. While the rats are in the hyperoxia cabin, the rats were fed food and water. After the rats were removed from the cabin, tadalafil was dissolved in water and the amount of a dose of 10 mg/kg was given to the rats. Ten days later, having taken some blood samples for biochemical analysis from each group, surgery was performed under general anesthesia to take out the kidneys for histopathological examination.
Creation of hyperoxia
The oxygen was given to the rats in a cabin via an inlet hole at one end and exhaled air was expelled from the outlet hole from the other end. While the exit hole of the cabin was left open, the entrance hole was connected to the oxygen cylinder with an extension hose. Oxygen was continuously supplied to the cabin, so that the FiO2 = 0.96 (96%) value of the air inside the cabin remained constant. The FiO2 value in the exit hole of the cabin was checked every 2 h with an oxygen monitor (Datex-Ohmeda™ 5120 oxygen monitor and oxygen sensor; BOC Healthcare, Chalfont St. Giles, UK) calibrated with the help of a cannula.
Surgical method
Sevoflurane (Sevorane, Abbott USA) was administered at 1–2% volume concentration in 4 lt/min O2 flow using Dräger Titus (Lübeck, Germany) brand device to provide anesthesia. First, a thoracotomy incision was made to access the heart. After anesthesia was achieved, both kidneys were removed and the heart was reached by thoracotomy incision for biochemical analysis, and the rats were sacrificed after blood samples were taken. The excised tissues were kept for histopathological examination in a 10% formaldehyde solution.
Biochemical examination
The samples were centrifuged at 2500 g for 10 min, and serum aliquots were stored at − 20 °C until the analysis. Serum nitric oxide (NO) levels were measured by an enzyme-linked immunosorbent assay (ELISA) kit for rats (Bioassay Technology Laboratory, Shangai, China) in accordance with the manufacturer’s instructions. Samples were placed into wells coated with a rat NO antibody used as a catcher. Biotinylated rat NO antibody was added and bound to NO in the sample. Then, streptavidin–horseradish peroxidase was added. Following the sandwich formation, the excess unbound conjugate was removed and substrate solution was added. The absorbances were determined with an ELISA reader at 450 nm (FLASH Scan S12, Analytic Jena AG, Germany). Levels of serum NO concentrations were shown as µmol/L. Rat serum SOD1 levels were determined by ELISA kit (Bioassay Technology Laboratory, China) according to the principle mentioned above. SOD1 levels were expressed as ng/mL.
Histopathologic examination
The kidneys were fixed in 10% neutral buffered formalin for 24 h. Tissues were paraffinized following graded ethyl alcohol dehydration. Paraffin blocks were cut to obtain 4-micron ribbons. Ribbons were transferred to slides and stained with routine hematoxylin and eosin staining technique after deparaffinization. Each slide was examined and graded under a light microscope (Olympus CX41) semi-quantitatively. The sections were photographed directly using a stereomicroscope at 400 × high-power fields.
The glomeruli and convoluted tubules in the cortex of the kidneys were examined. Endothelial and epithelial cell proliferation of the capillaries cluster in the glomerulus was scored between 0 and 3 (0: no, 1: mild, 2: moderate, 3: severe). Parenchymal degeneration of the tubules was evaluated. It was scored according to the degree of narrowing of the tubular lumen and swelling of the epithelial cell tissue in the tubule (0: no, 1: mild, 2: moderate, 3: severe). Fifteen glomeruli and tubules were examined in the cortical region, and the average was taken. The mean of the scores given by evaluating 15 glomeruli in each rat was calculated.
Statistical analysis
Data were analyzed using SPSS-22 for Windows (SPSS, Inc., Chicago, IL, USA) software. The continuous data were not normally distributed. Statistical analysis of the medians of continuous variables was performed with the Kruskal–Wallis test. Results were given as median, minimum and maximum values. Categorical variables were analyzed with a Chi-square test, and p < 0.05 was considered statistically significant.
Results
Table 1 shows the biochemical data and analysis results of the study groups (Table 1). While our findings showed that SOD levels were significantly different among the control and experimental groups and within the experimental groups, no statistical difference was found in terms of NO levels among the groups (Table 1). According to our findings, the SOD level was the highest in the hyperoxia group and the lowest in the hyperoxia + tadalafil group (Table 1).Table 1 Biochemical data analysis results in the groups
Groups
Control (C) Hyperoxia (H) Hyperoxia + Tadalafil (H + T) P
n 8 8 8
SOD 4.1 5.5 3.7 < 0.001 H vs HT, p < 0.05
C vs H, p < 0.05
C vs HT, p > 0.05
NO 48.9 56.4 51.6 0.238
Data are presented as mean ± SD. SOD; superoxide dismutase, ng/mL
NO nitric oxide, µmol/L
p ≤ 0.05 was considered to be statistically significant
Analysis results of histopathological findings in study groups are summarized in Tables 2 and 3. Glomerular and tubular injuries were higher in Group H and H + T than in Group C. The photo of the histopathologic findings related to normal structures of the glomeruli and tubule for the control group is shown in Fig. 1. In Fig. 2A, B pathologic findings related to the proliferation of endothelial and epithelial cells and filling of the Bowman’s space in glomeruli for the hyperoxia group (H) and tadalafil group (H + T) are shown and in Fig. 3A, B hydropic degeneration of the tubular epithelium and the enlargement of the tubule.Table 2 Glomerular damage degree and epithelial proliferation status with analysis results, a p value of < 0.05 was accepted significant, R Pearson’s R value
Groups Glomerular injury
No Mild Moderate Severe P
Control (C) 16
100%
0 0 0 C vs H < 0.05
C vs H + T < 0.05
H vs H + T > 0.05
Hyperoxia (H) 0 1
6.25%
5
31.25%
10
62.5%
Hyperoxia + Tadalafil (H + T) 0 1
6.25%
8
50%
7
43.75%
Table 3 Renal tubular damage and parenchymal degeneration degree with analysis results, a p value of < 0.05 was accepted significant, R Pearson’s R value
Groups Renal tubular injury P
No Mild Moderate Severe
Control (C) 10
62.5%
6
37.5%
0 0 C vs H < 0.001
C vs HT < 0.001
H vs HT = 0.545
Hyperoxia (H) 0 2
12.5%
7
43.75%
7
43.75%
Hyperoxia + Tadalafil (H + T) 0 4
25%
7
43.75%
5
31.25%
Fig. 1 C group—normal glomeruli and tubule structures, H&E, 20
Fig. 2 A H group—proliferation of endothelial and epithelial cells in glomeruli and narrowing of Bowman's space, H&E, 20×. B HT group—filling the Bowman’s space by proliferation of endothelial and epithelial cells, H&E 10×
Fig. 3 A H group—hydropic degeneration of tubular epithelium and enlargement of tubules, H&E, 20×. B HT group—mild tubular deformation, H&E, 4×
While the glomerular and tubular injury was higher in the H group and the H + T group than in the control group (p < 0.001), as a result of the rate of severe glomerular and tubular injury in the hyperoxia group (H), was 62.5% and 43.8% and in the group given tadalafil (H + T) was 43.8% and 31.3%, respectively (Table 2).
Discussion
Oxygen therapy is required in 15–20% of hospitalized patients and hyperoxia occurs in approximately half of them [19], and also currently, the high prevalence of hyperoxia was shown during the COVID-19 pandemic, as oxygen therapy was given to 75% of patients hospitalized for Covid 19 infection [20]. Damiani and Girardi reported that more oxygen therapy without hypoxia causes oxidative stress (OS) (which is defined as the deterioration of the balance between ROS and antioxidants in favor of ROS [21]) and inflammation, especially in patients with myocardial infarction (MI), stroke, cardiac arrest, and sepsis [22]. Increased OS can trigger oxidative changes that can lead to molecular and cellular dysfunction by damaging proteins, lipids, and nucleic acids which provide intracellular signaling in the cell thereby disrupting intracellular signaling pathways [23] and inducing proinflammatory responses [24].
Based on our research, we can say that among the studies reporting that exposure to hyperoxia causes damage to various organs by increasing oxidative stress, and exposures related to the kidneys, especially in the neonatal period [2]. Human and animal studies have shown that neonatal hyperoxia increases OS and can adversely affect glomerular and tubular maturity, leading to renal corpuscle enlargements, renal tubular necrosis, interstitial inflammation, and fibrosis. [2]. According to studies conducted on rats in the neonatal period, it has been reported that this situation may pave the way for hypertension in adult life [2]. However, we can say that there is only one study related to the adult kidney [25], and according to the conclusion of this study, among critically ill patients with 2 or more SIRS criteria, treatment with a low-normal Pao2 target compared with a high-normal Pao2 target did not result in a statistically significant reduction in organ dysfunction. However, the authors have mentioned that their study may have had limited power to make more clear this issue more, as well [25].
Normally, the body's defense and protection system aims to protect the integrity of the cell for normal function of the cell by activating the antioxidant mechanism against ROS up to a certain threshold. Therefore, the cell tries to deal with the OS in this way [1]. Once this threshold point is exceeded in favor of ROS, the cell needs exogenous antioxidants to maintain its integrity and thus function, and this suggests that the balance between SOD and ROS is achieved by some kind of feedback mechanism [1]. The increase in SOD, which is one of the antioxidant markers that increase in response to ROS, can be used as a biochemical indicator of the increased OS [21].
In this experimental study, we aimed to show the effects of hyperoxia in adult rat kidneys by increasing OS with the hyperoxia model, both histopathologically and by measuring SOD, which is one of the antioxidant biochemical markers. In addition, we aimed to investigate the contradictory results about the effect of tadalafil, which has been used as an exogenous antioxidant in many studies, with the same method.
In our study, based on the analysis of the studied biochemical parameters, while there was a difference in SOD levels between the hyperoxia group and the control group, this difference disappeared with the use of tadalafil in the hyperoxia + tadalafil group. In this work, while OS increases with the hyperoxia method, SOD increases to compensate for its formed damage, and when an exogenous antioxidant (which in our study is tadalafil) is used, the SOD level decreases in this group as the need for SOD will also decrease. In their experimental study examining the relationship between hyperoxia–ischemic reperfusion and SOD, Cai et al. reported that the SOD level was higher in the normobaric hyperoxia + ischemia + reperfusion group than in the ischemic–reperfusion group alone [26]. And thus, the authors concluded that normobaric hyperoxia intervention may have protective effects on rats’ kidneys having a renal ischemia–reperfusion injury, probably by means of the increased SOD level in that group [26]. The difference between our study and the study of Cai et al. was that the level of SOD was measured in serum, while they measured it in kidney tissue. However, we found that the NO level was not affected by the increased OS level and its damage. This may be related to the duration and/or amount of exposure to hyperoxia and thus to the level of OS damage. To further clarify this issue experimental and clinical randomized controlled studies are needed.
Tubular damage rate, which is one of the histopathological findings of the hyperoxia effect in our study (Figs. 1, 2 and 3), was also shown in some studies in the literature [2, 27–29] in neonatal rat kidneys, but unlike these studies, we preferred adult rats in our study. Xu et al. stated that inhibited expression of tight junction proteins (claudin-4, occludin, and ZO-1) in proximal tubules might be a potential mechanism of neonatal hyperoxia-induced nephrogenic disorders [28]. In the analysis of the histopathological findings of our study, we see that the state of hyperoxia causes glomerular and tubular damage due to increased oxidative stress, and this damage rate does not improve much with the use of tadalafil (Tables 2 and 3). Although the difference in both glomerular and tubular damage rates was not significant in the experimental groups in total, when we analyzed according to the degree of damage, we found that tadalafil reduced the degree of severe damage in the tadalafil-treated group. In our study, we attributed the different pathological findings in both kidneys of a rat due to hyperoxia, to the fact that these two kidneys may have different anatomical structures and this structure may give different responses to rat metabolism. In a similar study in the literature, but in the rat ovaries, it was reported that tadalafil at a dose of 5 mg/kg decreased the rate of ovarian damage and accelerated the recovery rate [15]. In the current study, the insignificance of the difference in the degree of injury may be related to the tadalafil dose and treatment duration in the tadalafil-treated group, and an important difference from the study by Arikan et al. [15] is that it was a study on kidneys. In the study of Sugimoto et al. [17] investigating the effect of tadalafil at a dose of 10 mg/kg for 30 days in the rat prostate in a nonbacterial prostatitis model, it was reported that tadalafil showed anti-inflammatory and stromal anti-proliferative effects at this dose and time. In addition these, in the study of Özmerdiven et al., it was reported that tadalafil had positive effects on post-ESWL renal damage [18].
Since the histopathological findings of oxidative stress induced by hyperoxia and the analysis results of biochemical parameters show the parallelism between the groups, they confirm each other in terms of their results.
Study Limitation
To investigate the significance of only SOD and NO parameters first as the antioxidant markers due to the cost-effective approach, taking into account the budget of our clinic without measuring the levels of other antioxidant markers and failure to investigate ROS levels are the limitation of our study.
Conclusion
Based on our study results, exposure to hyperoxia condition causes renal glomerular and tubular damage, and tadalafil does not show a protective effect on this damage according to the dose and exposure time of our study. Hence, there is a need for randomized controlled experimental and clinical studies to determine the effect of tadalafil on this issue and, if any, at what dose and duration.
Author contributions
Contributed substantially to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work: NK, AD, SY, OK, AS, MD Drafted the work or revised it critically for important intellectual content: NK, AD, MCC, SY and OK.
Funding
This study was not funded by any grant.
Declarations
Conflict of interest
The authors have no conflicts of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Yeni S Demir A Kilicarslan N Cicek MC Saricetin A Dirican M Tadalafil against hyperoxia-induced oxidative stress; an experimental study Andrologia 2022 54 e14494 10.1111/and.14494 35676072
2. Jiang JS Chou HC Yeh TF Chen CM Neonatal hyperoxia exposure induces kidney fibrosis in rats Pediatr and Neonatal 2015 56 4 235 241 10.1016/j.pedneo.2014.11.003
3. Esteban A Anzuento A Alia I Gordo F Apezteguia C Palizas F Cide D How is mechanical ventilation employed in the intensive care unit? an international review Am J Respir Crit Care Med 2000 61 5 1450 1458 10.1164/ajrccm.161.5.9902018
4. Ni YN Wang YM Liang BM Liang ZA The effect of hyperoxia on mortality in critically ill patients: a systematic review and meta analysis BMC Pulm Med 2019 19 53 10.1186/s12890-019-0810-1 30808337
5. Nielsen JK Bonnesen B Hansen EF Jensen JS Lapperre TS Weinreich UM Hilberg O Guideline for the management of COVID-19 patients during hospital admission in a non-intensive care setting Eur Clin Respir J 2020 7 1761677 10.1080/20018525.2020.1761677 33224450
6. Bhandari V Maulik N Kresch M Hyperoxia causes an increase in antioxidant enzyme activity in adult and fetal rat type II pneumocytes Lung 2000 178 1 53 60 10.1007/s004080000006 10723720
7. Barazzone C White CW Mechanisms of cell injury and death in hyperoxia: role of cytokines and Bcl-2 familiy proteins Am J Respir Cell Mol Biol 2000 22 5 517 519 10.1165/ajrcmb.22.5.f180 10783120
8. Verratti V Di Giulio C Berardinelli F Tiboni GM Pellicciotta M Brunetti L Ferrante C Pampiniform plexus and oxidative stress during chronic hypoxia and hyperoxia Int J Immunopathol Pharmacol 2008 21 2 353 357 10.1177/039463200802100212 18547479
9. Mostafa T Anis TH Ghazi S El-Nashar AR Imam H Osman IA Reactive oxygen species and antioxidants relationship in the internal spermatic vein blood of infertile men with varicocele Asian J Androl 2006 8 4 451 455 10.1111/j.1745-7262.2006.00172.x 16763721
10. de Jonge E Peelen L Keijzers PJ Joore H de Lange D van der Voort PHJ Bosman RJ Association between administered oxygen, arterial partial oxygen pressure and mortality in mechanically ventilated intensive care unit patients Crit Care 2008 12 6 R156 10.1186/cc7150 19077208
11. Freeman BA Crapo JD Free radicals and tissue injury Lab Invest 1982 47 5 412 426 6290784
12. Lobo V Patil A Phatak A Chandra N Free radicals, antioxidants and functional foods: Impact on human health Pharmacogn Rev 2010 4 8 118 126 10.4103/0973-7847.70902 22228951
13. Dröge W Free radicals in the physiological control of cell function Physiol Rev 2022 82 1 47 95 10.1152/physrev.00018.2001
14. Valko M Leibfritz D Moncol J Cronin MT Mazur M Telser J Free radicals and antioxidants in normal physiological funcitons and human disease Int J Biochem Cell Biol 2007 39 1 44 84 10.1016/j.biocel.2006.07.001 16978905
15. Arikan DC Bakan V Kurutas EB Sayar H Coskum A Protective effect of tadalafil on ischemia/reperfusion injury of rat ovary J Pediatr Surg 2010 45 11 2203 2209 10.1016/j.jpedsurg.2010.07.011 21034945
16. Morelli A Comeglio P Filippi S Sarchielli E Vignozzi L Maneschi E Cellai I Mechanism of action of phosphodiesterase type 5 inhibition in metabolic syndrome-associated prostate alterations: an experimental study in the Rabbit Prostate 2013 73 428 441 10.1002/pros.22584 22996758
17. Sugimoto M Zhang X Ueda N Tsunemori H Taoka R Hayashida Y Hirama H A phosphodiesterase 5 inhibitor, tadalafil, suppresses stromal predominance and inflammation in a rat model of nonbacterial prostatitis BMC Urol 2019 19 99 10.1186/s12894-019-0525-x 31646996
18. Ozmerdiven G Vuruskan BA Kaygisiz O Vuruskan H Protective effects of diltiazem and tadalafil on shock wave-induced kidney injury in rats Bratisl Lek Listy 2017 118 4 228 232 10.4149/BLL_2017_045 28471234
19. Hale KE Gavin C O’Driscoll BR Audit of oxygen use in emergency ambulances and in a hospital emergency department Emerg Med J 2008 25 11 773 776 10.1136/emj.2008.059287 18955625
20. Wiersinga WJ Rhodes A Cheng AC Peacock SJ Prescott HC Pathophysiology, transmission, diagnosis, and treatment of coronavirus disease 2019 (COVID-19): A Review JAMA 2020 324 8 782 793 10.1001/jama.2020.12839 32648899
21. Halliwell B Biochemistry of oxidative stress Biochem Soc Trans 2007 35 5 1147 1150 10.1042/BST0351147 17956298
22. Damiani E Donati A Girardis M Oxygen in the critically ill: friend or foe? Curr Opin Anaesthesiol 2018 31 2 129 135 10.1097/ACO.0000000000000559 29334496
23. Lopez-Alarcona C Denicola A Evaluating the antioxidant capacity of natural products: a review on chemical and cellular-based assays Anal Chim Acta 2013 763 1 10 10.1016/j.aca.2012.11.051 23340280
24. Xu S Touyz RM Reactive oxygen species and vascular remodeling in hypertension: still alive Can J Cardiol 2006 22 11 947 951 10.1016/S0828-282X(06)70314-2 16971980
25. Gelissen H de Grooth HJ Smulders Y Wils EJ de Rujter W Vink R Smit B Effect of low-normal vs high-normal oxygenation targets on organ dysfunction in critically ill patients: a randomized clinical trial JAMA 2021 326 10 940 948 10.1001/jama.2021.13011 34463696
26. Cai SY Pei J Yan B Liu ZY Chen Y Sima CY Su CJ Effects of normobaric hyperoxia intervention on renal ischemia-reperfusion injury in rats Zhonghua Yi Xue Za Zhi 2021 101 14 1036 1040 33845544
27. Chen CM Chou HC Maternal inflammation exacerbates neonatal hyperoxia-induced kidney injury in rat offspring Pediatr Res 2019 86 2 174 180 10.1038/s41390-019-0413-9 31086293
28. Xu X, Zhang X, Gao L, Liu C, You K. (2020) Neonatal hyperoxia downregulates claudin-4, occludin, and ZO-1 expression in rat kidney accompanied by impaired proximal tubular development. Oxid Med Cell Longev, p 2641461. 10.1155/2020/2641461
29. Ali MF Venkatarayappa SKB Benny M Rojas C Yousefi K Shehadeh KA Kulandavelu S Effects of Klotho supplementation on hyperoxia-induced renal injury in a rodent model of postnatal nephrogenesis Pediatr Res 2020 88 4 565 570 10.1038/s41390-020-0803-z 32059229
| 36443608 | PMC9707269 | NO-CC CODE | 2022-12-01 23:20:22 | no | Int Urol Nephrol. 2022 Nov 29;:1-7 | utf-8 | Int Urol Nephrol | 2,022 | 10.1007/s11255-022-03416-w | oa_other |
==== Front
Drug Deliv Transl Res
Drug Deliv Transl Res
Drug Delivery and Translational Research
2190-393X
2190-3948
Springer US New York
36443634
1254
10.1007/s13346-022-01254-y
Review Article
A comprehensive review on the COVID-19 vaccine and drug delivery applications of interpenetrating polymer networks
http://orcid.org/0000-0003-3627-7805
Aldaais Ebtisam A. [email protected]
grid.411975.f 0000 0004 0607 035X Department of Radiological Sciences, Imam Abdulrahman Bin Faisal University, P.O. Box 2435, Dammam, 31441 Saudi Arabia
28 11 2022
119
21 10 2022
© Controlled Release Society 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Abstract
An interpenetrating polymer network (IPNs) is a concoction of two or more polymers (natural, synthetic, and/or a combination of both) in which at least one polymer is synthesized or crosslinked in the intimate presence of the other. These three-dimensional networked systems have gained prominence in a series of biomedical applications, especially in the last two decades. The last decades witnessed a surge in the meaningful applications of interpenetrating polymer networks, especially in drug delivery as simple IPN systems advanced and resulted in the formation of highly efficient microspheres, nanoparticles, nanogels, and hydrogels, intelligent enough to sense and respond to changes in external stimuli such as temperature, pH, and ionic strength. The structure of the polymers, crosslinking agents, crosslinking density, and polymerization method play an integral role in determining the properties and application of IPNs in drug delivery. This review article is a modest effort to highlight the importance and applications of different types of interpenetrating polymer networks for the sustained, site-specific drug delivery of various therapeutic formulations, as witnessed in scientific research literature over the past 22 years (2000–2022). A special section of the manuscript is devoted to studying the efficacy of network polymers in vaccine delivery and highlighting the future scope (if any) of incorporating the IPN system in COVID-related vaccine/drug delivery.
Graphic abstract
Four key focus areas in this review article [1, 2].
Keywords
Interpenetrating polymer networks (IPNs)
Drug delivery
Crosslinked structure
Controlled release
COVID
==== Body
pmcIntroduction
Incorporating polymers in bio-medicinal applications is one of the most exciting and inevitably rewarding facets of polymer chemistry in the modern scientific world. Interpenetrating polymer networks (IPNs) is a term used to define such a polymeric network. One polymer chain permeates another cross-linked polymeric chain/s without forming any covalent bonds between the two. The polymeric chains are so strongly interlaced in an IPN that it is not theoretically possible to separate them without detaching some chemical bonds [3]. IPNs garnered significant scientific interest and appreciation in the last two decades owing to their unique characteristic of combining the valuable properties of two or more individual polymers into a single polymeric network. As a result of this combination, in some cases, an interpenetrating polymer network is reported to possess entirely new properties previously not found in any of its single constituent polymers. It is analogous to synthesizing a new chemical compound with entirely new and more valuable characteristics than its constituent atomic elements. However, in this particular case, no covalent bonds are involved.
Interpenetrating polymer networks are viable for controlled and targeted drug delivery as their three-dimensional networked structure allows easy encapsulation of the drugs. A single polymer itself cannot regulate the complex parameters associated with drug delivery applications; crosslinking two or more polymer networks is a better approach. Natural polymers can deliver drugs at a controlled rate for extended periods with greater bioavailability. On the other hand, synthetic polymers can offer high mechanical strength. IPNs offer a means to preserve and combine these meaningful features of each polymer type to form an interlocked structure with improved stability for sustained/controlled drug release. The shortcomings witnessed in the literature about therapeutic dose administration by conventional routes can be overcome by using IPNs. A drug passing through the body without any external encapsulation can undergo biodegradation in the body via a series of chemical and enzymatic reactions. This may result in reduced therapeutic efficiency of the drug when it reaches its site of action. Drug entrapment in three-dimensional polymeric networks can prevent this, ensuring a specific amount of drug is released at a specific site for a controlled period. The “IPNs in drug delivery applications” section contains a detailed discussion of the application and efficacy of a wide range of IPN systems in drug delivery operations.
Other than drug delivery, the scientific community has also reported widespread applications of interpenetrating polymer networks as wound dressing formulations [4–6], for tissue engineering to design artificial scaffolds [7, 8] and injectable supports [9] and bone graft substitutes [10], etc. Incorporating natural and/or biopolymers even of synthetic origin is believed to offer a sustainable approach to IPN synthesis owing to the low toxicity profiles of such polymer subunits. Polymers like chitosan, alginate, and natural gums [11, 12] are biocompatible and biodegradable, making them good candidates for synthesizing IPN systems that could be utilized in future drug delivery and tissue engineering applications.
Stimuli-sensitive IPNs such as pH-sensitive and thermosensitive IPN systems are emerging as excellent options for controlled and targeted therapeutic delivery. Such IPNs are also called “smart” polymeric multicomponent systems as they can intelligently respond to an exogenous and/or an internal stimulus like variations in pH, temperature, and pressure. These intelligent polymers are sensitive and can be modified easily in terms of shape and volume to respond to the selected stimuli and to perform their drug delivery tasks at specific sites in the body in a controlled manner. These specific stimuli responsiveness characteristics also offer additive protection to the encapsulated drug from the surrounding biological environment. IPN hydrogels and nanogels fall under the intelligent IPNs category.
Effective drug delivery in therapeutic and preventive vaccines is another primary concern of the scientific research community, especially since the onset of the coronavirus pandemic in the world. The literature cites evidence of the use of interpenetrating polymer networks in the safe and efficient delivery of vaccines as well. Polymeric nanomaterials such as hydrogels and nanogels have been used for the safe encapsulation of different types of vaccines aimed at the fine-tuned modulation of the immune system against a variety of viral infections. A thorough discussion on this aspect of polymeric drug delivery and its potential extension to COVID-related drugs and vaccines using interpenetrating polymer networks is another primary goal and a novelty of this comprehensive review article discussed in the “IPNs for vaccine delivery” section.
A previously published document by Raina et al. [3] is an excellent review regarding the efficacy of interpenetrating polymer networks as an enterprising drug delivery system. This perspective, although not exhaustive, is developed by going through an extensive literature survey on the subject matter and thus can be consulted as a comprehensive source of information on the drug delivery applications of IPNs. Most of the existing literature has discussed mainly the synthesis, preparation, and applications of drug delivery system of IPNs. This study has discussed synthesis, preparation, applications, disadvantages, limitations, and large-scale production of IPNs comprehensively. The main novelty of this article is that it has focused on a COVID-related vaccine delivery system mainly which is lacking in existing literature.
A brief historical perspective
The preparation of the first interpenetrating polymer network dates back to 1914 [13]. Aylsworth developed this network polymer based on phenol and formaldehyde, vulcanizing it with rubber and sulfur to provide the polymer with extra mechanical strength and thickness [13, 14]. Another such high-impact polymer network was then synthesized in 1974 by Amos. It was developed by graft copolymerization of styrene-butadiene rubber with polystyrene. The research in this domain further led to an ion exchange resin synthesis, the name interpenetrating polymer network. IPNs were employed for the first time by John Miller around 1960 when he prepared a homo IPN via a styrene suspension polymerization reaction [3, 15]. Afterward, there was no looking back. A series of IPNs were developed by polymer chemists worldwide, ranging from simple IPN systems to innovative and stimuli intelligent IPNs nanomaterials that are being utilized for many applications today.
IPN synthesis and characterization
The structural fabric of an interpenetrating polymer network consists of two or more polymers firmly entwined to form a three-dimensional network wherein one polymer must be either crosslinked or be synthesized in the intimate presence of the other. Its synthesis is quite different from any other polymer blend, graft, and/or block. It is sometimes known as a polymeric alloy. In IPN synthesis, polymers are blended to create a polymer mixture with reduced phase separation ability. Interpenetration then improves the compatibility of the polymers present in this blend as crosslinking creates a permanent entanglement between the polymer subunits. IPNs can be synthesized the same way as other types of polymers: from synthetic, semi-synthetic, natural, or a mix of synthetic and natural monomers. The most commonly employed polymers in an interpenetrating polymer network are as shown in Fig. 1. IPN synthesis generally includes three main steps: Three chemical immiscible phase formations
Coating agent deposition
Rigidization of coating agent
Fig. 1 Popularly employed natural and synthetic polymers in preparing interpenetrating polymer networks
Based on these three steps, an interpenetrating polymer network can be fabricated using any of the following synthetic techniques. (I) Casting evaporation: This method is based on the heating of the polymers followed by casting to produce an intercalated polymeric structure [16]. It is a widely utilized technique for creating cross-linked polymer networks. This procedure incorporates heating the whole polymer until it is completely dissolved, then mixing it with the cross-linker solution. The cross-linker solution receives polymer-I addition as part of the sequential procedure. In both instances, the solution was heated, mixed, cast, and dried. The creation of IPNs gels is accomplished via the casting evaporation process [17].
(II) Emulsification crosslinking: The emulsion cross-linking method, which is mostly a single emulsion cross-linking method, is based on w/o emulsion. However, w/w emulsion method is also used to make IPN today [18]. It is a phase separation technique; emulsions such as water-in-oil (w/o) emulsion or an oil-in-water (o/w) emulsion are crosslinked to produce an interpenetrating polymer network. The w/o emulsion can be prepared by dispersing water-soluble polymers in aqueous media, followed by the addition of the oil phase. The o/w emulsion, on the other hand, is prepared by the reverse of the previous method. Both emulsification methods follow a final crosslinking step to yield an IPN system as the final product.
(III) Mini-emulsion/inverse mini emulsion: In this method, an o/w emulsion is prepared by dispersing hydrophobic monomeric droplets in an aqueous medium. The hydrophobic monomeric droplets are initially prepared by applying shear stress and sonicating the polymers with the required initiators [19]. Crosslinkers are then added to crosslink the polymer and produce the IPNs. The inverse mini emulsion method is the exact opposite of the mini emulsion, where hydrophilic monomeric droplets are dispersed in a hydrophobic phase, i.e., forming a w/o emulsion followed by crosslinking [3].
(IV) Coacervation phase separation: The method involves a partial desolvation of the polymer by adding solvents and/or thermodynamic changes. Two or more oppositely charged polymers can then react to form a complex IPN system.
(V) Radiation polymerization: This method is based on polymer processing by irradiating with gamma rays. One advantage of the method is that it usually does not require any chemical initiators and/or crosslinking agents; it is a single-step process all the way from preparation to sterilization [20].
Other techniques such as those incorporating centrifugal force, shear stress, enzymatic-induced crosslinking, ionic-induced crosslinking [21], pan coating, spray drying, and air suspension coating have also been reported as effective strategies for the IPN synthesis. Sterilization is a post-polymer synthesis requirement whenever an interpenetrating polymer network is prepared for biomedical applications and the transport of sensitive agents such as pharmaceutical drugs. A series of sterilization techniques can be employed for the said purpose, including chemical treatment such as treatment with ethylene oxide, decontamination by irradiation, autoclaving at high temperature (121 ∘C), and using moisture or dry heat, etc. [3]. Irradiation with high-energy gamma (gamma) and beta (beta) rays is recommended for polymer networks because it increases the mechanical stability of the polymers [10].
Characterization is another crucial parameter after the synthesis and preparation of IPNs like all other polymers. The word characterization is a cumulative term encompassing morphological, spectroscopic, and thermal characterization. The morphology of an interpenetrating polymer network can be studied using microscopic techniques such as scanning electron microscopy (SEM) [22] and transmission electron microscopy (TEM) [23]. Factors such as shape, intensity, the extent of crosslinking, mixing, and phase distribution are essential in investigating an IPN architecture. The thermodynamic stability of the polymeric network is evaluated via thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). Infra-red spectroscopy (FTIR) and NMR provide information about the microstructure and miscibility of IPNs [24, 25]. The mechanical properties of the IPNs like hardness, tensile strength, and elongation at breaking are tested by a compression test, esthesiometry and tearing tests, etc. Differential mechanical analysis (DMA) can also be employed by applying a strain or stress onto a polymer sample at controlled frequencies and recording the phase angle and deformation data response. Having detailed information on the structural characteristics of the IPNs is essential to understanding the conditions necessary for its formation and inspecting the effect of any possible chemical variations on the interpenetrated polymers [3].
IPN structure-property relationship
The chemical compatibility between constituent polymers in an interpenetrating polymer network significantly affects its morphology, which can have a pronounced effect on IPN properties. Other factors such as the synthesis/polymerization method, crosslinking density, and interfacial tension between component units can also influence the properties of a network polymer. IPNs are considered superior to simple polymer mixtures owing to their valuable properties, such as a higher mechanical and thermodynamic stability and no toxicity. Also, biodegradability in natural IPNs is specifically beneficial for applications in the biomedical field [26]. Thermodynamic incompatibility obtained from small entropy gain during the mixing of simple polymer mixtures results in their multiphase morphology, thus poor stability [27]. In the case of the union of a natural polymer with a synthetic polymer in the IPNs, eminently enlarged phase stability is expected in the end product. The biological acceptability and the mechanical strength of different polymers intercalated into a single networked system take its chemistry and applications to an improved level.
IPN classification
Interpenetrating polymer networks can be classified based on the types of chemical bonding and/or the method of synthesis.
IPN classification based on synthesis method
Interpenetrating polymer networks can be classified into the following eight main types based on the synthesis method. (I) Sequential IPNs: Sequential IPNs are based on the synthesis of a polymer network A by the reaction of the monomer with the initiator and the crosslinking agent. This polymer network A is then allowed to swell in the presence of a monomer B containing the initiator and the crosslinking agent to form a polymer network B. Crosslinking density governs the structural morphology of a sequential IPN.
(II) Simultaneous IPNs: In simultaneous IPNs, both polymer networks A and B are synthesized simultaneously by independent and/or non-interfering chemical ways. The monomers, catalysts, and crosslinkers for both A and B are mixed simultaneously to execute the individual polymerization reactions. Simultaneous IPN offers the highest entanglement with minimal phase separation in the final polymer network [28].
(III) Full IPNs: A full interpenetrating polymer network synthesis method is based on an individual crosslinking of IPN components without any induced crosslinking between individual polymers A and B.
(IV) Semi-IPN: One polymer (A or B) is present as a crosslinked network in a semi-interpenetrating polymer network. In contrast, the other polymer (B or A) either exists in a linear or a branched form.
(V) Gradient IPNs: A gradient IPN synthesis is based on the partial swelling of polymer network A by a monomer B followed by a quick polymerization reaction until a diffusional equilibrium develops. This method provides enhanced structural solidarity to the interpenetrating polymer network.
(VI) Graft/joined IPNs: This type of IPN synthesis involves the grafting of polymer A on polymer B without any homopolymer crosslinking. This slight grafting of polymers allows improved interfacial bonding in the polymeric network.
(VII) Thermoplastic IPNs: Thermoplastic interpenetrating polymers combine IPNs with polymer blends in the presence of physical crosslinking agents such as ionomers. These physical crosslinkers join polymeric chains such that at low temperatures, these IPNs behave as thermosets (maintaining their shape and size). At the same time, they are free to flow, i.e., to break their physical crosslinks and behave as thermoplastics.
(VIII) Functional silicone fluids with thermoplastic resin IPNs: This IPN type is a subclass of thermoplastic IPNs prepared by blending molten thermoplastic resins with functional silicone fluids.
Other IPNs related materials include latex interpenetrating elastomeric networks: linear polymers coagulated with subsequent crosslinking by different reactions; conterminously grafted copolymers: grafted polymeric network formed by attaching two ends of polymer B with individual polymer A molecules; non-bonded crosslinked IPNs: a polymerized large ringed structure consisting of a double bond; and latex IPNs: a network structure consisting of a micro-IPNs enclosed in each particle.
IPN classification based on chemical bonding
Interpenetrating polymer networks can be classified based on chemical bonding into two main types [16]. I Non-covalent IPNs: This type is based on incorporating inorganic material into the organic network of IPNs. Two distinct phases in a non-covalent IPNs lead to restricted molecular interactions, thus lowering the stability of the networked structure. However, this phase interaction can be improved by hydrogen bond formation between suitable entities such as organic polyacrylates and silicon alkoxides.
II Covalent IPNs: Covalent IPNs are formed by incorporating silicon alkoxides along the polymeric backbone through free radical polymerization. It allows better interfacial tension and molecular interaction between constituent polymeric units.
IPN microspheres
Microspheres are microscopic particles varying from 1 to 1000 μm [15]. It is reported that IPN microspheres can be used to deliver proteins and nucleic acids to the targeted biological sites. When these microspheres are used as biomaterials, the fact that they are round and porous makes it easier to load them with drugs. Several research groups have reported preparing and applying interpenetrating polymer network microspheres and microbeads for targeted oral drug delivery [29]. A stable IPN microsphere was prepared by emulsion crosslinking of acrylamide grafted locust bean gum (a high molecular weight branched polysaccharide composed of galactose and mannose) polyvinyl alcohol using glutaraldehyde as the crosslinking agent. Buflomedil hydrochloride was encapsulated into the IPN microspheres. A high drug entrapment efficiency (51–73%) was reported [30]. A controlled drug release was witnessed for a prolonged time without collapsing the polymeric matrix. The authors reported crosslinker amount, polymeric blend ratio, and the amount of loaded drug as essential factors in controlling the drug release profile. These IPN microspheres were regarded as promising drug-carrying vehicles, especially for highly water soluble, short half-life drugs. A microporous interpenetrating polymer network based on carrageenan and guar gum was also reported for drug delivery applications [31]. Carrageenan is a high molecular weight polysaccharide with a helical structure of D-galactose and 3,6-anhydrogalactose subunits. The sulfate groups extended outwards from a spiral chain of carrageenan, allowing adequate crosslinking in the polymeric network. Guar gum, a non-ionic polysaccharide composed of D-mannopyranosyl linear chain units having α-D-galactopyranose at the branch points, enabled topical entanglement and helix-helix aggregation in the IPNs. Microwave irradiation was employed for crosslinking purposes. Metronidazole (an antiprotozoal) was encapsulated as the model drug in the IPN microspheres. A 72% drug release profile was recorded for 12 h. Analytical characterization techniques showed that the IPNs, which has a honeycomb-like structure, have high mechanical and thermal stability.
IPN nanomaterials
IPN nanoparticles offer numerous advantages in drug delivery applications, including better in vivo stability due to a lower chance of interaction with the biological surroundings, thus adequate drug protection, easy application and/or removal without invasive procedures, direct administration to the targeted tissue, a predominantly higher surface for drug absorption, and lastly, a highly reproducible method of preparation. Polymeric nanoparticles’ chemical formulation and physical properties can be invariably modified following their applications. This feature adds to the value of IPNs nanomaterials as drug carriers as they can easily overcome biological barriers, such as clearance through kidney glomeruli and prevent the risk of nonspecific accumulation in different body organs [32], otherwise a major concern and cause of cell toxicity in the incorporation of polymers in drug delivery applications.
Polymeric nanomaterials can be shaped as hydrogels, nanogels, nanotubes, nanocages, etc. IPN hydrogels can form aqueous solutions with high colloidal stability, entrapping natural macromolecules [33]. This property offers an additive benefit for controlled drug delivery applications. The natural polymers that have been explored for IPN hydrogel development include protein, peptide chains, and polysaccharides, such as salecan, cellulose, dextrose, chitosan, alginate, and xyloglucan [34, 35]. A gel composition based on interpenetrating polymer networks at the nanoscale offers increased elasticity and enhanced mechanical strength. Different types of polymeric networks entwined in a single system offer a synergistic effect incorporating more than one phase at a time. For instance, a rubbery phase and a glass phase in the IPN hydrogel exhibit properties of both but remain separated even when subjected to shear stress. Diversity in IPNs may also exist by varying the number of crosslinks present in the network [36]. A polymeric solution can be converted into a gel via chemical crosslinking or photopolymerization. The hydrophilic functional moieties of the polymer chains offer the hydrogels the ability to retain water and swell in an aqueous media, facilitating the drug unloading process [37]. Hydrogels can also be synthesized with stimuli-responsive polymers to respond to environmental stimuli, including temperature, pH, or the ionic strength of the media, disintegrating and releasing the drug at the targeted site [38–40].
Nanogels and/or hydrogel nanoparticles are a sub-class of hydrogels which are nanometer in size (ranging from 1 to 1000 nm) and can also be employed as efficient drug delivery systems. Nanogels are swollen networks made up of amphiphilic and/or hydrophilic polyionic polymers that can be natural or synthetic [28]. The preparation of IPNs dendrimers is also a major advancement in IPN nanomaterials. Dendrimers are nanosized hyperbranched, globular structures having a precise molecular weight and multivalent surface functionalities. These are highly ordered structures with an inner core and an outer shell. The internal cavities present in an IPN dendrimer can be used to encapsulate different therapeutic agents, including drugs, proteins, nucleic acids, and antibodies [41]. Literature evidence also infers the possibility of engineering IPN nanomaterials for upgradation and/or downregulation of the immune system, i.e., stimulating an immune response against infectious diseases and allergies while suppressing it in the case of an autoimmune disorder [42], thus, widening the scope of applying interpenetrating polymer networks in clinical applications.
“Smart” polymeric multicomponent systems
Smart polymeric multicomponent systems means that a small change in an outside stimulation of any polymeric solvent can cause a big change in a system’s function, structure, or stability [43]. Because these systems are smart and can respond quickly, they are used to make advanced functional drugs. The application of the smart polymers is increasing continuously with the development of new drugs. As scientists work to create ever-more-potent pharmaceuticals, they are paying more and more attention to the methods of administration now in usage. Traditional methods of medication administration cause an initial increase in blood drug levels, followed by a peak and subsequent decrease. Commercially available controlled release devices have been shown to be effective in either maintaining drug levels within the therapeutic range with a single dosage or targeting medication to a specific location while reducing systemic drug levels [44]. The delivery of insulin to patients with diabetes mellitus, anti-arrhythmics to patients with heart rhythm disorders, gastric acid inhibitors for ulcer control, and nitrates to patients with angina pectoris are just a few clinical scenarios where the approach of a constant drug delivery rate is insufficient. Another clinical scenario where a responsive delivery system of anti-inflammatory medications might be helpful is the beginning of infection on medical device or biomaterial surfaces. Furthermore, research in the field of chronopharmacology suggests that the timing of the start of several illnesses is strongly influenced by the circadian rhythm. As a result, time-dependent drug delivery is necessary for medications like blockers, birth control, general hormone replacement, vaccination, and cancer treatment. The abovementioned treatment problems have been optimized by developing drugs with smart and responsive materials [45]. A stimulus-sensitive “smart” IPN hydrogel was prepared by radical polymerization using a temperature-sensitive poly(N-isopropylacrylamide) and a pH-sensitive hyaluronic acid. The hydrogel loaded with luteolin was tested for transdermal drug delivery against hyperproliferation of keratinocytes in the epidermal layer of the skin [22]. Luteolin is a potent therapeutic agent, often used to alleviate skin conditions such as psoriasis. A drug incorporation efficiency of 42.8% was reported using the particular hydrogel, and the highest drug release was observed at 25 ∘C, at pH 5.5. Another novel UV-triggered IPN drug delivery system was reported [46]. The interpenetrating polymer network was prepared using silicone as the host polymer and a spiropyran functionalized guest polymer. The authors reported that the hydrophobic spiropyran moieties transformed to hydrophilic merocyanine upon photoirradiation, releasing the drug without degrading the drug delivery system. A stimulus-responsive drug release of dopamine, L-dopa, and prednisone was reported to be up to 90–95% with the IPNs prepared in this study.
IPNs in drug delivery applications
Drug delivery refers to a method or a process of administering a pharmaceutical compound to achieve specific therapeutic benefits in animals and humans [47]. A successful drug delivery operation can be divided into two distinct steps: spatial placement and temporal delivery. Spatial placement refers to targeting the drug to specific organs, tissues, and/or specific sub-cellular compartments. In contrast, on the other hand, temporal delivery refers to controlling the rate at which the drug is released to these targeted sites [48].
Various obstacles can hinder a successful drug delivery operation, including low drug solubility in physiological fluids, environmental or enzymatic degradation of the drug and non-specific toxicity, and inability to cross biological membrane barriers, resulting in low bioavailability overall. Polymeric structures such as interpenetrating polymer networks can be promising drug delivery carriers. Their three-dimensional crosslinked morphology supports compact drug encapsulation, safe intracellular transport, and sustained release at the targeted site [49]. Interpenetrating polymer networks can be used as effective delivery vehicles to transport active constituents of a drug in the form of a tablet, capsule, microsphere, hydrogel film, sponges, and/or transdermal patches. Transdermal IPN drug delivery facilitates the delivery of drugs in lower doses for a more extended time, avoiding the need to first pass through any metabolic route. This section explores the specific application of IPNs in delivering different types of drugs.
Researchers highlight one possible limitation of using the IPNs for drug delivery over its numerous advantages. Strong interpenetration of different polymers as present in the IPNs can make it difficult for drugs to escape [3]. This difficulty can be handled by manipulating the design and chemistry of the interpenetrating polymer networks at the synthesis and drug encapsulation stages. Stimuli-responsive nanosized IPNs, as discussed earlier, are at a lower risk of facing this issue because of increased sensitivity to external stimuli and their increased surface area. The drug loading/encapsulation efficiency of an IPN can be determined by using the relative formula given in Eq. (1). The actual drug content can be measured via UV-Vis spectrophotometry [29]. The drug release profile of an IPN system is also noted as a percent drug release [50].1 %Encapsulationefficiency=ActualdrugcontentTheoreticaldrugcontent×100%
Delivery of anticancer drugs
Researchers have reported successful interpenetrating polymer networks for encapsulation and delivery of chemotherapeutic drugs for many different types of cancers, including metastatic breast cancer, colorectal cancer, and gastric cancer. Agnihotri et al. [51] studied and reported a controlled, in vitro gastrointestinal release of a drug encapsulated in a semi-IPN matrix. The polymeric network was synthesized by a free radical polymerization reaction of chitosan with polyethylene oxide and polyacrylamide. Ceric ammonium nitrate was used as the initiator in this case. Similarly, Rokhade et al. [52] synthesized a chitosan-based IPN hydrogel microsphere using the emulsification crosslinking method. Glutaraldehyde was employed as a crosslinker, and the IPNs obtained were loaded with 5-fluorouracil, an antimetabolite helpful drug for cancer treatment. A sustained drug release was witnessed over 24 h of encapsulation in the IPN microsphere.
Another stimulus-responsive semi-IPN hydrogel was prepared using a thermo sensitive N-isopropylacrylamide and a pH-sensitive N-ethylmaleamic acid (NEMA) [2]. Free radical polymerization was carried out in the presence of N, N’-methylenebisacrylamide as the crosslinking agent. Doxorubicin hydrochloride was encapsulated as the model drug, and the in vitro drug profile was tested in a stimulating tumor environment (phosphate buffer pH 6.5 and 37 ∘C temperature, comparable to human physiological conditions). The authors deduced that incorporating pH-sensitive monomers into a thermosensitive polymer network usually weakens the latter’s thermosensitivity in copolymerized hydrogels as the incoming monomers interrupt the continuous sequence of the pendant groups in the parent polymer chain.
NEMA was proposed as an alternative choice, claiming that N-ethylmaleamic acid consists of both an amide (-CONH) functional group and a weakly acidic carboxylic (-COOH) group; thus, it introduced pH sensitivity without disturbing the continuity of amide hanging groups in the N-isopropylacrylamide hydrogel network. A rapid drug release profile was recorded without leakage on its way to their targeted site, establishing the research protocol successful. A recent study [53] prepared an interpenetrating polymer network hydrogel based on polyethylene glycol-diacrylate and modified gelatin. The IPN hydrogel was loaded with gold (III) porphyrin (AuP), a lipophilic metal compound considered useful for anticancer applications. The authors claimed a sustained AuP release from the IPN hydrogel following first-order kinetics in an in vitro experimental protocol performed on mice lung cancer cells. AuP visibly inhibited tumor growth by up to 60% and reduced angiogenesis. No cytotoxic signs were observed in any of the treatment groups, thus rendering the polymeric hydrogel biocompatible.
Similarly, a semi-IPN nanoparticle hydrogel was prepared and tested for drug delivery of a widely popular chemotherapeutic drug, paclitaxel (PTX) [54]. The semi-interpenetrating polymer network was developed by irradiation polymerization using poly (ethylene glycol) diacrylate and hyaluronic acid. The researchers reported a sustained drug release for up to 13 days in mice used as model animals. The tumor growth was effectively inhibited without causing any harmful effects. The IPN hydrogel thus exhibited good biocompatibility and biodegradability at the same time.
The IPN nanogel composed of natural gelatin and poly (acrylamidoglycolic acid) was synthesized by simple free radical emulsion polymerization [55]. The nanogel was tested for anticancer drug delivery by loading a hydrophobic curcumin drug. A 42–48% encapsulation efficiency was exhibited by the nanogel, facilitating targeted drug release in a phosphate buffer solution. Curcumin encapsulated nanogel also exhibited pH-sensitive properties and released the drug to the targeted colorectal cancer cell line in a sustained manner. Full IPN and semi-IPN systems have also been explored as sustained drug delivery systems for drugs that treat non-melanoma-type skin cancer [56]. A better dose control minimized side effects, and high cell specificity by using IPNs for drug delivery is reported primarily to treat skin-related ailments.
Delivery of antibiotics
Antibiotic drugs can be entrapped for delivery in interpenetrating microgels. IPN microgels are intermolecularly crosslinked polymeric gel particles homogeneously distributed in a solvent system. Complete dissolution of an antibiotic drug can be retarded up to and above 10 h when caged in an IPN microgel matrix. Based on their colloidal size, IPN microgel particles can also exhibit conformational changes in response to variations in environmental parameters [57, 58]. A pH-sensitive interpenetrating microgel system was developed based on chitosan, acrylamide grafted polyvinyl alcohol (PVA), and a hydrolyzed version of this acrylamide grafted PVA [58]. The microgel was tested for the controlled release of cefadroxil, an antibiotic drug commonly used against bacterial infections of the skin and throat. Analytical characterization techniques ensured a uniform drug distribution inside the polymeric matrix. Hydrophilic interactions between chitosan and polyacrylamide modified PVA assisted the formation of a uniform microgel blend. The incorporation of PVA lends elastic properties to the microgel beads. The IPN matrix exhibited promising characteristics for controlled drug release, a prolonged drug release witnessed for more than 10 h.
In another study, interpenetrating hydrogel microspheres prepared by a water-in-oil (w/o) emulsion crosslinking method for sustained release of ciprofloxacin hydrochloride, a quinoline antibiotic that is used to treat bacterial fevers such as pneumonia, typhoid, and diarrhea, were reported [59]. A xantham gum-based superabsorbent polymer and polyvinyl alcohol were used to prepare these IPN microspheres. Hydrophilic interaction again played an integral role in the hydrogel preparation. The in vitro drug release profile was tested in both an acidic (pH 1.2) and an alkaline (pH 7.4) medium. The researchers reported an inverse relationship between the polymeric network density and drug release rate. The denser the networked structure by adding a higher amount of crosslinker, the more sustained the drug release.
The drug release profile of a penicillin-based antibiotic, amoxicillin, was also tested in an earlier study using a semi-IPN [60]. The semi-interpenetrating polymer network was prepared by crosslinking chitosan with polyvinylpyrrolidone (PVP). Both porous freeze-dried hydrogels (pore diameter: 39.20±2.66μm) and non-porous air-dried hydrogels were prepared, exhibiting a pH-dependent swelling behavior. However, the porous semi-IPN hydrogels displayed a superior swelling, and drug release potential than the non-porous ones. Seventy-three percent amoxicillin was released in 3 h at pH 1 from freeze-dried polymeric networks and 33% drug release reported from air-dried polymeric hydrogels.
Delivery of antihypertensive drugs
Different types of interpenetrating polymer networks have also been incorporated into the safe delivery of hypertensive drugs. IPN microspheres based on gellan gum and polyvinyl alcohol were used to encapsulate and deliver an antihypertensive drug called carvedilol [61]. Carvedilol is a non-cardio selective β-blocker that is frequently used in conditions such as angina pectoris. The microspheres were prepared by an emulsion crosslinking method. The researchers reported a strong effect of the size and morphology of IPN microspheres on drug encapsulation efficiency and its liberation. PVA is an easy to process pH and thermostable polymer, while gellan gum is a natural polysaccharide. The incorporation of both polymers in an interpenetrating polymer network provides the IPN structure with high mechanical strength as well as biocompatible and non-toxic properties. A crystalline dispersion of the drug was witnessed in the polymeric matrix with a drug encapsulation efficiency of up to 87%. A prolonged drug release profile was recorded up to 12 h in vitro in the simulated gastric fluid and the intestinal fluid.
In another study [62], the IPN hydrogel composed of sodium alginate and polyacrylic acid was reported for the transdermal delivery of prazosin hydrochloride, an antihypertensive drug specifically used to treat abrupt blood pressure fluctuations. The hydrogel exhibited a reversible swelling behavior, sensing physiological and environmental pH changes leading to an oscillatory drug release pattern. Semi-IPN microspheres prepared by w/o emulsification crosslinking of chitosan and N, N’-dimethylacrylamide were used to deliver chlorothiazide, a diuretic drug formulation [62]. Uniform molecular dispersion of the drug in the polymeric matrix followed by a sustained release for up to 12 h was reported rendering the IPN drug delivery protocol successful.
In an earlier study [63], a unique IPN hydrogel synthesis was reported and tested for antihypertensive drug delivery applications. The IPN hydrogel was prepared by networking polyvinyl alcohol with polyacrylic acid but without the addition of any crosslinker. The fabricated hydrogel’s swelling behavior and drug delivery were tested in vitro in a stimulated intestinal fluid by encapsulating diltiazem hydrochloride. Seventy-nine percent drug entrapment efficiency and a controlled drug release up to 24 h were reported in rats as model animals. Reduction in blood pressure recorded up to 40.1%. The clinical study claimed a superior drug carrying potential of IPN hydrogels compared to standard hydrogels. Thacharodi et al. [64] explored the potential of a collagen-chitosan-based IPN hydrogel patch for delivering propranolol hydrochloride transdermally. Five milligrams of drug was released from the patch in 24 h, establishing the possibility of utilizing an interpenetrating polymer network as transdermal patches for drug delivery.
Delivery of anti-inflammatory drugs
Anti-inflammatory drugs hold immense therapeutic significance in relieving pain, bringing down the high temperature by reducing other short-term and long-term inflammatory symptoms, including sprains, strains, and arthritis. These drugs can be packaged in interpenetrating polymer networks for effective delivery at specific pain points. A targeted drug delivery of ketoprofen, a gastroprotective anti-inflammatory drug, was reported in IPN microbeads [65]. The polymeric microbeads were prepared by graft copolymerization of polyacrylamide (PAAm) and gum ghatti (GG) in the presence of sodium alginate. The IPN microbeads demonstrated excellent pH sensitivity, a significant drug release profile reported under gastrointestinal/alkaline (pH 7.4) conditions compared to the amount of drug release witnessed under acidic (pH 1.2) conditions in the stomach in in vivo clinical trials performed using Wistar rats as model animals. An alkaline pH leads to the alkaline hydrolysis of -CONH2- group to -COO- in PAAm-grafted-GG, releasing the entrapped drug. This example demonstrates the significant impact of an IPN structural morphology on its properties.
IPN microstructures based on gelatin and carboxymethyl tamarind gum (CTG) were explored to deliver aceclofenac, a non-steroidal, anti-inflammatory drug often used to treat arthritic conditions [27]. A high drug encapsulation efficiency of IPN microbeads was recorded as high as 96%. A slow drug release rate (less than 10% in 2 h) was reported in an HCl solution (pH 1.2), while an extended-release (up to 60%) was observed in phosphate buffer (pH 6.8). The researchers observed a prominent influence of the functional groups present in the IPNs on polymeric chain swelling and relaxation, controlling its drug release profile. Negatively charged carboxyl functional groups present on gelatin undergo electrostatic forces of repulsion at high pH levels, causing extended swelling of the networked structure, thus a faster drug release. Physical crosslinking/blending of gelatin with a natural polymer like CTG enhanced its mechanical strength and formed a solid IPN-drug biocomposite. The anti-inflammatory activity of the released drug lasted for 7 h in albino rats, marking the success of the in vivo clinical study.
Delivery of antiviral drugs
Temperature and pH-sensitive drug delivery systems are specifically beneficial for the site-specific transport of antiviral drugs [66]. Polymer chemists reported the preparation of dual responsive interpenetrating polymer network hydrogel microbeads [67] by functionally modifying guar gum by co-polymerization using N-vinylcaprolactum and sodium alginate. These IPN hydrogel microbeads were then used for a colon-specific drug delivery. The hydrogel was used to encapsulate an anti-HIV drug called zidovudine in it. Poly (N-vinyl caprolactam) grafting on guar gum appreciably improved its intrinsic viscosity. A maximum of 68% drug entrapment was reported, and a high drug release rate was recorded at pH 7.4 at 37 ∘C. Encapsulation in IPN microbeads enhanced the drug release time up to 34 h. Similarly, IPN-based anti-HIV drug delivery was reported using semi-IPN microspheres [68] prepared by w/o emulsification crosslinking using PVA and dextran-grafted-acrylamide. Abacavir sulfate was the model drug encapsulated in the IPN microspheres; the in vitro drug release studies reported a pH-dependent drug release from the polymeric network.
Even after considering all these efficient drug delivery sources for antiviral drugs, we cannot neglect the fact that viral infections can be dealt with better with a prophylactic approach, i.e., disease prevention rather than treatment after falling sick and what could be a better prophylactic approach than vaccination and immunization regimens. Interpenetrating polymer networks can serve as potential vaccine carriers if designed intelligently. The proceeding section discusses the promising application of network polymers in vaccine delivery and its associated pros and cons.
Other than those discussed in the “IPNs in drug delivery applications” section, interpenetrating polymer network-based drug delivery systems have also been applied, particularly for transporting anti-malarial [69], anti-diabetic [70], anti-asthmatic [71], and anticonvulsant [72] drugs. In another relevant study [73], researchers reported the targeted delivery of insulin in an interpenetrating polymer network hydrogel. The IPN hydrogel was prepared using polyacrylic acid co-acrylamide and carboxymethyl chitosan. Insulin was entrapped in it and orally delivered to rats used as model animals. The authors observed a pH-sensitive insulin release, an enhanced rate recorded on increasing the pH from 1 to 6.2. The carboxylic acid functional group in polymeric chains is protonated under acidic conditions, leading to the shrinkage of the hydrogel. The formation of hydrogen bonds between insulin and the polymeric chains restricted insulin release under acidic conditions.
On the other hand, an elevated pH led to ionization and swelling of the polymeric chains, thus easing the insulin release. A 60–99% drug release profile was reported within 2 h. This example again highlights the importance of the structural variations present in an IPN on its properties and potential applicability in drug delivery. Kurakula et al. [74] highlighted the future application of povidone-iodine (PVP-I) as an antiviral agent against COVID-19. They left a question mark on whether this drug-loaded polymer can be used as a medical aid in these pandemic-affected times. This initiative can be supported further by interpenetrating PVP, a synthetic polymer with natural polymers, to prepare biocompatible, least toxic IPNs as a defensive action against the deadly virus.
IPNs for vaccine delivery
Vaccination is a cost-effective mechanism to prevent and fight against diseases. It could either be prophylactic or therapeutic. Vaccination against viral infections involves a preventative strategy, while vaccination against cancer falls under the therapeutic category. A fundamental principle of prophylactic subunit vaccination is the induction of an antigen-specific immune response. Antigens (weakened particles mimicking intracellular pathogens) are enclosed in vaccines as the active pharmaceutical ingredients. When these antigens are released into the body cells through vaccines, antigen-presenting cells (APCs) recognize, carry, and present these antigens to the T-cells; the white blood cells are integral for immunogenicity. APCs release co-stimulatory factors called cytokines and activate the T-cells. T-cell differentiation occurs to induce antigen-specific cytotoxicity. The exact immunization mechanism is followed in case of a pathogenic attack where the T-cells are activated in the presence of APCs triggered by ligands characteristic of the invading pathogens. Immunity acquired through vaccines can help fight the body against any actual pathogenic attack.
Polymeric entities can be employed as vehicles to ensure the effective transport of these vaccines to targeted sites. Incorporating interpenetrating polymer networks in vaccine delivery is a relatively new idea that can improve vaccine performance and potency. Unlike conventional vaccine administration, sustained drug release from IPNs can ensure solid and long-lasting antibody responses. This may alleviate the need for multiple and frequent vaccine doses and boosters instead of a conventional vaccination regimen. Polymeric nanogels have proved the most fruitful in this regard so far. Polymeric nanogels can act as “integrated adjuvants,” combining the immunomodulatory properties with targeted antigen delivery functions [75, 76]. An adjuvant is a term used in immunological studies to define any substance that can accelerate, prolong, or enhance an antigen-specific immune response [77]. The biodegradable, biocompatible, and mucoadhesive properties of IPNs nanogels can make them promising adjuvants to enhance the efficiency of subunit vaccines [77].
APCs prefer particulate antigen uptake over soluble antigen uptake. Antigen-loaded IPN nanoparticles can act as antigen depots, slowly releasing antigen for prolonged bioavailability and better immune response quality. In vitro clinical trials have revealed that as vaccine delivery systems, polymeric nanoparticles work by activating dendritic cells followed by co-stimulation of cytokines. Activated dendritic cells migrate to the regional lymph nodes, presenting antigens to the T-cells, turning on a humoral immune response in the body [78]. The intrinsic properties of the polymeric nanogels, including their chemical composition, shape, and size; any charge present on their surface; and/or their interaction with water and other solvents (hydrophilicity and hydrophobicity), however, play determining roles in their incorporation in vaccine delivery operations and in shaping the induced immune response as a result. Polymer research studies claim that the encapsulation of a vaccine in polymeric networks prevents the in vivo degradation of its constituent antigens. Enclosure in polymeric systems can also increase a vaccine’s active targeting specificity [32].
Nonetheless, very few examples of vaccine delivery applications and efficacy of interpenetrating polymer networks are available in the scientific literature; thus, this end is always open to research. This particular review article is a little effort to open a new windowpane in this area by relating the research studies conducted on vaccine delivery using different polymer systems in general and how those can be extended to incorporating IPNs in vaccine delivery operations.
Protein-based vaccines
The weakened and/or inactive foreign particles that make up the chemical composition of most vaccines available against viral infections are called antigens. These antigens are essentially proteins; thus, all the antigen vaccines fall under the protein-based vaccine category [79]. Continuous antigen delivery for long periods is considered effective in inducing immunity. Encapsulation of vaccine components in three-dimensional polymer networks like IPNs can fulfill the said task. Prolonged antigen exposure allows enough time for affinity maturation and antibody isotype switching to generate a robust immune memory [80]. Polymeric nanogels are specifically helpful for delivering protein-based vaccines. They offer an active surface for protein absorption; antigens from IPN nanogels can form surface conjugated complexes with antibody molecules, thus making them faster and more effective immunomodulatory mechanisms expected.
Factors such as molecular weight and chemical composition, including the types of co-polymers, crosslinkers, and crosslinking density, control vaccines’ targeting release mechanism. For instance, a polymer network with a higher molecular weight results in slower drug release. Similarly, the functional groups present at the surface of a polymer network say a nanogel can be modified with targeting moieties to achieve site-specific vaccine delivery. Stimuli-responsive IPN nanogels such as those sensitive to acid hydrolysis (acetal or hydrazide bonds) or those containing functional groups susceptible to reduction can serve as vaccine carriers even better. The timely release of antigens in response to slight variations in pH and temperature also increases the residence time of the antigen at the specific site, eliminating the need for multiple vaccinations [81]. These functional group moieties can additionally aid in strengthening the immune system by activating a complementary system, i.e., triggering a series of proteins and enzymes capable of promoting inflammatory and phagocytotic reactions [76]. Sustained antigen presentation to the T cells from a few hours to a few days is worthwhile. It improves T cell expansion and differentiation leading to an enhanced immune response. On the other hand, excessive antigen presentation for weeks and/or months may lead to T cell death. Instead, thus, polymer design and drug encapsulation strategy are two critical preliminary considerations when developing polymeric networks for vaccine delivery.
Two different techniques can reportedly encapsulate subunit vaccines in polymers: physical entrapment and chemical conjugation. Physical entrapment of antigens involves loading the components into the core and/or onto the surface of the polymer via polymerization, emulsification, self-assembly, or complexation method. In order to achieve a sustained protein release from an IPN matrix, the diffusion of the protein through the pores of the polymeric network is suggested while keeping the matrix intact [82]. Natural polymers such as alginate and chitosan are recommended for designing porous polymer networks. A supercritical enhanced atomization spray drying technique was used to develop a polymeric vaccine [83]. Aqueous solutions containing inactive infectious particles and chitosan were prepared and converted into nanospheres by employing intense shear stresses generated by depressurization of liquid-gas (supercritical CO2-N2) mixtures to break up the liquid phase into sub-micron droplets. The immunization efficiency of the developed vaccine was tested against Strangles infection using a mice model. Chemical conjugation for preparing vaccine-loaded polymeric systems involves covalent binding of the antigen/protein into the polymeric particles using conjugation techniques such as ligation and metal-catalyzed cycloaddition click reaction.
Other than chitosan, polyacrylates such as poly (methyl methacrylate) and poly (tert-butyl acrylate), hyaluronic acid, alginate [84], and cyclodextrin [85] have also been explored for designing polymeric adjuvants for vaccine delivery. Poly(lactide-co-glycolide)-based micro- and nanoparticles have also been explored for vaccine delivery applications [86].
DNA vaccines
Vaccines can also be developed using hereditary materials such as nucleic acids: DNA and RNA. Compared to conventional vaccines, DNA vaccines hold a higher disease-preventing potential as they can induce both an antibody and a cell-mediated immune response [87]. However, choosing a suitable vector for carrying DNA and RNA-based vaccines is more challenging than designing a delivery system for protein-based vaccines. A perfect drug delivery vector for vaccines based on nucleic acids must fulfill four conditions: (i) interact effectively with DNA, (ii) be stable in body fluids, (iii) target specific cells, and (iv) be able to cross membranes in order to release the drug. Network polymers satisfy all the necessary conditions. DNA can be encapsulated in polymeric networks in either condensed or non-condensed form [88]. Cationic polymers are usually employed as DNA condensing agents through electrostatic forces of attraction. The integrity of the DNA-polymer depends on many factors, including complex size, stability, toxicity, and protection against DNase degradation during intracellular trafficking. Adsorption of nucleic acid onto cationic nanoparticles can facilitate the immediate release of DNA at the targeted site. Researchers have reported the application of chitosan-based nanoparticles encapsulating a cocktail of DNA encoding nine immunogenic antigens of a respiratory virus using a mice model. High levels of cytotoxic T cells with antiviral action were observed post vaccine administration in the pulmonary cells of the animal [89]. Oral delivery of another chitosan-DNA-based vaccine encoding a mite dust allergen with improved proinflammatory immune response was reported [90].
RNA vaccines
Messenger RNA (mRNA) can also be incorporated as a principal component in formulating effective vaccines. mRNA-based vaccines offer a higher safety profile and more precise and reproducible immune response, and most importantly, they are cost-effective and offer an added incentive of faster production [91]. The first report of injecting an mRNA into animals for vaccination purposes was received in the 1990s, encoding an influenza virus nucleoprotein [92]. mRNA vaccines are packaged with single-stranded genetic sequences that instruct the host cells to produce proteins, yielding an immune response against suspected pathogens. Despite their numerous advantages over conventional antigen and DNA-based vaccines, RNA-based vaccines have a few shortcomings.
mRNA molecules have a large molecular size and are negatively charged, hydrophilic, and fragile, making the passage of a naked mRNA molecule through the lipophilic cell membranes quite complex. Network polymers can facilitate intracellular transport of these mRNA molecules by encapsulation and/or complexation. Cationic polymers can efficiently complex anionic mRNA through electrostatic forces of attraction to form a polyplex or a micelleplex. Polymer-based vectors have been shown to be more stable than traditional lipid-based vectors in RNA-based vaccine delivery. Polymeric networks have the potential to act as cytoplasmic delivery vehicles, crossing various extracellular barriers for carrying the vaccine components to their target sites, i.e., to ribosomes inside the cells. Most vaccines designed for combating the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are also mRNA-based vaccines.
The prospective application of an IPN in COVID vaccines
Coronavirus is a spherically shaped virus with spiked-shaped immunogenic proteins present on its surface. The virus has an RNA genome and a lipid bilayer surrounding it. It attacks the lung cells by disintegrating its outer envelope and inserting this RNA genome into the target cells. mRNA-based vaccines trigger an immune response in the body by mimicking the RNA genome of the virus itself. The characteristics studied for developing polymeric networks like IPNs to deliver RNA vaccines can be applied as an extension to prepare a COVID vaccine.
There will always be a need to consider a series of factors when designing interpenetrating polymer networks for carrying COVID vaccines, including the molecular weight, shape, and size of the polymeric carriers. These factors, in turn, control the rate of glomerular filtration, governing effective removal and clearance of the polymeric carriers from the body by excretion. The size preferable for interaction with APCs and transportation to lymph nodes may not be suitable for efficient renal clearance [93]. IPNs composed of slowly degrading and/or non-biodegradable polymeric subunits may pose a risk of bodily accumulation, resulting in toxicity; designing biodegradable IPNs primarily from natural polymers is thus a more viable option.
The onset of disease conditions such as infections or inflammation causes changes in the pH of the human organs. pH-sensitive polymers can be manipulated to respond to these changes and specifically release the enclosed vaccine at the targeted sites. Ionizable polymers with pKa values ranging from 3 to 10 can be used to design effective IPN systems. Polyacids containing acidic functional groups such as carboxylic acid, polybases based on basic entities such as amines, and natural polymers such as dextran, hyaluronic acid, and chitosan (containing NH2 and OH groups) can be used for this purpose [94]. Usually, an inflammatory action leads to acidosis, i.e., a reduced pH. The primary amines of IPNs-friendly polymers such as chitosan become positively charged by accepting protons. This idea can be further investigated for designing IPN systems for COVID vaccine delivery.
The messenger RNA is a negatively charged molecule. The hydrophilic and polar nature of polymers present in an interpenetrating polymer network can also prove beneficial for developing strong interactions with and enclosing the mRNA molecules of the vaccine. Natural polymers can be derivatized, such as the derivatization of chitosan to carboxymethyl chitosan to introduce an additional -COO group to interact with the vaccine better. The review written by Mallakpour et al. [84] is an exciting lead in using natural polymers as adjuvants in vaccine delivery systems. A comprehensive review on incorporating nanoparticles in COVID-related therapeutics is provided by Sharma et al. [95]. It can be read as an additional reference to take the initiative of designing IPN nanomaterials for vaccine delivery further.
Disadvantages and limitations of Interpenetrating Polymer Networks (IPNs)
There are some disadvantages and limitations to IPNs in clinical adoption. It is a very important factor of a drug delivery system to deliver the active drug within the correct site and speed. Sometimes, it may be very difficult to extract the active drug due to excessive penetration among the polymers [16]. In addition, several factors during production can affect the final polymer’s quality, including the reactor type, reaction mechanism, and operating circumstances [16]. There is another problem associated with IPNs which is the lack of effective interface. The surface energy phenomena and the insufficient chemical engagement between phases may contribute to the problem of ineffective interfaces [36]. The lack of an efficient interface is problematic for the non-covalent system, and it also threatens the covalent system. The mechanical property of IPNs is an important parameter in context of pharmaceutical applications which can be hampered at the time of crosslinking. The proper integrity of the drug delivery system means the polymers must be maintained to work properly in the pharmaceutical applications and throughout their whole lifetime. The therapeutic agent must be protected by the drug delivery system until it is discharged from the system, and the drug delivery system must be able to preserve its integrity until then. Though increasing the degree of cross-linking may improve the mechanical properties of IPNs, high degree of cross-linking can make the brittle structure of IPNs [96].
Another major issue of IPNs is large-scale production of drug delivery systems. The manufacture of IPNs on a wide scale is the biggest obstacle to drug delivery research and development. For ultimate commercialization, laboratory or pilot technology must always be scaled up. Though it is easy to improve the performance of IPNs in laboratory on a small scale, it is a very hard and complex procedure to maintaining the better performance at a large and global scale. It might be difficult to maintain the concentration and composition of polymers on a big scale. Even though IPN drug delivery technologies have received several patents and research grants, commercialization is still in its early stages. This is largely because academic scholars conduct the majority of the research investigations. However, more work has to be done to transition IPN-based drug delivery devices from the experimental stage to pilot scale manufacturing and expand their useful applications. This can be done by taking a number of things into account, such as increasing selectivity without sacrificing biocompatibility and stability, optimizing polymer modification techniques, using the right engineering configurations, understanding how transport works, and using materials and methods that do not cost too much. After analyzing the whole information about IPN, this study has summarized the brief description of IPN in a single graphical image. Figure 2 represents a brief graphical summary of IPN.Fig. 2 A brief graphical summary of IPN
Challenges and future outlook
Successful vaccination of diseases such as polio and smallpox has dramatically improved the health sector worldwide. However, traditional immunization approaches somehow lag in treating ailments such as malaria, tuberculosis, and HIV. The encapsulation of vaccines in three-dimensional polymeric networks, such as interpenetrating polymer networks, may provide an auspicious opportunity to combat these threats while also instilling new hope in improving the efficacy of the COVID vaccination protocol. New IPNs can be designed to encapsulate and deliver coronavirus vaccines for sustained, site-specific release of its components, reducing the need for subsequent booster shots. However, this idea is still at its juvenile stages and requires extensive research and contemplation in the upcoming years. The World Health Organization (WHO) reports frequent mutations of the SARS-CoV-2 virus [87].
These ever-new variants make the synthesis and application of any IPN design for vaccine delivery a very challenging task. Hit and trial experiments with different polymers, crosslinking agents, and synthesis techniques can be employed to propose a single IPN formulation capable of performing the drug delivery task for the mRNA vaccine components against the coronavirus disease.
Further development is also required in the analytical characterization of drug-loaded IPN carriers to study chemistry and possible interactions between a proposed polymeric network and different vaccine components. Modifying the structural morphology to synthesize novel IPNs is relatively more straightforward at the lab scale, but maintaining a uniform chemical composition and concentration simultaneously is still a considerable challenge in the large-scale production of such designs. Commercialization of biocompatible IPNs is at an early stage and requires further development.
Concluding remarks
To summarize, incorporating an interpenetrating polymer network as a drug delivery vehicle is a promising approach for site-specific, controlled drug release. Proper placement of any drug at the targeted site maximizes drug availability and therapeutic efficiency and minimizes any associated side effects. Designing new IPNs with biodegradable and biocompatible polymers can improve drug delivery characteristics. Considering the previous relevant points, effective IPN systems can be developed for vaccine delivery, specifically mRNA vaccine delivery. Multiple sustained doses for extended periods can also alleviate the possible side effects of a vaccine expected from a single, prompt heavy dose. Nevertheless, this area inevitably requires rigorous research and scientific effort to acquire any desirable outcome soon. Table 1 represents examples of IPNs in different drug delivery applications since the last 12 years. Table 1 Literature review of some representative examples of interpenetrating polymer networks in different drug delivery applications since the last decade
Drug delivery application type: anticancer
SL no. Network type Monomers and polymers Initiator Crosslinker Method of preparation Encapsulated drug Ref
01 Dual responsive semi-IPN hydrogel N-iso propylacrylamideethylmaleamic acid and sodium alginate Potassium per sulfate (K2S2O8) N, N’-methylenebis (acrylamide) (BIS) Free radical polymerization Doxorubicin hydrochloride (Dox) [2]
02 IPN microspheres Sodium alginate and locust bean gum - Calcium chloride (CaCl2) Ionotropic gelation Capecitabine (CAP) [90]
03 Semi-IPN hydrogel Salecan and poly (methacrylic acid) Ammonium persulfate N, N‘- methylenebis (acrylamide) (BIS) Free radical polymerization Doxorubicin (Dox) [91]
04 Semi-IPN microspheres Acrylamide, polyvinyl alcohol, and sodium alginate Benzophenone Ferric chloride (FeCl3) Graft copolymerization in the presence of UV irradiation 5-uoro-uracil (5-FU) [92]
Drug delivery application type: antibiotic
05 IPN hydrogel microspheres Xanthan gum, polyacrylic acid, polyvinyl alcohol - Glutaraldehyde Water-in-oil (w/o) emulsion polymerization Ciproflaxin hydrochloride (CIPRO) [52]
06 IPN microgel Chitosan, acrylamide graft polyvinyl alcohol and hydrolyzed acrylamide grafted polyvinyl Ceric ammonium nitrate (CAN) Glutaraldehyde Graft copolymerization Cefadroxil [51]
07 IPN hydrogel Alginate, gelatin and hydroxyapatite - Glutaraldehyde Composite polymerization Ciproflaxin hydrochloride [93]
Drug delivery application type: antihypertensive
08 IPN microspheres Gellan gum and polyvinyl alcohol (PVA) - Glutaraldehyde Water-in-oil (w/o) emulsication Carvedilol [54]
09 Semi-IPN microspheres Chitosan and N, N’ dimethylacrylamide Potassium persulfate (K2S2O8) Glutaraldehyde Water-in-oil (w/o) emulsication Chlorothiazide [55]
10 IPN hydrogel Polyvinyl alcohol (PVA) and polyacrylic acid (PAA) Benzoyl peroxide - Modified emulsion method without crosslinker Diltiazem hydrochloride (DL) [56]
Drug delivery application type: anti-inflammatory
11 IPN microbeads Polyacrylamide (PAAm), gum ghatti, and sodium alginate Ceric ammonium nitrate Ca2+ ions and glutaraldehyde Graft copolymerization in the presence of microwave irradiation followed by dual crosslinking Ketoprofen [58]
12 IPN microstructures Gelatin and carboxymethyl tamarind gum (CTG) - Glutaraldehyde Bio-composite formation by physical crosslinking Aceclofenac [24]
Drug delivery application type: anti-viral
13 IPN hydrogel microbeads Guar gum, poly (N-vinyl caprolactam), and sodium alginate - Glutaraldehyde Graft copolymerization Zidovudine [60]
Drug delivery application type: vaccine carrier
14 Hybrid hydrogel Caprolactone and lactide Polyethylene glycol (PEG) Stannous octoate catalyst Ring opening polymerization p-DNA vaccine [94]
Drug delivery application type: vaccine
15 Nanoparticles Chitosan Sodium acetateacetic acid - Complex coacervation method DNA [83]
Acknowledgements
The author acknowledges the publishers who have approved the use of some of their figures in this article.
Author contributions
The author has participated in the analysis and interpretation of the data, drafting the article or revising it critically for important intellectual content, and approving the final version.
Data availability
The article contains two figures that have been used from other publications after getting the publishers’ approval. The approvals are attached at the end of the submitted manuscript.
Declarations
Ethics approval and consent to participate
As a solo author for this review article, there was no need for other participant approval.
Consent for publication
As a solo author for this review article, there was no need for a consent statement.
Competing interests
The author declares no competing interests.
Highlights
• The three-dimensional network of IPN systems is particularly valuable for drug encapsulation and a site-specific, targeted drug delivery.
• The structure-property relationships of these polymeric systems can be utilized for novel biomedical applications.
• A novel application is the targeted delivery of protein-based and mRNA-based vaccines such as a COVID-19 vaccine proposed as a novel approach in this perspective.
• Network polymers can ensure a sustained vaccine delivery at a specific site for a prolonged time period.
• It can maximize the availability and efficacy of the vaccine while reducing the need for multiple booster shots.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Silverstein MS. Interpenetrating polymer networks: so happy together? Polymer. 2020;207:122929. Available from: https://www.sciencedirect.com/science/article/pii/S0032386120307540.
2. Dadfar SMR Pourmahdian S Tehranchi MM Dadfar SM Novel dual-responsive semi-interpenetrating polymer network hydrogels for controlled release of anticancer drugs J Biomed Mater Res, Part A 2019 107 10 2327 2339 10.1002/jbm.a.36741
3. Raina N Rani R Khan A Nagpal K Gupta M Interpenetrating polymer network as a pioneer drug delivery system: a review Polym Bull 2020 77 9 5027 5050 10.1007/s00289-019-02996-5
4. Pulat M Kahraman AS Tan N Gümüşderelioğlu M Sequential antibiotic and growth factor releasing chitosan-PAAm semi-IPN hydrogel as a novel wound dressing J Biomater Sci Polym Ed 2013 24 7 807 819 10.1080/09205063.2012.718613 23594070
5. Huang Y Zhao X Zhang Z Liang Y Yin Z Chen B Degradable gelatin-based IPN cryogel hemostat for rapidly stopping deep noncompressible hemorrhage and simultaneously improving wound healing Chem Mater 2020 32 15 6595 6610 10.1021/acs.chemmater.0c02030
6. Yang C Xue R Zhang Q Yang S Liu P Chen L Nanoclay cross-linked semi-IPN silk sericin/poly(NIPAm/LMSH) nanocomposite hydrogel: an outstanding antibacterial wound dressing Mater Sci Eng C Mater Biol Appl 2017 81 303 313 10.1016/j.msec.2017.08.008 28887976
7. Crosby CO, Stern B, Kalkunte N, Pedahzur S, Ramesh S, Zoldan J. Interpenetrating polymer network hydrogels as bioactive scaffolds for tissue engineering. Rev Chem Eng. 2022;38(3):347–361. Available from: 10.1515/revce-2020-0039 [cited 2022-05-08].
8. Fares MM, Shirzaei Sani E, Portillo Lara R, Oliveira RB, Khademhosseini A, Annabi N. Interpenetrating network gelatin methacryloyl (GelMA) and pectin-g-PCL hydrogels with tunable properties for tissue engineering. Biomater Sci. 2018;6:2938–2950. Available from: 10.1039/C8BM00474A.
9. Shojarazavi N Mashayekhan S Pazooki H Mohsenifard S Baniasadi H Alginate/cartilage extracellular matrix-based injectable interpenetrating polymer network hydrogel for cartilage tissue engineering J Biomater Appl 2021 36 5 803 817 10.1177/08853282211024020 34121491
10. Avadhanam V Ingavle G Zheng Y Kumar S Liu C Sandeman S Biomimetic bone-like composites as osteo-odonto-keratoprosthesis skirt substitutes J Biomater Appl 2021 35 8 1043 1060 10.1177/0885328220972219 33174770
11. Kaity S Ghosh A Carboxymethylation of locust bean gum: application in interpenetrating polymer network microspheres for controlled drug delivery Ind Eng Chem Res 2013 52 30 10033 10045 10.1021/ie400445h
12. Hajikhani M, Khanghahi MM, Shahrousvand M, Mohammadi-Rovshandeh J, Babaei A, Khademi SMH. Intelligent superabsorbents based on a xanthan gum/poly (acrylic acid) semi-interpenetrating polymer network for application in drug delivery systems. Int J Biol Macromol. 2019;139:509–520. Available from: https://www.sciencedirect.com/science/article/pii/S0141813019336542.
13. Aylsworth J. Best invention ever; US Patent 1111284, 1914.
14. Aylsworth A Jiang SX Desbois A Hou ST Characterization of the role of full-length CRMP3 and its calpain-cleaved product in inhibiting microtubule polymerization and neurite outgrowth Exp Cell Res 2009 315 16 2856 2868 10.1016/j.yexcr.2009.06.014 19559021
15. Jana S Jana S Interpenetrating polymer network: biomedical applications 2020 Singapore Springer
16. Lohani A, Singh G, Bhattacharya SS, Verma A. Interpenetrating polymer networks as innovative drug delivery systems. J Drug Deliv. 2014;2014: 583612.
17. Kosmala JD Henthorn DB Brannon-Peppas L Preparation of interpenetrating networks of gelatin and dextran as degradable biomaterials Biomaterials 2000 21 20 2019 2023 10.1016/S0142-9612(00)00057-0 10966010
18. Bhattacharya SS Shukla S Banerjee S Chowdhury P Chakraborty P Ghosh A Tailored IPN hydrogel bead of sodium carboxymethyl cellulose and sodium carboxymethyl xanthan gum for controlled delivery of diclofenac sodium Polym-Plast Technol Eng 2013 52 8 795 805 10.1080/03602559.2013.763361
19. Koul V Mohamed R Kuckling D Adler HJP Choudhary V Interpenetrating polymer network (IPN) nanogels based on gelatin and poly (acrylic acid) by inverse miniemulsion technique: synthesis and characterization Colloids Surf B 2011 83 2 204 213 10.1016/j.colsurfb.2010.11.007
20. Elbarbary AM, Ghobashy MM. Phosphorylation of chitosan/HEMA interpenetrating polymer network prepared by γ-radiation for metal ions removal from aqueous solutions. Carbohydr Polym. 2017;162:16–27. Available from: https://www.sciencedirect.com/science/article/pii/S0144861717300139.
21. Du M, Lu W, Zhang Y, Mata A, Fang Y. Natural polymer-sourced interpenetrating network hydrogels: fabrication, properties, mechanism and food applications. Trends Food Sci Technol. 2021;116:342–356. Available from: https://www.sciencedirect.com/science/article/pii/S0924224421004763.
22. Kim AR, Lee SL, Park SN. Properties and in vitro drug release of pH- and temperature-sensitive double cross-linked interpenetrating polymer network hydrogels based on hyaluronic acid/poly (N-isopropylacrylamide) for transdermal delivery of luteolin. Int J Biol Macromol. 2018Oct;118(Pt A):731–40.
23. Alizadeh N, Barde M, Minkler M, Celestine AD, Agrawal V, Beckingham B, et al. High-fracture-toughness acrylic–polyurethane-based graft-interpenetrating polymer networks for transparent applications. Polym Int. 2021;70(5):636–647. Available from: https://onlinelibrary.wiley.com/doi/abs/10.1002/pi.6149.
24. Thomas S, Grande D, Cvelbar U, Raju KVSN, Narayan R, Thomas SP, et al. Micro- and nano-structured interpenetrating polymer networks: from design to applications. Wiley; 2016.
25. Pramanik S Dutta J Chakraborty P Development of pH-responsive interpenetrating polymer networks of polyacrylamide-g-gum arabica and sodium alginate for gastroprotective delivery of gabapentin Indian J Pharm Sci 2021 07 83 473 482
26. Alokour M Yilmaz E A polymer hybrid film based on poly(vinyl cinnamate) and poly(2-hydroxy ethyl methacrylate) for controlled flurbiprofen release J Polym Res 2021 28 4 137 10.1007/s10965-021-02493-2
27. Jana S, Banerjee A, Sen KK, Maiti S. Gelatin-carboxymethyl tamarind gum biocomposites: In vitro characterization & anti-inflammatory pharmacodynamics. Mater Sci Eng C Mater Biol Appl. 2016;69:478–85.
28. Matricardi P, Di Meo C, Coviello T, Hennink WE, Alhaique F. Interpenetrating Polymer Networks polysaccharide hydrogels for drug delivery and tissue engineering. Adv Drug Deliv Rev. 2013;65(9):1172–1187. Polysaccharide-based systems in drug and gene delivery. Available from: https://www.sciencedirect.com/science/article/pii/S0169409X13000628.
29. Ghosal K, Adak S, Agatemor C, G P, Kalarikkal N, Thomas S. Novel interpenetrating polymeric network based microbeads for delivery of poorly water soluble drug. J Polym Res. 2020;27(4):98.
30. Banerjee S, Chaurasia G, Pal D, Ghosh AK, Ghosh A, Kaity S. Investigation on crosslinking density for development of novel interpenetrating polymer network (IPN) based formulation. J Sci Ind Res. 2010;69:777–84.
31. Swain S, Bal T. Carrageenan-guar gum microwave irradiated micro-porous interpenetrating polymer network: a system for drug delivery. Int J Polym Mater Polym Biomater. 2019;68(5):256–265. Available from: 10.1080/00914037.2018.1443931.
32. Ferreira SA, Gama FM, Vilanova M. Polymeric nanogels as vaccine delivery systems. Nanomed Nanotechnol Biol Med. 2013;9(2):159–73.
33. Dror M Elsabee MZ Berry GC Interpenetrating polymer networks for biological applications Biomater Med Devices Artif Organs 1979 7 1 31 39 10.3109/10731197909119370 454781
34. Biswas A, Mondal S, Das SK, Bose A, Thomas S, Ghosal K, et al. Development and characterization of natural product derived macromolecules based interpenetrating polymer network for therapeutic drug targeting. ACS Omega. 2021Nov;6(43):28699–709.
35. Zou Z Zhang B Nie X Cheng Y Hu Z Liao M A sodium alginate-based sustained-release IPN hydrogel and its applications RSC Adv 2020 10 39722 39730 10.1039/D0RA04316H 35515393
36. Singh P Senthil Kumar S Keerthi T Tamizh Mani T Getyala A Interpenetrating polymer network (IPN) microparticles an advancement in novel drug delivery system: a review Pharma Science Monitor. 2011 3 1826 1837
37. Zhu C, Tang N, Gan J, Zhang X, Li Y, Jia X, et al. A pH-sensitive semi-interpenetrating polymer network hydrogels constructed by konjac glucomannan and poly (γ-glutamic acid): synthesis, characterization and swelling behavior. Int J Biol Macromol. 2021;185:229–239. Available from: https://www.sciencedirect.com/science/article/pii/S0141813021012411.
38. Park JC, Hwang YS, Lee JE, Park KD, Matsumura K, Hyon SH, et al. Type I atelocollagen grafting onto ozone-treated polyurethane films: cell attachment, proliferation, and collagen synthesis. J Biomed Mater Res. 2000;52(4):669–77.
39. Zhao J, Zhao X, Guo B, Ma PX. Multifunctional interpenetrating polymer network hydrogels based on methacrylated alginate for the delivery of small molecule drugs and sustained release of protein. Biomacromol. 2014;15(9):3246–52.
40. Gachuz EJ, Castillo-Santillán M, Juarez-Moreno K, Maya-Cornejo J, Martinez-Richa A, Andrio A, et al. Electrical conductivity of an all-natural and biocompatible semi-interpenetrating polymer network containing a deep eutectic solvent. Green Chem. 2020;22:5785–5797.
41. Hsu HJ, Bugno J, Lee SR, Hong S. Dendrimer-based nanocarriers: a versatile platform for drug delivery. Wiley interdisciplinary reviews Nanomedicine and nanobiotechnology. 2017 Jan;9(1).
42. Dobrovolskaia MA McNeil SE Immunological properties of engineered nanomaterials Nat Nanotechnol 2007 2 8 469 478 10.1038/nnano.2007.223 18654343
43. Mukherji D Marques CM Kremer K Smart responsive polymers: fundamentals and design principles Annual Review of Condensed Matter Physics. 2020 11 271 299 10.1146/annurev-conmatphys-031119-050618
44. Langer R, Tirrell DA. Designing materials for biology and medicine. Nature. 2004;428(6982:487–492.
45. Schmaljohann D Thermo-and pH-responsive polymers in drug delivery Adv Drug Deliv Rev 2006 58 15 1655 1670 10.1016/j.addr.2006.09.020 17125884
46. Ghani M, Heiskanen A, Kajtez J, Rezaei B, Larsen NB, Thomsen P, et al. On-demand reversible UV-triggered interpenetrating polymer network-based drug delivery system using the spiropyran-merocyanine hydrophobicity switch. ACS Appl Mater Interfaces. 2021;13(3):3591–604.
47. Ward MA, Georgiou TK. Thermoresponsive polymers for biomedical applications. Polymers. 2011;3(3):1215–1242. Available from: https://www.mdpi.com/2073-4360/3/3/1215.
48. Kaity S Ghosh A Facile preparation of acrylamide grafted locust bean gum-poly(vinyl alcohol) interpenetrating polymer network microspheres for controlled oral drug delivery J Drug Delivery Sci Technol. 2016 33 1 12 10.1016/j.jddst.2016.02.005
49. Smith GN, Brok E, Schmiele M, Mortensen K, Bouwman WG, Duif CP, et al. The microscopic distribution of hydrophilic polymers in interpenetrating polymer networks (IPNs) of medical grade silicone. Polymer. 2021;224:123671. Available from: https://www.sciencedirect.com/science/article/pii/S0032386121002949.
50. Rehmani S, Ahmad M, Minhas MU, Anwar H, Zangi MIud, Sohail M. Development of natural and synthetic polymer-based semi-interpenetrating polymer network for controlled drug delivery: optimization and in vitro evaluation studies. Polym Bull. 2017;74(3):737–761.
51. Agnihotri SA, Aminabhavi TM. Novel interpenetrating network chitosan-poly(ethylene oxide-g-acrylamide) hydrogel microspheres for the controlled release of capecitabine. Int J Pharm. 2006;324(2):103–15.
52. Rokhade AP, Shelke NB, Patil SA, Aminabhavi TM. Novel hydrogel microspheres of chitosan and pluronic F-127 for controlled release of 5-fluorouracil. J Microencapsul. 2007;24(3):274–88.
53. Lee P, Lok CN, Che CM, Kao WJ. A Multifunctional hydrogel delivers gold compound and inhibits human lung cancer xenograft. Pharm Res. 2019;36(4):61.
54. Wang Y, Li Q, Zhou JE, Tan J, Li M, Xu N, et al. A photopolymerized semi-interpenetrating polymer networks-based hydrogel incorporated with nanoparticle for local chemotherapy of tumors. Pharm Res. 2021Apr;38(4):669–80.
55. Madhusudana Rao K, Krishna Rao KSV, Ramanjaneyulu G, Ha CS. Curcumin encapsulated pH sensitive gelatin based interpenetrating polymeric network nanogels for anti cancer drug delivery. Int J Pharm. 2015Jan;478(2):788–95.
56. Jimenez-Rosales A Flores-Merino MV A brief review of the pathophysiology of non-melanoma skin cancer and applications of interpenetrating and semi-interpenetrating polymer networks in its treatment Regenerative Engineering and Translational Medicine. 2018 4 4 187 205 10.1007/s40883-018-0061-3
57. Thorne JB Vine GJ Snowden MJ Microgel applications and commercial considerations Colloid Polym Sci 2011 289 5 625 10.1007/s00396-010-2369-5
58. Krishna Rao KSV, Vijaya Kumar Naidu B, Subha MCS, Sairam M, Aminabhavi TM. Novel chitosan-based pH-sensitive interpenetrating network microgels for the controlled release of cefadroxil. Carbohydr Polym. 2006;66(3):333–344. Available from: https://www.sciencedirect.com/science/article/pii/S0144861706001561.
59. Bhattacharya SS Mazahir F Banerjee S Verma A Ghosh A Preparation and in vitro evaluation of xanthan gum facilitated superabsorbent polymeric microspheres Carbohyd Polym 2013 98 1 64 72 10.1016/j.carbpol.2013.05.011
60. Risbud MV, Hardikar AA, Bhat SV, Bhonde RR. pH-sensitive freeze-dried chitosan–polyvinyl pyrrolidone hydrogels as controlled release system for antibiotic delivery. J Control Release. 2000;68(1):23–30. Available from: https://www.sciencedirect.com/science/article/pii/S016836590000208X.
61. Agnihotri SA, Aminabhavi TM. Development of novel interpenetrating network gellan gum-poly(vinyl alcohol) hydrogel microspheres for the controlled release of carvedilol. Drug Dev Ind Pharm. 2005;31(6):491–503.
62. Hosseinzadeh H Interpenetrating network polymer hydrogels of sodium alginate and poly(acrylic acid) for controlled release of prazosin hydrochloride Orient J Chem 2012 03 28 349 352 10.13005/ojc/280144
63. Ray D Gils P Mohanta G Sahoo P Comparative delivery of diltiazem hydrochloride through synthesized polymer: hydrogel and hydrogel microspheres J Appl Polym Sci 2009 01 116 959 968
64. Thacharodi D, Panduranga Rao K. Collagen-chitosan composite membranes controlled transdermal delivery of nifedipine and propranolol hydrochloride. Int J Pharm. 1996;134(1):239–241. Available from: https://www.sciencedirect.com/science/article/pii/0378517396044535.
65. Boppana R, Krishna Mohan G, Nayak U, Mutalik S, Sa B, Kulkarni RV. Novel pH-sensitive IPNs of polyacrylamide-g-gum ghatti and sodium alginate for gastro-protective drug delivery. Int J Biol Macromol. 2015;75:133–43.
66. Rani P, Sen G, Mishra S, Jha U. Microwave assisted synthesis of polyacrylamide grafted gum ghatti and its application as flocculant. Carbohydr Polym. 2012;89(1):275–281. Available from: https://www.sciencedirect.com/science/article/pii/S0144861712002202.
67. Eswaramma S, Rao KSVK. Synthesis of dual responsive carbohydrate polymer based IPN microbeads for controlled release of anti-HIV drug. Carbohydr Polym. 2017;156:125–134. Available from: https://www.sciencedirect.com/science/article/pii/S0144861716310773.
68. Sullad AG, Manjeshwar LS, Aminabhavi TM. Novel semi-interpenetrating microspheres of dextran-grafted-acrylamide and poly(vinyl alcohol) for controlled release of abacavir sulfate. Ind Eng Chem Res. 2011;50(21):11778–84.
69. Krishna Rao KSV Espenti C Kummara MR Eswaramma S Raju R Development of gelatin-lignosulfonic acid blend microspheres for controlled release of an anti-malarial drug (pyronaridine) Indian Journal of Advances in Chemical Science. 2015 01 3 25 32
70. Kulkarni RV, Patel FS, Nanjappaiah HM, Naikawadi AA. In vitro and in vivo evaluation of novel interpenetrated polymer network microparticles containing repaglinide. Int J Biol Macromol. 2014;69:514–22.
71. Rokhade AP, Shelke NB, Patil SA, Aminabhavi TM. Novel interpenetrating polymer network microspheres of chitosan and methylcellulose for controlled release of theophylline. Carbohydr Polym. 2007;69(4):678–687. Available from: https://www.sciencedirect.com/science/article/pii/S0144861707001154.
72. Prajapati VD, Gandhi AK, Patel KK, Patel BN, Chaudhari AM, Jani GK. Development and optimization of modified release IPN macromolecules of oxcarbazepine using natural polymers. Int J Biol Macromol. 2015;73:160–9.
73. Yin L, Ding J, Fei L, He M, Cui F, Tang C, et al. Beneficial properties for insulin absorption using superporous hydrogel containing interpenetrating polymer network as oral delivery vehicles. Int J Pharm. 2008;350(1–2):220–9.
74. Kurakula M, Rao GSNK. Pharmaceutical assessment of polyvinylpyrrolidone (PVP): as excipient from conventional to controlled delivery systems with a spotlight on COVID-19 inhibition. J Drug Delivery Sci Technol.. 2020;60.
75. De Temmerman ML, Rejman J, Demeester J, Irvine DJ, Gander B, De Smedt SC. Particulate vaccines: on the quest for optimal delivery and immune response. Drug Discov Today. 2011;16(13–14):569–82.
76. Look M, Bandyopadhyay A, Blum JS, Fahmy TM. Application of nanotechnologies for improved immune response against infectious diseases in the developing world. Adv Drug Deliv Rev. 2010;62(4–5):378–93.
77. Rice-Ficht AC, Arenas-Gamboa AM, Kahl-McDonagh MM, Ficht TA. Polymeric particles in vaccine delivery. Curr Opin Microbiol. 2010;13(1):106–12.
78. Broos S, Lundberg K, Akagi T, Kadowaki K, Akashi M, Greiff L, et al. Immunomodulatory nanoparticles as adjuvants and allergen-delivery system to human dendritic cells: implications for specific immunotherapy. Vaccine. 2010;28(31):5075–85.
79. Skwarczynski M, Toth I. Peptide-based synthetic vaccines. Chem Sci. 2016;7(2):842–54.
80. Lofthouse S. Immunological aspects of controlled antigen delivery. Adv Drug Deliv Rev. 2002;54(6):863–70.
81. Schijns VEJC, Lavelle EC. Trends in vaccine adjuvants. Expert Rev Vaccines. 2011;10(4):539–50.
82. George M, Abraham TE. Polyionic hydrocolloids for the intestinal delivery of protein drugs: alginate and chitosan-a review. Journal of Controlled Release: Official Journal of the Controlled Release Society. 2006;114(1):1–14.
83. Rodrigues MA, Figueiredo L, Padrela L, Cadete A, Tiago J, Matos HA, et al. Development of a novel mucosal vaccine against strangles by supercritical enhanced atomization spray-drying of Streptococcus equi extracts and evaluation in a mouse model. Eur J Pharm Biopharm. 2012;82(2):392–400. Available from: https://www.sciencedirect.com/science/article/pii/S0939641112002305.
84. Mallakpour S, Azadi E, Hussain CM. Chitosan, alginate, hyaluronic acid, gums, and β-glucan as potent adjuvants and vaccine delivery systems for viral threats including SARS-CoV-2: A review. Int J Biol Macromol. 2021;182:1931–40.
85. Petitjean M García-Zubiri IX Isasi JR History of cyclodextrin-based polymers in food and pharmacy: a review Environ Chem Lett 2021 19 4 3465 3476 10.1007/s10311-021-01244-5 33907537
86. Shae D, Postma A, Wilson JT. Vaccine delivery: where polymer chemistry meets immunology. Ther Deliv. 2016;7(4):193–196.
87. Zhang M, Hong Y, Chen W, Wang C. Polymers for DNA vaccine delivery. ACS Biomater Sci Eng. 2017;3(2):108–25.
88. Bolhassani A Javanzad S Saleh T Hashemi M Aghasadeghi MR Sadat SM Polymeric nanoparticles: potent vectors for vaccine delivery targeting cancer and infectious diseases Hum Vaccin Immunother. 2014 10 2 321 332 10.4161/hv.26796 24128651
89. Kumar M, Behera AK, Lockey RF, Zhang J, Bhullar G, De La Cruz CP, et al. Intranasal gene transfer by chitosan-DNA nanospheres protects BALB/c mice against acute respiratory syncytial virus infection. Hum Gene Ther. 2002;13(12):1415–25.
90. Li GP, Liu ZG, Liao B, Zhong NS. Induction of Th1-type immune response by chitosan nanoparticles containing plasmid DNA encoding house dust mite allergen Der p 2 for oral vaccination in mice. Cell Mol Immunol. 2009;6(1):45–50.
91. Wu Z, Li T. Nanoparticle-mediated cytoplasmic delivery of messenger RNA vaccines: challenges and future perspectives. Pharm Res. 2021;38(3):473–8.
92. Pardi N, Hogan MJ, Weissman D. Recent advances in mRNA vaccine technology. Curr Opin Immunol. 2020;65:14–20.
93. Nevagi RJ, Skwarczynski M, Toth I. Polymers for subunit vaccine delivery. Eur Polym J. 2019;114:397–410. Available from: https://www.sciencedirect.com/science/article/pii/S0014305718320676.
94. Carreira AS, Gonçalves FAMM, Mendonça PV, Gil MH, Coelho JFJ. Temperature and pH responsive polymers based on chitosan: applications and new graft copolymerization strategies based on living radical polymerization. Carbohydr Polym. 2010;80(3):618–630. Available from: https://www.sciencedirect.com/science/article/pii/S0144861710000093.
95. Sharma A, Kontodimas K, Bosmann M. Nanomedicine: a diagnostic and therapeutic approach to COVID-19. Front Med. 2021;8. Available from: https://www.frontiersin.org/article/10.3389/fmed.2021.648005.
96. Pal K Banthia A Majumdar DK Polymeric hydrogels: characterization and biomedical applications Des Monomers Polym 2009 12 3 197 220 10.1163/156855509X436030
| 36443634 | PMC9707272 | NO-CC CODE | 2022-12-01 23:20:22 | no | Drug Deliv Transl Res. 2022 Nov 28;:1-19 | utf-8 | Drug Deliv Transl Res | 2,022 | 10.1007/s13346-022-01254-y | oa_other |
==== Front
Econ Change Restruct
Economic Change and Restructuring
1573-9414
1574-0277
Springer US New York
9458
10.1007/s10644-022-09458-5
Article
Is there any recovery power for economic growth from green finance? Evidence from OECD member countries
Huang Leping [email protected]
1
Cao Yuning [email protected]
2
Zhu Yingfu [email protected]
1
1 grid.464478.d 0000 0000 9729 0286 School of Foreign Languages, Tianjin University of Commerce, Tianjin, 300134 China
2 grid.464478.d 0000 0000 9729 0286 School of Economics, Tianjin University of Commerce, Tianjin, 300134 China
28 11 2022
118
14 7 2022
4 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
This paper investigates the dynamic interactions between green finance, economic growth, and green energy consumption for the Organization of Economic Cooperation and Development (OECD) members. The econometric analysis is conducted on annual data gathered throughout 2010–2020 using different estimation techniques of the Vector Autoregressive model, causality, and co-integration approaches. The main results confirmed a positive bi-directional relationship between GDP and green energy consumption. In addition, there is a two-way relationship between the volume of green bond issuance and the use of green energy in OECD countries. The recommended practical policy recommendations are establishing a unified green bonds market among OECD member states, prioritizing green projects to support the issued green bonds, improving the financial system, and financing rural electrification and electric vehicle transition by green bonds.
Keywords
Green recovery
Decarbonization economic policy
COVID-19
Causality approach
Green finance
JEL Classification
Q41
K31
C32
Humanity and Social Science Youth Foundation of Ministry of Education of ChinaNo.19YJC740020 Huang Leping
==== Body
pmcIntroduction
The emergence of the coronavirus disease at the end of 2019 and its transformation into a pandemic in the world in 2020 severely disrupted economic growth globally. Meyer et al. (2022) declare that the COVID-19 pandemic has caused severe exogenous supply and demand shocks for firms and businesses. It has changed the expectations of businesses for future situations of economic markets. Teng et al. (2022) argue that COVID-19 has altered the traditional global supply chain causing unprecedented shock to economic sectors. In order to control the rapid spread of COVID-19 and reduce the deaths caused by this pandemic, governments were forced to order the closure of cities, economic enterprises, hotels, and domestic and international travel (Nicolar et al., 2020). These measures lead to the reduction of economic activities, stagnation, the spread of poverty and unemployment, the reduction of people purchasing power, and the bankruptcy of many economic enterprises. In such a way, the years 2020 and 2021 can be called dark and stagnant years for the world economy.
With the discovery of the COVID-19 vaccine, the threat caused by this pandemic is fading gradually, and the economy of many countries can return to their regular activity. However, to compensate for the economic stagnation of the previous years, the governments have announced an economic growth program with all the power and capacities under the concept of economic growth recovery in the world economic literature (Wang and Zhang 2021). The recovery of economic growth can neutralize COVID-19's adverse effects and increase government budget income. Improve job creation and put the welfare level of society in a more favorable situation.
Achieving economic growth as a priority for the world's countries is a logical and correct matter (Economic growth is the basis of development, job creation, poverty alleviation, and reduction of investment risk in the world.). Because in the shadow of such economic growth, the post-COVID-19 era can be accompanied by a decrease in unemployment and flourishing of human development. Another issue that should be considered is achieving eco-friendly economic growth that can put countries on the path of sustainable development indicators. The fact that countries should think about protecting their environment has been a concern of the United Nations for the last few decades, and countries have always been invited to cooperate comprehensively to eliminate the threats of global warming and environmental pollution. Many experts (e.g., Elavarasan et al. 2021) have considered the post-COVID era a global opportunity to advance sustainable development indicators. While another group of experts (e.g., see Ray et al. (2022), who confirmed the temporary impact of COVID-19 on the reduction of CO2 emissions, has called countries' lack of attention to environmental issues and focusing only on revitalizing economic engines a severe threat to the path of sustainable development in the post-COVID-19 era.
One of the issues in achieving green economic recovery is the lack of sufficient capital to promote green projects in countries. It is crucial to open a special field for finance called green finance (Taghizadeh-Hesary and Yoshino, 2019; Zhang et al., 2022a; Taghizadeh-Hesary et al., 2022; Zhao et al. 2022). On the other hand, assessing the rule of green finance instruments in filling the finance gap of green projects is essential.
Developing green financing tools are a globally accepted solution for developing green projects in different countries (Sun et al., 2022). It can help increase investment in clean energy development projects, reduce energy intensity and increase energy effectiveness. Green financing tools can help attract capital from the private sector and abroad for developing and implementing green energy projects, with favorable guarantees from the issuer and a reasonable investment return rate. One of the newest and most effective green financing tools is green bonds. In recent decades, many countries have been developing their green bond markets and using this financing tool to develop green energy consumption and reduce dependence on fossil energy. Historically, the World Bank issued the first green bond in November 2008 following the intentions of Swedish Pension Funds to invest in clean environment projects in 2007. Figure 1 shows the development trend of the green bond market in different regions of the world. According to the data, the volume of green bonds issued worldwide has increased from 37 billion US dollars in 2014 to more than 500 billion US dollars in 2021, which shows a 13-fold increase in this market globally. Among the different regions of the world, Europe and Asia–Pacific are the two leaders in developing the global green bond market. In such a way, the number of green bonds issued in Europe in 2021 was about 265 billion US dollars, and Asia–Pacific has more than 130 billion US dollars. This degree of green bond market development shows the importance of this green financing tool for the countries of the world, especially Europe and Asia–Pacific. Countries that can issue green bonds with proper legislation and transparency of the market will surely enjoy its positive effects during the path of energy transition and environmental protection.Fig. 1 Issued green bonds in different regions, 2014–2021, billion US dollars.
Source: Authors from Climate bonds (https://www.climatebonds.net/market/data/)
Studying how GDP changes, issued green bonds, and clean energy demand can provide essential and effective strategic policies. Especially in the post-COVID-19 era, policymakers are doing their best to use different programs and tools to increase the consumption of renewable energy compared to the consumption of fossil energy and increase their economic growth to a greater share of the consumption of green energy. However, the issue of the existence/absence of nexus between economic growth and the issuance of green bonds, between economic growth and clean energy consumption, and the issuance of green bonds—clean energy consumption, requires more comprehensive and detailed research.
The main objective of this paper is to explore the causal relationship between issued green bonds, economic growth, and green energy consumption for 38 OECD member countries throughout 2010–2020. The main reasons for choosing OECD as the case study are: OECD member countries have strong economies and suitable infrastructures for developing green energy. Therefore, it is essential to investigate the issue of issuing green bonds and its effect on economic variables. According to the OECD Economic Outlook (2022), member countries have experienced an unprecedented economic recession due to the spread of COVID-19. The predictions are that their economic power will be restored to the level before the spread of COVID-19. Another reason is the developing of the green financing market in OECD member countries. The issuing of green bonds started in 2007–8 and reached over 3 billion US dollars in 2011, 95 billion US dollars in 2016, and around 3 trillion US dollars in 2021.
The contributions of this paper to the previous literature are from the following aspects: The triple causal relationship between green bond issuance, economic growth, and clean energy consumption has been considered in this research. Moreover, a panel of OECD countries is analyzed through the panel-VAR technique, which considers all variables as exogenous, providing practical results. In addition, this article discusses the main obstacles to developing the green bond market in OECD countries in the post-COVID-19 era.
To fill in the literature gap and contribute to existing literature, the research structure is determined as below: Sect. 2 provides a brief discussion about the previous studies to recognize the literature gap to fill in. The following Section clarifies the data information and the econometric modeling strategy. Section 4 argues the empirical findings and estimations. Section 5 expresses a discussion with the aim of clarification developing barriers to the green bonds market in the OECD member countries. Section 6 provides the robustness check results. The final section concludes the paper and provides policy implications and recommendations for future research.
Literature review
To discuss the literature review, two separate approaches can be considered. The first approach to the subject literature is to review studies that have addressed the importance of green financing, especially green bonds. In a study, Uddin et al. (2022) expressed that green bonds have spillover impacts different aspects of local and global economies. It can provide a better investment climate for green projects and help countries achieve sustainable development. Flammer (2021) addressed green bonds as an efficient instrument to promote climate-friendly investments in countries. In another study, Tang and Zhang (2020) argued that green bonds could provide significant advantages for shareholders leading to a better green capital market. Several studies have focused on the fact that to achieve sustainable development. It is necessary to establish and develop green bonds market. Bhutta et al. (2022) studied the impacts of green bonds and concluded that promoting green bonds is helpful in combating environmental degradation. In another study, Wang et al. (2022) analyzed monthly data from 2011 to 2021 of different economies and found that issued green bonds significantly contribute to sustainable development. Tolliver et al. (2020) expressed that in line with the Paris Agreement, all countries try differently to promote green financing tools, and green bonds are among the most appropriate instruments. Voica et al. (2015) declared that promoting green investments is essential for countries to increase the share of consumption of green energy in the total energy consumption basket. The other fresh studies about the importance of green financing instruments are summarized in Table 1 as follows:Table 1 Summarization of fresh studies on green bonds' role in sustainable development.
Source: Authors
Author/s Year of research Main findings
Teti et al. (2022) Issued green bonds are efficient instruments for investors in green projects
Ning et al. (2022) Green bonds are practical tools for promoting sustainable green financing
Baldacci and Possamai (2022) Providing incentives by governments may lead to a more efficient green bonds market, causing better green projects' promotion
Lin and Hong (2022) Green bonds market helps China to promote green projects and energy transformation
Chai et al. (2022) Green bonds are appropriate tools in the COVID-19 era in order to increase clean energy consumption
Khan et al. (2022) In Asia, green finance reduces environmental pollution
Mamun et al. (2022) Green bonds accelerate the speed of the decarbonization process around the world
Nguyen et al. (2022) Developing economies can use green bonds to access sufficient capital to invest in green projects
Loffler et al. (2021) Green bonds help countries to conduct greenium concept, which means environment-friendly economic activities
Chen and Zhao (2021) Environment-related projects in China are positively affected by green bonds
In the second literature review approach, there are studies focusing on the relationship between green bonds, green energy consumption, and economic growth. Ntsama et al. (2021) expressed that issued green bonds have different economic and social advantages improving good governance indicators and social responsibilities, which are two success wings for sustainable development. Lie t al. (2021) studied the interactions between green energy and green finance in the case of China from 1990 to 2020. The Wavelet Power Spectrum technique findings revealed that green finance positively impacts green energy development. Similarly, Zhou and Li (2022) explored the relationship between green finance, green energy, and sustainable development in China. They found out the crucial role of clean energy consumption and green finance tools on environmental sustainability. Zhang et al. (2022b) argued that as green finance tools can enhance private participation in green projects, it also causes GDP growth and environmental protection. In another study, Yang et al. (2022) concentrated on G7 economies and denoted green financing tools' significant roles in green economic growth and sustainable development. Rasoulinezhad and Taghizadeh-Hesary (2022) applied the STIRPAT (the stochastic impact by regression on the population, affluence, and technology) to analyze the impacts of green finance in 10 economies. The major findings confirmed green bonds' positive and statistically significant role on CO2 emissions reduction and green project promotion. Huang et al. (2022) emphasized green finance's role in promoting green innovation, which can solve the problem of high energy intensity and low energy efficiency in non-developed economies. Focusing on the linkage between green fiancé and green economic performance, Feng et al. (2022a) analyzed a group of countries in the Belt and Road Initiative (BRI) throughout 2008–2018. They concluded that government policies are crucial to reach efficient green financing tools to go toward green economic recovery in the post-COVID era. In another fresh study, Kung et al. (2022) investigated the impacts of issued green bonds on bioenergy development in China. The results of conducting a recourse model confirmed the positive role of issued green bonds on biopower and biofuels in Taiwan. Cui et al. (2022) studied the green bonds market under COVID-19. They concluded that the market of green bonds was affected harshly by the consequences of COVID-19. However, the market has the potential to recover and positively helps countries promote green economic activities.
Reviewing the collection of the latest literature on green bonds, economic growth, and clean energy consumption shows that, so far, no comprehensive and in-depth research has been done on the causal relationship of these three variables for OECD countries. Therefore, this article tries to solve this literature gap and can have diverse and new strategic and practical results for experts and policymakers. The causal relationship between these three variables is more important for the post-COVID-19 era, when the countries of the world seek to achieve the recovery of economic growth and prioritize sustainable development indicators.
Strategy and method of research
Data sources and characteristics
This paper tries to find out the causal interactions between issued green bonds, economic growth, and renewable energy consumption for the case of 38 OECD member countries. The countries are experiencing the situation of economic growth recovery in the post-COVID era with consideration of sustainable development indicators achievements. To do the empirical analysis, the annual data of variables are gathered from different global databases (the World Bank, the initiative Climate Bonds; BP Statistical Review of World Energy 2021; OECD Main Economic Indicators (MEI)) throughout 2010–2020 (observation numbers for panel data are 418).
Regarding three selected variables in the OECD member countries, Table 2 reports the correlation among these series:Table 2 Correlations between three variables of the model.
Source: Author's compilation from SPSS 20
Variables Economic growth Issued green bonds Green energy consumption
Economic growth 1 – –
Issued green bonds 0.014 1 –
Green energy consumption 0.081 0.321 1
The results of pairwise correlation among three variables depicted the positive relationship between economic growths, issued green bonds, and green energy consumption in the 38 OECD member countries, meaning that all these variables may have significant impacts on each other.
Estimation process strategy
Choosing an appropriate estimator is a crucial step in doing research. In this paper, to explore how three variables of economic growth, issued green bonds and green energy consumption, are related, the Vector Autoregressive model (VAR) in the panel data framework for 38 OCED member states is employed. The estimation technique of VAR has various advantages (Grossmann et al. 2014), including exogenous observations, and can be written as Eq. 1:1 yi,t=β0+∑j=1pβiyi,t-j+μi+ei,t
i=1,…,37,t=2010-2020
In the above Equation, y denotes dependent variables. β and μ are the coefficient vector and heterogeneity, respectively. e represents a random error.
A possibility that can be done in the framework of the panel-VAR approach model is to implement variance decomposition, which shows a percentage change in the variable's value from which variable and how much.
According to the three variables of issued green bonds, economic growth, and green energy consumption, the paper will examine the three following Eqs. (2–4):2 ΔEGROi,t=fΔEGROi,t-1,ΔGBONi,t-1,ΔGECi,t-1
3 ΔGBONi,t=fΔEGROi,t-1,ΔGBONi,t-1,ΔGECi,t-1
4 ΔGECi,t=fΔEGROi,t-1,ΔGBONi,t-1,ΔGECi,t-1
where Δ shows the first variable difference in the model.
In order to ensure the correctness of using the Vector Autoregression method, it is necessary to perform unit root tests. In this paper, the cross-sectional dependency of variables is checked by the Pesaran Cross-Sectional test. Then the second-generation panel unit root test (Pesaran CIPS Pesaran 2007) is conducted. In the next step, the Westerund co-integration test is used to find out the existence of a long-term relationship among series. The Granger Wald causality test is employed to ascertain the existence and direction of causality relationship. The conceptual framework of the empirical modelling strategy is shown in Fig. 2 as follows:Fig. 2 Conceptual framework of empirical modeling.
Source: Authors’ depiction
Estimation results and discussions
In the first step, the existence of the unit root in the series is checked. Pesaran's cross-sectional test is employed to clarify whether the variables are interdependent. The results of this test are represented in Table 3:Table 3 Pesaran's cross-sectional test.
Source: Authors’ calculations Pesaran's cross-sectional
– CD-test Prob Mean ρ Mean abs (ρ)
Economic growth 44.394 0.00 0.70 0.80
Issued green bonds 39.402 0.00 0.69 0.71
Green energy consumption 38.711 0.00 0.54 0.68
The findings, reported in Table 3, confirmed the dependency between variables. Therefore, the second generation of panel data unit root tests can be employed to find out whether exists a unit root in series. To this end, the CIPS (Table 4) is conducted and depicts the non-stationarity of series at level.Table 4 CIPS unit root test.
Source: Authors’ calculations
Variable CIPS Critical value
10% 5% 1%
Economic growth − 1.510 − 2.06 − 2.18 − 2.35
Issued green bonds − 2.023 − 2.06 − 2.18 − 2.35
Green energy consumption − 2.429 − 2.06 − 2.18 − 2.35
Since the three variables of our model are not stationary, the existence of co-integration nexus between series is checked employing the Westerlund's panel co-integration test. The test's results, reported in Table 5, confirmed the absence of co-integration linkage between three variables.Table 5 Westerlund's co-integration test.
Source: Authors’ calculations
Stat Prob
H0: No co-integration 0.7983 0.2193
According to the above carried preliminary tests, it is not possible to do an estimation with error correction. As a popular solution, we can transform all variables into the logarithmic form which may lower the challenge of heterogeneity of variables. Of course, it should be pointed out that because the growth rate cannot be converted into a logarithm, we are forced to change the economic growth variable. The best alternative for this variable is the amount of GDP.
The findings of re-conducting the Pesaran panel unit root test, listed in Table 6, confirmed data non-stationarity for issued green bonds and GDP. In contrast, the CIPS statistics for green energy consumption are close to the critical value causing the need to test stationarity for the variables in the form of logarithm and first different approach.Table 6 CIPS unit root test (logarithmic variable).
Source: Authors’ calculations
Variable CIPS Critical value
10% 5% 1%
GDP − 1.942 − 2.06 − 2.18 − 2.35
Issued green bonds − 2.288 − 2.06 − 2.18 − 2.35
Green energy consumption − 2.351 − 2.06 − 2.18 − 2.35
The results of CIPS for the first difference of the logarithmic series, represented in Table 7, depicted the stationarity of all three series.Table 7 CIPS unit root test (first difference).
Source: Authors’ calculations
Variable CIPS Critical value
10% 5% 1%
GDP − 2.863 − 2.1 − 2.21 − 2.45
Issued green bonds − 2.895 − 2.1 − 2.21 − 2.45
Green energy consumption − 3.562 − 2.1 − 2.21 − 2.45
After determination of stationarity of variables, the VAR estimation can be conducted. However, prior to estimation, it is necessary to recognize an appropriate number of lags. To this end, the Hansen and Singleton (1982) and Andrews and Lu (2001) criteria are employed (Table 8):Table 8 Lags selection.
Source: Authors’ calculation
Lag CD J J prob MBIC MAIC MQIC
1 0.983 34.53 0.100 − 95.014 − 18.54 − 50.032
2 0.984 18.119 0.443 − 70.021 − 18.65 − 39.992
3 0.821 12.493 0.179 − 31.481 − 5.390 − 16.011
By determining the appropriate lags number, the panel single-lag VAR technique is conducted to evaluate the impacts of variables. The results of the estimation are reported in Table 9:Table 9 Panel VAR results.
Source: Authors’ calculations
Coefficient Standard error z P > z
dln (GDP)
dln (GDP) L1 0.132 0.062 3.442 0.000
dln (GBON) L1 0.031 0.042 1.142 0.319
dln (GEC) L1 0.042 0.011 2.103 0.032
dln (GBON)
dln (GDP) L1 0.142 0.173 1.243 0.219
dln (GBON) L1 0.041 0.063 1.381 0.201
dln (GEC) L1 0.162 0.031 − 3.618 0.000
dln (GEC)
dln (GDP) L1 0.391 0.131 2.449 0.000
dln (GBON) L1 0.182 0.072 2.229 0.019
dln (GEC) L1 0.184 0.058 1.504 0.283
dln shows the first difference of the logarithm of variables
The major empirical estimation results confirmed that the lagged GDP and green energy consumption positively impact the GDP variable. The green energy demand affects the growth rate of issued green bonds has statistically significant value. Therefore green energy consumption level of OECD in the previous period impacts the development of green bonds in the present period. Green energy demand in the examined OECD member countries is influenced by the previous values of GDP and issued green bonds. Both growth rates have the impact of increasing the level of consumption of green energy in these economies.
The causal relationship among the three variables is checked in the next step using the Granger-Wald test. The results are reported in Table 10:Table 10 Granger causality Wald test.
Source: Authors’ calculations
Chi2 Df Prob > Chi2
dln (GDP)
dln (GBON) 1.513 1 0.281
dln (GEC) 4.421 1 0.028
All 6.318 2 0.024
dln (GBON)
dln (GDP) 1.295 1 0.313
dln (GEC) 13.193 1 0.010
All 14.048 2 0.000
dln (GEC)
dln (GDP) 7.221 1 0.000
dln (GBON) 5.459 1 0.025
All 10.582 2 0.009
dln indicates the first difference of the logarithm of variables
According to Table 10, it can be expressed that:i. There is a bi-directional linkage between green energy demands toward GDP.
ii. A bi-directional causality relationship exists between issued green bonds and the volume of consumption of green energy sources.
iii. There is not any significant causal relationship between issued green bonds and GDP.
The above findings from the results of the causality test of the relationships between the three variables of GDP, published green bonds, and clean energy consumption can be shown in Fig. 3:Fig. 3 GDP -Green bonds green energy consumption relationship in OECD.
Source: Authors’ depiction
The remarks from conducting the causality test are as follows: (i) a bi-directional causality relationship presents between green bonds and green energy consumption, which corresponds to the findings of other studies about green energy (the main role of issued green bonds is the improvement of lowering carbon-projects causing a larger contribution of renewable energy sources in a country), (ii) There is also bi-directional connectivity between GDP and green energy consumption, (iii) A causal linkage was not confirmed between green bonds and GDP, (iv) the concept of "neutrality hypothesis" in the relationship between GDP and issued green bonds exists in OECD member countries. It is a crucial finding, highlighting that policies related to green bonds market development may not retard GDP size in OECD countries. It can be expressed that any development in the green bonds market may increase the volume of green energy consumption in these economies, which can be defined as a major element for long-run economic growth recovery beyond the shadow of threats of COVID-19.
To ascertain the reliability of estimation results, the stability test, as reported in Table 11, is conducted:Table 11 Stability test.
Source: Authors’ calculations
Eigenvalue Modulus
Real Imaginary
− 0.274291 0 0.274291
0.182939 0 0.182939
0.089313 0 0.089313
The variance decomposition can be implied to clarify the contributions of variables to change of each other. As reported in Table 12, nearly 0.40% of the variation in GDP is explained by changes in the volume of consumption of renewable energy sources, whereas the volume of issued green bonds has a contribution of nearly 1.05% to the variation of GDP in OECD member countries. Analyzing variance decomposition of issued green bonds, it can be highlighted that renewable energy consumption and GDP have contributions of nearly 0.40% and 3.43%, respectively to the variation in issued green bonds. In addition, 17.5% and 1.32% of variation in renewable energy consumption are explained by issued green bonds and GDP, respectively.Table 12 Variance decomposition.
Source: Authors’ calculations
Response of Impulse from
dln (GDP) dln (GBON) dln (GEC)
dln (GDP)
1 1.00 0.000 0.000
10 0.981 0.0040 0.0105
15 0.981 0.0040 0.0105
dln (GBON)
1 0.000 0.999 0.000
10 0.0041 0.985 0.0343
15 0.0041 0.985 0.0343
dln (GEC)
1 0.1517 0.000 0.8751
10 0.1754 0.0134 0.8549
15 0.1754 0.0134 0.8549
Discussion
As it is clear from the research findings, developing the green bond market as one of the most important green financing tools can play an important role in developing economic growth and clean energy consumption in OECD countries. In addition, developing green financing tools can help these economies to pave the path to sustainable development more easily and quickly. Considering the economic challenges caused by COVID-19 and the war between Russia and Ukraine, this issue is vital for the OECD countries seeking economic recovery based on protecting the environment (OECD, 2020). Historically, the OECD member countries are pioneers in issuing green bonds worldwide. For example, Japan represented its first issued green bonds in October 2014, South Korea sold its first green bonds through the Korea Export–Import Bank (KEXIM) in 2013, and Canada and Italy announced its first issuing green bonds in 2020 and 2021, respectively. Generally, OECD countries have announced various programs to go toward sustainable development. The most famous programs are the 2050 Zero Carbon Emissions in Japan, Clean Power Roadmap for Atlantic Canada, Offshore Wind Energy Roadmap by 2030 in the Netherlands, Turkey's Decarbonization Roadmap by 2050, National Energy Strategy in Italy, and the US Sustainable Finance Roadmap. In all these programs, the issue of green financing or strategies to solve the insufficiency of capital to promote green projects has been drawn big attention. However, barriers to develop green financing tools such as green bonds exist in OECD. For instance, the green bonds market's transparency remains a crucial challenge. In the absence of transparency in the green bond market, investors will be afraid to participate and bring their capital. Therefore there will be little trust in the green bond market without transparency. Russia's military tension with Ukraine has increased the risk of investment in many OECD countries. Therefore green projects in these countries are not the priority of investment by private sector capitalists. Despite the fading of the threat of COVID-19, many life and business patterns have changed, and it seems that there is a need to develop information technology infrastructure in the markets of green financing tools. The result of such action will be the emergence of digital green financing tools such as digital green bonds. Feng et al. (2022b) argued that the global green energy markets need IT-based financing tools to attract global capital for low-carbon transition plans beyond geographical borders.
Robustness check
To investigate the robustness of the results of estimations, additional analysis is conducted. To this end, we expand the econometric model by adding two control variables of inflation rate and FDI collected from the World Bank database. The new estimation is done through a panel co-integration estimator, Fully Modified Ordinary Least Squares (FMOLS). The estimation results are provided in Table 13:Table 13 Robustness check (FMOLS estimation)
Dependent variable Explanatory variable Coefficient Standard error P value
GDP Issued green bonds 0.021 0.024 –
Green energy consumption 0.173 0.059 ***
Inflation rate 0.19–3 0.031 **
FDI 0.118 0.026 –
Issued green bonds GDP 0.231 0.006 –
Green energy consumption 0.127 0.023 ***
Inflation rate − 0.052 0.045 –
FDI 0.106 0.039 ***
Green energy consumption Issued green bonds 0.231 0.142 ***
GDP 0.021 0.041 ***
Inflation rate − 0.036 0.041 ***
FDI 0.100 0.031 –
*** and ** denote statistically significant at 1% and %, respectively
According to Table 13, it can be expressed that the primary estimation results are robust to the use of control variables.
In addition, another robustness check is done for a sample group of OECD countries. To this end, the interconnections among issued green bonds, green energy consumption, and GDP in the EU members in OECD is analyzed through the Granger-Wald causality test. The results of the second robustness check are reported in Table 14, revealing the same results in Table 10.Table 14 Second robustness check (sub-sampling).
Source: Authors’ calculations
Chi2 Df Prob > Chi2
dln (GDP)
dln (GBON) 1.817 1 0.301
dln (GEC) 4.153 1 0.054
dln (GBON)
dln (GDP) 1.401 1 0.282
dln (GEC) 13.006 1 0.021
dln (GEC)
dln (GDP) 7.632 1 0.049
dln (GBON) 5.028 1 0.020
dln indicates the first difference of the logarithm of variables
Conclusions and policy recommendations
This paper studied the issue of green economic recovery in the post-COVID-19 era in OECD member countries. To this end, the relationship between economic growths, issued green bonds, and green energy consumption was explored through econometric techniques. Regarding the aforementioned empirical estimation findings, the following concluding remarks can be expressed:There is a positive bi-directional causal linkage between GDP and renewable energy consumption. Therefore, it can be pointed out that the further expansion of the use of clean energy has a positive and meaningful contribution to the economic growth of the OECD countries. Economic growth in these countries leads to more investment in green projects and, as a result increase the share of renewable energy sources in the aggregate energy consumption portfolio of OECD countries.
A bi-directional linkage between the growth rate of issued green bonds and the consumption volume of renewable energy sources exists. In another sense, there is a two-way relationship between the volume of green bond issuance and the use of green energy in OECD countries.
No causal relationship exists between GDP and issued green bonds in OCED member countries, expressing the lack of efficiency in issued green bonds.
Analyzing variance decomposition of issued green bonds, it can be highlighted that renewable energy consumption and GDP have contributions of nearly 0.40% and 3.43%, respectively to the variation in issued green bonds. In addition, 17.5% and 1.32% of variation in renewable energy consumption are explained by issued green bonds and GDP, respectively.
OECD countries need to implement policies to have more effect on their economic growth. The higher the share of green bonds issued in the growth of the GDP of these countries, it indicates the further development of the use of clean energy in agriculture and various industries.
As practical policy recommendations, financial system improvement (to increase the credibility and reputation of the green bond issuers), financing rural electrification (to increase the life quality of the rural population), and electric vehicle transition by green bonds are highly recommended. Moreover, OECD countries need to conduct different policies like tax incentives and an increase in the rate of return of green bonds to attract the private sector to green bond markets. The countries should also collect carbon tax from the polluting industries and inject it into green bonds as seed money to increase the rate of return of green projects (Yoshino et al. 2021). In addition, market integrity and enhancing risk-return profile may be addressed as two practical policy implications for these nations to make the green bond market an accelerator for GDP growth. Attention to these policies should be paid more in the COVID-era and the post-COVID era since this pandemic has reduced the volume of investments in green energy projects that would endanger climate-related goals. These applied policies may increase the effectiveness of issued green bonds so that their impact on the growth of the OECD countries' GDP is more significant. Another essential practical policy is the development of infrastructure for the application of digital green financing tools. The digital issuance and use of green financing tools will lead to more manageable development of the market of these types of tools around the world, greater transparency, and better adaptation to the business climate in the post-COVID-19 era.
Although this paper has significant contributions to the earlier literature, further exploration of the green finance-income-renewable energy relationship at the country level gives more detailed policies to OECD economies. Comparing practical experiments of countries to develop green bonds market under the condition of COVID-19 would represent a further finding for the relationship between these variables. Another suggestion for future research is quantifying the COVID-19's effects and its inclusion as an important variable in the econometric model. This will calculate the direct and indirect effect of COVID-19 on the relationship between the model variables, which will provide policymakers and experts with more accurate strategic and practical policies.
Declarations
Conflict of interest
We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all the authors. We declare that there is no conflict of interest.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Andrews DWK Lu B Consistent model and moment selection procedures for GMM estimation with application to dynamic panel data models J Econom 2001 101 1 123 164 10.1016/S0304-4076(00)00077-4
Baldacci B Possamai D Governmental incentives for green bonds investment Math Financ Econom 2022 16 539 585 10.1007/s11579-022-00320-w
Bhutta U Tariq A Farrukh M Reza A Iqbal M Green bonds for sustainable development: review of literature on development and impact of green bonds Technol Forecast Soc Chang 2022 175 121378 10.1016/j.techfore.2021.121378
Chai S Chu W Zhang Z Li Z Abedin M Dynamic nonlinear connectedness between the green bonds, clean energy, and stock price: the impact of the COVID-19 pandemic Ann Oper Res 2022 10.1007/s10479-021-04452-y
Chen Y Zhao Z The rise of green bonds for sustainable finance: global standards and issues with the expanding Chinese market Curr Opin Environ Sustain 2021 52 54 57 10.1016/j.cosust.2021.06.013
Cui T Suleman M Zhang H Do the green bonds overreact to the COVID-19 pandemic? Financ Res Lett 2022 49 103095 10.1016/j.frl.2022.103095 36119916
Elavarasan R Pugazhendhi R Jamal T Dyduch J Arif M Kumar N Shafiullah G Chopra S Nadarajah M Envisioning the UN sustainable development goals (SDGs) through the lens of energy sustainability (SDG 7) in the post-COVID-19 world Appl Energy 2021 292 116665 10.1016/j.apenergy.2021.116665
Feng H Liu Z Wu J Iqbal W Ahmad W Marie M Nexus between government spending's and green economic performance: role of green finance and structure effect Environ Technol Innov 2022 27 102461 10.1016/j.eti.2022.102461
Feng S Zhang R Li G Environmental decentralization, digital finance and green technology innovation Struct Chang Econ Dyn 2022 61 70 83 10.1016/j.strueco.2022.02.008
Flammer C Corporate Green Bonds J Financ Econ 2021 142 2 499 516 10.1016/j.jfineco.2021.01.010
Grossmann A Love I Orlov A The dynamics of exchange rate volatility: a panel VAR approach J Int Financ Mark Inst Money 2014 33 1 27 10.1016/j.intfin.2014.07.008
Hansen L Singleton K Generalized instrument variables estimation of nonlinear rational expectations models Econometrica 1982 50 5 1269 1286 10.2307/1911873
Huang Y Chen C Lei L Zhang Y Impacts of green finance on green innovation: a spatial and nonlinear perspective J Clean Prod 2022 365 132548 10.1016/j.jclepro.2022.132548
Khan M Riaz H Ahmed M Saeed A Does green finance really deliver what is expected? An Empirical Perspective Borsa Istanb Rev 2022 22 3 586 593 10.1016/j.bir.2021.07.006
Kung S Li H Li S Zhang L Kung C Effects of green bonds on bioenergy development under climate change: a case study in Taiwan province China Adv Clim Chang Res 2022 13 1 97 106
Lin L Hong Y Developing a green bonds market: lessons from China Eur Bus Organ Law Rev 2022 23 1 143 185 10.1007/s40804-021-00231-1
Li M Hamawandy N Wahid F Rjoub H Bao Z Renewable energy resources investment and green finance: evidence from China Resour Policy 2021 74 102402 10.1016/j.resourpol.2021.102402
Loffler K Petreski A Stephan A Drivers of green bond issuance and new evidence on the "greenium" Eurasian Econ Rev 2021 11 1 24 10.1007/s40822-020-00165-y
Mamun A Boubaker S Nguyen D Green finance and decarbonization: evidence from around the world Financ Res Lett 2022 46 Part B 102807 10.1016/j.frl.2022.102807
Meyer B Prescott B Sheng X The impact of the COVID-19 pandemic on business expectations Int J Forecast 2022 38 2 529 544 10.1016/j.ijforecast.2021.02.009 35185229
Nguyen A Hoang T Nguyen D Nguyen L Doan D The development of green bond in developing countries: insights from Southeast Asia market participants Eur J Dev Res 2022 10.1057/s41287-022-00515-3
Nicola M Alsafi Z Sohrabi C Kerwan A Al-Jabir A Loifidis C Agha M Agha R The socio-economic implications of the coronavirus pandemic (COVID-19): a review Int J Surg 2020 78 185 193 10.1016/j.ijsu.2020.04.018 32305533
Ning Y Cherian J Sial M Alvarez-Otero S Comite U Zia-Ud-Din M Green bond as a new determinant of sustainable green financing, energy efficiency investment, and economic growth: a global perspective Environ Sci Pollut Res 2022 10.1007/s11356-021-18454-7
Ntsama U Yan C Nasiri A Mboungam A Green bonds issuance: insights in low- and middle-income countries Int J Corp Soc Responsib 2021 10.1186/s40991-020-00056-0
OECD. Economic Outlook. 2022. URL: https://www.oecd.org/economic-outlook/
Pesaran M A simple panel unit root test in the presence of cross-section dependence J Appl Econ 2007 22 2 265 312 10.1002/jae.951
Ray R Singh V Singh S Acharya B He Y What is the impact of COVID-19 pandemic on global carbon emissions? Sci Total Environ 2022 816 151503 10.1016/j.scitotenv.2021.151503 34752864
Sun Y Guan W Cao Y Bao Q Role of green finance policy in renewable energy deployment for carbon neutrality: evidence from China Renew Energy 2022 197 643 653 10.1016/j.renene.2022.07.164
Taghizadeh-Hesary F Yoshino N The way to induce private participation in green finance and investment Financ Res Lett 2019 31 98 103 10.1016/j.frl.2019.04.016
Taghizadeh-Hesary F Li Y Rasoulinezhad R Mortha A Long A Lan Y Zhang Z Li N Zhao X Wang Y Green finance and the economic feasibility of hydrogen projects Int J Hydrog Energy 2022 47 58 24511 24522 10.1016/j.ijhydene.2022.01.111
Tang D Zhang Y Do shareholders benefit from green bonds? J Corp Financ 2020 61 101427 10.1016/j.jcorpfin.2018.12.001
Teng B Wang S Shi Y Sun Y Wang W Hu W Shi C Economic recovery forecast under impacts of COVID-19 Econ Model 2022 110 105821 10.1016/j.econmod.2022.105821 35261424
Teti E Baraglia I Dallocchio M Mariani G The green bonds: empirical evidence and implications for sustainability J Clean Prod 2022 366 132784 10.1016/j.jclepro.2022.132784
Tolliver C Keeley A Managi S Drivers of green bond market growth: the importance of nationally determined contributions to the paris agreement and implications for sustainability J Clean Prod 2020 244 118643 10.1016/j.jclepro.2019.118643
Uddin, G., Jayasekara, R., Park, D., Luo, T., and Tian, S. 2022. Go green or stay black: Bond market dynamics in Asia. Int Rev Financ Anal
Voica M Panait M Radulescu I Green investments—between necessity, fiscal constraints and profit Procedia Econ Financ 2015 22 72 79 10.1016/S2212-5671(15)00228-2
Wang Q Zhang F What does the China's economic recovery after COVID-19 pandemic mean for the economic growth and energy consumption of other countries? J Clean Prod 2021 295 126265 10.1016/j.jclepro.2021.126265 33589853
Wang K Zhao Y Jiang C Li Z Does green finance inspire sustainable development? Evidence from a global perspective Econ Anal Policy 2022 75 412 426 10.1016/j.eap.2022.06.002
Yang Q Du Q Razzaq A Shang Y How volatility in green financing, clean energy, and green economic practices derive sustainable performance through ESG indicators? A sectoral study of G7 countries Resour Policy 2022 75 102526 10.1016/j.resourpol.2021.102526
Yoshino N Rasoulinezhad E Taghizadeh-Hesary F Economic impacts of carbon tax in a general equilibrium framework: empirical study of Japan J Environ Assess Policy Manag. 2021 23 01n02 2250014 10.1142/S1464333222500144
Zhang D Mohsin M Taghizadeh-Hesary F Does green finance counteract the climate change mitigation: asymmetric effect of renewable energy investment and R&D Energy Econ 2022 113 106183 10.1016/j.eneco.2022.106183
Zhang Y Liu Z Baloch Z Combining effects of private participation and green finance for renewable energy: Growth of economy as mediating tool Renew Energy 2022 195 1028 1036 10.1016/j.renene.2022.05.086
Zhao J Taghizadeh-Hesary F Dong K Dong X How green growth affects carbon emissions in China: the role of green finance Econ Res Ekon Istraž 2022 10.1080/1331677X.2022.2095522
Zhou M Li X Influence of green finance and renewable energy resources over the sustainable development goal of clean energy in China Resourc Policy 2022 78 102816 10.1016/j.resourpol.2022.102816
| 0 | PMC9707273 | NO-CC CODE | 2022-12-01 23:20:22 | no | Econ Change Restruct. 2022 Nov 28;:1-18 | utf-8 | null | null | null | oa_other |
==== Front
Manag Rev Q
Management Review Quarterly
2198-1620
2198-1639
Springer International Publishing Cham
309
10.1007/s11301-022-00309-1
Article
Data monetization: insights from a technology-enabled literature review and research agenda
http://orcid.org/0000-0002-2001-3621
Ofulue Joan [email protected]
Benyoucef Morad
grid.28046.38 0000 0001 2182 2255 Graduate Studies Office, Faculty of Engineering, University of Ottawa, SITE 1024, 800 King Edward Ave, Ottawa, ON K1N 6N5 Canada
29 11 2022
145
1 6 2022
8 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Data is one of the most valuable resources in the world. The new data economy has led to several data use cases and one of them is monetization. Given how recent the concept of data monetization is, this study aims to investigate it, better understand its meaning, implications and issues in the academic literature and provide guidance for practitioners as well as direction for further research. A systematic literature review is conducted on English literature from 2013 to 2022. 54 articles were identified from where a topic and sub-topic categorization and a conceptual framework is developed. The paper makes contributions in three areas: (1) providing a holistic understanding of data monetization models through the extension of a framework from the literature, (2) categorizing of the key topics and trends in data monetization, and (3) using a systematic approach to identify managerial implications and a future research agenda.
Keywords
Data monetization
Big data monetization
Data commercialization
Data monetization model
Data monetization strategy
Data monetization infrastructure
Data business model
JEL Classification
M150
Y800
O390
==== Body
pmcIntroduction
The concept of data monetization is new in the academic literature. However, practitioner firms like Gartner (Moore 2015), EY (EY Global 2018), Deloitte (Deloitte AI Institute 2021), KPMG (Mohasseb 2015), and academic institutions such as MIT (Barbara and Jeanne 2017), (Wixom, Cashing in on your Data 2014), (Moore 2015) have published several articles on the concept. As companies continue to generate massive amounts of data and handle existing historical data, they are turning to consulting companies to help them understand how to maximize the value of their data. Data monetization is the commercialization of data and information assets. Data monetization occurs when organizations exchange data and information assets for financial return or something with equivalent value (Buff et al. 2014). According to Prakash (2014), data monetization refers to the organization's ability to create additional revenue from existing data sources (internal and external), to create useful information, insights, and observations. Fortune 500 organizations such as Amazon, Facebook, and Apple focus on data-driven business models to develop new products and services and improve the customer experience while generating additional revenue streams. This concept focuses on data as a product and as such is managed that way (Marcinkowski and Gawin 2020).
Data monetization goes beyond selling raw or processed data therefore the concept of “monetization” may be misleading. Beyond selling the data directly for cash, data monetization occurs when organizations use data to create value driven products, convert data and analytics into financial returns and other tangible benefits such as supplier funded advertising and discounts, or by avoiding costs that could come from operational inefficiencies.
Data monetization either aims to reduce operational costs by leveraging internal data and/or generate revenue through other models such as selling data and wrapping data around products and services.1
The data monetization global market is estimated to grow from US$2.1 billion in 2020 to US$15.5 billion in 2030 (compound annual growth rate of 22.1%) (Kanhaiya et al. 2022). This will be driven by the increasing magnitude of generated data, awareness of data monetization, emerging technology opportunities and trends (Moore 2015) such as Business Intelligence and Analytics (BI&A), cloud computing, blockchain, Internet of Things (IoT), social networks and post-COVID-19 pandemic business approaches and strategies (Mordor Intelligence 2022).
Organizations wanting to develop a successful data monetization strategy will require a good understanding of the different data monetization models, their implications, opportunities, and limitations. Given the recency of data monetization as a discipline, efforts have been dedicated towards producing academic research across different areas, however there is still work to be done. The first published Systematic Literature Review (SLR) was conducted by Liu and Chen in 2015 to raise awareness, within the academic community, on the potential of data monetization research. The authors contributed to understanding data monetization by providing use cases, guiding principles and a framework that combines Analytics 3.0 (advanced analytics) and BI&A 3.0 (mobile and sensor-based analytics) to better understand the subject. Analytics 3.0 is a stage where organizations realize measurable business impacts from the combination of traditional analytics, big data and powerful data gathering and analysis methods to a company’s offerings, embedding data smartness into the company’s products and services. This review was limited to a short period (i.e., from 2010 to 2015) and only considered articles that contain “data monetization” or “monetization” within their titles thereby excluding papers that discuss data business models that are not explicitly titled “monetization”. In 2016, Thomas and Leiponen conducted an SLR on data commercialization based on 51 articles. They chose the term commercialization rather than monetization to differentiate the trade in data through commercial transactions. They identified monetization challenges and models beyond the internal use case for monetization. The identified models included data suppliers, data managers, data custodians, application developers, service providers and data aggregators. While the authors considered broader terms beyond “data monetization”, the review was limited to a short review period (i.e., from 2010 to 2013) and narrowly focused on monetization models through the lens of the players involved and therefore did not provide insights beyond a single construct (players). Hanafizadeh and Nik (2020), using an SLR, proposed a configuration model called “Data monetization configuration” (DaMoc) and tested it with a real application (i.e., Cardlytics). They identified the following global themes categorized into different layers: the monetization layer (theme consisting of trading model, goods, end consumer), the data refinement process layer (theme consisting of assets, data driven operation and value), the base layer (theme consisting of resources and supplies) and the accessing and processing restrictions layer (theme consisting of privacy, legal and ethical issues of data processing). Similar to Liu and Chen (2015), the review only considered articles that contained “data monetization” within their title thereby significantly limiting their scope. In addition, despite the longer publication period (i.e., from 2006 to 2018), the authors had a small paper sample of only 18 papers. Faroukhi et al. (2020a, b) used a longer publication period (i.e., from 2000 to 2020) and a larger paper sample (i.e., 97 papers) to conduct an SLR using a Big Data Value Chain (BDVC) framework. The BDVC framework describes steps for administering an organization’s data related processes. The steps range from data generation to data exposition. They further proposed two monetization models: a reduced data monetization model and a full data monetization model. The reduced data monetization model aims to monetize data only through the storage and visualization phases. The full data monetization model is more generic, expensive and supports monetization along the whole BDVC. Unfortunately, this study does not show how BDVC and data monetization are integrated, and it does not provide a reconciliation to existing literature.
While the abovementioned SLRs have contributed to a domain-based review on data monetization, there continues to be a notable lack of comprehensive work in the academic literature that seeks to systematically map the literature in a way that consolidates and structures knowledge in the domain for both the academic and practitioner community, highlights areas in need of further research, and guides the planning of new research while supporting claims of relevance and novelty.
This paper conducts a technology empowered SLR. The data is electronically sourced from online scientific databases, which is then analyzed through a combination of VOSviewer2 and manual effort. The SLR provides a thorough evaluation of the literature, going beyond monetization models that focus on trading data or data monetization in the context of data value chains. Indeed, the study finds, identifies, selects, analyzes, evaluates, and systematically synthesizes work that focuses on different models of data monetization. The work described in this paper is not limited by industry and geography. It consolidates the existing literature, develops a content categorization and a corresponding conceptual framework, and provides a structure for exploring specific research areas in data monetization that cater to a variety of stakeholders such as academic researchers, business managers and decision makers. The paper further provides practitioners with insights into existing data monetization models, which may serve as a starting point for various data monetization initiatives.
The literature review seeks to answer the following research question: What are the main subjects, challenges, and opportunities of data monetization in the academic literature and what are the corresponding implications for practitioners and academic researchers?
To achieve this objective, the study describes the areas of data monetization, maps the literature on the subject, proposes a categorization and corresponding conceptual framework highlighting the literature which has been produced till date, proposes new research questions capable of increasing the quality and relevance of the academic literature and proposes managerial implications for organizations.
The paper is structured as follows. The conceptual background is presented in Sect. 2, Sect. 3 defines the research method, followed by a discussion of the findings, managerial implications and research agenda in Sect. 4. Section 5 presents the conclusion, limitations, and considerations for future work.
Conceptual background
Data
Data is a public good that is consumed by people but whose supply is not affected by people’s consumption (non-rivalrous) (Thomas and Leiponen 2016). Non-rivalrous means that multiple actors can exploit a single dataset, it is replicable and using it does not make its value disappear however once the data is revealed its value drops. Data value varies by the way one uses it, combines it and whether one can make it available at the right time (Parvinen et al. 2020). Most organizations now acknowledge data as a strategic asset and many practitioners have gone as far as calling data, not oil, the most valuable resource in the world thereby making data one of the most important assets for digital transformation.
As the world becomes more digital, the volume of data continues to increase, and the notion of big data is becoming a widespread phenomenon. Big data is a major enabler of data monetization and has quickly gained popularity among industries that own huge data assets. Gartner in its glossary defines big data as a “high volume”, “high velocity”, “high variety” information asset that requires a cost-effective, innovative form of information processing that enables enhanced insights, decision making and process automation.3 According to Faroukhi et al. (2020a, b) “high volume” indicates large amounts of generated data that cannot be processed through traditional processing and storage means. “High velocity” indicates the speed at which data can be generated. “High variety” refers to the characteristics of data that come in different formats including structured formats such as traditional relational database values, semi structured formats like XML, and unstructured formats which could include email and IoT data. While the benefits of big data is real, its characteristics are often constraining. Big data often has data quality issues, which make its use challenging. It must be well arranged and free from gaps and erroneous records to enable efficient data-based decisions (Marcinkowski and Gawin 2020). In addition, the violation of data privacy leads to several ethical and legal issues (Faroukhi et al. 2020a, b).
The analysis of this data is so rampant and has become one of the four technology trends of the decade together with mobile, cloud, and social business (Saynajoki et al. 2017). Organizations are sitting on large amounts of data (historical and transactional) with no clue on how to maximize its value. The new data economy has led to several data use cases and one of them is monetization. Since the term Data and Big Data monetization are used interchangeably throughout the academic literature, this paper will use data to mean both big data and data.
Data monetization
Data monetization is an evolution from Business Intelligence (BI), which has become a strategic tool for many organizations in the last two decades. BI is the process, strategy and technology involved in using business information by organizations for data driven analysis to extract usable and shareable information. During the 2000s, data was used for descriptive analytics, which allowed organizations to extract, process and aggregate data for internal use within the organization (Zakaria et al. 2020). With the proliferation of digital technologies such as big data analytics, IoT, Cloud, Machine Learning, and Artificial Intelligence (AI), data monetization has become a formal discipline and models beyond traditional BI have been developed. This is demonstrated when Liu and Chen (2015) developed an analytics framework for data monetization by adapting two evolution models: the evolution of Business Intelligence and Analytics (BI&A) and Analytics 3.0 framework.
With this, it can be said that organizations have been performing internal data monetization by improving internal operations through traditional analytics capabilities and data mining approaches to improve their bottom line returns (Alfaro et al. 2019), meet business needs and solve internal problems (Najjar 2013). In the literature, Alfaro et al. (2019) investigated the monetization journey of BBVA (a global financial group) which improved pre-existing internal monetization activities, and pursued new approaches by establishing a data science center of excellence and getting the center to collaborate with the business on data monetization projects. Marcinkowski and Gawin (2020) shared insights from a facility that evolved from “service based” to “data driven”, which led to a strong and loyal customer base as well as additional collaborative opportunities. Internal data monetization focuses on reducing operational cost and improving business operations (existing processes and products). It informs strategic business decisions and refines business processes by inputting data into the management process (Schroeder 2016). Internal monetization is usually the first stage for organizations as they deal with limited organizational and technological resources (Lange et al. 2021).
Data monetization as a concept has been heavily studied by MIT’s Center for Information Systems Research (MIT CISR). One of their very first case studies was Owens and Minor (OM). In the early 2000 OM, a distributor of healthcare supplies, created information offerings based on data from products and services distributed from hundreds of suppliers to several thousand hospitals. These offerings, called “Spend Analytics”, were sold to hospitals and supplier organizations that needed to make better decisions about procurement and product distribution/market penetration respectively. OM later evolved by creating a separate solution that focused on its information-based offerings (Wixom, Cashing in on your Data 2014). By 2014, MIT CISR started to produce more data monetization research in the form of research briefings. This was an early case of external data monetization.
Several papers in the literature have tackled external data monetization from different perspectives. In terms of holistic case studies, Najjar and Kettinger (2013) studied the data monetization journey between retailers, suppliers, and a supplier portal. The paper identified three pathways to monetization and followed the third pathway where the retailer built a technical data infrastructure (supplier platform) and leveraged the analytics capabilities of the suppliers. Parvinen et al. (2020) studied 24 companies to provide a wide-ranging view of data-monetization practices in large and middle-sized companies. They identified three models: selling data, selling analysis, and selling data-based services. They created a matrix of these models against customer types (current customers, actors in the current value chain, anyone), devised a path using this matrix and came up with steps on the path to monetization. De Reuver et al. (2015) studied the case of 3cixty (a data platform for mobile context-aware travel applications) which developed a multi-sided platform to serve multiple users (advertisers, app developers, government organization and end-users). The objective of a multi-sided platform is to facilitate the transactions between different user groups such as consumers and app developers. The platform provides apps and services for city visitors. The paper explored different revenue models from the end-user perspective. The authors discovered that the more willing a user is to share data, the less likely the user is willing to pay for the app. De La Vega et al. (2018) introduced data monetization in the context of IoT based on the study of two companies; Company A provides smart city services, Company B owns a smart building, they propose a data marketplace with a peer-to-peer architecture powered by blockchain to enforce trust and non-repudiation among peers.
Another traditional business model that has influenced data monetization is a two-sided market, defined as a trading platform dealing with two distinct user groups that provide each other with benefits. The trading platform involves a data broker, a data provider, and a data consumer. The data broker acts as an intermediary which connects two or more market participants via the platform and simplifies their interactions. Large companies operate their own data platforms to manage regular data interactions with third parties, while smaller companies tend to exchange via neutral platforms (Spiekermann 2019). The two-sided platform is a component of a data marketplace where firms and individuals can buy, sell or trade second or third-party data. Examples of data marketplaces include: Salesforce’s Data Studio, Oracle’s BlueKai, Adobe’s Audience Marketplace (Sinha 2019) and Snowflake Data Marketplace. The two-sided markets were studied by Agarwal et al. (2019), Bataineh et al. (2020a, b) and Saleh et al. (2021).
With nascent digital technologies such as Blockchain, data monetization continues to evolve. Al-Zahrani (2020) proposed a subscription-based data sharing model that not only leverages blockchain but also Data as a Service (DaaS), data centers in the cloud concept. In this model users subscribe to a data provider for a specific period and pay for the data access based on the selected subscription plan. Javaid et al.(2020), Abubaker et al. (2022), Madinen et al. (2022), Khezr et al. (2022a; b) leveraged IoT with blockchain technology to provide trustful trading through automatic review systems for monetizing IoT data using Ethereum smart contracts.
Research method
A systematic literature review (SLR) is a rigorous research methodology, not just to gather, organize and analyze existing research on the subject but to help researchers develop evidence-based guidance for research in their area of study. This SLR follows Kitchenham’s 2004 approach for conducting a systematic literature review, which consists of three stages: (1) identification and selection of studies; (2) data collection and extraction; and (3) data synthesis and interpretation.
Scopus, Web of Science Core and ABI/INFORM Global were selected as the databases of choice. The search strategy was developed by first identifying key terminologies and synonyms from the research topic, which were then translated into Boolean queries. These words are “data monetization model”.
The initial search was conducted in Scopus to assess the validity of the query at a high-level. From the results, a paper was selected based on how close the title was to the research topic. Additional terms were identified such as “information model”, “data business model”, “data commercialization”, etc. The paper in question was: Advancing data monetization and the creation of data-based business models (Parvinen, et al. 2020). This approach, called Pearl growing (also known as ‘Citation mining’ or ‘Snowballing’) ensures that all relevant literature has been identified. This approach was also leveraged to extract papers manually.
The query results were extracted into Mendeley4 to support export to Covidence.5 All papers from the three databases were exported into Covidence for de-duplication. The abstracts were initially screened in Covidence to exclude all papers that are either not relevant or do not meet the inclusion criteria by identifying papers as either “Yes”, “No” or “Maybe”. The results were exported back from Covidence to Mendeley for the papers to be extracted for a full text review.
In addition to the papers identified by the initial query, four papers were manually identified through “backward snowballing”.
The query used is broken into three blocks. The first block captures “data” related terms, and the second block captures “monetization” related terms. These two sets of words are separated by a proximity operator to capture articles where the terms appear a few words apart. The third and last block of words captures “model” related terms. The query is shown below:((data OR insight OR information OR "digital business" OR "business Intelligence") W/4.
(moneti* OR commerciali* OR "revenue generat*")) AND.
(model OR strategy OR approach* OR offering).
Exclusion criteria
Results prior to 2013 were excluded given the recency of the topic. The practice of data monetization, although common since 2000 (Wikipedia 2014) has only been published since 2013. This is evident from the initial results of our query where we searched the database to identify when papers on “data monetization” were first published.
Inclusion criteria
The systematic literature review prioritizes peer reviewed journal and conference papers. Selected papers should explicitly describe at least one model for data monetization and only work published in the English language was considered. The paper selection steps are represented in the PRISMA results flow diagram in Fig. 1Fig. 1 Systematic review selection and review procedure adapted from PRISMA (2009)
Content analysis
To answer to the research questions, both thematic and co-occurrence analyses were performed. Given that the literature on data monetization is nascent and the final number of articles selected for analysis was only 54, the thematic analysis was performed manually and supported by the co-occurrence analysis using a tool.
Firstly, the authors independently identified key themes from all 54 papers. Secondly, to aid the development of key themes, the authors developed a term co-occurrence map based on the titles and abstracts of all 54 papers (see Fig. 2) using VOS Viewer’s6 enrichment technique called co-word network analysis. Co-word analysis assumes that words that frequently appear together have a thematic relationship with one another. Thematic analysis is an analysis that requires researchers to systematically extract qualitative data (e.g., text) from a collection of documents (e.g., articles, interviews) for identifying, analyzing, and reporting on a theme. Thematic analysis empowers researchers with autonomy in dealing with the themes that manifest themselves from the research data (Naveen et al. 2021), (Marc et al. 2022).Fig. 2 A network visualization of terms based on keyword co-occurrence in 54 data monetization articles using full counting and the 75 most relevant terms. The size of the nodes indicates the frequency of the keyword. In this case, a minimum threshold of 4 is set as a frequency. The link between nodes indicates the similarity of keywords, i.e., the closer the node, the greater the similarity. Red (Cluster 1) = challenges, blue (Cluster 3) = business model, and yellow (Cluster 6) = marketplace, light blue (Cluster 4) and purple (Cluster 5) = revenue and players, green (Cluster 2) = platform
Each color in Fig. 2, below, represents a thematic cluster and each node in a network represents an entity (e.g., article, author, country, institution, keyword, journal), and in the case of Fig. 2, a keyword in the title and abstract of all 54 papers. The size of the node indicates the number of times that the keyword is used. The bigger the node, the greater the occurrence of the keyword. Each circle in Fig. 2 represents a term that appears at least four times in the titles and abstracts of all 54 papers. Setting the criterion that keywords are included when they have appeared in a minimum of four articles helps curate a pragmatic set of clusters for network visualization based on how prolific or prominent the keywords in the clusters are in the corpus (Marc et al. 2022). The link between the nodes represents the keywords that co-occur or occur together. The thickness of the link signals the number of times that the keywords co-occur or occur together (Satish et al. 2022).
The network map produced six clusters. The authors then reconciled the themes identified from the manual exercise conducted independently. The explanation for each cluster was developed and the themes were identified manually, wherein keywords in each cluster were organized to convey a coherent narrative that explains the essence and scope of the similarities within a cluster. As a result, the map was further refined and re-categorized and coded as can be seen in Appendix A: Themes mapped to clusters and paper count which shows the number of articles that discuss each concept. Where discrepancies occurred between the authors, there was elaborate discussions until a mutual agreement was reached. As a result, three final categories where identified: Data Monetization Strategy (DMS), Data Monetization Infrastructure (DMI) and Data Monetization Challenges (DMC).
DMS was extracted from several clusters. Clusters 3 and 6 (blue and yellow networks) were refined to form the data monetization models, while Cluster 5 (purple network) and Cluster 4 (light blue network) were for the players and the revenue models. DMI was identified based on Cluster 2 (green network) with terms such as Internet, platform, IoT, blockchain, and device. DMC was identified based on Cluster 1 (red network) with terms such as challenge, gap, security, and privacy. Using this as a foundation, the authors included the additional themes that were identified and agreed upon from the manual exercise.
Hence DMS includes operating model, players, and revenue model and DMI includes cloud, blockchain, sensors and IoT. For DMC related terms, challenges related to data monetization were extracted from all 54 papers. Using a word art, the top seven challenges were identified as security, privacy, pricing, contract design, data quality, beliefs, and data skills. The total number of papers that discuss the identified themes can be found in Appendix A: Themes mapped to clusters and paper count. The breakdown of papers by themes can be found in Appendix B: Categorization of papers by themes identified.
Findings and proposed research agenda
Basic characteristics of the literature
A total of 54 papers were collected for data extraction and synthesis. In Fig. 3, we illustrate the trend of publications with a significant spike in 2020 and 2021, accounting for about 50% of the total publications. In Appendix C: Overview of studies, year, journal and paper methodology, we illustrated the breakdown of papers by journal and methodology. About 35% of publications were made in an IEEE/data related journal or conference with about 20% leveraging the case study approach, 23% literature review, 14% mathematical approach and the other 40% some form of qualitative and exploratory approaches. Given that the query was last executed in Oct 2022 there is a risk of missing newly published papers that have not been captured.Fig. 3 Count of papers by year and article type
In the following subsections, and based on the network visualization map generated through VOS Viewer and a review of all 54 papers, we answer the research question: What are the main subjects, challenges, and opportunities of data monetization in the academic literature and what are the corresponding implications for practitioners and academic researchers? We do so by identifying categorization and subcategorization areas and proposing areas for further research at the categorization level. We also point to the managerial implications of our findings.
Data monetization strategy (DMS)
The data monetization framework presented in Fig. 4 was devised to visualize the main components of a data monetization strategy. The framework includes the models to the left, the players to the right, and the revenue models to the top. This framework builds upon existing classifications in the literature and enhances them by introducing the players organized by the value they add to the monetization ecosystem, as well as the overarching revenue models. The mapping of the revenue models to the monetization players is represented in Table 1. In addition, the studies that discuss the DMS category and subcategories are shown in Appendix B: Categorization of papers by themes identified.Fig. 4 Data Monetization Framework. “Adapted from Barbara and Jeanne (2017), Susan Moore (2015), Parvinen et al. (2020) and Zakaria et al. (2020)”
Table 1 Mapping of revenue models to monetization players. Adapted from Spiekermann (2019), Kemppainen et al. (2018), Thomas and Leiponen (2016)
Value generated Players Freemium Advertising Subscription Pay-per-use Transaction based Model Service based model Licensing
Generate Data Data Providers ✓ ✓ ✓ ✓ ✓ ✓ ✓
Combine Data Data Aggregators ✓ ✓ ✓ ✓ ✓ ✓ ✓
Improve Data Data Managers ✓ ✓ ✓ ✓ ✓ ✓ ✓
Define and Enforce Data Standards Data Regulators
Custodians of Data Data Bank ✓ ✓ ✓ ✓ ✓ ✓ ✓
Facilitate Data Transactions Data Brokers ✓ ✓ ✓ ✓ ✓ ✓ ✓
Data Facilitators ✓ ✓ ✓ ✓ ✓ ✓ ✓
Tool Providers ✓ ✓ ✓ ✓ ✓ ✓
Enrich Monetization Ecosystem Service Providers ✓ ✓ ✓ ✓ ✓ ✓
App Developers ✓ ✓ ✓ ✓ ✓ ✓
Consultant ✓ ✓ ✓ ✓ ✓
Consume Data Data Consumers ✓ ✓ ✓ ✓ ✓ ✓ ✓
In Sects. 4.2.1, 4.2.2 and 4.2.3 the authors describe the findings and in Sect. 4.2.4 and 4.2.5, the authors present managerial implications and a research agenda respectively.
Figure 4 is the first attempt to visualize the monetization models identified from the literature. The categorization by Wixom (2014) (data wrapping, bartering, and selling) and Parvinen et al. (2020) continue to be a basis in the data monetization literature. We contribute to the literature by enhancing the existing models with critical elements such as the revenue model and players based on the value generated within the ecosystem.
Operating model (OM)
The operating model is depicted to the center left of the framework in Fig. 4. The first dimension is the type of data monetization (internal, indirect, and external data monetization). Internal data monetization is when a company has data assets, extracts value from them, and does not wish to share those assets with other parties. Internal data monetization has existed long before the concept was introduced. Any company using data to improve its performance can be considered as involved in internal data monetization.
Internal data monetization seeks to reduce operational cost, improve business operations, and improve its reputation (Alfaro, et al. 2019). Marcinkowski and Gawin (2020), Alfaro et al. (2019), Najjar (2013), Schroeder (2016), Lange et al. (2021), Firouzi et al. (2020), Quach et al. (2022) all describe an internal data monetization model. Examples of internal data monetization would include efforts to help banks optimize the placement of bank branches, data from sensors in building to complement service driven operations leading to decreasing property utilization cost, developing business intelligence and analytical capabilities to meet business needs and solve internal problems.
External data monetization is when a company shares its data assets with other parties such as suppliers and customers. External monetization can take various forms. Susan Moore (2015) proposed direct and indirect data monetization types. In direct monetization, the trade in data is through commercial transactions that involve monetary rewards. The indirect method uses data, refines it, and produces information assets, services or products that can be sold. Indirect monetization can, for instance, help identify new customer needs and create new revenue opportunities. Direct monetization involves selling data directly and indirect monetization involves data wrapping and bartering.
Barbara and Jeanne (2017) defined three business models within external data monetization, namely data wrapping, bartering, and selling. Data wrapping involves wrapping core offerings with an analytics feature. Here you are making money by distinguishing offerings with features and experiences. An example is when commercial banks create a financial tool for customers that automatically categorizes their transactions into common budgeting categories (Alfaro, et al. 2019), (Firouzi, et al. 2020), (Quach et al. 2022). In a 2014 poll by MIT CISR, 73% of executives chose wrapping as the data monetization approach that offers the greatest future potential for their companies (Wixom, Cashing in on your Data 2014). The desired outcome of wrapping includes increased value, market share, product price cost effective services.
Data bartering involves providing data in exchange for non-monetary rewards such as reports, favorable terms, free services, benchmark metrics, or analytics software (Wixom, Cashing in on your Data 2014). This is popular amongst social media companies such as Facebook who provide free access to social media platforms in exchange for user data. The key concern with bartering is that organizations or individuals may not recognize the true worth of the information they are giving up. Therefore, data regulations need to be established to protect both parties.
Data selling is the most common form of external data monetization. Common methods involve retailers selling Point of Sale (POS) transaction data to consumer research firms like IQVIA, Kantar, Nielsen, etc. Given that selling raw data poses privacy concerns and questions contractual obligations, companies are developing alternate revenue streams by selling information from reports and analytics (Wixom, Cashing in on your Data 2014). Data selling can take various forms, from a marketplace model where data providers can offer their data (Grubenmann et al. 2017), (Faroukhi et al. 2020a, b), (Rao and Ng 2016), (Kemppainen, et al. 2018), (Spiekermann 2019), (Firouzi, et al. 2020), (Lange et al. 2021); to watching targeted ads for rewards (Yu et al. 2020), (Trzaskowski 2022); to subscription based data sharing models (Al-Zahrani 2020) or merchant models where a third party collects data from its owners, processes it and sells the information to consumers (Saleh, et al. 2021).
As part of data selling, Parvinen et al. (2020), Firouzi et al. (2020) and Calvin et al. (2021) identify three business models that align with the information offering consumption path identified by Buff et al. (2014). These are described below.
Selling data (data offering) This involves selling raw and prepared data directly.
Selling analysis (insights offering) This involves selling analysis and restricting access to the original data. Given that data doesn’t change hands, privacy and security concerns are mitigated. The less versatile use of the analysis given that buyers cannot combine it with other data sources brings to light questions around value. Criteria such as data quality and business context relevance also play a critical role in determining the value of insights (Rix et al. 2021a, b).
Selling data-based services (action offerings) This involves creating a new service that can provide customers with relevant signals on the business environment, help scale how data is delivered using multi-sided business models and can help customers act on insight. An example of a business model that provides customers with relevant signals on the business environment include Facebook’s sale of advertising space that enables publishers to target their specific user groups based on their user data (Matsakis 2018). Business models that help customers act on insights could include consulting, onsite support, process automation and process outsourcing. This is similar to services provided by major management consulting firms such as KPMG, Deloitte, PWC etc.
Faroukhi et al. (2020a, b) introduced data monetization from a Big Data Value Chain (BDVC) perspective. The BDVC describes steps that aim to administer organizations’ data related processes. These steps are: data, acquisition, preprocessing, storage, analysis, visualization and exposition. Given the importance of the data lifecycle, the steps are mapped against the data consumption path (selling data, selling analysis, and selling data-based services). Data monetization can occur at any step. Selling raw data occurs in the initial value chain steps (data generation to data storage). As data becomes more refined, the rest of the steps facilitate selling analysis.
Players (P)
The data monetization ecosystem consists of a variety of players as identified to the right of Fig. 4. Players can be individuals, organizations, or systems. To make an effective external data monetization strategy, a data consuming party and a data providing party must exist. A data monetization player can take on a variety of roles which adds complexity to the ecosystem. The roles could range from public cloud platform providers to third party brokers, facilitators, and consultants (Najjar 2013), (Rix et al. 2021a, b). Thomas and Leiponen (2016) describe data monetization players as models of value creation within data ecosystems. They include data suppliers, data managers, data custodians, application developers, service providers and data aggregators. Faroukhi et al. (2020a, b) proposed to build data business models around four pillars, namely data users, data providers, data facilitators and data aggregators. Rix et al. (2021a, b) proposed 10 roles for the data ecosystem. As per the literature, the following are data monetization players and the value they provide to the monetization ecosystem.
Data providers These are the originators/owners of the data (Grubenmann et al. 2017) (Spiekermann 2019) (Thomas and Leiponen 2016) (Rix et al. 2021a, b). They can also be called data suppliers as they generate the data leveraged in the ecosystem. It could be smart phone users or individuals having some personal data to sell (Bataineh et al. 2020a, b).This could include user generated, IoT sensor generated or company data. Data providers may play multiple roles such as all the other roles defined below. For a multi-sided market, stability of the data providers in the ecosystem and the nature of the domains and parties involved in collecting and sharing data determine the success of this market (Bataineh et al. 2020a, b), (Parvinen et al. 2020).
Data aggregators They combine the data and provide users with aggregated services and data, thereby enabling them to produce a targeted advertising business model. They also perform data crawling and visualization. The most common data aggregators are price comparison services such as the travel search engine Kayak (Faroukhi et al. 2020a, b), (Gopalkrishnan 2013), (Hartmann 2016). There are also aggregators such as Meta, Google, and Twitter where the user does not pay to use the services, but the aggregator monetizes the service through advertising (Robinson 2017).
Data managers They improve the data. These are organizations that catalogue, clean, and parse data that is not in an easily usable format or improve the value of the data with additional context. They add value to data by improving the efficiency, interpretability, and the overall functionality of the data (Thomas and Leiponen 2016), (Klaus 2011), (Saynajoki et al. 2017).
Data regulators They define and help enforce data standards. These organizations recommend and ensure the security, privacy, and ethical use of data. They define standard data technologies and standardization of data transfer (Faroukhi et al. 2021).
Data banks They are the custodians of the data. Data banks are organizations that enable the reuse and resale of data by providing a ‘trusted’ infrastructure (Thomas and Leiponen 2016), (Schwab 2011). According to Saynajoki et al. (2017) data banks orchestrate external data distribution between companies. They also reassure end users and data consumers through provenance validation and certification and auditing services to ensure that the integrity and the quality of data is maintained from sourcing to use (Perrin 2013).
Data brokers They collect and bundle data for prospective buyers. The broker is a platform equipped with the needed infrastructure to store and share data. They provide services that enable providers and consumers to perform data selling and buying transactions (Bataineh et al. 2020a, b), (Lin et al. 2020), (Schroeder 2016). They can be referred to as orchestrators (Rix et al. 2021a, b). Examples of such platforms are Snowflake and the Azure data marketplace. Lin et al. (2020) proposed that adopting brokers has three advantages. First, brokers can continue the trading process when both sides are offline. Second, brokers facilitate a refund thereby protecting the rights of participants. Third, parties do not need to reveal sensitive information because of a decentralized architecture. Brokers resolve privacy issues as they are the bridges that link data providers and consumers.
Data facilitators They are the typical third parties with the required capabilities to share data with data consumers. Facilitators do not own the data but provide services such as data cleaning, data analytics and consulting services (Najjar 2013). Data facilitators could correspond to a technical platform based on tools for data collection, integration, processing, storage, analysis, and visualization (Faroukhi et al. 2020a, b). They provide the physical architecture and the provision of outsourced analytics services.
Tool Providers Schroeder (2016) and Calvin et al. (2021) identified this player category. Hardware and software infrastructures are a significant facilitator of data monetization. From IOT to cloud to on-premises tools, the producers of these tools have a significant contribution to make as they enable all players in the monetization ecosystem. Examples include but are not limited to Microsoft, AWS, and Google who provide both software and hardware solutions. Data brokers, facilitators and tool providers facilitate transactions within the data monetization ecosystem.
Service providers Service providers develop new services for data, distinct to the resale, analysis or repackaging of data or the development of specific applications (Perrin 2013), (Saynajoki et al. 2017).
Application developers They are organizations and software entrepreneurs that design, build and sell applications that enable data monetization (Hammell 2012), (de Reuver et al. 2015). They design and build tools to analyze data (Saynajoki et al. 2017).
Consultants They demonstrate the value of data monetization to data providers and support them in developing strategies (Rix et al. 2021a, b). They can also provide data sourcing and consultation services to help buyers find the right data according to their use case (Luch Kelly 2022). The complexity of the data monetization ecosystem brings about new questions such as what are the use cases for data monetization? What is the best and most scalable architecture that supports a chosen strategy? How can the organization’s structure be designed to successfully deploy a chosen strategy? What are the privacy and ethical considerations to be made? How can existing processes be optimized based on findings from the data? etc. Answering these questions and guiding organizations through their data monetization strategy and execution has brought about a myriad of services offered by consulting firms such as IBM, KPMG, Deloitte, etc. Service providers, application developers and consultants enrich the data monetization ecosystem with their products and services.
Data consumers These are the parties that need and consume the data. They are individuals, businesses or systems that use collected data from various sources such as product usage, behaviors, preferences, Internet activities, IoT, etc. (Faroukhi et al. 2020a, b) and are willing to buy real-time data streams (Lin et al. 2020). Data consumer requirements vary in terms of the type, quality, and amount of data based on their scope and the applications they need (Bataineh et al. 2020a, b). Al-Zahrani (2020) refers to these players as data subscribers.
Revenue model (RM)
The data monetization ecosystem consists of several revenue models as identified at the top of Fig. 4. The identified operating models leverage one or more revenue models, which serve to provide compensation to the players within the ecosystem. A revenue model is seen as one fee or a combination of fees for different players (Kemppainen et al. 2018). The revenue model determines how players will be charged/rewarded for the value they receive/provide in the monetization ecosystem. Data monetization models can use a combination of revenue models to achieve the desired objectives. Spiekermann (2019) using eight revenue models and Kemppainen et al. (2018) using 14 revenue models created a revenue and price model taxonomy which serves as a conceptual starting point. For any revenue model, users’ willingness to pay for and share personal data is critical to success. De Reuver et al. (2015) discovered that the more a user is willing to share data, the less likely they are to pay for an application. As per our literature review, the following consolidated data monetization revenue models have been identified.
Free of charge A strategy to attract users and build a community. This is sometimes referred to as a freemium model where businesses give away basic data to encourage further engagement and charge a premium for access to more detailed data (Thomas and Leiponen 2016), (de Reuver et al. 2015). Popular music and movie streaming platforms such as YouTube, Spotify, Alexa, and Hulu music allow users access to a broad collection of music selection and movies, attracting a large user base and restricting the services offered to freemium members.
Advertising Revenue is mainly generated from advertisers. The competitive advantage for models relying on advertising as the main source of revenue lies in platforms, enabling better ways to gather and evaluate information (Tucker 2014). Kemppainen et al. (2018) propose that when adopting a human centered approach to personal data management, no advertising policy serves as the foundation of a revenue model. The no advertising model reflects the changing attitudes towards personal data usage, individual rights to privacy and companies’ need for finding alternative revenue models.
Subscription (membership) Several subscription-based models exist in the literature. Subscriptions can either be free of charge or fee based to be renewed periodically. Organizations utilize package levels where basic level data (raw data) is the most basic level while more refined, aggregated data is the top level (Najjar 2013), (Spiekermann 2019). In an advertising and subscription-based revenue model, the key drivers of revenue are the number of users and their willingness to pay (Kemppainen, et al. 2018).
Pay-per-use A price is charged per unit of data consumed with this unit needing to be defined. This option is popular for Application Programming Interface (API) access.
Transaction based model Consists of a transaction fee that is time or volume based. The platform operator facilitates data transactions between the stakeholders (Kemppainen et al. 2018).
Service based model Consists of a service fee, a connection fee, and a membership fee. The platform operator generates revenue by offering value-adding services on the platform or charging for the usage of the platform (Kemppainen et al. 2018).
Licensing Data marketplaces often provide standardized licensing models as well as regulations regarding data access and usage (Spiekermann 2019).
Managerial implications of DMS
Today, due to access to a wealth of information, almost every business can aim to be a data business. For an organization to effectively develop a data monetization strategy, there are key considerations to be made. The first step is to access the current ecosystem to identify opportunities, gaps and risks. Organizations need to understand where they are at in the monetization journey and where they want to end up. This evaluation requires careful consideration of the data asset inventory, characteristics of the data that are central to realizing benefits, and metrics for assessing the value of data and return on investment (Quach et al. 2022). In addition, organizations need to determine the value of their data as not all data is of equal value, what insights it can produce, who would be interested (entities internal or external to the organization), how to deliver the information in the most useful format, how much can be paid (revenue model), when to deliver this data as some data may be required in real-time, and finally how to process the data to add value. These questions need to be answered through both internal assessment and competitive market research.
Furthermore, the structure of the organization as well as its analytical and technical capabilities will determine the most successful pathways to data monetization.
The evaluation of the organization’s structure involves answering questions such as: How is the personnel organized to successfully deliver on data mandates? What is the organization’s attitude towards innovation and disruption? Are there dedicated resources for data monetization? Analytical capabilities involve evaluating data skillsets and identifying skill gaps, while technical capabilities involve evaluating the technical data infrastructure (digital platform). Organizations can decide to build inhouse platforms which can be expensive or leverage data marketplaces which provide productive and transparent means for data monetization. Data marketplaces provide a platform to sell datasets, data services or APIs. Data marketplaces enable data monetization by providing an access to a network of data buyers, avoiding costly data integration operations, and enabling small companies to grow data monetization capabilities. A data marketplace offers three benefits. Firstly, it empowers individuals and organizations to monetize rich data that is automatically generated and has become rampant due to the advent of IoT. Secondly, it allows non-technical users such as business managers to easily navigate the complex world of data as these marketplaces are designed like regular everyday websites. Thirdly, it can thrive as a result of big data and the network effect of a two-sided model that brings data producers and consumers together.
Research agenda on DMS
The following three DMS areas require further research. (1) To understand the factors to be considered in an effective data monetization strategy. Such factors could range from establishing a data monetization center of excellence as signified by Alfaro et al. (2019) to developing a data monetization strategy that is part of the organization’s broader strategy. (2) With the myriad of players, there is a need for the academic community to further investigate the interdependencies between multiple roles that players can take on, the value co-creation process, as well as how the overall data monetization ecosystem is governed. (3) There is a need to understand data monetization revenue models based on business models and players within the data monetization ecosystem. Although Kemppainen et al. (2018) studied revenue models at a conceptual level by looking at business models that are suitable for other multi-sided markets, there is insufficient literature on revenue models for data monetization.
Data monetization infrastructure (DMI)
As per the literature, we identified the Cloud, Blockchain, Sensors and IoT as data monetization infrastructures. In Sects. 4.3.1, 4.3.2 and 4.3.3 we describe the findings and in Sect. 4.3.4 and 4.3.5 we present the managerial implications and the research agenda respectively. In addition, the studies that discuss the DMI category and subcategories are presented in Appendix B: Categorization of papers by themes identified.
Cloud (C)
Cloud computing delivers on-demand computing on a pay-as-you-go model via the Internet. It enables organizations to switch from a CAPEX (capital expenditures) model to an OPEX (operating expenses) model for Information Technology resources. Data monetization relies on distributed architectures such as cloud computing and a trustless (i.e., involved participants do not need to know or trust each other) data trading infrastructure. Cloud computing remains a suitable solution to provide a secure, comprehensive, robust, scalable, and elastic ecosystem to host data monetization. It also provides an efficient model for data monetization as a service (Faroukhiet al. 2020a, b). In the data economy, most data and IoT services reside in the cloud. Massive amounts of data is being generated with the growth of IoT and the value of the data needs to be extracted by a supportable ecosystem such as IoT-Cloud that solves the problem of network resource occupation, high latency, and additional network load by distributing the execution of the computing task in a balanced manner to maximize the benefits of the system (Yu et al. 2020a, b). Cloud computing enables the provision of complimentary data for AI-driven services. Complimentary data is data formed by integrating multiple data types from multiple sources (Saleh et al. 2021). Cloud computing enables data platforms that implement the most valuable data monetization business models. The platforms are not only for selling data but also for delivering various data products and services (Lange et al. 2021).
Cloud computing can be deployed either as Infrastructure as a service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS). It supports various data models such as Analytics as a Service (AaaS), Data as a Service (DaaS), etc.
Analytics as a Service (AaaS) provides a data analytics platform service in the cloud. With huge volumes of data being available, people want answers, not more data. Moreover, the cost of the in-house infrastructure that supports data analytics continues to rise. AaaS, sometimes referred to as ‘agile analytics’, is defined as generating insights from data wherever this data may be located and turning a general-purpose analytical platform into a shared utility (Demirkan and Delen 2013). AaaS is a multifaceted concept that can be offered as SaaS when presented as a reporting application for business end-users. It can also be offered as PaaS that provides data scientists with a data analysis suite for their development. Finally, it can be offered as IaaS that provides virtualized resources to host vast amounts of data (Naous et al. 2017).
Data as a Service (DaaS) is a data management framework provided through services in the cloud to bring data storage, integration, processing, analysis services, security, availability, elasticity, and quality directly to the consumer. DaaS provides data centers on the cloud. It enhances data accessibility through different channels, allows the cleansing and enriching of data to occur in a centralized place and eliminates geographical and scalability limitations. This data is offered to different systems, applications, or users with elastic access to data, scalability, high availability, and system performance by demand (based on service level agreements—SLAs) regardless of geographical or organizational separation of the network (Rajesh 2012).
Blockchain (BC)
Blockchain is a distributed and decentralized ledger with the main purpose of removing third parties. It is a series of data blocks, produced and joined chronologically. It consists of a consensus method, distributed ledger, smart contracts, peer-to-peer network and blocklist containing a cryptographic hash that guarantee reliable transactions by executing a decentralized consensus protocol (Al-Zahrani 2020). The technology uses digital networks in which different types of users can interact and share data (Xie 2020). For data monetization that relies on decentralized peer-to-peer architecture and IoT, this monetization technique is highly effective given there is no need for a third party and high interoperability exists between fog nodes (Khezr et al. (2022a; b). Fog computing is also referred to as edge computing. In a fog computing architecture, companies can make data available to other companies in a peer-to-peer fashion, without needing a cloud intermediary, thereby maximizing the locality of the processing, and avoiding bottlenecks. In this architecture, data processing, filtering and stream-based event generation is done in a fog node. Blockchain allows relationships, commercial agreements, data delivery, access control and access logs to be performed directly between data producers and consumers without the need for mutual trust or a central entity (De La Vega, et al. 2018) (Kolade 2022). Data security is established, and the data users have some confidence in the quality of the data (Javaid, et al. 2020) because poor data quality does not only have financial impacts, but it also has a negative impact on the productivity and the businesses reputation. Blockchain provides a secure, transparent, anonymous, cost effective and decentralized solution for IoT data (Javaid, et al. 2020), (Khezr et al. (2022a; b). It reduces the risk of privacy incidence and avoids disputes with transactions (Xie 2020). In blockchain driven data monetization, ownership rights and identity authentication, the performance of the blockchain network, pricing, security, privacy, and transparency (Al-Zahrani 2020) all contribute to the effectiveness of this monetization infrastructure. Blockchain can help address privacy concerns by offloading the computation over sensitive data to an external network where it may be broken into different nodes and apply cryptographic techniques (Shrobe et al. 2018).
The design features of blockchain such as immutability, transparency, and traceability are being applied to several fields such as medicine, economics, IoT, etc.
In 2022, there has been an uptick in the literature on blockchain technology for data monetization as researchers continue to find solutions to resolve the dominance of storage and delivery networks by cloud providers. This has mainly been due to the successful utilization of technologies such as Non-Fungible Tokens (NFT). NFT is a digital asset that offers ownership guarantees for every asset added on the blockchain network. Blockchain solutions promise not only to provide users with the ability to control their data but also alternatives to challenges found in centralized frameworks (e.g., security and availability) (Madine et al. 2022), (Khezr et al. (2022a; b).
Sensors (S) and IoT
The proliferation of sensors and IoT based devices has led to Analytics 3.0, allowing organizations to make data driven decisions and unlock value through data monetization (Faroukhi et al. 2020a, b). The goal of IoT is to increase the connectedness of people and things. Sensors drive the IoT ecosystem as they detect and measure changes in position, temperature, light, etc. Sensors turn objects into data-generating mediums that often interact with their environment. Infrastructure is required to support data collection, transmission, processing, analysis, reporting and advanced querying. The use of sensors is common in industries such as Energy and Mining, Power and Utilities, Healthcare, Transportation and Vehicles, Industrial Internet, Hospitality, Technology, Financial Services, and retail. Lengyel et al. (2015) proposed a Sensor Hub framework set up as Platform as a Service (PaaS) that serves as an enabler for data monetization. The solution enables collecting sensor data, transmitting, processing, analyzing, and supporting the utilization of data. For smart buildings, for instance, knowledge gleaned is used in optimizing cleaning and waste management processes, preserving heating, cooling, and lighting energy (Saynajoki et al. 2017). IoT is typically enabled by distributed and decentralized architectures such as cloud computing and blockchain, which can offer a secure and dependable way for monetizing IOT data.
Managerial implications of DMI
Given the variety of data sources for data monetization, organizations need to have the right technical infrastructure to retrieve, store, share and track data. Therefore, infrastructure is the technological facilitator for data monetization. There are several infrastructure configurations that support data monetization, ranging from the most fundamental leveraging of a web plugin that controls the access of ad platforms to a user’s browser profile (Parra-Arnau 2017) to data management platforms, to cloud network environments via interfaces and communication protocols (Faroukhi et al. 2020a, b), to data trading platforms that build a secure, reliable and scalable data sharing infrastructure (Lin et al. 2020), (Madine et al. 2022), (Abubaker et al. 2022). In a data marketplace, platform architecture can be based on a centralized or decentralized approach. In a centralized approach data products are offered by different providers via a central location which could be a cloud infrastructure. This central location contains semantified (restructured and optimized to capture contextual relationships) and reconciled data that application developers can access via an application programming interface (API) (de Reuver et al. 2015). In a decentralized approach, the data products remain with the data provider and examples of such framework include blockchain. A data monetization strategy must take infrastructure into consideration since a technical data infrastructure can be developed in-house, outsourced, or delivered as a service (Najjar 2013). Organizations need to consider the following questions: what data is required and how will it be acquired? In what way will the data be processed? In what way will the data be distributed? (Marcinkowski and Gawin 2020).
Infrastructure considerations must include the following. (1) Data-as-a Service (DaaS) for providing raw and anonymized data. Such direct data monetization strategy is indicated when the organization lacks sufficient infrastructure and analytics capabilities. (2) Insights-as-a-Service (IaaS) for when the organization has the capability to aggregate both internal and external data to produce analytical insights and visualization. (3) Analytics-as-a-Service (AaaS) for when the organization not only provides analytical insights but empowers the data consumers with BI tools requiring zero setup and maintenance. Notice that this is similar to cloud models such as IaaS, PaaS and SaaS and therefore has a huge infrastructure burden on the providing organization. (4) An indirect strategy such as data-driven business models that leverage existing data to improve productivity and increase efficiency (Trianz 2022).
With the many infrastructure opportunities come challenges with security, legal and privacy issues, as well as the need for suitable standards. Organizations must avoid the tendency of using an existing infrastructure to enable data monetization as existing infrastructures may be unable to fulfil storage, bandwidth, processing and security requirements. Organizations must plan for a dedicated infrastructure that is secure, scalable, accessible, and well governed (Trianz 2022).
Today, data marketplaces are platforms that allow organizations to share their data with internal and external partners as well as the public. Studies show that organizations that leverage next-generation data marketplaces will gain a competitive digital edge because data marketplaces are the best demand generation platforms and the easiest route to data monetization. A data marketplace can be personal (because it allows consumers to get paid for sharing their data), B2B and IoT based, with B2B marketplaces understandably making up the majority.
Data marketplaces can offer large volumes of actionable data and APIs without having to complete complex transformations. They can be offered both as centralized and decentralized platforms (Luch Kelly 2022), although there has been an increased interest in decentralized platforms due to their promise to address security and privacy challenges.
Research agenda on DMI
The following three DMI areas require further research. (1) Academic research still needs to develop seamless, elegant architectures that can address models for data monetization. Consideration should be given to architectures that support plug-and-play of the components based on the specific monetization model under consideration. Also, additional research needs to be dedicated to addressing how the cloud as a data monetization infrastructure can help resolve challenges such as data quality, security, and privacy. (2) Research on the application of blockchain to data monetization remains highly theoretical and its potential remains untapped due to integrating challenges with existing technologies (Bechtsis et al. 2021). Academic research on blockchain for data monetization infrastructure must progress to the applicability stage. (3) The integrated application of both centralized (i.e., cloud computing) and decentralized (i.e., blockchain) technologies needs to be further explored.
Data monetization challenges (DMC)
The top seven challenges across all 54 papers are discussed below. Privacy/trust/security and contract design/pricing are the most re-occurring themes in the data monetization literature. 24 papers identified privacy as a major challenge to data monetization given that external data monetization involves distributing raw data or data insights.In Sects 4.4.1–4.4.6 we describe the findings and in Sects. 4.4.7 and 4.4.8 we present the managerial implications and research agenda respectively. In addition, the studies that discuss the DMC category and subcategories are defined in Appendix B: Categorization of papers by themes identified.
Security (S) and privacy (PV)
Data monetization cannot be discussed without security and privacy. A major obstacle to data sharing is a lack of trust and security. Data security refers to the process of keeping data confidential and protecting it from theft, errors, and accidental destruction (Parvinen et al. 2020). Earlier research on privacy suggests that people make trade-offs between utility, price, and privacy (de Reuver et al. 2015). Even though consumers value their privacy, they tend to provide their information for a monetary value or a service (e.g., users of online services such as Google and Facebook) (Sánchez 2022). There is an increasing regulatory and security concern into the behavior of organizations that sell personal data (Thomas and Leiponen 2016). Security and privacy issues prevent data owners from sharing data amongst themselves despite the profitability from data sharing.
Users’ perception of privacy infringement will continue to pose a risk to the free flow of data between data monetization players. Empirical studies reveal that there has been a dichotomy in human behavior that continues to baffle privacy experts and has been a major hurdle in the development of models that put a price on privacy. Parra-Arnau (2017) attempted to resolve privacy in a web tracking scenario by creating a privacy model that allows for the optimal trade-off between economic reward and privacy. The user’s privacy is ensured by a means of collaborative masking. Rao and Ng (2016) introduced the idea of obfuscation of user information to protect user privacy. Individual personal identifiable information (PII) is stripped off or noise is introduced to the data before the data is sold.
Effective contract designs can help alleviate security and privacy concerns through the establishment of appropriate assurance practices (Najjar 2013). Regulatory complexity and the absence of a legal framework may lead to considerable legal uncertainty with regards to trading data. The current regulatory environment does not have a cohesive and comprehensive set of laws to support a data monetization ecosystem. The ones in existence are siloed given they are created in an ad hoc manner for different institutional purposes (Spiekermann 2019). For example, in the US, there is no single regulation to protect personal privacy. There are a set of laws and regulations for sectors of activity or regions such as California Consumer Privacy Act (CCPA) and Health Insurance Portability and Accountability Act (HIPAA). However, in the EU, the General Data Protection Regulation (GDPR) exists as a single body of rules protecting privacy and personal data (Perrin 2013).
Pricing (PC)
Putting a price tag on data is not an easy task. The reason is that in order to accommodate diverse demands, data sellers devise different plans and pricing schemes for their buyers because buyers can obtain varying utility from the same data. This could be because data users have different uses of a particular data, different skillsets and varying complementary knowledge (Sinha 2019) (Rix 2021a; b). The players involved must mutually agree on the valuation of the traded data. The following characteristics of data by Agarwal et al. (2019) further make this a unique problem: data can be replicated at zero marginal cost, its value to a firm is combinatorial (i.e., the value of a particular dataset to a firm may depend on other datasets available), and the authenticity and usefulness of data is difficult to verify a priori without first applying it.
Rao and Ng (2016) proposed an information market for Internet users to enable the exchange of data. Their model gives users an idea about the value of their information using the concept of Shannon’s information theory, which is a measure of uncertainty of information. Using this, one can estimate the value and price of the information type in the information market. While Shannon’s information theory helps understand the amount of information that has been divulged, there is a need to understand the demand in the information market from the buyers interested in the information. So, the buyers state the amount they would be willing to pay for an information category and the average is used to determine the demand in the information market. Another technique identified by Chao Li (2013) includes linking the price of the data with the amount of noise added to the data by a third party called a “market marker”. In this scenario, the market maker can be prone to act maliciously since they have the unperturbed data. Al-Zahrani (2020) proposes a subscription-based data-sharing model where the users subscribe to a data provider for a specific period and pay for the data access based on the selected subscription plan. Thomas and Leiponen (2016) argued that data packaging and pricing models must be considered to identify what data can be made available, in what mode and at what prices while taking into consideration associated costs.
Recently, there has been more focus on pricing data, but there is still work to be done. Calvin et al. (2021) using a topology formation derived three pricing models for manufacturing from 11 features, namely price determination, price discovery, measurement unit, payment flow, timing of price determination, bundle component, bundling type, degree of integration, differentiation, price dynamics, and value creation. While this topology can be applied beyond manufacturing, academic research needs to consider quantitative and practical applications.
Stein Hannah et al. (2021) proposed a framework that provides four approaches: criteria-based for internal qualitative valuation, reporting-based for external qualitative valuation, cost-based for internal quantitative valuation and transaction-based for external quantitative valuation. This was tested using a case study in the manufacturing context, yet it needs to be tested in a broader context using multiple industries.
Monteiro et al. (2021) identified the need to focus on the value dimension of the Vs of big data. They acknowledge that academic research on value is lacking compared to the three classical dimensions (Volume, Velocity and Variety) and existing studies do not agree on the right way to measure and define this value.
Contract design (CD)
Data is non-rivalrous and only partially exclusive. Non-rivalry means that the same data can be used by many and partially exclusive implies that data is only exclusive within a specified type of use. These characteristics of data emphasize how critical it is to have clear contractual agreements (Thomas and Leiponen 2016). A contract is a legal agreement that states how parties must interact and fulfil their obligations. Contracts involve NDAs (non-disclosure agreements), data sharing and purchase contracts.
Given that data monetization involves strategy designs that involve multiple players and revenue structures, designing an optimal and fair contract agreeable by all parties is critical. Designing contracts helps address IP (intellectual property), privacy, and security issues by ensuring data sold or shared is used for the intended and agreed upon purpose (Najjar 2013). Sinha et al. (2019) propose a contract-theoretical framework to accommodate heterogeneous honest buyers as well as adversarial types. The framework proposes that the seller add noise to data query answers, charge more for lower noise, and thwart rational adversaries by levying fines.
Further academic research needs to consider answering questions that tackle how contracts should be designed to cater for IP protection, pricing concerns, regulatory complexity, data reuse/licensing and data quality.
Data quality (DQ)
The quality of data plays a major role in data monetization. Data quality addresses issues such as accuracy, completeness, consistency, interpretability, and reliability (Thomas and Leiponen 2016). Depending on the quality of data, organizations can choose to only be involved in specific models for data monetization. For low quality data and organizations not willing or able to process data, data monetization may simply be selling raw datasets. Mature organizations with the right infrastructure can sell more than just raw data. They can sell insights, data-based products, and other refined data outcomes. According to Faroukhi et al. (2020a, b) data users often scramble with low quality data, diverse data sources, data management, regulated strategies, and the violation of data privacy. They propose a data management platform to ensure end-to-end integrity of all the processes within an organization to be able to exploit valuable information. Javaid et al. (2020) propose a review system based on blockchain technology that holds the review of users who have used IoT data so that other users can trust the data they are using. The system provides confidence to users that the quality of data is satisfactory. IoT data is heterogenous in nature and therefore create compatibility issues on different platforms (Al-Zahrani 2020). Poor quality data cost business an average of $15 million of losses per year (Moore 2018). Poor quality data has a negative impact on customer trust, product reliability and ultimately business reputation (Al-Zahrani 2020).
Beliefs (B)
Perceptions are difficult to change given they are inherent deep-rooted beliefs of individuals and organizations, hence the need to nurture trust between the parties involved. The lack of trust and security can cause data providers to fear that competitors could benefit from disclosures of in-house data (Spiekermann 2019). It could also discourage data owners from participating as there is a tendency for organizations who have economic benefit to optimize surveillance and manipulation tactics (Trzaskowski 2022). No doubt that contracts can help alleviate concerns. However, having shared values is required to give players in the ecosystem a chance. For a collaborative mutually beneficial relationship, demonstrated trustworthiness, inter-organizational coordination to establish governance mechanisms and successful and repeatable interactions demonstrate reliability (Najjar 2013).
Data skills (DS) and other challenges
Having the right skillset can make or break a data monetization agenda. The right data skillset includes both the technical skills required to orchestrate data from data providers to data consumers. Organizations need to develop strategies to hire and retain the talent required to deliver an end-to-end data strategy (Alfaro et al. 2019).
Other DMCs identified throughout the literature include identifying a trade-off between information transparency and risk of losing information advantage to data consumers (Najjar 2013), the organization’s position in the value network, organization type and culture (Parvinen et al. 2020), IP protection (Thomas and Leiponen 2016), poor infrastructure (Bram et al. 2015), willingness of users to share personal data with app developers and pay for platform applications (de Reuver et al. 2015), lack of demand for data (Spiekermann 2019), regulatory complexity (Najjar 2013), data provenance (Schroeder 2016), standards and accessibility (Schroeder 2016), and internal politics (Schroeder 2016). In the context of Big Data Value Chain (BDVC) and cloud, challenges are related to deployment, scalability, exposition, networking, and enormous resources (Faroukhi, El Alaoui, et al. 2021).
Managerial implications for DMC
Organizations looking to monetize data must deploy security systems such as centralized authentication and authorization, role and data based access control, encryption and data anonymization. The owners of data monetization infrastructure must consider the legal risks, data protection barriers, competitive barriers, data availability problems, and data delivery methods. Data marketplaces can address many of these challenges as they rely on privacy assured, transaction secured and transparent platforms. They remove the effort of finding data providers and foster trustworthy transactions (Luch Kelly 2022).
Given the risks of cybersecurity incidents and the reputational implications of such incidents, many industries (e.g., the health sector) choose not to monetize data. But privacy and data protection laws provide the tools required to ensure individual data is protected and organizations are transparent (i.e., they reveal their commercial practices) (Trzaskowski 2022).
With regards to pricing, the million-dollar question remains what are the most effective means to determine price equilibrium for all the players? Since data is experienced goods, how can pricing mechanisms function if there is less willingness to pay given that buyers do not recognize the value of data because it has not been fully disclosed (Spiekermann 2019), (Rix 2021a; b)? How can pricing models be developed to consider the cost of collecting, maintaining, and making data available? In a data marketplace, how do you determine pricing that satisfies consumers and covers the cost for providers? Without financial incentives, datasets will be poorly maintained. Can someone get the same (or better) data for free somewhere else? Finally, can organizations ensure that data is accurate, updated and obtained through ethical means? These questions still need to be further explored.
Research agenda for DMC
The following DMC related research questions have been identified. (1) Investigate how data monetization can be designed with issues such as privacy and security at the forefront. Designing data-based services with security and privacy in mind is called privacy by design and is particularly important since legal developments are outpaced by technological developments. Privacy by design is an approach that takes privacy into account in the designing of a data product or service. There are reports and principles that provide such design guidance. Examples of such principles include privacy as the default, end-to-end security, avoiding false dichotomies such as privacy vs. security, etc. (2) Contribute to the academic literature on pricing models for data products and develop pricing packages and contract designs with security and privacy in mind. (3) Conduct research on developing data standards that improve quality, accessibility, and combinatorial insights. (4) Conduct research on developing regulation and policies around different types of data such as open data, proprietary data and social media data given that these three types of data barely overlap, and have different sources, uses and implications (Schroeder 2016). (5) Conduct research on developing contracts that are designed to cater for IP protection, pricing concerns, regulatory complexity, data usage/licensing and data quality.
Conclusion and limitations
This paper contributes to improving the understanding of data monetization in three ways. First, it provides a holistic understanding of areas within data monetization using a framework derived from the literature. The framework outlines the existing business models based on the research of Wixom (2014), Parvinen et al. (2020), Faroukhi et al. (2020a, b) and enhances it by introducing the players based on value contribution to the monetization ecosystem and the revenue models. The framework categorizes the models based on identified dimensions. The models include internal monetization, indirect monetization, data wrapping, data bartering and data selling with most of the literature describing a model for selling data. The framework goes further by mapping the data selling models against the BDVC phases. Second, the paper systematically derives a broad categorization and sub-categorization for the key themes in data monetization. The categories are Data Monetization Strategy (DMS), Data Monetization Infrastructure (DMI), and Data Monetization Challenges (DMC). The literature review identifies challenges such as privacy, data management, pricing, contract agreement and security which can serve as input for industry as they carve out their data monetization strategy. Third, the paper highlights managerial implications and future research agendas based on the proposed categorization. For DMS, the paper proposes that academic researchers focus on understanding factors to be considered in designing an effective data monetization strategy, developing data monetization revenue models based on chosen business models and players within the monetization ecosystem and on the interdependencies between multiple roles players can take on, the value co-creation process as well as how the overall ecosystem is governed. Organizations need to understand the structure as well as the analytical and technical capabilities that can determine the pathway to data monetization. As stated by Hartmann et al. (2016), there is a need to understand factors that impact data monetization strategies from an ecosystem perspective. This includes the characteristics of data and technological interdependencies that impact data monetization. There is a need to understand how different factors such as data quality impact the choice of a monetization model. For DMI, the paper proposes the development of a cloud and blockchain architecture that supports data monetization models as well as practical applicability of cloud/blockchain to address DMC issues. For DMC, the paper proposes that future research and practice should consider how data monetization can be designed with privacy and security at the forefront, developing data standards that improve quality, accessibility, and combinatorial insights, developing regulation and policies around different types of data and developing contracts that alleviate data monetization concerns.
To the best of our knowledge, this study is the first that considers all the data monetization models that currently exist in the academic literature. Note that the selected literature addresses data monetization from a business perspective, treating data as a resource to generate revenue. Current academic literature does not address monetization from a social perspective where data is treated as a public good or for social initiatives (e.g., Open Data initiative). Insights from data monetization have several sociological and cultural aspects that require research exploration (Thomas and Leiponen 2016).
Inevitably, the work has limitations due to the research design and exploratory nature of content analysis. From a research design perspective, the determination of the sample based on the search string, selection of timelines, database identification and criteria for paper selection (via inclusion and exclusion criteria) contribute to this limitation. Also, since the research follows an exploratory approach, the authors acknowledge the subjectivity of the outcome. Given the novelty of the research area, the authors do not anticipate that changes to these parameters may alter the overall findings.
For future work, we suggest expanding the search criteria by introducing grey literature. The data monetization framework can be further refined to improve the validity of the findings beyond academia. Researchers can also take up alternative methods such as semantic analysis to search for new concepts and better categorization or even to validate the findings of this review. Finally, identified research areas can be further explored to improve the discipline of data monetization.
Appendix A: Themes mapped to clusters and paper count
Theme Category—Number of papers that discuss Cluster #- Color
Data monetization strategy (DMS)—54
1.1 Operating model (OM)- 19 Cluster 3, 6- Blue
1.2 Players (P)—29 Cluster 4- Yellow, Cluster 5- Purple
1.3 Revenue model (RM)—10
Data monetization infrastructure (DMI)—24
2.1 Cloud (C)- 10 Cluster 2- Green
2.2 Blockchain (BC)—14
2.3 Sensors (S) and IoT—9
Data monetization challenges (DMC)—33
3.1 Security (S)—13 Cluster 1-Red, Cluster 7- Orange
3.2 Privacy (PV)—24
3.3 Pricing (PC)—15
3.4 Contract design (CD)—3
3.5 Data quality (DQ)—7
3.6 Beliefs (B)—6
3.7 Data skills (DS)—4
Appendix B: Categorization of papers by themes identified
Appendix C: Overview of SLR studies, year, journal and paper methodology
Title Authors Year Journal Paper methodology
Data Monetization: Lessons from a retailer's journey Najjar, M.S., Kettinger, W.J 2013 MIS Quarterly Executive Case study
A review of data monetization: Strategic use of big data Liu, C.-H.; Chen, C.-L 2015 International Conference on Electronic Business (ICEB) Literature review
Designing viable multi-sided data platforms: The case of context-aware mobile travel applications de Reuver, M; Haaker, T; Nikayin, F; Kosman, R 2015 Lecture Notes in Computer Science Case study with survey
How much is your information worth—A method for revenue generation for your information Rao, D; Ng, W K 2015 IEEE International Conference on Big Data Deductive: Markovian decision process
SensorHUB: An IoT driver framework for supporting sensor networks and data analysis Lengyel, L; Ekler, P; Ujj, T; Balogh, T; Charaf, H 2015 International Journal of Distributed Sensor Networks Descriptive and case study
Utilization and Monetization of Healthcare Data in Developing Countries Bram, J T; Warwick-Clark, B; Obeysekare, E; Mehta, K 2015 Big Data Exploratory
A User-Centric Approach to Pricing Information Rao, D; Ng, W K 2016 IEEE 2nd International Conference on Big Data Computing Service and Applications Shannons information theory
Big data business models: Challenges and opportunities Ralph Schroeder 2016 Cogent Social Sciences Interview
Big data commercialization Thomas, L D W; Leiponen, A 2016 IEEE Engineering Management Review Systematic literature review
Capturing value from big data—a taxonomy of data-driven business models used by start-up firms Hartmann, P.M., Zaki, M., Feldmann, N., Neely, A 2016 International Journal of Operations and Production Management DDBM frawework with clustering algorithm on 100 companies
Monetizing Personal Data: A Two-Sided Market Approach Bataineh, A S; Mizouni, R; El Barachi, M; Bentahar, J 2016 Procedia Computer Science Experimental analysis
Monetizing the user's information asset in internet information market Rao, D; Ng, W K 2016 IEEE International Congress on Big Data Mathematical analysis: Information pricing model
Some remarks and ideas about monetization of sensitive data Piotrowska, A M; Klonowski, M 2016 Lecture Notes in Computer Science Analyze monetization protocol developed by Bilogrevic et al
Data commercialisation: Extracting value from smart buildings Säynäjoki, A; Pulkka, L; Säynäjoki, E.-S.; Junnila, S 2017 Buildings Exploratory: Literature and qualitative analysis
Decentralizing the Semantic Web: Who will pay to realize it? Grubenmann, T; Dell'Aglio, D; Bernstein, A; Moor, D; Seuken, S 2017 CEUR Workshop Proceedings Exploratory
Pay-per-tracking: A collaborative masking model for web browsing Parra-Arnau, J 2017 Information Sciences Experimental analysis
A Peer-to-Peer Architecture for Distributed Data Monetization in Fog Computing Scenarios De La Vega, F; Soriano, J; Jimenez, M; Lizcano, D 2018 Wireless Communications and Mobile Computing Case study
Emerging Revenue Models for Personal Data Platform Operators: When Individuals are in Control of Their Data Kemppainen, Laura; Koivumäki, Timo; Pikkarainen, Minna; Poikola, Antti 2018 Journal of Business Models Qualitative questionnaire
A marketplace for data: An algorithmic solution Agarwal, A; Dahleh, M; Sarkar, T 2019 2019 ACM Conference on Economics and Computation Descriptive: Mathematical model
BBVA's data monetization journey Alfaro, E; Bressan, M; Girardin, F; Murillo, J; Someh, I; Wixom, B H 2019 MIS Quarterly Executive Case study
Data Marketplaces: Trends and Monetisation of Data Goods Spiekermann, M 2019 Intereconomics Exploratory and qualitative
A Fully Decentralized Infrastructure for Subscription-based IoT Data Trading Lin, C.-H.V.; Huang, C.-C.J.; Yuan, Y.-H.; Yuan, Z.-S.S 2020 IEEE International Conference on Blockchain Exploratory
Advancing data monetization and the creation of data-based business models Parvinen, P; Pöyry, E; Gustafsson, R; Laitila, M; Rossi, M 2020 Communications of the Association for Information Systems Qualitative: Exploratory and descriptive
An adaptable Big Data Value Chain (BDVC) framework for end-to-end big data monetization Faroukhi, A Z; Alaoui, I E; Gahi, Y; Amine, A 2020 Big Data and Cognitive Computing Systematic literature review
An Intelligent Game based Offloading Scheme for Maximizing Benefits of IoT-Edge-Cloud Ecosystems Yu, M; Liu, A; Xiong, N N; Wang, T 2020 IEEE Internet of Things Journal Mathematical analysis
Big data monetization throughout Big Data Value Chain: a comprehensive review Faroukhi, A Z; El Alaoui, I; Gahi, Y; Amine, A 2020 Journal of Big Data Systematic literature review
Data-driven business model development †“ insights from the facility management industry Marcinkowski, B; Gawin, B 2020 Journal of Facilities Management Qualitative case study
Monetizing Mobile Data via Data Rewards Yu, H; Wei, E; Berry, R A 2020 IEEE Journal on Selected Areas in Communications 2 stage Sackelberg game
Reputation System for IoT Data Monetization Using Blockchain Javaid, A; Zahid, M; Ali, I; Khan, R.J.U.H.; Noshad, Z; Javaid, N 2020 Lecture Notes in Networks and Systems Exploratory
Subscription-Based Data-Sharing Model Using Blockchain and Data as a Service Al-Zahrani, F A 2020 IEEE Access Inductive: Model creation
Toward monetizing personal data: A two-sided market analysis Bataineh, A S; Mizouni, R; Bentahar, J; El Barachi, M 2020 Future Generation Computer Systems Mathematical analysis
A Novel Approach for Big Data Monetization as a Service Faroukhi, A Z; El Alaoui, I; Gahi, Y; Amine, A 2021 Advances in Intelligent Systems and Computing Systematic literature review
Big Data Monetization: Platforms and Business Models Monteiro, D.S.M.P., Meira, S.R.L., Ferraz, F.S 2021 Iberian Conference on Information Systems and Technologies, CISTI Systematic literature review
Cloud as platform for monetizing complementary data for AI-driven services: A two-sided cooperative game Bataineh, A.S., Bentahar, J., Wahab, O.A., Mizouni, R., Rjoub, G 2021 IEEE International Conference on Services Computing, SCC Modelling
Conceptualizing Data Ecosystems for Industrial Food Production Calvin, R., Hannah, S., Qiang, C., Jana, F., Wolfgang, M 2021 IEEE 23rd Conference on Business Informatics, CBI Design data ecosystem and Case study
Data-driven secure, resilient and sustainable supply chains: gaps, opportunities, and a new generalised data sharing and data monetisation framework Bechtsis, D., Tsolakis, N., Iakovou, E., Vlachos, D 2021 International Journal of Production Research Literature review and case studies
From Qualitative to Quantitative Data Valuation in Manufacturing Companies Stein, H., Holst, L., Stich, V., Maass, W 2021 IFIP Advances in Information and Communication Technology Case study & exploratory action research
Ideation is Fine, but Execution is Key': How Incumbent Companies Realize Data-Driven Business Models Lange, H.E., Drews, P., Hoft, M 2021 IEEE 23rd Conference on Business Informatics, CBI 19 experts interviews and literature review
Insight monetization intermediary platform using recommender systems Hanafizadeh, P; Barkhordari Firouzabadi, M; Vu, K M 2021 Electronic Markets Design science: Literature review and model creation
Pricing Models for Data Products in the Industrial Food Production Rix, C., Frank, J., Stich, V., Urban, D 2021 IFIP Advances in Information and Communication Technology Exploratory following the procedure of typology formation by Welter
Towards data markets in renewable energy forecasting Goncalves, C; Pinson, P; Bessa, R J 2021 IEEE Transactions on Sustainable Energy Mathematical analysis
Untangling the Open Data Value Paradox: How Organizations Benefit from Revealing Data Enders, T., Benz, C., Satzger, G 2021 Lecture Notes in Information Systems and Organisation Semi structured experts interviews
Blockchains and the disruption of the sharing economy value chains Kolade, O., Adepoju, D., Adegbile, A 2022 Strategic Change Conceptual paper
Data-driven value extraction and human well-being under EU law Trzaskowski, Jan 2022 Electronic Markets Exploratory
Cloud Computing as a Platform for Monetizing Data Services: A Two-Sided Game Business Model Bataineh, Ahmed Saleh, Jamal Bentahar, Rabeb Mizouni, Omar Abdel Wahab, Gaith Rjoub, and May El Barachi 2021 IEEE Transactions on Network and Service Management Modelling
AI-Driven Data Monetization: The Other Face of Data in IoT-Based Smart and Connected Health Firouzi, Farshad, Bahar Farahani, Mojtaba Barzegari, and Mahmoud Daneshmand 2020 IEEE Internet of Things Journal Conceptual, reference architecture and case study
A Scalable, Standards-Based Approach for IoT Data Sharing and Ecosystem Monetization Figueredo, Ken, Dale Seed, and Chonggang Wang 2020 IEEE Internet of Things Journal Conceptual, reference architecture and case study
User incentives for blockchain-based data sharing platforms Jaiman, Vikas, Leonard Pernice, and Visara Urovi 2022 Plos one 17 Architecture proposal and evalution
Trustful data trading through monetizing IoT data using BlockChain based review system Abubaker, Zain, Asad Ullah Khan, Ahmad Almogren, Shahid Abbas, Atia Javaid, Ayman Radwan, and Nadeem Javaid 2022 Concurrency and Computation: Practice and Experience Exploratory
Blockchain and NFTs for Time-bound Access and Monetization of Private Data Madine, Mohammad, Khaled Salah, Raja Jayaraman, Ammar Battah, Haya Hasan, and Ibrar Yaqoob 2022 IEEE Access Exploratory
Towards a secure and dependable IoT data monetization using blockchain and fog computing Khezr, Seyednima, Abdulsalam Yassine, and Rachid Benlamri 2022 Cluster Computing Exploratory and evaluation
An Edge Intelligent Blockchain-based Reputation System for IIoT Data Ecosystem Khezr, Seyednima, Abdulsalam Yassine, Rachid Benlamri, and M. Shamim Hossain 2022 IEEE Transactions on Industrial Informatics Exploratory and evaluation
A General Approach on Privacy and its Implications in the Digital Economy Sánchez, Mariola 2022 Journal of Economic Issues Exploratory
Digital technologies: tensions in privacy and data Quach, Sara, Park Thaichon, Kelly D. Martin, Scott Weaven, and Robert W. Palmatier 2022 Journal of the Academy of Marketing Science Exploratory
Appendix D: Summary of players and value generated
Value Generated Players Definition
Generate data Data providers Originators/owners of the data/data suppliers as they generate the data leveraged in the ecosystem. It could be smart phone users or individuals having some personal data to sell. It could be user generated, IoT sensor generated or company data
Combine data Data aggregators Combines the data and provides aggregated services and data, thereby enabling them to produce a targeted advertising business model. They also perform data crawling and visualization. Common data aggregators are price comparison services such as the travel search engine Kayak. Others include Meta, Google, and Twitter
Improve data Data managers These organizations catalogue, clean, and parse information that is not in an easily usable format or improve the value of the data with additional context. They add value to data by improving the interpretability and the overall functionality of the data
Define and enforce data standards Data regulators Define and help enforce data standards. These organizations recommend and ensure the security, privacy, and ethical use of data
Custodians of data Data bank Custodians of data that enable the reuse and resale of data by providing a ‘trust’ infrastructure
Facilitate data transactions Data brokers Collect and bundle data for prospective buyers. The broker is an online platform or cloud platform equipped with the needed infrastructure to store and share data. They provide services that enable the data provider and data consumers to perform data selling and buying transactions. They can be referred to as the orchestrators
Data facilitators Have the capabilities to share data with data consumers. Facilitators do not own the data but provide services such as data cleaning, data analytics and consulting services. Data facilitators could correspond to a technical platform based on tools for data collection, integration, processing, storage, analysis, and visualization. They provide the physical architecture and the provision of outsourced analytics services
Tool providers Provide hardware and software infrastructure for data monetization. Examples include but are not limited to Microsoft, AWS, and Google who provide both software and hardware solutions
Enrich monetization ecosystem Service providers Develop new services for data, distinct to the resale, analysis or repackaging of data or the development of specific applications
App developers Design, build and sell applications that enable data monetization
Consultant Demonstrates the value of data monetization to data providers and support them in developing strategies
Consume data Data consumers Consume/subscribe to the data. They are individuals, businesses or systems that use collected data
Author contributions
All authors contributed to the study’s conception and design. Protocol preparation, execution of query and application of inclusion and exclusion criteria were performed by JO. The final papers selected was validated by MB. JO and MB identified themes from the selected papers. The first draft of the manuscript was written by JO and was reviewed and updated by MB. All authors read and approved the final manuscript.
Funding
This research is funded, in part, by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canadas (NSERC).
Data availability
All data analyzed is from published, secondary sources, available in the public domain.
Declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
1 Wrapping is when products are augmented with analytics features or experiences to increase the value and price of the product, market share, or customer satisfaction and loyalty.
2 https://www.vosviewer.com/.
3 https://www.gartner.com/en/information-technology/glossary/big-data.
4 www.mendeley.com.
5 www.covidence.org.
6 https://www.vosviewer.com/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Abubaker Z Khan AU Almogren A Abbas S Javaid A Radwan A Javaid N Trustful data trading through monetizing IoT data using BlockChain based review system Pract Exp 2022 34 5 e6739
Agarwal A, Dahleh M, Sarkar T (2019) A marketplace for data: an algorithmic solution. In: ACM EC 2019—proceedings of the 2019 ACM conference on economics and computation. 10.1145/3328526.3329589
Alfaro E Bressan M Girardin F Murillo J Someh I Wixom BH BBVA's data monetization journey MIS Q Exec 2019 10.17705/2msqe.00011
Al-Zahrani H Subscription-based data-sharing model using blockchain and data as a service IEEE Access 2020 10.1109/ACCESS.2020.3002823
Bataineh AS Mizouni R Bentahar J El Barachi M Toward monetizing personal data: a two-sided market analysis Future Gener Comput Syst 2020 111 435 459 10.1016/j.future.2019.11.009
Bataineh A Mizouni R Bentahar J Barachi M Toward monetizing personal data: a two-sided market analysis Future Gener Comput Syst 2020 10.1016/j.future.2019.11.009
Bechtsis D Tsolakis N Iakovou E Vlachos D Data-driven secure, resilient and sustainable supply chains: gaps, opportunities, and a new generalised data sharing and data monetisation framework Int J Prod Res 2021 60 4397 4417 10.1080/00207543.2021.1957506
Bram JT Warwick-Clark B Obeysekare E Mehta K Utilization and monetization of healthcare data in developing countries Big Data 2015 3 59 66 10.1089/big.2014.0053 26487984
Buff A, Wixom B, Tallon P (2014) Foundations for data monetization. A SAS Best Practise White Paper
de la Vega F, Soriano J, Jimenez M, Lizcano D (2018) A peer-to-peer architecture for distributed data monetization in fog computing scenarios. Wireless Communications and Mobile Computing
de Reuver M, Haaker T, Nikayin F, Kosman R (2015) Designing viable multi-sided data platforms: the case of context-aware mobile travel applications. In: Lecture notes in computer science (including subseries lecture notes in artificial intelligence and lecture notes in bioinformatics). 10.1007/978-3-319-25013-7_28
Deloitte AI Institute (2021) Data as a strategic asset. Deloitte AI Institute. https://www2.deloitte.com/us/en/pages/consulting/articles/data-strategic-asset.html
Demirkan H Delen D Leveraging the capabilities of service-oriented decision support systems: Putting analytics and big data in cloud Decis Support Syst 2013 55 412 421 10.1016/j.dss.2012.05.048
Donthu N Kumar S Mukherjee D Pandey N Lim WM How to conduct a bibliometric analysis: an overview and guidelines J Bus Res 2021 133 285 296 10.1016/j.jbusres.2021.04.070
EY Global (2018) How to monetize IoT data to create value. https://www.ey.com/en_gl/consulting/how-to-monetize-iot-data-to-create-value
Faroukhi AZ, El Alaoui I, Gahi Y, Amine A (2020) Big data monetization throughout big data value chain: a comprehensive review. J Big Data
Faroukhi AZ El Alaoui I Gahi Y Amine A Big data monetization throughout big data value chain: a comprehensive review J Big Data 2020 7 1 22 10.1186/s40537-019-0281-5
Faroukhi AZ Alaoui IEL Gahi Y An adaptable big data value chain (BDVC) framework for end-to-end big data monetization Big Data Cognit Comput 2020 10.3390/bdcc4040034
Faroukhi AZ, El Alaoui I, Gahi Y, Amine A (2021) A novel approach for big data monetization as a service. Advances in Intelligent Systems and Computing
Firouzi F, Farahani B, Barzegari M, Daneshmand M (2020) AI-driven data monetization: the other face of data in IoT-based smart and connected health. IEEE Internet of Things Journal
Gartner Glossary (2013) https://www.gartner.com/en/information-technology/glossary/big-data
Gopalkrishnan V, Steier D, Lewis H, Guszcza J, Lucker J (2013) Big data 2.0: new business strategies from big data
Grubenmann T, Dell'Aglio D, Bernstein A, Moor D, Seuken S (2017) Decentralizing the semantic web: Who will pay to realize it? In: CEUR workshop proceedings. https://openreview.net/forum?id=ryrkDpyIW
Hammell R, Lewis H, Perricos C, Branch D (2012) Open growth: Stimulating demand for open data in the UK
Hartmann PM, Zaki M, Feldmann N, Neely A (2016) Capturing value from big data—a taxonomy of data-driven business models used by start-up firms. International Journal of Operations and Production Management
Javaid A, Zahid M, Ali I, Khan RJ, Noshad Z, Javaid N (2020) Reputation system for IoT data monetization using blockchain. Lecture Notes in Networks and Systems
Kanhaiya et al. (2022) Data monetization market research, 2030. Allied Market Research. https://www.alliedmarketresearch.com/data-monetization-market
Kelly L (2022) Data marketplaces: the ultimate guide. https://about.datarade.ai/data-marketplaces
Kemppainen L Koivumaki T Pikkarainen M Poikola A Emerging revenue models for personal data platform operators: when individuals are in control of their data J Bus Models 2018 10.5278/ojs.jbm.v6i3.2053
Khezr S, Yassine A, Benlamri R, Hossain MS (2022a) An edge intelligent blockchain-based reputation system for IIoT data ecosystem. IEEE transactions on industrial informatics
Khezr S, Yassine A, Benlamri R (2022b) Towards a secure and dependable IoT data monetization using blockchain and fog computing. Cluster Computing 1–14
Kitchenham B (2004) Procedures for performing systematic reviews. Keele University Technical Report TR/SE-0401. https://www.inf.ufsc.br/~aldo.vw/kitchenham.pdf
Klaus K (2011) On the importance of data quality in services: an application in the financial industry. In: International conference on emerging intelligent data and web technologies, pp 148–152
Kolade O Adepoju D Adegbile A Blockchains and the disruption of the sharing economy value chains Strateg Change 2022 31 137 145 10.1002/jsc.2483
Kumar S Sahoo S Lim WM Dana LP Religion as a social shaping force in entrepreneurship and business: Insights from a technology-empowered systematic literature review Technol Forecast Soc Change 2022 10.1016/j.techfore.2021.121393
Lange HE, Drews P, Höft M (2021) Ideation is fine, but execution is key: how incumbent companies realize data-driven business models. In: IEEE 23rd conference on business informatics (CBI). 10.1109/CBI52690.2021.00030
Lim M Rasul T Kumar S Ala M Past, present, and future of customer engagement J Bus Res 2022 10.1016/j.jbusres.2021.11.014
Lin CH, Huang CC, Yuan YH, Yuan ZS (2020) A Fully decentralized infrastructure for subscription-based IoT data trading. In: Proceedings - 2020 IEEE international conference on blockchain, Blockchain 2020
Liu CH, Chen CL (2015) A review of data monetization: strategic use of big data. In: Proceedings of the international conference on electronic business (ICEB) (Proceedings of the International Conference on Electronic Business (ICEB))
Madine M, Salah K, Jayaraman R, Battah A, Hasan H, Yaqoob I (2022) Blockchain and NFTs for time-bound access and monetization of private data. pp 94186–94202
Marcinkowski B Gawin B Data-driven business model development: insights from the facility management industry J Facil Manag 2020 10.1108/JFM-08-2020-0051/full/html
Matsakis L (2018) Facebook's targeted ads are more complex than it lets on. WIRED. https://www.wired.com/story/facebooks-targeted-ads-are-more-complex-than-it-lets-on/.
Mohasseb S (2015) Framing a winning data monetization strategy. KPMG. https://home.kpmg/mu/en/home/insights/2015/10/framing-a-winning-data.html
Moore S (2015) How to monetize your customer data. https://www.gartner.com/smarterwithgartner/how-to-monetize-your-customer-data/
Moore S (2018) How to create a business case for data quality improvement. https://www.gartner.com/smarterwithgartner/how-to-create-a-business-case-for-data-quality-improvement/
Mordor Intelligence (2022) Data monetization market - growth, trends, COVID-19 impact, and forecast (2022–2027). Mordor Intelligence. https://www.mordorintelligence.com/industry-reports/data-monetization-market
Naghizadeh P, Sinha A (2019) Adversarial contract design for private data commercialization. In: The 20th ACM conference on economics and computation. Phoenix, AZ, 19
Najjar MS, Kettinger WJ (2013) Data monetization: lessons from a retailers journey. MIS Quarterly Executive 12. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.researchgate.net/profile/Mohammad-Najjar-2/publication/273447946_Data_Monetization_Lessons_from_a_Retailer's_Journey/links/58beb8d7458515dcd28e1cdc/Data-Monetization-Lessons-from-a-Retailers-
Naous D, Schwarz J, Legner C (2017) Aanalytics as a service: cloud computing and the transformation and business analytics business models and ecosystems. European Conference on Information Systems
Parra-Arnau J (2017) Pay-per-tracking: a collaborative masking model for web browsing. Information Sciences
Parvinen P, Poyry E, Gustafsson R, Laitila M, Rossi M (2020) Advancing data monetization and the creation of data-based business models. Communications of the Association for Information Systems. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://aisel.aisnet.org/cgi/viewcontent.cgi?article=4212&context=cais
Perrin B, Naftalski F, Houriez R (2013) Cultural behaviour and personal data at the heart of the big data industry. E&Y and Forum D'Avignon
Prakash SD (2014) Big data led big monetization. Vidwat 7(1):6. https://login.proxy.bib.uottawa.ca/login?url=https://www-proquest-com.proxy.bib.uottawa.ca/scholarly-journals/big-data-led-monetization/docview/1542653322/se-2?accountid=14701
PRISMA (2009) PRISMA transparent reporting of systematic reviews and meta-analyses. https://www.prisma-statement.org/
Quach S, Thaichon P, Martin KD, Weaven S, Palmatier RW (2022) Digital technologies: tensions in privacy and data. J Acad Mark Sci 1–25
Rajesh S, Swapna S, Reddy P (2012) Data as a service (daas) in cloud computing. Global Journal of Computer Science and Technology
Rao D, Ng WK (2016) A user-centric approach to pricing information. Proceedings - IEEE 2nd international conference on big data computing service and applications. 10.1109/BigDataService.2016.46
Rix C, Frank J, Stich V, Urban D (2021a) Pricing models for data products in the industrial food production. In: IFIP advances in information and communication technology. 10.1007/978-3-030-85914-5_59
Rix C, Hannah S, Qiang C, Jana F, Wolfgang M (2021b) Conceptualizing data ecosystems for industrial food production. In: IEEE 23rd conference on business informatics (CBI). 10.1109/CBI52690.2021b.00031
Robinson B (2017) How banks can become better aggregators than Google. October. https://www.linkedin.com/pulse/how-banks-can-become-better-aggregators-than-google-ben-robinson/
Saleh BA Bentahar J Wahab OA Mizouni R Rjoub G Cloud as platform for monetizing complementary data for AI-driven services: a two-sided cooperative game IEEE Int Conf Serv Comput (SCC) 2021 10.1109/SCC53864.2021.00064
Sánchez M A general approach on privacy and its implications in the digital economy J Econ Issues 2022 56 244 258 10.1080/00213624.2022.2025729
Saynajoki A Pulkka L Saynajoki E-S Junnila S Data commercialisation: extracting value from smart buildings Buildings 2017 10.3390/buildings7040104
Schroeder R Big data business models: Challenges and opportunities Cogent Soc Sci 2016 10.1080/23311886.2016.1166924
Schwab K, Marcus A, Oyola JO, Hoffman W, Luzi M (2011) Personal data the emergence of a new asset class. World economic forum. http://www3.weforum.org/docs/WEF_ITTC_PersonalDataNewAsset_Report_2011.pdf
Shrobe H, Shrier DL, Pentland A (2018) New Solutions for Cybersecurity. MIT Press
Spiekermann M Data marketplaces: trends and monetisation of data goods Intereconomics 2019 10.1007/s10272-019-0826-z
Thomas LDW Leiponen A Big data commercialization IEEE Eng Manag Rev 2016 10.1109/EMR.2016.2568798
Trianz (2022) Data monetization strategies for revenue generation. April. https://www.trianz.com/insights/data-monetization-strategies-for-revenue-generation
Trzaskowski (2022) Data-driven value extraction and human well-being under EU law. Electr Mark 1–13
Tucker CE (2014) Social networks, personalized advertising, and privacy controls. J Mark Res
Wikipedia (2014) Data monetization. https://en.wikipedia.org/wiki/Data_monetization
Wixom BH (2014) Cashing in on your data. MIT Center for Information Systems Research. https://cisr.mit.edu/publication/2014_0801_DataMonetization_Wixom
WixomB, Jeanne R (2017) How to monetize your data. MIT Sloan Management Review, 58.3. https://sloanreview.mit.edu/article/how-to-monetize-your-data/
Xie S, Zheng Z, Chen W, Wu J, Dai HN, Imran M (2020) Blockchain for cloud exchange: a survey. Comput Electr Eng 81
Yu H Wei E Berry RA Monetizing mobile data via data rewards IEEE J Sel Areas Commun 2020 10.1109/JSAC.2020a.2971867
Yu M, Liu A, Xiong NN, Wang T (2020b) An intelligent game based offloading scheme for maximizing benefits of IoT-edge-cloud ecosystems. IEEE Internet Things J
| 0 | PMC9707275 | NO-CC CODE | 2022-12-01 23:20:22 | no | Manag Rev Q. 2022 Nov 29;:1-45 | utf-8 | null | null | null | oa_other |
==== Front
Public Organiz Rev
Public Organization Review
1566-7170
1573-7098
Springer US New York
686
10.1007/s11115-022-00686-x
Article
How Joint Governance is Possible: Political Factors in COVID-19 Prevention and Control of Beijing
Xu Yiwen [email protected]
Yiwen Xu
Ph. D candidate, school of government, Peking University, research interest: law-based government.
Zeng Yang [email protected]
Yang Zeng
Ph. D candidate, school of government, Peking University, research interest: politics of disaster.
Shang Junying [email protected]
Junying Shang
Ph. D candidate, school of government, Peking University, research interest: politics of revolution.
He Zengke [email protected]
He Zengke He
Professor at the research center for Chinese politics, Peking University, research interest: corruption and anti-corruption, government innovation and political reform, national governance, the rise and fall of great powers. Communication address: school of government, Peking University, Beijing, P. R. China, 100871.
grid.11135.37 0000 0001 2256 9319 School of Government, Peking University, Beijing, 100871 People’s Republic of China
28 11 2022
119
3 11 2022
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
In the modern risk society, the general sense of collaborative governance is facing a profound crisis—poor governance efficiency due to the lack of authority. The paper introduces a case from China and finds a new mechanism in COVID-19 prevention and control which is called joint governance. Instead of the relatively equal status among multiple actors, joint governance is characterized by a dominant power forming alliances with other stakeholders. The political advantages that Beijing enjoys make it work. Our study suggests that joint governance is a new perspective of collaborative governance which brings politics back to administrative domains.
Keywords
Joint Governance
Political Factors
Transboundary Crisis Management
COVID-19
Beijing
http://dx.doi.org/10.13039/100000010 Ford Foundation
==== Body
pmcIntroduction
Over the last three decades, collaborative governance has become a new strategy for governments to replace some traditional forms of governance in policy making and implementation. The definition of collaborative governance is multi-actor cooperation between the governmental sectors and stakeholders of non-governmental sectors or among various departments within the government based on a common goal of realizing public interests (Ansell & Gash, 2008; Emerson et al., 2012; Bingham, 2020). According to this definition, the essential characters of collaborative governance are as follows: multi-actor participation in decision-making process, equal consultation among agents, high level of mutual trust and broad consensus on governance goals. However, it has some limitations when collaborative governance is used in reality (Bianchi, et al., 2021; Bingham, 2011; Huxham et al., 2000). The most noticeable problem is poor efficiency because of equitable distribution of public authority/leadership in all stakeholders (Purdy, 2012; Newman et al., 2004), which has gotten increasing attentions by many academics in the research area of transboundary crisis management (Boin et al., 2005; Boin & Rhinard, 2008; ‘t Hart & Sundelius, 2013; Clercy & Ferguson, 2016).
The paper introduces joint governance, a new form of collaborative governance that one dominant actor forms an alliance with other stakeholders when facing the same problem. The definition of joint governance has four essential criteria: (1) self-centered. The dominant agent uses its’ political strengths to behave like a leader rather than a collaborator when it allies with other actors. (2) high degree of consensus. All actors face the same problem that cannot be solved alone, which motivates them to ally with each other. (3) shared responsibility. All actors must bear its responsibility in joint governance, while the dominant actor must undertake more risks. (4) all-round alliance. This characteristic includes three dimensions: vertically intragovernmental alliance; horizontally intergovernmental alliance; alliance between governmental sectors and non-governmental sectors. To describe the details of joint governance, we take a close look at Beijing, the capital of China, in COVID-19 prevention and control, and find that Beijing has successfully leveraged its political advantages to ally with multiple stakeholders including the central government departments, the surrounding governments at the same administrative level, enterprises and social forces within its jurisdiction. Due to joint governance, the epidemic situation in Beijing has always been under control even facing Omicron virus, which also makes Beijing a case more worthy of scholar’s attention.
Therefore, the research questions of the paper are as follows: what are the core characteristics of governance mechanism that Beijing applies to prevent and control COVID-19 and why is Beijing able to do so? To answer these questions, we have collected a large number of primary and secondary materials of Beijing over the past two years, including the number of confirmed cases and the death toll, typical events, news reports, official policies and academic papers. Besides, the method of individual observation is also used to further support our findings. All the collaborative researchers in this study have been living in Beijing during the period and have gained the first-hand experience of Beijing’s response to the epidemic. These personal experiences help us better understand the relevant materials collected.
The paper has two potential contributions: firstly, we use joint governance to describe the features of Beijing’s governance mechanism of epidemic prevention and control, which contributes to the existing literature of collaborative governance. Previous studies contend that the theoretical basis of collaborative governance is the theory of polycentric governance which believes that multi-subject participation, decentralization of authority and equality of status are keys to promote the efficiency of cooperation among multiple organizations (Ansell & Gash, 2008; Emerson et al., 2012; Posthuma & Rossi, 2017). We suggest that joint governance is a special subtype of multi-actor governance. Secondly, the paper may also contribute to the research on transboundary crisis management. Current studies point out that transboundary crisis has become a new form of global challenge due to its nature of crossing geographical and functional boundaries. Scholars argue that an effective strategy to deal with transboundary crisis is to build a resilient emergency management system, which includes high levels of public communication and policy transparency, as well as public empowerment and public–private collaboration (Boin & Lodge, 2016; Boin & Rhinard, 2008; Boin et al., 2014). However, research on transboundary crisis management mainly focus on European cases with few attentions to other regions outside the EU. We would like to provide a case from China for research in this field.
Case Analysis and Empirical Findings: How Joint Governance Works?
Different from the territorial jurisdiction of most administrative regions and from the collaborative governance of mutual benefit and risk sharing among multiple agents, Beijing’s practice of COVID-19 prevention and control shows obvious characteristics of leverage. As a local government, Beijing has leveraged the support and influence of many departments of the central government, which continuously introduced favorable policies. Beijing has also leveraged the governance resources of two neighboring provincial-level regions—Tianjin and Hebei—to make contributions or even sacrifices for Beijing. Outside the government’s system, state-owned enterprises have poured massive resources into Beijing and a large number of volunteers have been mobilized and organized. As Figure 1 shows, in the mode which is called “joint governance”, individuals and organizations of different sectors serve Beijing without bargaining, which significantly expands and strengthens the power and efficacy of Beijing in epidemic prevention and control, far beyond the average level of local government and beyond its own daily state of governance.Fig. 1 Schematic diagram of joint governance enhancing Beijing’s governance capacity
Political Leverage from the Central Government
On February 21, 2020 (8 days after Wuhan’s lockdown), Beijing established a joint prevention and control coordination mechanism for strict entry of the capital under the leadership municipal party secretary and the mayor with 14 central ministries including public security, transportation, civil aviation and railway, etc., laying the organizational foundation for a series of follow-up anti-epidemic measures.
In terms of preventing the risk of imported cases, the joint activities between Beijing and central ministries are mainly reflected in the diversion of international passenger flights to Beijing by the Civil Aviation Administration of China (CAAC). Data showed that Beijing saw 95 imported cases in six days from March 18 to March 23 and its first domestic case associated with imported cases on March 23, indicating that there was a huge risk of local outbreaks caused by foreign cases. On the same day, the CAAC required all international passenger flights bound for Beijing to enter the country from 12 designated cities, known as “first points of entry”. The move made an immediate effect on the reduction of imported cases. As of April 15, only 36 cases were reported in Beijing with the average daily number dropping to 1.57.
In terms of preventing the risk of domestic cases, the joint governance between Beijing and central ministries has also played an important role. When serious clustered cases occurred in other regions, Beijing was able to coordinate timely with relevant central ministries to restrict or prevent people from those regions to enter the capital. For example, after the outbreak of the epidemic at Nanjing Lukou Airport in July 2021, China Railway Corporation immediately imposed restrictions on passengers departing from 23 stations including Nanjing and suspended the sale of tickets to Beijing. In December 2021, a cluster of imported cases of Omicron virus strain were reported in Ningbo, Zhejiang Province. In accordance with the joint prevention and control mechanism of the capital, Beijing and the central railway department jointly suspended the sale of tickets from Ningbo to Beijing.
In addition, Beijing has received special support from central government in many aspects of epidemic prevention and control. For example, after the clustered outbreak in Xinfadi market on June 11, 2020, Beijing had traced the source of the transmission chain with 24 h through epidemiological investigation and big data monitoring with the assistance of the National Center for Disease Control and Prevention (CDC). On June 18, 2020, the chief epidemiologist of the CDC attended the press conference in Beijing and announced that the outbreak in Xinfadi market had been brought under control. Another example is to promote coronavirus vaccine project, Beijing’s biopharmaceutical industry base established a special work team to actively coordinate with relevant central ministries for guidance and acceptance check to assist Sinovac to complete the preliminary procedures within only 25 days and the construction of the vaccine production workshop with a capacity of 100 to 300 million doses in 100 days. In order to ensure the transport of emergency supplies to Beijing, on January 13, 2022, the Ministry of Transport, the Ministry of Public Security and the State Post Bureau jointly required that vehicles with permits should pass through quickly in accordance with the principle of “no necessity, no blocking”.
Alliance with Surrounding Provinces and Cities
From the perspective of geographical location, the only province and city adjacent to Beijing are Tianjin and Hebei. In recent years, with the deepening coordination of Beijing-Tianjin-Hebei area, the three provinces and cities have become increasingly close in policymaking, economic development, personnel exchanges and other aspects. As the central city of this area, Beijing has taken the initiative to form a cross-regional joint governance of epidemic prevention and control with Tianjin and Hebei, which has become a “key move” to strengthen its capabilities in epidemic surveillance and early warning, risk control, material support and economic recovery.
Epidemic surveillance and early warning are the key link in crisis management system and can nip the emergency in the bud. For the prevention and control of infectious diseases, the capability of surveillance and early warning means the early detection, the early reporting, the early isolation and the early treatment, especially the former two. By the exchange of epidemic information in Beijing-Tianjin-Hebei area, the joint governance mechanism has improved Beijing’s capability of surveillance and early warning. When the epidemic occurs in any part of the three, the local CDC will immediately notify the other two places and try its best to avoid the spread of cases by exchanging flow information and sharing of travel tracks of close contacts through local big data centers and government resource platforms.
Imported risk is the Achilles heel of epidemic prevention and control in megacities because of the high population mobility. The consequent high risk of imported case needs to be controlled by joint governance mechanism. The high intensity of local prevention and control in Tianjin and Hebei, especially the control of highway access to Beijing, has blocked the risk of the spread of the epidemic from neighboring areas. This is fully reflected in the relevant statements of “making a good political moat for the capital” in the speeches of major party and government leaders in Tianjin and Hebei.
Material support is not only related to public health security in the process of fighting against the epidemic, but also affects people’s cooperation and support for the policies of epidemic prevention and control. Thus, Tianjin and Hebei have made great contributions to Beijing’s material supplies. For example, during the outbreak in Xinfadi market in the summer of 2020, Tianjin & Hebei coordinated a supply of more than 1,000 tons of vegetables to Beijing. Hebei set up a special fund of 50 million yuan for supply protection, which provided corresponding subsidies for some products in short supply in Beijing.
Coordinating epidemic prevention and control with economic and social development is one of the biggest challenges facing every local government. The cooperation in Beijing-Tianjin-Hebei area has provided a lot of convenience for the economic recovery in Beijing. First, the exchange of health information in Beijing-Tianjin-Hebei area provides convenience for commuters around Beijing. When the epidemic first broke out in early 2020, commuters around Beijing were required to be quarantined and observed for 14 days. This policy was later adjusted to allow people to enter Beijing with a negative nucleic acid certificate within 48 h and a green code of “Beijing health kit”. Second, the coordination among governments in Beijing-Tianjin-Hebei area has provided targeted services for enterprises to resume work and production, and the problem of “breakpoints” in regional industrial and supply chains has been effectively resolved. Third, the ports and sea transportation in Tianjin and Hebei have opened a new transportation channel for the resumption of work and production.
Cooperation with Enterprises and Social Forces
The participation of multiple agents is origin meaning of the word “governance”. The government, enterprises and social entities have all made significant contributions to the epidemic prevention and control in Beijing. However, in joint governance, the roles of enterprises and social forces (especially volunteers) are not only participants in the general sense, but more subordinate under the leadership of political parties and governments.
According to media reports, by September 2020, a total of 256,000 volunteers in Beijing had participated in the services of publicity of epidemic prevention and control, psychological counseling for vulnerable groups, etc. Many of these volunteers were mobilized through the CCP’s organizational system. In the first half of 2020, a total of 74,000 officials were dispatched to serve in communities, and 189,000 incumbent party members were mobilized to help fight the epidemic. Following the outbreak in Xinfadi market, Beijing launched a second round of mobilizations by government officials and employees of state-owned companies to serve voluntarily in communities. The participation of a large number of volunteers strengthened Beijing’s comprehensive capabilities of governance at the community level, especially the capabilities of epidemic surveillance and early warning and material support.
Enterprises, especially state-owned companies, have also shown strong support to epidemic prevention and control in Beijing. During the outbreak in Xinfadi market, state-owned companies responded quickly and carried out actions at all costs in the construction of nucleic acid testing laboratories, the construction of a big data system for epidemic surveillance, and the provision of medical supplies and daily necessities. Some of these companies are directly controlled by State-owned Assets Supervision and Administration Commission of the State Council, whose administrative level are equal to those of Beijing’s deputy leaders. Their support for the epidemic prevention and control in Beijing is comparable to the nationwide mobilization of “Wuhan defense battle” in early 2020. Besides, it is worth mentioning that Beijing has seized opportunities and made full use of the support of the central government and its own industrial foundation to promote the research of nucleic acid testing reagents and vaccines and to develop the biopharmaceutical industry. It not only provides an important material guarantee and scientific and technological support for China’s disease prevention and control, but also played a significant role in stimulating the economic recovery of Beijing.
In general, with the help of the central government, the surrounding provinces and cities, enterprises and social forces, Beijing leverages the political powers, human and material resources outside itself, which have strengthened its capabilities in epidemic surveillance and early warning, imported risk control, material support and economic recovery. As a result, Beijing can balance multiple governance objectives, including public security, economic growth and people’s welfare. The mode of joint governance has distinct characteristics and is significantly different from other megacities in China. Shanghai was once well-known for “balanced governance” dominated by science (Gao, 2022), but hardly compared with Beijing in leveraging external resources. Table 1 summarizes the stages and measures of joint governance in Beijing. How to grasp the characteristics of joint governance theoretically is the topic to be discussed in the next section.Table 1 The stages and measures of joint governance in Beijing
Epidemic stages/waves Wuhan-linked and local-spread cases Imported cases from abroad Clustered outbreaks in Xinfadi market The normalization of epidemic prevention and control The Omicron BA.2 variant related cases
Durations Jan 2020 to Mar 2020 Apr 2020 to May 2020 Jun 2020 to Jul 2020 Aug 2020 to Mar 2022 Apr 2022 to present
Joint governance mechanism Preliminary build Diversify the risks of imported cases Enrichment and development Control the risks of domestic transmission
With the central government Joint prevention and control coordination mechanism for strict entry of the capital International passenger flights bound for Beijing are required to enter the country from 12 designated cities Epidemiological investigations and information disclosure with the assistance of the CDC Strict measures are taken by the railway and civil aviation departments to control the entry to Beijing; the transport of material supplies is ensured All measures are used together
With Tianjin and Hebei Joint prevention and control mechanism in Beijing-Tianjin-Hebei area; mutual recognition of commuter policies in surrounding areas of Beijing; resumption of work and production in cross-regional industrial and supply chains Tianjin and Shijiazhuang are the two “first points of entry” for Beijing to divert international passenger flights Traffic control into Beijing; joint epidemiological investigation; allocation and supply of materials for people’s livelihood Strengthened traffic control into Beijing as soon as an outbreak occurs; information exchange and sharing The stringency of local epidemic prevention and control in Tianjin and Hebei has been further enhanced to prevent the spread of highly contagious strain to Beijing
With enterprises and social forces Many enterprises have invested in the production of epidemic prevention materials; more than 100,000 volunteers are mobilized State-owned enterprises respond quickly; more party members and volunteers mobilized to the front lines of communities Blowout growth in the biopharmaceutical industry stimulates Beijing’s economic recovery and development; voluntary services related to the epidemic become regular
Joint Governance: A New Perspective of Collaborative Governance
Transboundary Crisis and Governance Dilemma of Megacities
Modern society is increasingly characterized as a “risk society” (Beck, 1992). Anthony Giddens believes that modernity leads to the emergence of risks and further accelerates the accumulation and amplification of risks. When the potential risks of modern society accumulate to a certain extent, they will quickly turn into crises and cause huge impacts on the whole society (Giddens, 1991). Crisis refers to “an event that poses a serious threat to the core values and basic framework of a social system and requires rapid decision-making under time pressure and uncertain conditions” (Rosenthal et al., 1989). Among many crises, transboundary crisis is increasingly becoming the main manifestation of modern social crisis. The so-called transboundary crisis refers to a crisis type that has the nature of crossing the boundary of time and space and the boundary of public policy (Boin, 2019). As a result, the governance of transboundary crisis often goes beyond the disposal scope of a single local government and requires the joint participation of multiple entities.
In addition to the cross-boundary nature of the crisis, the size of cities also affects the difficulty of crisis governance. In China, megacities are defined as those with a permanent urban population of 10 million or more. By this measure, there were seven megacities in China (Beijing, Tianjin, Shanghai, Guangzhou, Shenzhen, Chongqing and Chengdu) by the end of 2021. Due to the characteristics of large population density, high spatial aggregation, intensive industrial groups, rich network resources and high level of internationalization, megacities are often faced with huge risks of crisis management. Once a public crisis occurs, megacities are more likely to suffer from negative impacts on personnel, economy, city image and other aspects, posing great challenges to the credibility of local governments (Yu, 2021).
It is obvious that the COVID-19 is a transboundary crisis. Under the double superposition of cross-boundary nature and urban scale, megacities are faced with the governance dilemma of dynamic balance of multiple objectives, such as maintaining public security, promoting economic development and ensuring people’s welfare (See Figure 2). First, the larger the city, the greater the risk and its social amplification effect, and the more severe the test of governance capability. Policy makers need to consider what can be done to achieve “dynamic zeroing” in their jurisdictions at a lower cost. Second, policy makers in megacities often face a dilemma: if they want to contain the epidemic early, strict measures (such as requiring home quarantine, restricting business activities, etc.) must be adopted, but those strict measures may conversely interfere the normal market order or block economic activities. Third, given the large population and its complex composition, megacities may also face the problem of livelihood deficit, which has been fully demonstrated in Shanghai through the Omicron epidemic in the first half of 2022.Fig. 2 The triangle dilemma of epidemic prevention and control in megacities
Collaborative Governance: The Theoretical Basis of Transboundary Crisis Management
Due to the cross-boundary nature of the COVID-19 crisis, the traditional territorial jurisdiction mechanism is often unable to respond effectively. On the one hand, COVID-19 is a major public crisis which requires the joint participation of the government, the market and the society. On the other hand, there is a high possibility of spillover effects from the outbreak in megacities. In view of this, it is an effective path for the epidemic prevention and control in megacities to build a transboundary crisis management mechanism based on the theory of collaborative governance.
Generally speaking, there are three types of transboundary crisis management mechanisms: the coordination of government-market-society sectors, the coordination of horizontal and vertical governments, and the coordination among multiple agencies within a single government (Zhang, 2021). For example, in 2013, six central ministries including the Ministry of Environmental Protection and the National Development and Reform Commission jointly issued The Implementation Rules of the Action Plan for Air Pollution Prevention and Control in the Beijing-Tianjin-Hebei Region and its Surrounding Areas, based on which a coordination group was established. It is one of the earliest fields of transboundary crisis management mechanisms (Liu and Peng, 2015).
Collaborative governance is a new theory evolved from polycentric governance theory, which has attracted extensive attention from academic circles in recent years. The concept of “collaboration” was first proposed by H. Haken, a German theoretical physicist, and originally meant “mutual coordination and interaction”. The concept was later taken by social scientists and refers to “a mutually beneficial relationship between two or more organizations based on common goals” (Mattessich, 1992). “Governance” means the use of authority to maintain order and meet the needs of the public within a given area (Yu, 1999). The goal of governance is to improve the welfare of people and finally achieve “good governance”. Therefore, collaborative governance can be defined as a multi-actor relationship of cooperation between the governmental sectors and stakeholders of non-governmental sectors or between various departments within the government based on a common goal of realizing public interests (Bingham, 2020). The preconditions of collaborative governance are as follows: multi-actor participation in decision-making process, equal consultation among agents, high level of mutual trust and broad consensus on governance goals (Emerson et al., 2012).
To be specific, when it comes to transboundary crisis management, the realization of collaborative governance needs to meet the following conditions: (1) Diversification of governance subjects. The subjects involved in transboundary crisis management include not only a certain level of government, but also governments at the same level and the upper or lower levels, as well as many social organizations (such as NGOs, private enterprises, communities and families). (2) Diversity of governance authorities. The power of decision-making is not controlled by a certain level of government but scattered among various subjects. In other words, all parties involved have the right to express their views and the right to vote in decision-making process. (3) Equality of status and voluntary cooperation. Each subject is in a relationship of equality and mutual assistance, and the communication among them does not rely on coercive power but establishes a partnership based on consultation and dialogue. (4) Common goal and agreement to achieve that goal. All subjects have a high degree of consensus on eliminating public crisis and are willing to contribute to this end (Danahue, 2004).
Joint Governance: A Solution to the Lack of Authority in Collaborative Governance
Although collaborative governance can improve the efficiency of transboundary crisis management in megacities, it also has some limitations in practical situations. For example, megacities like Guangzhou, Shenzhen, Chongqing and Chengdu are more inclined to adopt the collaborative governance mechanisms of government-market-society sectors and multiple agencies within a single government, rather than the coordination of horizontal or vertical governments. Shanghai, on the other hand, has established a joint prevention and control mechanism with its neighboring Jiangsu and Zhejiang provinces but is limited to screening and transport of overseas entry.
The main reason for this phenomenon is the lack of authority in collaborative governance. Studies show that cross-regional collaboration is more difficult than the coordination among sectors within one city (Wang and Zhang, 2019). It is because that governments of different regions may have different opinions on leadership attribution, benefit distribution, governance goals, responsibility sharing, mutual trust and other aspects, leading to many obstacles in the coordination and cooperation of cross-regional governance. For example, the air pollution control in Beijing-Tianjin-Hebei area faces problems such as legal barriers, development gaps, and unequal allocation of power and responsibility (Du, 2018). In epidemic prevention and control, the lack of authority limits the maximization of crisis management efficiency in megacities.
In our case, Beijing has developed a new mechanism of joint governance on the basis of collaborative governance, which effectively solved the lack of authority in transboundary crisis management. It must be pointed out that the theoretical basis of this mechanism does not go beyond the scope of collaborative governance. It can be seen as a new form of collaborative governance theory in a specific period, a specific region and a specific field (such as crisis management). However, in practice, there are some characteristics of joint governance, which make it different from the general sense of collaborative governance. Firstly, the most prominent feature of the joint governance mechanism is “self-centered” in the dimension of leadership attribution. Secondly, the establishment of joint governance mechanism often requires a high degree of consensus. Thirdly, joint governance mechanism can effectively reduce the coordination cost and realize the sharing of power, responsibility and benefit. Finally, the joint network not only covers the traditional collaborative governance mode, but also include the linkage of governments at the same level and the linkage of central-local governments, forming an omni-directional integrated governance pattern.
The four characteristics of joint governance are based on distinct political factors. Beijing’s joint governance has accomplished the combination of three dimensions: the combination with the central government to set up a joint prevention and control coordination mechanism for strict entry of the capital, the combination with neighbors to establish a joint prevention and control mechanism for the Beijing-Tianjin-Hebei region, and the combination with enterprises and social forces within its jurisdiction. Practice has proved that the joint governance mechanism has not only helped Beijing go through several epidemic waves since the beginning of 2020, but also achieved a dynamic balance among multiple objectives. More importantly, it has overcome the lack of authority in collaborative governance, leaving a profound impact on the upgrading of governance capacities and the reshaping of the central-local relationship and state-society relationship.
Discussion: Political Factors Underpinning Joint Governance
Political Capital
The most important reason for Beijing to adopt joint governance is that as the capital city of China, Beijing has some political capitals that no other cities can match. First, China is a unitary state. This form of political organization gives Beijing higher bargaining power than other local governments in the process of interacting with the central government. Although in unitary or federal countries, the capital city is the seat of the central government which undertakes state functions. However, as the former is characterized by centralization and the latter by checks and balances, the political status of the unitary capital is much higher than that of the federal capital.
Second, in addition to the unitary political organization form, the regime type of China is often classified as authoritarian state, which means a lower level of democracy and rule of law. When democracy and rule of law are relatively absent, the political process is more likely to be disturbed by non-institutional factors. As the capital city of China, Beijing provides office space for many central government ministries and commissions, which makes it easier for leaders of Beijing to get closer to the core of the state power and have more opportunities to develop personal relationships with leaders of the central government. This kind of first-mover advantage gives Beijing the chance to bypass formal institutional constraints and influence the central government’s policy through informal mechanisms.
Third, top leaders of Beijing enjoy a higher political status than other government officials of the same rank. China is a party-state ruled by the Communist Party of China (CCP). The political status of government officials is divided not only by the administrative level, but also by the ranking of status in CCP. The administrative level corresponds to the party status in general, so in most cases, the administrative level can distinguish the political status of government officials. However, when government officials are of the same administrative level, the key factor determining their political status is their party status. Beijing’s party secretary has a higher political status than some major leaders of central ministries or leaders of neighboring provinces and cities. According to public sources, CAI Qi, the current secretary of the Beijing Municipal Party Committee, is a member of the 19th CCP Central Committee and the Political Bureau, indicating that CAI Qi has entered the highest level of power in CCP, while most major leaders of central ministries are not in that position.
Political Support
Another key factor that enables Beijing to unite central ministries and neighboring provinces and cities is that the policies of epidemic prevention and control adopted by Beijing are supported by the top leaders of the central government who provide the political endorsement for the joint governance in Beijing-Tianjin-Hebei area. One of the main difficulties facing joint governance activities is how the initiator establishes its leadership authority without its jurisdiction. Despite Beijing’s top leaders have a higher political status than most major leaders of central ministries or leaders of neighboring provinces and cities, because of the limitation of its jurisdiction, Beijing has no administrative power over various agents in Tianjin and Hebei. Under this circumstance, political support of the top provides a strong impetus for Beijing to break through the above institutional constraints. President Xi Jinping stressed that the governments of Beijing-Tianjin-Hebei region should establish a working mechanism for joint prevention and control of COVID-19 in a research activity in Beijing as early as February 10, 2020 (twenty days after Wuhan’s lockdown). And on 16 February 2020, three local governments officially signed and published The Working Mechanism for Joint Prevention and Control of Novel Coronavirus Infections in the Beijing-Tianjin-Hebei Region to detail the above requirements of President Xi. In Chinese political context, top leaders’ demands usually have the importance as soft law, and local governments need to firmly obey and implement these demands. It is the president’s support for joint governance policy that enables Beijing to take the lead in the process of epidemic prevention and control in Beijing-Tianjin-Hebei region.
On the one hand, political support of top leaders stems from the high priority given to COVID-19 prevention and control by the top decision-making level of the central government. This priority can be attributed to lessons learned from the SARS epidemic. Since the Spring Festival of 2003, SARS first broke out in Guangdong province and quickly spread across the country. Beijing became the second hardest-hit area next to Guangdong. Due to the lack and delay of real information, the central government underestimated the severity of SARS and overestimated the ability of Guangdong and Beijing to fight the epidemic. It was not until April 2003 that the central leaders began to realize the severity of SARS and take actions to strengthen the supervision and control over local governments (Hongyi, 2004). One of the lessons that top leaders have learned from SARS is the necessity for the central government to intervene in the early stages of a public health emergency. After 17 years, when COVID-19 broke out in Wuhan city, President Xi chaired a meeting of Standing Committee of the Political Bureau of the CCP Central Committee and made arrangement for epidemic prevention and control on the first day of the Spring Festival.
On the other hand, political support of top leaders also stems from the President’s high trust in Cai Qi, the Party Secretary of Beijing. This trusting relationship between the upper and lower political leaders has laid a solid political foundation for Beijing to obtain the authorization of the central government. In China’s political ecology, local officials will have more opportunities to obtain political resources if they can establish a personal relationship or even a protective relationship with the superior leaders (Ma, 2022). Cai Qi is a former subordinate of President Xi and the two have high overlap in time and space from their political resumes. In epidemic prevention and control, the trust of top leaders empowers Beijing to have more discretionary space and provide strong political backup for Beijing to take the lead in promoting joint governance with the central ministries and neighboring provinces and cities.
Political Responsiveness
In addition to vertical and horizontal government-to-government cooperation, the joint governance mechanism during the epidemic is also manifested as the cooperation between Beijing municipal government and enterprises and social forces. The driving force behind this kind of “government-led, market-oriented and socially-coordinated” governance is not political capital or political support from the top, but the high level of political reputation that Beijing municipal government has accumulated. Different from the government response based on the principal-agent relationship in western countries, the responsiveness in China’s political system is mainly reflected in the initiative of government officials to collect and respond to social demands (Su & Meng, 2016; Meng et al., 2017). In the absence of a high level of democracy and rule of law, local government’s initiatives to respond to the society are an alternative to institutional change. This alternative can greatly improve social expectations of the government and mobilize the enthusiasm for political participation (He & Warren, 2011; Hsu, 2009).
The high level of political responsiveness aroused the enthusiasm of enterprises and social forces to cooperate with the Beijing municipal government in epidemic prevention and control. On the one hand, Beijing has long established a formal channel of 12,345 citizen service hotline, which provides an effective way for enterprises, social organizations and citizens to reflect their demands. Meanwhile, Beijing also introduced measures of governance reform centering on “immediate response” mechanism since 2019. The reform integrated more than 100 governmental affairs hotlines into one, laying an organizational and technical foundation for the government to collect and respond to people’s demands. The “immediate response” mechanism has played an important role in COVID-19 prevention and control. According to statistics, in the early days of the outbreak, Beijing recruited more than 500 operators for hotlines and kept 24 h available. In 2020 alone, the 12,345 citizen service hotline received more than 1.65 million pieces of complaints and requests about the epidemic, with a resolution rate of 93.75% and satisfaction rate of 94.18%.
On the other hand, Beijing also actively uses cyberspace to expand the channels for the expression of social demands. The “Message Board for Leaders” of the People’s Daily is an influential online platform for political inquiry in China. It was established in August 2006, providing convenience for Chinese netizens to express their demands, reflect their problems and make suggestions to the government. At present, most local government officials in China have opened their own sections under the message board with assistants responsible for collecting and sorting out netizens’ appeals. According to statistics from People’s Daily Online, In the first half of 2022, Beijing residents were the most active on the “Message Board for Leaders” with 10.2 cases per 10,000 people. Meanwhile, the Beijing municipal government also had the highest response rate of 9.3 cases per 10,000 people.
Political Trust
In epidemic prevention and control, the effective implementation of policies formulated by the government depends on the willingness of the public to comply with these policies. If the public is willing to cooperate with the government policies, the government will be more likely to use the people’s power to control the epidemic in a short time (Bargain & Aminjonov, 2020; Fancourt et al., 2020). In the early stages of the COVID-19 outbreak, physical methods such as wearing masks, social distancing, reducing the frequency of gatherings, quarantining at home, and collecting travel codes are effective means of suppressing the spread of novel coronavirus. As the technology matures, nucleic acid testing and vaccination have become technological means in addition to physical means. However, the effectiveness of those measures varies widely by country. People in some western countries are reluctant to follow government policies and even against policies of quarantine and vaccination through demonstrations. In China, there are few social protests against the government policies. Most Chinese people voluntarily wear masks, maintain social distance and cooperate with nucleic acid testing and vaccination.
The main reason for this phenomenon is the high level of political trust that Chinese people have in their government. People in many western countries do not have a high level of political trust in their central government (Ronald, 1997; Nye & Joseph, 1997), leading them to doubt the legitimacy of government policies about the epidemic. In contrast, Chinese people’s high level of political trust in the central government is strongly linked to traditional Chinese culture. Under the influence of Confucianism, worship and obedience to power have become one of the main reasons for the majority of Chinese people to trust their government (Shi, 2001). Although Chinese people’s political trust shows a hierarchical pattern of “strong in the central, weak in the local”, the political trust in local governments various significant by region (Cary and Wilkes, 2018). In general, governments in economically developed areas such as Beijing enjoy a higher level of political trust (Lyu & Li, 2018). In the process of epidemic prevention and control, Beijing made full use of political trust to mobilize a large number of volunteers to participate in the front-line work.
Conclusion: A Double-Edged Sword to Bring “Politics” Back
Reviewing the process of COVID-19 prevention and control in Beijing in the past two years, the study finds that Beijing has constructed a joint governance mechanism based on the leverage from the central government, the alliance with surrounding provinces and cities, and the cooperation with enterprises and social forces. As a result, its governance capabilities of epidemic surveillance and early warning, risk control of imported cases, material support, economic recovery and development are strengthened, achieving a dynamic balance among multiple governance objectives. The reason that Beijing makes joint governance possible is mainly due to the political factors—political capital, political support, political responsiveness, and political trust—that empower it to overcome the lack of authority in collaborative governance. The key difference between joint governance and collaborative governance is the relationship among multiple participants. Under collaborative governance, the participants are roughly equal; while under joint governance, there is a dominant power which is obviously above the others.
Joint governance is a double-edged sword to the governance under the epidemic. On the positive side, joint governance constitutes an extension to the theory and practice of collaborative governance. It is important to distinguish between two sets of concepts: firstly, compared with the general sense of collaborative governance, joint governance emphasizes more on the centralization of power, which means there is a power center instead of diversified and loose alliance. It is not a subvert of collaborative governance, but a beneficial attempt to carry out collaborative governance in transboundary crisis and emergencies. Secondly, joint governance does not mean to go back to the traditional mode of territorial jurisdiction. Although the power is concentrated in certain regions or in certain periods of time, local governments still need to give play to their own subjective initiatives, divide their functions according to different resource endowments, and explore countermeasures in line with local conditions facing the complex and changeable situations. Only in this way can the “joint force” be greater than the simple sum of their respective dispersed forces.
However, with the long-lasting impact of COVID-19 and the normalization of crisis management, the negative side of the politicization of epidemic prevention and control slowly rises to the surface. It is alarming to the possibility of a new “totalitarianism” which may damage the legitimacy of a regime in the long term. In fact, the cost of joint governance can be higher than its benefits. For instance, the surrounding provinces and cities of Beijing are in a relatively weak position in the process of joint governance. They cannot and should not bear the cost alone because it does not accord with the principle of equality and proportionality. In the long run, unequal and disproportionate relationships in transboundary governance will erode social solidarity and threaten political stability.
The greatest contribution of joint governance to the theory of collaborative governance is to bring “politics” back to the administrative domains. Politics and administration are supposed to be dichotomized since The Study of Administration of Thomas Woodrow Wilson. The former lays stress on legitimacy while the latter put more emphasis on efficiency. From this point of view, joint governance is an art of balancing and trade-off. Future studies of transboundary crisis management must have a composite perspective of politics and administration. This argument applies in crisis as well as in normal times.
Funding
Granted by the Ford Foundation as a sub-project of the “COVID-19 Pandemic and National Governance” project.
Declarations
Conflict of Interest
None.
Informed consent
None (Research methods such as questionnaires, interviews and focus groups for specific respondents were not used in this paper).
Ethical approval
None (The data used in this paper were all from official statistics and media reports, and do not involve ethical issues in the collection of primary data).
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
English Literature
Ansell C Gash A Collaborative governance in theory and practice Journal of Public Administration Research and Theory 2008 18 4 543 571 10.1093/jopart/mum032
Bargain O Aminjonov U Trust and compliance to public health policies in times of COVID-19 Journal of Public Economics 2020 192 104316 10.1016/j.jpubeco.2020.104316 33162621
Beck U Risk Society: Toward a New Modernity 1992 Sage
Bianchi C Nasi G Rivenbark WC Implementing collaborative governance: Models, experiences, and challenges Public Management Review 2021 23 11 1581 1589 10.1080/14719037.2021.1878777
Bingham LB The Next Generation of Administrative Law: Building the Legal Infrastructure for Collaborative Governance Wisconsin Law Review 2020 10 289 350
Boin A The Transboundary Crisis: Why we are unprepared and the road ahead Journal of Contingencies and Crisis Management 2019 27 1 94 99 10.1111/1468-5973.12241
Boin A Busuioc M Groenleer M Building European Union capacity to manage transboundary crises: Network or lead-agency model? Regulation & Governance 2014 8 4 418 436 10.1111/rego.12035
Boin A Lodge M Designing resilient institutions for transboundary crisis management: A time for public administration Public Administration 2016 94 2 289 298 10.1111/padm.12264
Boin A McConnell A Hart PT Governing after Crisis: The Politics of Investigation, Accountability and Learning 2008 Cambridge University Press
Boin A Rhinard M Managing transboundary crises: What role for the European Union? International Studies Review 2008 10 1 1 26 10.1111/j.1468-2486.2008.00745.x
Boin A ‘t Hart P Stern E Sundelius B The politics of crisis management: Public leadership under pressure 2005 Cambridge University Press
Bingham LB Collaborative governance 2011 Sage Publications Ltd 386 401
Cary Wu Wilkes R Local-National Political Trust Patterns: Why China is an Exception International Political Science Review 2018 39 4 436 454 10.1177/0192512116677587
Danahue J On Collaborative Governance 2004 Harvard University
De Clercy, C., & Ferguson, P. (2016). Leadership in precarious contexts: Studying political leaders after the global financial crisis. Politics and Governance, 4(2), 104–114. 10.17645/pag.v4i2.582.
Emerson K Nabatchi T Balogh S An Integrative Framework for Collaborative Governance Journal of Public Administration Research and Theory 2012 22 1 1 29 10.1093/jopart/mur011
Fancourt D Steptoe A Wright L The Cummings effect: Politics, trust, and behaviours during the COVID-19 pandemic The Lancet 2020 396 10249 464 465 10.1016/S0140-6736(20)31690-1
Giddens A Modernity and Self-Identity: Self and Society in the Late Modern Age 1991 Polity Press
He B Warren ME Authoritarian deliberation: The deliberative turn in Chinese political development Perspectives on Politics 2011 9 2 269 289 10.1017/S1537592711000892
Hongyi L Wong J Yonnian Z Local Management of SARS in China: Guangdong and Beijing The SARS Epidemic: Challenges to China’s Crisis Management 2004 World Scientific 90 95
Huxham C Vangen S Huxham C Eden C The challenge of collaborative governance Public Management an International Journal of Research and Theory 2000 2 3 337 358 10.1080/14719030000000021
Hsu SP In search of public accountability: The “Wenling model” in China Australian Journal of Public Administration 2009 68 S40 S50 10.1111/j.1467-8500.2009.00624.x
Lyu S Li L The dual effect of economic development on trust in the central government: Evidence from two waves of a national survey in China China Review 2018 18 2 65 86
Ma X Localized Bargaining: The Political Economy of China’s High-Speed Railway Program 2022 Oxford University Press
Mattessich PW Monsey BR Collaboration: What Makes It Word: A Review of Research Literature on Factors Influencing Successful Collaboration 1992 Amherst H. Wilder Foundation
Meng T Pan J Yang P Conditional receptivity to citizen participation: Evidence from a survey experiment in China Comparative Political Studies 2017 50 4 399 433 10.1177/0010414014556212
Newman J Barnes M Sullivan H Knops A Public participation and collaborative governance Journal of Social Policy 2004 33 2 203 223 10.1017/S0047279403007499
Nye, Joseph S., JR. (1997). “Introduction: the decline of confidence in government.” In Joseph S. Nye Jr., Philip D. Zelikov, and David C. King (eds), Why People Don’t Trust Government. Cambridge: Harvard University Press, 1–18.
Posthuma A Rossi A Coordinated governance in global value chains: Supranational dynamics and the role of the International Labour Organization New Political Economy 2017 22 2 186 202 10.1080/13563467.2016.1273342
Purdy JM A framework for assessing power in collaborative governance processes Public Administration Review 2012 72 3 409 417 10.1111/j.1540-6210.2011.02525.x
Ronald, R. (1997). Modernization and Post-modernization: Cultural, Economic, and Political Change in Forty-Three Societies. Princeton University Press.
Uriel Rosenthal Michael Charles Coping with Crises: The Management of Disasters, Riots and Terrorism 1989 Charles C. Thomas
Shi T Cultural Values and Political Trust: A Comparison of the People’s Republic of China and Taiwan Comparative Politics 2001 33 4 401 419 10.2307/422441
Su Z Meng T Selective responsiveness: Online public demands and government responsiveness in authoritarian China Social Science Research 2016 59 5 52 67 10.1016/j.ssresearch.2016.04.017 27480371
‘t Hart P Sundelius B Crisis management revisited: A new agenda for research, training and capacity building within Europe Cooperation and Conflict 2013 48 3 444 461 10.1177/0010836713485711
Chinese Literature
Du CB Ecological Compensation Mechanism in Synergy Governance of Haze–A Case Study of Beijing-Tianjin-Hebei Region Academic Journal of Zhongzhou 2018 12 29 34
Liu B Peng ZC Trans-boundary Crisis and Coordinated Emergency Plans: A Comparative Analysis of Emergency Plans for Haze Pollution in Beijing-Tianjin-Hebei Region Journal of Tongji University (Social Science Edition) 2015 26 04 67 75 10.3969/j.issn.1009-3060.2015.04.008
Wang XD Zhang DA The Realistic Predicament and Realization Path of China’s Regional Collaborative Governance Chinese Public Administration 2019 2 12 15 10.19735/j.issn.1006-0863.2019.06.02
Yu, K. P. (1999). An Introduction of Governance and Good Governance. Marxism & Reality, 37–41. 10.15894/j.cnki.cn11-3040/a.1999.05.008
Yu KP Important Issues on the Innovation of Chinese Urban Governance Based upon Case Studies of Megacities Wuhan University Journal (Philosophy & Social Science) 2021 74 3 88 99 10.14086/j.cnki.wujss.2021.03.009
Zhang HB Cross Regional Coordination in Emergency Management Journal of Nanjing University (philosophy, Humanities and Social Sciences) 2021 58 01 102 110
| 0 | PMC9707276 | NO-CC CODE | 2022-12-01 23:20:22 | no | Public Organiz Rev. 2022 Nov 28;:1-19 | utf-8 | null | null | null | oa_other |
==== Front
Sci China Life Sci
Sci China Life Sci
Science China. Life Sciences
1674-7305
1869-1889
Science China Press Beijing
36443513
2215
10.1007/s11427-022-2215-6
Review
Broad strategies for neutralizing SARS-CoV-2 and other human coronaviruses with monoclonal antibodies
Ling Zhiyang 12
Yi Chunyan 12
Sun Xiaoyu 12
Yang Zhuo 12
Sun Bing [email protected]
123
1 grid.9227.e 0000000119573309 State Key Laboratory of Cell Biology, Shanghai Institute of Biochemistry and Cell Biology, Center for Excellence in Molecular Cell Science, Chinese Academy of Sciences, Shanghai, 200031 China
2 grid.410726.6 0000 0004 1797 8419 University of Chinese Academy of Sciences, Beijing, 100049 China
3 grid.440637.2 0000 0004 4657 8879 School of Life Science and Technology, ShanghaiTech University, Shanghai, 201210 China
24 11 2022
121
23 6 2022
4 10 2022
© Science China Press and Springer-Verlag GmbH Germany, part of Springer Nature 2022
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Antibody therapeutics and vaccines for coronavirus disease 2019 (COVID-19) have been approved in many countries, with most being developed based on the original strain of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). SARS-CoV-2 has an exceptional ability to mutate under the pressure of host immunity, especially the immune-dominant spike protein of the virus, which is the target of both antibody drugs and vaccines. Given the continuous evolution of the virus and the identification of critical mutation sites, the World Health Organization (WHO) has named 5 variants of concern (VOCs): 4 are previously circulating VOCs, and 1 is currently circulating (Omicron). Due to multiple mutations in the spike protein, the recently emerged Omicron and descendent lineages have been shown to have the strongest ability to evade the neutralizing antibody (NAb) effects of current antibody drugs and vaccines. The development and characterization of broadly neutralizing antibodies (bNAbs) will provide broad strategies for the control of the sophisticated virus SARS-CoV-2. In this review, we describe how the virus evolves to escape NAbs and the potential neutralization mechanisms that associated with bNAbs. We also summarize progress in the development of bNAbs against SARS-CoV-2, human coronaviruses (CoVs) and other emerging pathogens and highlight their scientific and clinical significance.
antibody therapy
broad neutralizing antibody
human coronaviruses
SARS-CoV-2
==== Body
pmcAcknowledgements
This work was supported by grants from the Ministry of Science and Technology of China (2018YFA0507402), the Key International Partnership Program of the Chinese Academy of Sciences (153D31KYSB20180055), and the National Natural Science Foundation of China (32270991, 32100123, 32100751).
Compliance and ethics
The author(s) declare that they have no conflict of interest.
Contributed equally to this work
==== Refs
References
Amanat F Thapa M Lei T Ahmed SMS Adelsberg DC Carreño JM Strohmeier S Schmitz AJ Zafar S Zhou JQ SARS-CoV-2 mRNA vaccination induces functionally diverse antibodies to NTD, RBD, and S2 Cell 2021 184 3936 3948.e10 10.1016/j.cell.2021.06.005 34192529
Barnes CO Jette CA Abernathy ME Dam KMA Esswein SR Gristick HB Malyutin AG Sharaf NG Huey-Tubman KE Lee YE SARS-CoV-2 neutralizing antibody structures inform therapeutic strategies Nature 2020 588 682 687 10.1038/s41586-020-2852-1 33045718
Baum A Fulton BO Wloga E Copin R Pascal KE Russo V Giordano S Lanza K Negron N Ni M Antibody cocktail to SARS-CoV-2 spike protein prevents rapid mutational escape seen with individual antibodies Science 2020 369 1014 1018 10.1126/science.abd0831 32540904
Behring EKS Ueber das Zustandekommen der diphtherie-immunitat und der tetanus-immunitat bei thieren Dtsch Med Wschr 1890 16 1113 1114 10.1055/s-0029-1207589
Benton DJ Wrobel AG Xu P Roustan C Martin SR Rosenthal P B Skehel JJ Gamblin SJ Receptor binding and priming of the spike protein of SARS-CoV-2 for membrane fusion Nature 2020 588 327 330 10.1038/s41586-020-2772-0 32942285
Brekke OH Sandlie I Therapeutic antibodies for human diseases at the dawn of the twenty-first century Nat Rev Drug Discov 2003 2 52 62 10.1038/nrd984 12509759
Burton DR Poignard P Stanfield RL Wilson IA Broadly neutralizing antibodies present new prospects to counter highly antigenically diverse viruses Science 2012 337 183 186 10.1126/science.1225416 22798606
Cai Y Zhang J Xiao T Peng H Sterling SM Walsh RM Jr. Rawson S Rits-Volloch S Chen B Distinct conformational states of SARS-CoV-2 spike protein Science 2020 369 1586 1592 10.1126/science.abd4251 32694201
Cameroni E Bowen JE Rosen LE Saliba C Zepeda SK Culap K Pinto D VanBlargan LA De Marco A di Iulio J Broadly neutralizing antibodies overcome SARS-CoV-2 Omicron antigenic shift Nature 2022 602 664 670 10.1038/s41586-021-04386-2 35016195
Cao Y Wang J Jian F Xiao T Song W Yisimayi A Huang W Li Q Wang P An R Omicron escapes the majority of existing SARS-CoV-2 neutralizing antibodies Nature 2022 602 657 663 10.1038/s41586-021-04385-3 35016194
Cao Y Yisimayi A Jian F Song W Xiao T Wang L Du S Wang J Li Q Chen X BA.2.12.1, BA.4 and BA.5 escape antibodies elicited by Omicron infection Nature 2022 608 593 602 10.1038/s41586-022-04980-y 35714668
Carter PJ Lazar GA Next generation antibody drugs: pursuit of the ‘high-hanging fruit Nat Rev Drug Discov 2018 17 197 223 10.1038/nrd.2017.227 29192287
Case JB Mackin S Errico JM Chong Z Madden EA Whitener B Guarino B Schmid MA Rosenthal K Ren K Resilience of S309 and AZD7442 monoclonal antibody treatments against infection by SARS-CoV-2 Omicron lineage strains Nat Commun 2022 13 3824 10.1038/s41467-022-31615-7 35780162
Cerino A Mantovani S Mele D Oliviero B Varchetta S Mondelli MU Human monoclonal antibodies as adjuvant treatment of chronic hepatitis B virus infection Front Immunol 2019 10 2290 10.3389/fimmu.2019.02290 31608071
Cerutti G Guo Y Wang P Nair MS Wang M Huang Y Yu J Liu L Katsamba PS Bahna F Neutralizing antibody 5–7 defines a distinct site of vulnerability in SARS-CoV-2 spike N-terminal domain Cell Rep 2021 37 109928 10.1016/j.celrep.2021.109928 34706271
Cerutti G Guo Y Zhou T Gorman J Lee M Rapp M Reddem E R Yu J Bahna F Bimela J Potent SARS-CoV-2 neutralizing antibodies directed against spike N-terminal domain target a single supersite Cell Host Microbe 2021 29 819 833.e7 10.1016/j.chom.2021.03.005 33789084
Chan AC Carter PJ Therapeutic antibodies for autoimmunity and inflammation Nat Rev Immunol 2010 10 301 316 10.1038/nri2761 20414204
Correia BE Bates JT Loomis RJ Baneyx G Carrico C Jardine J G Rupert P Correnti C Kalyuzhniy O Vittal V Proof of principle for epitope-focused vaccine design Nature 2014 507 201 206 10.1038/nature12966 24499818
Corti D Purcell LA Snell G Veesler D Tackling COVID-19 with neutralizing monoclonal antibodies Cell 2021 184 4593 4595 10.1016/j.cell.2021.07.027 34416148
Costello SM Shoemaker SR Hobbs HT Nguyen AW Hsieh CL Maynard JA McLellan JS Pak JE Marqusee S The SARS-CoV-2 spike reversibly samples an open-trimer conformation exposing novel epitopes Nat Struct Mol Biol 2022 29 229 238 10.1038/s41594-022-00735-5 35236990
Creech CB Walker SC Samuels RJ SARS-CoV-2 vaccines JAMA 2021 325 1318 1320 10.1001/jama.2021.3199 33635317
Crowe JE Jr. Bispecific antiviral neutralizing antibodies are twice as nice Nat Immunol 2022 23 346 347 10.1038/s41590-022-01147-9 35228693
Dacon C Tucker C Peng L Lee CCD Lin TH Yuan M Cong Y Wang L Purser L Williams JK Broadly neutralizing antibodies target the coronavirus fusion peptide Science 2022 377 728 735 10.1126/science.abq3773 35857439
de Melo GD Hellert J Gupta R Corti D Bourhy H Monoclonal antibodies against rabies: current uses in prophylaxis and in therapy Curr Opin Virol 2022 53 101204 10.1016/j.coviro.2022.101204 35151116
Dimitrov DS The secret life of ACE2 as a receptor for the SARS virus Cell 2003 115 652 653 10.1016/S0092-8674(03)00976-0 14675530
Dong J Zost SJ Greaney AJ Starr TN Dingens AS Chen EC Chen RE Case JB Sutton RE Gilchuk P Genetic and structural basis for SARS-CoV-2 variant neutralization by a two-antibody cocktail Nat Microbiol 2021 6 1233 1244 10.1038/s41564-021-00972-2 34548634
du Plessis L McCrone JT Zarebski AE Hill V Ruis C Gutierrez B Raghwani J Ashworth J Colquhoun R Connor TR Establishment and lineage dynamics of the SARS-CoV-2 epidemic in the UK Science 2021 371 708 712 10.1126/science.abf2946 33419936
Edelman GM Antibody structure and molecular immunology Science 1973 180 830 840 10.1126/science.180.4088.830 4540988
Emmons C Hunsicker LG Muromonab-CD3 (Orthoclone OKT3): the first monoclonal antibody approved for therapeutic use Iowa Med 1987 77 78 82 3557906
Estcourt LJ Passive immune therapies: another tool against COVID-19 Hematology 2021 2021 1 628 641 10.1182/hematology.2021000299 34889410
Fedry J Hurdiss DL Wang C Li W Obal G Drulyte I Du W Howes SC van Kuppeveld FJM Förster F Structural insights into the cross-neutralization of SARS-CoV and SARS-CoV-2 by the human monoclonal antibody 47D11 Sci Adv 2021 7 eabf5632 10.1126/sciadv.abf5632 33958322
Focosi D McConnell S Casadevall A Cappello E Valdiserra G Tuccori M Monoclonal antibody therapies against SARS-CoV-2 Lancet Infect Dis 2022 22 e311 e326 10.1016/S1473-3099(22)00311-5 35803289
Fu D Zhang G Wang Y Zhang Z Hu H Shen S Wu J Li B Li X Fang Y Structural basis for SARS-CoV-2 neutralizing antibodies with novel binding epitopes PLoS Biol 2021 19 e3001209 10.1371/journal.pbio.3001209 33961621
Gao Q Bao L Mao H Wang L Xu K Yang M Li Y Zhu L Wang N Lv Z Development of an inactivated vaccine candidate for SARS-CoV-2 Science 2020 369 77 81 10.1126/science.abc1932 32376603
Graham BS Ambrosino DM History of passive antibody administration for prevention and treatment of infectious diseases Curr Opin HIV AIDS 2015 10 129 134 10.1097/COH.0000000000000154 25760933
Greaney AJ Loes AN Crawford KHD Starr TN Malone KD Chu HY Bloom JD Comprehensive mapping of mutations in the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human plasma antibodies Cell Host Microbe 2021 29 463 476.e6 10.1016/j.chom.2021.02.003 33592168
Gupta A Gonzalez-Rojas Y Juarez E Crespo Casal M Moya J Falci DR Sarkis E Solis J Zheng H Scott N Early treatment for Covid-19 with SARS-CoV-2 neutralizing antibody sotrovimab N Engl J Med 2021 385 1941 1950 10.1056/NEJMoa2107934 34706189
Hamre D Procknow JJ A new virus isolated from the human respiratory tract Exp Biol Med 1966 121 190 193 10.3181/00379727-121-30734
Hansen J Baum A Pascal KE Russo V Giordano S Wloga E Fulton BO Yan Y Koon K Patel K Studies in humanized mice and convalescent humans yield a SARS-CoV-2 antibody cocktail Science 2020 369 1010 1014 10.1126/science.abd0827 32540901
Harvey WT Carabelli AM Jackson B Gupta RK Thomson EC Harrison EM Ludden C Reeve R Rambaut A Peacock SJ SARS-CoV-2 variants, spike mutations and immune escape Nat Rev Microbiol 2021 19 409 424 10.1038/s41579-021-00573-0 34075212
Hoffmann M Kleine-Weber H Schroeder S Krüger N Herrler T Erichsen S Schiergens TS Herrler G Wu NH Nitsche A SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor Cell 2020 181 271 280.e8 10.1016/j.cell.2020.02.052 32142651
Hong Q Han W Li J Xu S Wang Y Xu C Li Z Wang Y Zhang C Huang Z Molecular basis of receptor binding and antibody neutralization of Omicron Nature 2022 604 546 552 10.1038/s41586-022-04581-9 35228716
Hoy SM Amubarvimab/romlusevimab: first approval Drugs 2022 82 1327 1331 10.1007/s40265-022-01759-3 35997943
Iketani S Liu L Guo Y Liu L Chan JFW Huang Y Wang M Luo Y Yu J Chu H Antibody evasion properties of SARS-CoV-2 Omicron sublineages Nature 2022 604 553 556 10.1038/s41586-022-04594-4 35240676
Jackson CB Farzan M Chen B Choe H Mechanisms of SARS-CoV-2 entry into cells Nat Rev Mol Cell Biol 2022 23 3 20 10.1038/s41580-021-00418-x 34611326
Jakobovits A Amado RG Yang X Roskos L Schwab G From XenoMouse technology to panitumumab, the first fully human antibody product from transgenic mice Nat Biotechnol 2007 25 1134 1143 10.1038/nbt1337 17921999
Jangra S Ye C Rathnasinghe R Stadlbauer D Krammer F Simon V Martinez-Sobrido L García-Sastre A Schotsaert M Alshammary H SARS-CoV-2 spike E484K mutation reduces antibody neutralisation Lancet Microbe 2021 2 e283 e284 10.1016/S2666-5247(21)00068-9 33846703
Jones BE Brown-Augsburger PL Corbett KS Westendorf K Davies J Cujec TP Wiethoff CM Blackbourne JL Heinz BA Foster D The neutralizing antibody, LY-CoV555, protects against SARS-CoV-2 infection in nonhuman primates Sci Transl Med 2021 13 eabf1906 10.1126/scitranslmed.abf1906 33820835
Ju B Zhang Q Ge J Wang R Sun J Ge X Yu J Shan S Zhou B Song S Human neutralizing antibodies elicited by SARS-CoV-2 infection Nature 2020 584 115 119 10.1038/s41586-020-2380-z 32454513
Kallewaard NL Corti D Collins PJ Neu U McAuliffe JM Benjamin E Wachter-Rosati L Palmer-Hill FJ Yuan AQ Walker PA Structure and function analysis of an antibody recognizing all influenza A subtypes Cell 2016 166 596 608 10.1016/j.cell.2016.05.073 27453466
Kaufmann SHE Emil von Behring: translational medicine at the dawn of immunology Nat Rev Immunol 2017 17 341 343 10.1038/nri.2017.37 28393925
Keller MA Stiehm ER Passive immunity in prevention and treatment of infectious diseases Clin Microbiol Rev 2000 13 602 614 10.1128/CMR.13.4.602 11023960
Kmietowicz Z Covid-19: monoclonal antibodies authorised in US as alternative to vaccines for certain groups BMJ 2021 375 n3064 10.1136/bmj.n3064 34893506
Köhler G Milstein C Continuous cultures of fused cells secreting antibody of predefined specificity Nature 1975 256 495 497 10.1038/256495a0 1172191
Konings F Perkins MD Kuhn JH Pallen MJ Alm EJ Archer B N Barakat A Bedford T Bhiman JN Caly L SARS-CoV-2 Variants of Interest and Concern naming scheme conducive for global discourse Nat Microbiol 2021 6 821 823 10.1038/s41564-021-00932-w 34108654
Labrijn AF Janmaat ML Reichert JM Parren PWHI Bispecific antibodies: a mechanistic review of the pipeline Nat Rev Drug Discov 2019 18 585 608 10.1038/s41573-019-0028-1 31175342
Levin EG Lustig Y Cohen C Fluss R Indenbaum V Amit S Doolman R Asraf K Mendelson E Ziv A Waning immune humoral response to BNT162b2 Covid-19 vaccine over 6 months N Engl J Med 2021 385 e84 10.1056/NEJMoa2114583 34614326
Levin MJ Ustianowski A De Wit S Launay O Avila M Templeton A Yuan Y Seegobin S Ellery A Levinson DJ Intramuscular AZD7442 (Tixagevimab-Cilgavimab) for prevention of Covid-19 N Engl J Med 2022 386 2188 2200 10.1056/NEJMoa2116620 35443106
Li C Zhan W Yang Z Tu C Hu G Zhang X Song W Du S Zhu Y Huang K Broad neutralization of SARS-CoV-2 variants by an inhalable bispecific single-domain antibody Cell 2022 185 1389 1401.e18 10.1016/j.cell.2022.03.009 35344711
Li M Wang H Tian L Pang Z Yang Q Huang T Fan J Song L Tong Y Fan H COVID-19 vaccine development: milestones, lessons and prospects Sig Transduct Target Ther 2022 7 146 10.1038/s41392-022-00996-y
Li Q Guan X Wu P Wang X Zhou L Tong Y Ren R Leung K SM Lau EHY Wong JY Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia N Engl J Med 2020 382 1199 1207 10.1056/NEJMoa2001316 31995857
Li T Huang T Guo C Wang A Shi X Mo X Lu Q Sun J Hui T Tian G Genomic variation, origin tracing, and vaccine development of SARS-CoV-2: a systematic review Innovation 2021 2 100116 33997827
Li Z Li S Zhang G Peng W Chang Z Zhang X Fan Z Chai Y Wang F Zhao X An engineered bispecific human monoclonal antibody against SARS-CoV-2 Nat Immunol 2022 23 423 430 10.1038/s41590-022-01138-w 35228696
Liu H Wu NC Yuan M Bangaru S Torres JL Caniels TG van Schooten J Zhu X Lee CCD Brouwer PJM Cross-neutralization of a SARS-CoV-2 antibody to a functionally conserved site is mediated by avidity Immunity 2020 53 1272 1280.e5 10.1016/j.immuni.2020.10.023 33242394
Liu L Iketani S Guo Y Reddem ER Casner RG Nair MS Yu J Chan JFW Wang M Cerutti G An antibody class with a common CDRH3 motif broadly neutralizes sarbecoviruses Sci Transl Med 2022 14 eabn6859 10.1126/scitranslmed.abn6859 35438546
Liu L Iketani S Guo Y Chan JFW Wang M Liu L Luo Y Chu H Huang Y Nair MS Striking antibody evasion manifested by the Omicron variant of SARS-CoV-2 Nature 2022 602 676 681 10.1038/s41586-021-04388-0 35016198
Liu Z VanBlargan LA Bloyet LM Rothlauf PW Chen RE Stumpf S Zhao H Errico JM Theel ES Liebeskind MJ Identification of SARS-CoV-2 spike mutations that attenuate monoclonal and serum antibody neutralization Cell Host Microbe 2021 29 477 488.e4 10.1016/j.chom.2021.01.014 33535027
Liu Z Xu W Chen Z Fu W Zhan W Gao Y Zhou J Zhou Y Wu J Wang Q An ultrapotent pan-β-coronavirus lineage B (β-CoV-B) neutralizing antibody locks the receptor-binding domain in closed conformation by targeting its conserved epitope Protein Cell 2022 13 655 675 10.1007/s13238-021-00871-6 34554412
Loo YM McTamney PM Arends RH Abram ME Aksyuk AA Diallo S Flores DJ Kelly EJ Ren K Roque R The SARS-CoV-2 monoclonal antibody combination, AZD7442, is protective in nonhuman primates and has an extended half-life in humans Sci Transl Med 2022 14 eabl8124 10.1126/scitranslmed.abl8124 35076282
Lu G Hu Y Wang Q Qi J Gao F Li Y Zhang Y Zhang W Yuan Y Bao J Molecular basis of binding between novel human coronavirus MERS-CoV and its receptor CD26 Nature 2013 500 227 231 10.1038/nature12328 23831647
Lu G Zhang Y Zhang H Ai J He L Yuan X Bao S Chen X Wang H Cai J Geriatric risk and protective factors for serious COVID-19 outcomes among older adults in Shanghai Omicron wave Emerging Microbes Infects 2022 11 2045 2054 10.1080/22221751.2022.2109517
Lv Z Deng YQ Ye Q Cao L Sun CY Fan C Huang W Sun S Sun Y Zhu L Structural basis for neutralization of SARS-CoV-2 and SARS-CoV by a potent therapeutic antibody Science 2020 369 1505 1509 10.1126/science.abc5881 32703908
Mair-Jenkins J Saavedra-Campos M Baillie JK Cleary P Khaw F M Lim WS Makki S Rooney KD Nguyen-Van-Tam JS Beck CR The effectiveness of convalescent plasma and hyperimmune immunoglobulin for the treatment of severe acute respiratory infections of viral etiology: a systematic review and exploratory meta-analysis J Infect Dis 2015 211 80 90 10.1093/infdis/jiu396 25030060
Morens DM Fauci AS Emerging pandemic diseases: how we got to COVID-19 Cell 2020 183 837 10.1016/j.cell.2020.10.022 33125895
Mücke MM Zeuzem S The recent outbreak of acute severe hepatitis in children of unknown origin—what is known so far J Hepatol 2022 77 237 242 10.1016/j.jhep.2022.05.001 35533802
Muecksch F Wang Z Cho A Gaebler C Ben Tanfous T DaSilva J Bednarski E Ramos V Zong S Johnson B Increased memory B cell potency and breadth after a SARS-CoV-2 mRNA boost Nature 2022 607 128 134 10.1038/s41586-022-04778-y 35447027
Nicholls AJ Brown CB Edward N Cuthbertson B Yap PL Mcclelland DBL Hyperimmune immunoglobulin for cytomegalovirus infections Lancet 1983 321 532 533 10.1016/S0140-6736(83)92221-3
Nikitin PA DiMuzio JM Dowling JP Patel NB Bingaman-Steele JL Heimbach BC Henriquez N Nicolescu C Polley A Sikorski EL IMM-BCP-01, a patient-derived anti-SARS-CoV-2 antibody cocktail, is active across variants of concern including Omicron BA.1 and BA.2 Sci Immunol 2022 7 eabl9943 10.1126/sciimmunol.abl9943 35771946
Nyakatura EK Soare AY Lai JR Bispecific antibodies for viral immunotherapy Hum Vaccines Immunother 2017 13 836 842 10.1080/21645515.2016.1251536
Ou X Liu Y Lei X Li P Mi D Ren L Guo L Guo R Chen T Hu J Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its immune cross-reactivity with SARS-CoV Nat Commun 2020 11 1620 10.1038/s41467-020-15562-9 32221306
Park YJ De Marco A Starr TN Liu Z Pinto D Walls AC Zatta F Zepeda SK Bowen JE Sprouse KR Antibody-mediated broad sarbecovirus neutralization through ACE2 molecular mimicry Science 2022 375 449 454 10.1126/science.abm8143 34990214
Pérez-Alós L Armenteros JJA Madsen JR Hansen CB Jarlhelt I Hamm SR Heftdal LD Pries-Heje MM Møller DL Fogh K Modeling of waning immunity after SARS-CoV-2 vaccination and influencing factors Nat Commun 2022 13 1614 10.1038/s41467-022-29225-4 35347129
Piccoli L Park YJ Tortorici MA Czudnochowski N Walls AC Beltramello M Silacci-Fregni C Pinto D Rosen LE Bowen JE Mapping neutralizing and immunodominant sites on the SARS-CoV-2 spike receptor-binding domain by structure-guided highresolution serology Cell 2020 183 1024 1042.e21 10.1016/j.cell.2020.09.037 32991844
Pinto D Park YJ Beltramello M Walls AC Tortorici MA Bianchi S Jaconi S Culap K Zatta F De Marco A Cross-neutralization of SARS-CoV-2 by a human monoclonal SARS-CoV antibody Nature 2020 583 290 295 10.1038/s41586-020-2349-y 32422645
Prévost J Finzi A The great escape? SARS-CoV-2 variants evading neutralizing responses Cell Host Microbe 2021 29 322 324 10.1016/j.chom.2021.02.010 33705702
Raju TN The Nobel chronicles. 1901: Emil Adolf von Behring (1854-1917) Lancet 1998 352 75 10.1016/S0140-6736(05)79565-9 9800790
Rappazzo CG Tse LV Kaku CI Wrapp D Sakharkar M Huang D Deveau LM Yockachonis TJ Herbert AS Battles MB Broad and potent activity against SARS-like viruses by an engineered human monoclonal antibody Science 2021 371 823 829 10.1126/science.abf4830 33495307
Rappuoli R Bottomley MJ D’Oro U Finco O De Gregorio E Reverse vaccinology 2.0: human immunology instructs vaccine antigen design J Exp Med 2016 213 469 481 10.1084/jem.20151960 27022144
Sandritter T Palivizumab for respiratory syncytial virus prophylaxis J Pediatr Health Care 1999 13 191 195 10.1016/S0891-5245(99)90039-1 10690084
Sauer MM Tortorici MA Park YJ Walls AC Homad L Acton O J Bowen JE Wang C Xiong X de van der Schueren W Structural basis for broad coronavirus neutralization Nat Struct Mol Biol 2021 28 478 486 10.1038/s41594-021-00596-4 33981021
Shi R Shan C Duan X Chen Z Liu P Song J Song T Bi X Han C Wu L A human neutralizing antibody targets the receptor-binding site of SARS-CoV-2 Nature 2020 584 120 124 10.1038/s41586-020-2381-y 32454512
Shiakolas AR Kramer KJ Johnson NV Wall SC Suryadevara N Wrapp D Periasamy S Pilewski KA Raju N Nargi R Efficient discovery of SARS-CoV-2-neutralizing antibodies via B cell receptor sequencing and ligand blocking Nat Biotechnol 2022 40 1270 1275 10.1038/s41587-022-01232-2 35241839
Shrestha LB Tedla N Bull RA Broadly-neutralizing antibodies against emerging SARS-CoV-2 variants Front Immunol 2021 12 752003 10.3389/fimmu.2021.752003 34646276
Smith K Garman L Wrammert J Zheng NY Capra JD Ahmed R Wilson PC Rapid generation of fully human monoclonal antibodies specific to a vaccinating antigen Nat Protoc 2009 4 372 384 10.1038/nprot.2009.3 19247287
Starr TN Greaney AJ Addetia A Hannon WW Choudhary MC Dingens AS Li JZ Bloom JD Prospective mapping of viral mutations that escape antibodies used to treat COVID-19 Science 2021 371 850 854 10.1126/science.abf9302 33495308
Starr TN Greaney AJ Hilton SK Ellis D Crawford KHD Dingens AS Navarro MJ Bowen JE Tortorici MA Walls A C Deep mutational scanning of SARS-CoV-2 receptor binding domain reveals constraints on folding and ACE2 binding Cell 2020 182 1295 1310.e20 10.1016/j.cell.2020.08.012 32841599
Sun X Liu C Lu X Ling Z Yi C Zhang Z Li Z Jin M Wang W Tang S Unique binding pattern for a lineage of human antibodies with broad reactivity against influenza A virus Nat Commun 2022 13 2378 10.1038/s41467-022-29950-w 35501328
Sun X Yi C Zhu Y Ding L Xia S Chen X Liu M Gu C Lu X Fu Y Neutralization mechanism of a human antibody with pan-coronavirus reactivity including SARS-CoV-2 Nat Microbiol 2022 7 1063 1074 10.1038/s41564-022-01155-3 35773398
Sun Y Wang L Feng R Wang N Wang Y Zhu D Xing X Yang P Zhang Y Li W Structure-based development of three- and four-antibody cocktails against SARS-CoV-2 via multiple mechanisms Cell Res 2021 31 597 600 10.1038/s41422-021-00497-7 33782529
Tada T Zhou H Dcosta BM Samanovic MI Chivukula V Herati RS Hubbard SR Mulligan MJ Landau NR Increased resistance of SARS-CoV-2 Omicron variant to neutralization by vaccine-elicited and therapeutic antibodies eBioMedicine 2022 78 103944 10.1016/j.ebiom.2022.103944 35465948
Takashita E Kinoshita N Yamayoshi S Sakai-Tagawa Y Fujisaki S Ito M Iwatsuki-Horimoto K Chiba S Halfmann P Nagai H Efficacy of antibodies and antiviral drugs against Covid-19 Omicron variant N Engl J Med 2022 386 995 998 10.1056/NEJMc2119407 35081300
Takashita E Yamayoshi S Simon V van Bakel H Sordillo EM Pekosz A Fukushi S Suzuki T Maeda K Halfmann P Efficacy of antibodies and antiviral drugs against Omicron BA.2.12.1, BA.4, and BA.5 subvariants N Engl J Med 2022 387 468 470 10.1056/NEJMc2207519 35857646
Tonegawa S Somatic generation of antibody diversity Nature 1983 302 575 581 10.1038/302575a0 6300689
Tortorici MA Czudnochowski N Starr TN Marzi R Walls AC Zatta F Bowen JE Jaconi S Di Iulio J Wang Z Broad sarbecovirus neutralization by a human monoclonal antibody Nature 2021 597 103 108 10.1038/s41586-021-03817-4 34280951
Tortorici MA Walls AC Lang Y Wang C Li Z Koerhuis D Boons GJ Bosch BJ Rey FA de Groot RJ Structural basis for human coronavirus attachment to sialic acid receptors Nat Struct Mol Biol 2019 26 481 489 10.1038/s41594-019-0233-y 31160783
Turoňová B Sikora M Schürmann C Hagen WJH Welsch S Blanc FEC von Bülow S Gecht M Bagola K Hörner C In situ structural analysis of SARS-CoV-2 spike reveals flexibility mediated by three hinges Science 2020 370 203 208 10.1126/science.abd5223 32817270
Tyrrell DAJ Bynoe ML Cultivation of a novel type of common-cold virus in organ cultures Brit Med J 1965 1 1467 1470 10.1136/bmj.1.5448.1467 14288084
V’kovski P Kratzel A Steiner S Stalder H Thiel V Coronavirus biology and replication: implications for SARS-CoV-2 Nat Rev Microbiol 2021 19 155 170 10.1038/s41579-020-00468-6 33116300
van der Hoek L Pyrc K Jebbink MF Vermeulen-Oost W Berkhout RJM Wolthers KC Wertheim-van Dillen PME Kaandorp J Spaargaren J Berkhout B Identification of a new human coronavirus Nat Med 2004 10 368 373 10.1038/nm1024 15034574
van der Straten K van Gils MJ de Taeye SW de Bree GJ Optimization of anti-SARS-CoV-2 neutralizing antibody therapies: roadmap to improve clinical effectiveness and implementation Front Med Technol 2022 4 867982 10.3389/fmedt.2022.867982 35419561
Vitiello A Ferrara F Auti AM Di Domenico M Boccellino M Advances in the Omicron variant development J Intern Med 2022 292 81 90 10.1111/joim.13478 35289434
Wang C Li W Drabek D Okba NMA van Haperen R Osterhaus ADME van Kuppeveld FJM Haagmans BL Grosveld F Bosch BJ A human monoclonal antibody blocking SARS-CoV-2 infection Nat Commun 2020 11 2251 10.1038/s41467-020-16256-y 32366817
Wang C van Haperen R Gutiérrez-Álvarez J Li W Okba NMA Albulescu I Widjaja I van Dieren B Fernandez-Delgado R Sola I A conserved immunogenic and vulnerable site on the coronavirus spike protein delineated by cross-reactive monoclonal antibodies Nat Commun 2021 12 1715 10.1038/s41467-021-21968-w 33731724
Wang K Jia Z Bao L Wang L Cao L Chi H Hu Y Li Q Zhou Y Jiang Y Memory B cell repertoire from triple vaccinees against diverse SARS-CoV-2 variants Nature 2022 603 919 925 10.1038/s41586-022-04466-x 35090164
Wang N Sun Y Feng R Wang Y Guo Y Zhang L Deng YQ Wang L Cui Z Cao L Structure-based development of human antibody cocktails against SARS-CoV-2 Cell Res 2021 31 101 103 10.1038/s41422-020-00446-w 33262454
Wang P Casner RG Nair MS Wang M Yu J Cerutti G Liu L Kwong PD Huang Y Shapiro L Increased resistance of SARS-CoV-2 variant P.1 to antibody neutralization Cell Host Microbe 2021 29 747 751.e4 10.1016/j.chom.2021.04.007 33887205
Wang S Sun H Zhang Y Yuan L Wang Y Zhang T Wang S Zhang J Yu H Xiong H Three SARS-CoV-2 antibodies provide broad and synergistic neutralization against variants of concern, including Omicron Cell Rep 2022 39 110862 10.1016/j.celrep.2022.110862 35594869
Wang W Sun X Li Y Su J Ling Z Zhang T Wang F Zhang H Chen H Ding J Human antibody 3E1 targets the HA stem region of H1N1 and H5N6 influenza A viruses Nat Commun 2016 7 13577 10.1038/ncomms13577 27910950
Wang Z Schmidt F Weisblum Y Muecksch F Barnes CO Finkin S Schaefer-Babajew D Cipolla M Gaebler C Lieberman JA mRNA vaccine-elicited antibodies to SARS-CoV-2 and circulating variants Nature 2021 592 616 622 10.1038/s41586-021-03324-6 33567448
Weinreich DM Sivapalasingam S Norton T Ali S Gao H Bhore R Musser BJ Soo Y Rofail D Im J REGN-COV2, a neutralizing antibody cocktail, in outpatients with Covid-19 N Engl J Med 2021 384 238 251 10.1056/NEJMoa2035002 33332778
Weiss GA Watanabe CK Zhong A Goddard A Sidhu SS Rapid mapping of protein functional epitopes by combinatorial alanine scanning Proc Natl Acad Sci USA 2000 97 8950 8954 10.1073/pnas.160252097 10908667
Westendorf K Žentelis S Wang L Foster D Vaillancourt P Wiggin M Lovett E van der Lee R Hendle J Pustilnik A LY-CoV1404 (bebtelovimab) potently neutralizes SARS-CoV-2 variants Cell Rep 2022 39 110812 10.1016/j.celrep.2022.110812 35568025
Wise J Covid-19: evusheld is approved in UK for prophylaxis in immunocompromised people Brit Med J 2022 376 o722 10.1136/bmj.o722 35301227
Wong AHM Tomlinson ACA Zhou D Satkunarajah M Chen K Sharon C Desforges M Talbot PJ Rini JM Receptor-binding loops in alphacoronavirus adaptation and evolution Nat Commun 2017 8 1735 10.1038/s41467-017-01706-x 29170370
Woo PCY Lau SKP Chu C Chan K Tsoi H Huang Y Wong B HL Poon RWS Cai JJ Luk W Characterization and complete genome sequence of a novel coronavirus, coronavirus HKU1, from patients with pneumonia J Virol 2005 79 884 895 10.1128/JVI.79.2.884-895.2005 15613317
Wu F Zhao S Yu B Chen YM Wang W Song ZG Hu Y Tao Z W Tian JH Pei YY A new coronavirus associated with human respiratory disease in China Nature 2020 579 265 269 10.1038/s41586-020-2008-3 32015508
Wu K Li W Peng G Li F Crystal structure of NL63 respiratory coronavirus receptor-binding domain complexed with its human receptor Proc Natl Acad Sci USA 2009 106 19970 19974 10.1073/pnas.0908837106 19901337
Xu S Wang Y Wang Y Zhang C Hong Q Gu C Xu R Wang T Yang Y Zang J Mapping cross-variant neutralizing sites on the SARS-CoV-2 spike protein Emerging Microbes Infects 2022 11 351 367 10.1080/22221751.2021.2024455
Yamasoba D Kimura I Nasser H Morioka Y Nao N Ito J Uriu K Tsuda M Zahradnik J Shirakawa K Virological characteristics of the SARS-CoV-2 Omicron BA.2 spike Cell 2022 185 2103 2115.e19 10.1016/j.cell.2022.04.035 35568035
Yao H Sun Y Deng YQ Wang N Tan Y Zhang NN Li XF Kong C Xu YP Chen Q Rational development of a human antibody cocktail that deploys multiple functions to confer Pan-SARS-CoVs protection Cell Res 2021 31 25 36 10.1038/s41422-020-00444-y 33262452
Yi C Ling Z Lu X Fu Y Yang Z Wangmo S Chen S Zhang Y Ma L Gu W A SARS-CoV-2 antibody retains potent neutralization against Omicron by targeting conserved RBM residues Cell Mol Immunol 2022 19 647 649 10.1038/s41423-022-00853-6 35352009
Yi C Sun X Lin Y Gu C Ding L Lu X Yang Z Zhang Y Ma L Gu W Comprehensive mapping of binding hot spots of SARS-CoV-2 RBD-specific neutralizing antibodies for tracking immune escape variants Genome Med 2021 13 164 10.1186/s13073-021-00985-w 34649620
Yi C Xia J He L Ling Z Wang X Yan Y Wang J Zhao X Fan W Sun X Junctional and somatic hypermutation-induced CX4C motif is critical for the recognition of a highly conserved epitope on HCV E2 by a human broadly neutralizing antibody Cell Mol Immunol 2021 18 675 685 10.1038/s41423-020-0403-1 32235917
Yu S Zheng X Zhou B Li J Chen M Deng R Wong G Lavillette D Meng G SARS-CoV-2 spike engagement of ACE2 primes S2′ site cleavage and fusion initiation Proc Natl Acad Sci USA 2022 119 e2111199119 10.1073/pnas.2111199119 34930824
Zeng G Huang J The recent outbreak of acute severe hepatitis in children of unknown origin J Hepatol 2022 77 1213 1214 10.1016/j.jhep.2022.05.017 35644435
Zhou H Chen Y Zhang S Niu P Qin K Jia W Huang B Zhang S Lan J Zhang L Structural definition of a neutralization epitope on the N-terminal domain of MERS-CoV spike glycoprotein Nat Commun 2019 10 3068 10.1038/s41467-019-10897-4 31296843
Zhou P Yuan M Song G Beutler N Shaabani N Huang D He W T Zhu X Callaghan S Yong P A human antibody reveals a conserved site on beta-coronavirus spike proteins and confers protection against SARS-CoV-2 infection Sci Transl Med 2022 14 eabi9215 10.1126/scitranslmed.abi9215 35133175
Zhou T Tsybovsky Y Gorman J Rapp M Cerutti G Chuang GY Katsamba PS Sampson JM Schön A Bimela J Cryo-EM structures of SARS-CoV-2 spike without and with ACE2 reveal a pH-dependent switch to mediate endosomal positioning of receptor-binding domains Cell Host Microbe 2020 28 867 879.e5 10.1016/j.chom.2020.11.004 33271067
Zhou Y Liu Z Li S Xu W Zhang Q Silva IT Li C Wu Y Jiang Q Liu Z Enhancement versus neutralization by SARS-CoV-2 antibodies from a convalescent donor associates with distinct epitopes on the RBD Cell Rep 2021 34 108699 10.1016/j.celrep.2021.108699 33485405
Zost SJ Gilchuk P Case JB Binshtein E Chen RE Nkolola JP Schäfer A Reidy JX Trivette A Nargi RS Potently neutralizing and protective human antibodies against SARS-CoV-2 Nature 2020 584 443 449 10.1038/s41586-020-2548-6 32668443
Zost SJ Gilchuk P Chen RE Case JB Reidy JX Trivette A Nargi RS Sutton RE Suryadevara N Chen EC Rapid isolation and profiling of a diverse panel of human monoclonal antibodies targeting the SARS-CoV-2 spike protein Nat Med 2020 26 1422 1427 10.1038/s41591-020-0998-x 32651581
| 36443513 | PMC9707277 | NO-CC CODE | 2022-12-01 23:20:22 | no | Sci China Life Sci. 2022 Nov 24;:1-21 | utf-8 | Sci China Life Sci | 2,022 | 10.1007/s11427-022-2215-6 | oa_other |
==== Front
Environ Sci Pollut Res Int
Environ Sci Pollut Res Int
Environmental Science and Pollution Research International
0944-1344
1614-7499
Springer Berlin Heidelberg Berlin/Heidelberg
36441331
24387
10.1007/s11356-022-24387-6
Research Article
Modeling the economic viability and performance of solar home systems: a roadmap towards clean energy for environmental sustainability
Ali Shahid 1
Yan Qingyou 12
Dilanchiev Azer 3
http://orcid.org/0000-0003-1446-583X
Irfan Muhammad [email protected]
456
Fahad Shah 7
1 grid.261049.8 0000 0004 0645 4572 School of Economics and Management, North China Electric Power University, Beijing, 102206 China
2 grid.261049.8 0000 0004 0645 4572 Beijing Key Laboratory of New Energy and Low-Carbon Development, North China Electric Power University, Beijing, 102206 China
3 grid.444114.0 0000 0004 0386 6395 International Black Sea University, Tbilisi, Georgia
4 grid.43555.32 0000 0000 8841 6246 School of Management and Economics, Beijing Institute of Technology, Beijing, 100081 China
5 grid.43555.32 0000 0000 8841 6246 Center for Energy and Environmental Policy Research, Beijing Institute of Technology, Beijing, 100081 China
6 grid.444859.0 0000 0004 6354 2835 School of Business Administration, ILMA University, Karachi, 75190 Pakistan
7 grid.459727.a 0000 0000 9195 8580 School of Economics and Management, Leshan Normal University, Leshan, 614000 China
Responsible Editor: Philippe Garrigues
28 11 2022
120
7 9 2022
20 11 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Energy is a necessary source of economic development and social prosperity, linked with primary production and consumption activities worldwide. In this regard, solar home systems (SHSs) are beneficial in two ways, i.e., saving vitality overheads and meeting the energy demand of small enterprises. The current study aims to evaluate the performance of adopting SHS to develop the small-scale industry in Pakistan. An inclusive questionnaire survey was conducted, and respondents were selected using the purposive sampling method. As a step further, we scrutinize the moderating role of awareness and understanding of technology between the node of adopting SHS and the monetary enactment of small-scale industry. We authenticate the model using a sample of 357 respondents by applying the partial least square structural equation modeling (PLS-SEM) technique. The results indicate that low-cost energy through SHSs has a progressive and substantial linkage with the demonstration of small-scale industry and enhances the quality of energy supply in Pakistan. Similarly, awareness and understanding of SHS significantly moderate the relationships between enhanced energy supply through SHS, the quality of SHS, and the performance of the small-scale industry. These findings provide a valuable guideline to the regulation developing authorities that more attention is needed to focus on SHS to further improve the performance of small-scale industry.
Keywords
Solar home system
Low-cost energy
Enhanced energy supply
Energy economics
Quality of SHS
Small-scale industry
==== Body
pmcIntroduction
Energy is essential in facilitating daily activities and is considered a primary need in modern life (Baig et al. 2022; Ren et al. 2021). All sectors of life are associated with or dependent on energy (Deng et al. 2022; Ongan et al. 2022). The increasing energy needs require adopting and promoting energy sources that can meet the extensive energy requirements and save energy for future generations (Ali et al. 2021). Developing countries face a major challenge of energy poverty for men and women; approximately 1.2 billion people live without household electricity (Gray et al. 2018). Green technology positively impacts cleaner production and renewable energy (RE) projects. Currently, every country can increase the speed of its success by proficiently maintaining its demand and supply of energy (Iqbal et al. 2018). Solar power Europe has determined that the global solar capacity will reach 900 GW by the end of 2021.
Moreover, the grid-connected solar power capacities will reach 1870 GW by 2025. Under optimal conditions, the world can operate photovoltaic (PV) generation plant capacity, with 2147 at the end of 2025 (IRENA 2020). The present study is a contribution of SHSs to the success of small-scale enterprises in the country’s economy. Solar PV and wind turbine renewable electricity generation is a cheaper energy source. Project costs of both energy sources are estimated to decrease significantly in the future. Pakistan has expensive hydropower and thermal plants (Global solar council 2021). Compared with all other alternative energy sources, solar energy is considered robust, active, and available proceeding the Earth matched to bases (Nicholas and Buckley 2018). Countrywide, Pakistan has a solar isolation of 5.5 Wh m−2 day−1 and an annual mean sunlight duration of 8–10 h day−1. In the coastal parts of Baluchistan and Sindh, the wind speed ranges from 5 to 7 m s−1 (Ali et al. 2022a; Ghafoor et al. 2016). The average cost of installing a solar panel with a small capacity is between PRs. 8000 and PRs. 13,599. Another cost associated with solar panels is their breakdown. In Pakistan, the prices for 410-W solar panels, 540-W solar panels, 485-W solar panels, and 250-W mono solar panels are PKR18,900, 23,750, 22,800, and 14,000, respectively. You can operate without difficulty installing a 5-kW solar system. A 5-kW solar system may generate up to 16 to 20 kWh per day, which equates to approximately 500 to 600 kWh per month. A 10-kW solar system can generate approximately 1100 to 1200 power each month, or 36 to 40 units per day.
In contrast, solar systems have variable pricing, such as 5 kW for 90,000 PKR and 10 kW for 125,000 PKR including the Green Meter. The financial benefit of solar energy is immense significant savings on electricity bills, solar panel components require zero maintenance, residential solar system installations can increase property value, residential solar system owners are eligible for government incentives, and solar home systems enable small-scale businesses to generate their electricity and reduce their reliance on the grid. RE is a cost-effective energy source, such as SHSs, which can bring high profitability to small-scale businesses due to low-cost energy sources (Yang et al. 2022). Table 1 compares fossil fuel-based energy cost/unit and RE-based cost/unit.Table 1 Per unit energy cost comparison
Households and selected businesses Fossil fuel-based energy cost/unit RE-based cost/unit SHS energy-based cost/unit
Households 25.40 PKR per kWh 15.20 PKR per kWh 12.25 PKR per kWh
Furniture manufacturing 41.5 PKR per kWh 21.35 PKR per kWh 17.50 PKR per kWh
Food bakeries 42.5 21.35 17.50
Book making sector 44.5 21.35 17.50
Fast food making sector 40.5 21.35 17.50
Hotel industry 43.5 21.35 17.50
The sustainable development of solar energy projects requires assessing green financing and COVID-19 risks to promote the small-scale industry in the country. There is a need to adopt a solar home system (SHS) to develop small-scale industries in Pakistan. The government needs to improve and develop SHSs for the solid financial performance of the small-scale industry. The energy demand is increasing, which is considered one of the world’s decisive complications (Awan and Knight 2020). The performance is needed to measure SHSs regarding low-cost energy and enhanced profitability for small-scale industries in Pakistan. It is necessary to conduct studies that focus on low-cost-based energy. The energy shortfall seriously affects non-professional and professional activities and personal lives (Rafique and Rehman 2017). The governments of developing countries have focused on policy strategies to expand and facilitate renewable energy.
Like other underdeveloped countries, massive power is required to support industry and the large population in Pakistan (NEPRA 2018). Pakistan’s electricity demand and supply gap have been uncontrolled in the past few years. In Pakistan, policymakers need to adopt and support those RE sources that can increase the energy demand by replacing fossil fuel-based energy such as oil, gas, and coal. Energy poverty mitigation is inspired by the low-carbon energy revolution, energy efficiency, affordability, clean energy consumption, and reduced poverty under energy availability (Abdelhady 2021). The unexpected or expected fluctuations can be mitigated through adverse impacts on neighboring markets’ natural gas and oil (Wu et al. 2021a, b). The current study discussed the RE policy-level problems. The country needs to provide an uninterrupted energy supply to its citizens. Current energy production sources in Pakistan are insufficient to meet the country’s rising energy demands. A reliable energy supply can play an active part in developing the financial position through industry progress. Proper policy guidelines are necessary to adopt solar projects in the industrial sector. Several economic and other risk factors must be accommodated and adjusted with these guidelines (Elavarasan et al. 2021).
To ensure sustainable economic growth in Pakistan, conservative and conventional energy generation technologies must replace renewable energy sources. The RE policy of Pakistan has many positive features. However, the facts show that all these features cannot justify Pakistan’s vigorous and significant energy growth. The RE policy has failed to address significant issues effectively, even though some have been accepted in the policy framework. Still, RE policy is facing challenges in the country. One of the reasons for to slow acceptance of renewable energy technology (RET) is the lack of competition with conventional power generation in the country. The present study addresses the policy issue of lack of competition with conventional power generation due to the high capital cost of RE projects. The low installation cost of capital for solar power projects can attract investors and beat the economic comparisons. We expect the government to reduce the subsidy from conventional power generation and provide to the RE sector for small-scale business enterprises. The study has prodigious importance to the investors of Pakistan and the developing world because the study suggestions can play an energetic character in enhancing the concert of small-scale industries by adopting low-cost energy sources (SHS).
In this study, we assessed the performance of SHSs regarding low-cost energy and enhanced profitability for small-scale industries in Pakistan. It is necessary to conduct studies that focus on low-cost-based energy. The owners of small-scale industries face a heavy burden due to electricity bills. They even cannot afford this expensive source of electricity and looking for an affordable source of energy. We evaluate the impact of SHSs on small-scale enterprises for their betterment in profit, performance, and cost-related satisfaction. The study has explored that SHSs positively and significantly affect small-scale industries to increase their profit and performance in the study area. The main object of the study is to create awareness and cost affordability regarding SHSs for small-scale industry owners so that they can play an active part in the national economy. The owners of small-scale enterprises are looking for a low-cost and affordable energy source in Pakistan. An SHS is a valuable source to condense the abovementioned energy supply and demand gap. It is also beneficial for the household’s energy demand in remote areas and the small-scale industries located in the zones either connected or not connected with the grid. These industries include where the production, manufacturing, and rendering of services assume a micro or small scale. Our study examines the contribution of SHSs to the success of small-scale enterprises in the country’s economy. Pakistan is an emerging economy in Asia. It is a highly diverse economy. The small-scale industries cover 90% of the businesses in Pakistan, and they employ 83% of the workforce in the manufacturing industries. The central portion of the economy consists of small-scale established industries carried by single persons or partners in limited numbers. In the modern age, where electrification through the grid is not possible or minor, the introduction of SHSs has revolutionized. It is only valuable for households, but it has also made an outstanding contribution to the functioning and success of small-scale enterprises. SHS is the cheapest energy source, encouraged in small-scale industries with weak financial positions. Small-scale enterprises successfully achieve better performance because of their low-cost energy, enhanced energy supply, and durability (Duan et al. 2021). The present study is the first one that was conducted to evaluate the performance of SHSs to increase the profitability of the small-scale industry in Pakistan.
Many research studies have been conducted on the energy crisis (Dong et al. 2021; Hao et al. 2023; Wu et al. 2021a, b; Yang et al. 2021), solar energy site selections, and solar energy projects (Ali et al. 2022a; Rena et al. 2019). Still, the literature has a gap concerning adopting solar home systems (SHSs) to develop the small-scale industry. There is a lack of literature regarding the impact of SHSs on increasing the financial performance of the small-scale industry. In Pakistan, previous studies (Iqbal et al. 2018) concerning alternative energy have not focused on this issue primarily; they pin-pointed (i) sources of alternative energy generation, (ii) the direction of the upcoming energy sector, (iii) introduction of alternative energy with arguments, (iv) valuation of the whole energy subdivision in the nation-state, (v) energy combination or assortment, and (vi) demand and supply gap of energy. Although the previous researchers have taken a keen interest in the energy sector, all these studies have some specific gaps, e.g., (i) encouraging the adoption of a solar home system to increase the financial performance of the small-scale industry in Pakistan, (ii) the sustainable development of the local small-scale industry through adopting solar home systems, (iii) it is the alternative of electricity issues to adopting solar home systems to increase the performance of the small-scale or household industry, and (iv) there is a need for green investment through the solar home system to increase the financial performance of the small-scale industry in the country. Due to current study openings, the present research needs to discourse the subsequent queries for research contribution: (i) evaluate the importance of green investment in the solar home system to increase the financial performance and development of the small-scale industry in Pakistan, (ii) highlight the significance of green investment to adopt the solar home system for the development of the small-scale industry, and (iii) empirically explore the moderating part of awareness and understanding to adopt the solar home systems for the financial performance of the small-scale industry.
This research paper is categorized into sections: “Literature review” section discusses the literature review. “Formulation of hypotheses” section describes the formation of hypotheses with a theoretical background that reflects the research methodology. “Analysis and results” section defines data analysis with the results. “Discussion” section discusses. Finally, “Conclusions and policy implications” section summarizes the study, including limitations and future directions.
Literature review
Reliable energy resources are needed in modern life to perform routine activities (Dogan et al. 2020; Irfan et al. 2022; Isik et al. 2021); however, Pakistan’s energy sector faces a severe energy crisis, destroying the national economy. The expert and non-expert activities of the people of Pakistan are disturbing due to a severe energy shortage. The introduction of SHSs is evident in numerous industries and households in areas not connected to the electricity grid or where electrification through the grid is limited in benefits. The SHS helps run lower-power appliances with a 12 V direct current (D.C.). Some power conditioners/inverters change the 12/24 V power into 240 V alternating current (VAC) power and help to run larger power appliances. There is a need to adopt a solar home system (SHS) to develop small-scale industries in Pakistan. The government needs to improve and develop SHSs for the solid financial performance of the small-scale industry. The increasing importance of SHSs as an energy source in industry households forces researchers and scholars to make SHSs their topic of discussion and debate about the appliances such as heaters, lights, small plants, TVs, and computers; SHSs have an energy capacity of 5 h a day (GoP. government of Pakistan 2021). Scholars such as Ullah et al. (2020) have analyzed the contribution of SHSs to the operational and financial performance of small-scale industries with fewer investments in business functions. Our study examines the contribution of low-cost energy of SHSs, enhanced energy through SHSs, and the quality of SHSs to the performance of small-scale industries.
Solar home systems (SHSs) are the best energy source from this point of view. SHS is an impartial photovoltaic energy system that presents a cost-effective way of supplying a large amount of power for diverse appliances, such as lighting, to off-grid households in remote areas. In the rural areas where the electric grid is still unable to reach, SHSs can apply to meet household energy requirements, meeting fundamental electric needs. Globally, SHSs are a source of power to numerous homes in remote localities where electrification through the grid is not possible or limited (Jan et al. 2021). Cost occurs at installation or repair, but there is no periodical cost that the enterprises have to bear for operating electric appliances in business operations. Millions of SHSs have been installed for increasing electrification in the rural areas of Latin America, Asia, and Africa. Worldwide studies support installing SHSs to increase the electricity supply for rural areas at a low cost, which is a significant challenge (SAARC 2020). The SHS is needed worldwide to enhance prosperity and change the economic position of the rural population. The energy generated through SHSs is inexpensive as, in the case of installing solar panels to meet the energy requirements, power can be attained free of charge. The stand-alone rooftop SHSs can minimize energy costs. Rooftop SHSs can decrease the total cost of energy, and the per-unit cost is around USD 0.211, and there are no additional fuel charges. Installing standalone rooftop SHSs can earn an income within a few years. SHSs can reduce fossil fuel load pressure, minimize greenhouse gas emissions, and promote a green environment (Khan et al. 2005). RE is a cost-effective energy source, such as SHSs, which can bring high profitability to small-scale businesses due to low-cost energy sources.
The low-cost energy of SHSs associated with the performance of small-scale industries is checked in the study (Ahmad and Balkhyour 2020; Khattak et al. 2006). The SHSs have an essential role in the sustainable energy supply using microfinance in the poor rural areas of Myanmar and Bangladesh. The SHS is economically efficient in enhancing the electricity supply and has achieved a 15–18% renewable energy installed capacity goal by 2020 (Balkhyour et al. 2018). Consequently, the enhanced energy supply through SHSs enables small-scale industries to show better operational and financial performance (Mujaddad and Ahmad 2016). The RE policy has failed to address significant issues effectively, even though some have been accepted in the policy framework. Still, RE policy is facing challenges in the country. One of the reasons for to slow acceptance of renewable energy technology (RET) is the lack of competition with conventional power generation in the country (Ali et al. 2022c). The invention of the solar energy storage system has enhanced the energy supply through SHSs. This enhanced energy is used in small-scale industries to run diverse electric appliances and make their functions smooth and feasible. Their activities, such as operational and financial performance, have improved (Perpiña et al. 2020). Energy generation through SHSs is proven low-cost energy and can enhance the country’s economic position of small-scale industry and sustainable economic development.
The quality of SHSs and the performance of small-scale industries are excellent contributions to the literature. However, SHS information enables the managers to acquire quality solar panels and related instruments which offer enough energy to meet internal and external energy needs. The present study addresses the policy issue of lack of competition with conventional power generation due to the high capital cost of RE projects. The low installation cost of capital for solar power projects can attract investors and beat the economic comparisons. The awareness of SHS products is high, and the education level is high, but there is a massive gap between the perceived price of SHSs and the market price of SHSs in Uttar Pradesh, India (Kamran 2018). The quality of SHSs can enhance the satisfaction level of small-scale business owners and investors; the country’s government needs to ensure the quality of the SHS. A study (Adwek et al. 2020) demonstrates that awareness and understanding of the SHS can assist professionals in installing the panels. The information exploits them to obtain a lot of solar energy from the sun’s waves in the minimum time. Thus, the business processes can run smoothly without any hurdles and with great effectiveness as there is enough energy to run technologies. 60.3% of respondents are aware of their solar systems and their maintenance, but 76% are aware of nonfunctional solar systems in Nigeria (Rafique et al. 2020). The owners of small-scale industries face a heavy burden due to electricity bills. They even cannot afford this expensive source of electricity and looking for an affordable source of energy. We evaluate the impact of SHSs on small-scale enterprises for their betterment in profit, performance, and cost-related satisfaction. The study has explored that SHSs positively and significantly affect small-scale industries to increase their profit and performance in the study area. The awareness of the SHS in operational activities is essential for small-scale businesses to enhance profitability, which is also proved in the literature.
A study (Khan et al. 2020) posits that the awareness and understanding of business needs, installation, features, and working determines the quality of the solar panels applied for domestic or commercial use. This study (NEPRA 2018) expresses the process of controlling the capacity of SHSs, and the system operating knowledge can improve the speed and durability of SHSs. The durability and improved efficiency of the SHS can enhance the output of small-scale enterprises and business effectiveness by achieving competitive advantages over rivals. A study (Zubi et al. 2017) expresses that small-scale industries’ production, operation, and financial performance depend on SHSs. This awareness helps make decisions regarding the procurement of solar panels and related tools, their installation, and their usage to meet energy needs. In this case, the quality of the SHS improves and increases the effectiveness of the business. The low-cost and transparent policy encourages the general public to adopt SHS significantly and favorably (Ali et al. 2022b). Pakistan is experiencing a plain energy disaster, which consumes detrimental effects on the domestic budget. The energy provided through SHSs is a low-cost and profit-earning energy for small-scale industries. It ultimately plays an extraordinary role in the country’s sustainable economic development. Past studies such as Ali et al. (2022b) and Ali et al. (2022a) show the moderator effect of awareness and understanding of underlying factors in the performance of small-scale industries. Small-scale industries are divided into manufacturing and ancillary industries; other than these kinds of small-scale industries that have continued their professional for countless ages and are from diverse productions, exploration the furniture manufacturing, food bakeries, book making sector, fast food making sector, and hotel industry.
The main object of the study is to create awareness and cost affordability regarding SHSs for small-scale industry owners so that they can play a dynamic character in the nation’s economy. The financial resources of these industries are limited. The installation of SHSs proves beneficial to small-scale industries in achieving high operational and financial performance. This energy source keeps their business activities fluent, reduces energy expenses and per-unit costs, and has more profitability in operating results (Barman et al. 2017). The small-scale industry can boost up by adopting SHSs due to cost-effective energy sources with low capital costs when installing solar projects. The government needs to effect change in RE policy by the government to develop the small-scale industry. Adopting an SHS benefits the owner, including lower kerosene consumption and providing health and financial benefits to business and household women (López-vargas et al. 2021). Different SHSs have different qualities and features, including high quality, high durability, low maintenance requirements, high voltage power, and significant safety for health. The system can convert solar power into electric or thermal power fast and store solar energy for future needs, which is helpful in small-scale industries to achieve high operational and financial performance (Hossain et al. 2019). Full-service and plug-and-play SHSs offer high quality, while component-based match-and-mix systems offer low quality at a low cost (Charles et al. 2018). Our study aims to address the low-cost energy of SHSs, enhanced quality, profitability, and power supply as determiners of small-scale industries’ operational and financial performance.
The SHS provides energy at a low cost to small-scale industries; thus, it helps meet their energy needs on a minimum budget and improves overall performance. Similarly, the enhanced energy supply through SHSs meets the energy needs of small-scale industries, gives permanent survival, and helps them carry out their functions smoothly. High-quality SHS improves small-scale industries’ functioning quality and financial performance (Shaughnessy et al. 2019). Solar photovoltaic systems impact the lives of individuals and small-scale businesses, including the health of individuals, the education of children, social relations, livelihoods, and businesses’ financial position (Newcombe and Ko 2017). The energy produced through SHSs is an inexpensive source of energy that can change the economic position of the people living in rural areas and the owner of small-scale industries. The SHS can provide 100% rural electrification for telecommunication, lighting, and self-support business with the help of tontine financing. The tontine is a concept in which SHSs are delivered to the far-off rural areas of developing countries. The primary purpose of this tool is to provide low-cost electricity through SHSs, replacing kerosene lamps and candles, but not with the help of banks and savings (Khan 2019). Similarly, the excellent quality of SHSs can meet energy requirements in a better way and helps to save money, minimize maintenance costs, and provide health safety. It reduces the small-scale industries’ per-unit cost and increases their profitability and financial performance (Charles et al. 2018).
SHSs are not required to pay any bill on electricity generated through the application of SHSs as we are bound to pay monthly bills using electricity meters. In applying SHSs, small-scale industries with limited financial sources are better positioned to handle all financial matters (Narayan et al. 2018). The SHS is an attractive energy source for business owners who perform extra-earning activities such as mobile phones or mobile charging businesses. SHS can play a vital role in increasing energy access and reducing dependency on fossil fuels in Bangladesh. The installed SHS reduces energy costs and CO2 emissions (Adetona et al. 2020). It is not costly as it uses a natural source to generate energy instead of expensive raw material. Installing the solar energy system (SHS) requires money; afterward, it does not require much maintenance. We need to clean it only twice a year. We do not have to bear expenses or some billing as we have to bear in applying other energy sources. The amount saved in this way can be used in small-scale industries in other different business operations. Thus, installing an SHS improves small-scale industries’ financial position and performance (Gandini and Almeida 2017). In light of past studies, we can say that SHS is the best energy source that is inexpensive and cost-effective for small-scale industries and Punjab’s rural areas. This energy source is the best choice for small-scale industries to increase their profit and surviving period and increase sustainable development in the country.
The benefits of SHSs include a higher level of satisfaction for the business’s profit, increasing the quality of SHS equipment, and increased children’s study time (Ali et al. 2019). Electrification through SHSs provides direct and indirect benefits to small-scale business owners, such as reducing the use of kerosene and increasing the income of traditional fuel users due to converting solar light. The SHS electrification provides overall satisfaction for small-scale industries and comfort to the users (Mandal et al. 2018). This study has proposed a guideline to the regulation developing authorities that more attention is needed to focus on SHS usage and improve the performance of the small-scale industry. The SHSs can increase sustainability and economic viability through suitable social awareness, local technician availability, and influential energy committees, which positively and significantly affect small-scale businesses. Small-scale business owners have improved the productivity of their businesses by adopting SHSs for light and production purposes at night (Yadav et al. 2019).
The research study (Opiyo 2019) shows the advantages of SHSs in the small-scale industry. This study implies that SHSs generate electricity or heat to run appliances requiring a low power voltage. After installing an SHS, no periodical payments, daily or monthly, must be made. Thus, the energy generated through an SHS is low cost, and the money saved can be used in the business operation in another way. The healthy and helpful debate of this literature shows that adopting SHSs is very profitable and can remove the poverty level in developing countries due to the satisfactory performance of the small-scale industry. Solar energy is considered an incredible potential energy source due to zero levels of emission production. Due to a reliable energy source, various devices can be connected using clean energy. The solar home system encourages the adoption of RE to increase the financial performance of the small-scale industry in Pakistan, and it can enhance the sustainable development of the local small-scale industry. SHSs are the alternative to electricity issues to increase the performance of the small-scale or household industry. There is a need for green investment through the solar home system to increase the financial performance of the small-scale industry in the country. Figure 1 indicates the conceptual framework and moderation effect based on the above discussion. The current study has used three predictors with moderating relationship testing. The results of the hypotheses are shown in the data analysis section in Table 5. The hypotheses include awareness and understanding of SHS (AU), enhanced energy supply through SHS (EES), low-cost energy (LCE), the performance of the small-scale industry (PSSI), and quality of SHS (QSHS). As a result, the performance of small-scale industries is high.Fig. 1 Conceptual framework
Formulation of hypotheses
Low-cost energy and performance of the small-scale industry
The energy generated through SHS is low cost, as in the case of the installation of solar panels to meet the energy requirements, energy can be attained free of cost. We are not required to pay any bill for using electricity generated through the application of SHS as we are bound to pay monthly bills using electricity meters. With the application of SHS, small-scale industries with limited financial sources are in a better position to handle all their financial matters (Halder 2016). SHS is a renewable energy source. It takes energy from the sun and generates electricity (photovoltaic) and heat (solar thermal), which can run different appliances. It is not costly as it uses a natural source to generate energy instead of any costly raw material. Installing a solar energy system (SHS) requires money, and afterward, it does not require much maintenance (Ali et al. 2021). We need to clean it only twice a year. We do not have to bear expenses or some billing as we have to bear in case of applying other energy sources. The amount saved in this way can be used in small-scale industries in other business operations. Thus, installing SHS improves small-scale industries’ financial position and performance (Diouf and Avis 2019).H1: The low-cost energy of SHS has an optimistic association with the concert of small-scale industries.
Energy supply through SHS and performance of the small-scale industry
The acquisition of energy through SHS has enabled small-scale industries in remote off-grid areas or areas where the electricity through the grid is less supplied to meet the energy needs and run their business activities smoothly. Different kinds of technology requiring a low voltage can be run through the power generated through SHS, such as lightning appliances, fans, water pumps, heaters, computers, and printers. In small-scale industries, electric appliances requiring a low power voltage are mainly used to carry the business activities. Thus, the enhanced energy supply through SHS enables small-scale industries to show better operational and financial performance (Charles et al. 2018). As the world is getting modern with time, such solar panels are being introduced into the market, which can provide relatively high voltage power and better meet the energy requirements of small-scale, financially weak industries compared to grand-scale industries. Moreover, some well-known companies have introduced solar energy storage systems to complement the SHS. The invention of the solar energy storage system has enhanced the energy supply through SHS. This enhanced energy is used in small-scale industries to run diverse electric appliances and make their functions smooth and feasible. Their operational and financial performance has improved (Yadav et al. 2020). Thus:H2: Enhanced energy supply through SHS positively affects the performance of small-scale industries.
Quality of SHS and performance of the small-scale industry
Different SHSs have different qualities like durability, maintenance requirement, voltage power generation, speed to convert solar power into electric power, health safety, and the attachment of solar energy storage systems. The SHSs have high quality; in other words, the SHSs have high durability, low maintenance requirements, provision high voltage power, are highly safe for health, can convert solar power into electric or thermal power fast, and store solar energy for future needs, are helpful in small-scale industries to achieve high operational and financial performance (Chowdhury and Mourshed 2016). Suppose the applied SHS can generate high voltage power and convert the solar power into electric power fast. In that case, small-scale industries can perform business activities fast and respond to market requirements. If the SHS can store the energy, the small-scale industries do not worry about the energy needs and can preplan their future activities. Similarly, a good quality SHS better meets the energy requirements and helps save money, minimize maintenance costs, and have healthy safety. It reduces the per-unit cost of small-scale industries and increases their profitability and financial performance (Newcombe and Ko 2017). Hence:H3: The quality of SHS is linked with the performance of small-scale industries in a positive manner.
The awareness and understanding of SHS and the performance of the small-scale industry
In the areas where people have sufficient awareness and understanding of SHS, small-scale industries can benefit from SHS and improve their operational, production, and financial performance. The awareness and understanding in the personnel of small-scale industries about the inventions, needs, installation, working, and quality of SHS result in harnessing low-cost energy from SHS, more enhanced energy supply through SHS, and in the application of better quality solar systems and the achievement of the superior performance of small-scale industries (Sovacool 2018). The small industries have awareness and understanding that SHS can use solar energy systems for energy purposes in such a way that they do not require repairing again and again and give less costly energy. The inexpensive energy through SHS reduces the small-scale industries’ expenses and improves profitability and overall performance of small-scale industries. Similarly, the awareness and understanding of SHS enable small-scale industries to install such panels. In this way, these solar panels give high voltage power in a large amount that can better meet industries’ energy requirements. When the personnel of small-scale industries has better awareness and understanding of SHS, they can purchase better quality solar panels, giving more energy at a low cost, whose installation and further maintenance do not require considerable amounts and have solar energy storage system (Abdullah et al. 2017). The quality of the SHS improves the operational and financial performance of small-scale industries.H4: The awareness and understanding of SHS are significant moderators between the low-cost energy of SHS and the performance of small-scale industries.
H5: The awareness and understanding of SHS are significant moderators between the enhanced energy through SHS and small-scale industries’ performance.
H6: The awareness and understanding of SHS are significant moderators between SHS quality and the performance of small-scale industries.
Research methodology
Purposive or non-probability sampling was used in the current study for the performance of the small-scale industry through SHSs in Pakistan. In the research study, this sampling method did not provide an equal opportunity for all population members. Purposive sampling is utilized for specific demographic features and pilot, qualitative, or exploratory research. There are five primary non-probability sampling strategies: quota sampling, snowball sampling, purposive sampling, voluntary response sampling, and convenience sampling. We have selected small-scale industries as a part of this study on an SHS’s performance and quality. The study adopted specific criteria to present our sample of small-scale businesses in Pakistan: (i) include small-scale industries that have continued their business for many years and are from four cities (Lahore, Multan, Faisalabad, and Gujranwala), Pakistan. This research study includes diverse industries, including furniture manufacturing, food bakeries, bookmaking, fast food making, and hotel industries. A hierarchy was followed while selecting small-scale businesses where a minimum of five and a maximum of nineteen employees are working in the chosen industries (Stojanovski et al. 2017). (ii) The small-scale industries must have a 2.5 lac per day turnover in RMB. The authors surveyed from April to September (2021) to complete the research purpose; there was a fourth wave of coronavirus (COVID-19), named the delta variant virus, in Pakistan (Chowdhury and Mourshed 2016; Komatsu et al. 2013). It was challenging to obtain information from related respondents due to the pandemic situation, and mobile applications (e.g., WhatsApp, Facebook, and LinkedIn) were used to send the questionnaire.
Instruments and procedure
This study used questionnaires that comprised 23 measurement items and required a minor collection level of 115 (23*5) useful questionnaires. The past study (Groenewoudt et al. 2020) recommended the rule of thumb for the proposed sample size used in this study. In this regard, 570 questionnaires were distributed to small industry owners using SHSs, 487 were returned, and 357 were useable for analysis. The response rate was 62.63%; however, the researchers deleted 83 surveys because of unmatched and poor responses. The demographic characteristics of the respondents, including age, experience, income, and gender, as well as their various backgrounds, give the appropriate response in this study (see Appendix, Table A1). In addition, all delegates have diverse business experience and demographic characteristics. In addition, a purposive sample was used to recruit respondents for this study from four cities in Pakistan, including respondents with varied cultural and behavioral backgrounds. Purposive sampling is used to access the entire population and fits theoretical generalization (Lemaire 2018). The ongoing research investigates the impact of SHSs on the performance of the small-scale industry. The other primary aim of this research includes investigating the moderating effect of awareness and understanding of SHS among the links of low-cost energy through SHS, enhanced energy supply through SHS, quality of SHS, and performance of small-scale industry of Pakistan. The study followed quantitative data collection methods and received the data using questionnaires. The owners of the small-scale industry sector use the solar system for production. Respondents were selected based on purposive sampling. The present study utilized the usual 5-category scale, in which one always represents and five never expresses. The first half of the questionnaire pertains to the respondent’s personal information. In contrast, the second piece is concerned with the characteristics of the SHS, such as quality, cost, and energy supply. In addition, the last segment of the questionnaire connected to the changes in the small-scale industry’s performance due to SHSs. The current research has taken three predictors, such as low-cost energy through SHS (LCESHS) with four items, enhanced energy supply through SHS (EESSHS) with six items, and quality of SHS (QSHS) with four things. In addition, the performance of the small-scale industry (PSSI) engaged as a reliant on adaptable with five items and awareness and understanding of SHS (AUSHS) has taken four things as moderators. These constructs, along with the nexus, are shown in Table 2.Table 2 Convergent validity
Constructs Items Loadings Alpha CR AVE
Awareness and understanding of SHS AU(SHS)1 0.885 0.845 0.820 0.536
AU(SHS)2 0.670
AU(SHS)3 0.677
AU(SHS)4 0.675
Enhanced energy supply through SHS EES(SHS)1 0.912 0.944 0.956 0.782
EES(SHS)2 0.823
EES(SHS)3 0.905
EES(SHS)4 0.904
EES(SHS)5 0.912
EES(SHS)6 0.846
Low-cost energy through SHS LCE(SHS)1 0.880 0.909 0.936 0.785
LCE(SHS)2 0.879
LCE(SHS)3 0.888
LCE(SHS)4 0.896
Performance of small-scale industry PSSI1 0.808 0.889 0.918 0.692
PSSI2 0.818
PSSI3 0.854
PSSI4 0.817
PSSI5 0.863
Quality of SHS QSHS1 0.958 0.966 0.975 0.908
QSHS2 0.943
QSHS3 0.952
QSHS4 0.959
Measurement instrument and variables
In this study, all items from previous research have been incorporated. Items constructed on awareness and understanding were constructed from the study (Diouf 2016). Things regarding low-cost energy were adopted (Bisaga et al. 2019). Items related to enhancing energy supply through SHSs were adopted (Halder 2016). Objects related to SHS quality were constructed based on Hoque and Das (2013). Items that belong to the performance of small-scale industries were adopted (Diouf and Avis 2019). For this purpose, two professional translators were hired; first, questionnaires were converted into the native language, i.e., Urdu. Given the proposal of 62, a second translator translated these statements from the native language into English. There were no substantial discrepancies in the translation. However, a few grammatical errors were discovered and corrected. Finally, data collection questionnaires were issued in the local language.
Analysis and results
Structural equation modeling (SEM) was used for data analysis in the present investigation (Mondal and Klein 2011). We use this approach to examine the interpersonal proportions because it is considered factor attentive (Azimoh et al. 2015). Many studies have adopted PLS-SEM; we have used this method based on the evidence of these subsequent studies (Abdullah et al. 2017; Aberilla et al. 2019). The traditional statistical analysis approach is less important than structural equation modeling (SEM). It can facilitate the statistical analysis concerning convenience, accuracy, and efficiency (Ali et al. 2014; Ghafoor et al. 2016). SEM is a second-generation technique that solves the issues of first-generation studies. SEM is a multivariate investigation method that may investigate numerous variables concurrently. SEM can instantaneously deal with several compound interactions in business research; thus, it is continuously popular (Pakistan economic survey 2021). There are two categories of well-known SEM techniques: first, partial least square (PLS)-SEM or variance-based SEM (VB-SBM); second, most minor absolute deviation (LAD)-SEM, covariance-based SEM (CB-SBM). Measurement and structural models are two analytical techniques in PLS-SEM that display measurement findings in two phases (Akbar et al. 2019). The measurement assessment model has two tests: evaluation of the inner model or dependability and validity. On the other hand, the structural assessment model has hypotheses testing or evaluating the outer model. The current study has applied PLS 3.0 software for significant data analysis. Furthermore, covariance-based structural equation modeling has low potential than partial least square path modeling. PLS-SEM can break relationships among the variables and is more beneficial.
Measurement assessment model
Reliability and validity tests are required for each measurement assessment model’s provided constructs. The measuring model confirms the validity and reliability. All of the factor loadings for model are approved. The measurement evaluation model requires item reliability tests and internal consistency reliability. The model includes convergent and discriminant validity under validity tests (Amado and Poggi 2014). The average variance extracted (AVE) can measure convergent validity, the composite reliability (CR) can measure internal consistency, and outer loading can measure item reliability. The threshold value of all item loadings is well up to 0.5 (Sarker et al. 2020) (see Table 2). The averaged factor loadings in this investigation were more significant than 0.50; the analysis provided it, and each observation can contribute to a constructed variable (Urban and Heydenrych 2015). All values of AVE exceed the recommended assessment of 0.5. The average value surpasses the composite reliability under the threshold value of 0.7; the measurements are accurate, as the study indicated (Brislin 1970). In the present research, the values of results designate that completely the standards of AVE are 0.536 (awareness and understanding of SHS) and 0.908 (quality of SHS), and CR values are between 0.820 (awareness and understanding) and 0.975 (quality of SHS). All additional loading values are between 0.5 and 0.959.
Table 2 displays the measurement model’s validated validity and reliability results. The present mode shows how the variables are loaded with factors. The significance of each factor loading value is greater than (0.50). The measurement assessment model shows the convergent rationality of entire substances is effective. The path analysis was facilitated among the variables. The study results have confirmed that low-cost energy through SHS, enhanced energy supply through SHS, and quality of SHS have positive nexus with the small-scale industry of Pakistan. Hypotheses H1, H2, and H3 are accepted. The outcomes likewise revealed that awareness and understanding of SHS suggestively moderated amid the links enhanced energy supply through SHS, quality of SHS, and performance of small-scale industry of Pakistan and accepted H5 and H6. However, the findings also exposed that awareness and understanding of SHS are insignificantly moderated among the energy of the low-cost link through SHS and performance of small-scale industry of Pakistan and rejected H4. The measurement model first demonstrated the nexus’s convergent validity among the items. According to the results, loadings and AVE are greater than 0.50, whereas composite reliability (CR) values and alpha are more significant than 0.70. These numbers show that the items’ convergent validity—their strong correlation—is valid.
The outcome segment has similarly exposed, in Table 3, the discriminant rationality of the variables’ interconnection. Firstly, the discriminant validity was checked through Fornell-Larcker and cross-loadings. The values reflecting the variables’ links are higher than those with other variables in Fig. 2, which shows these values. The results reveal that the discriminant validity is sound, and there is little correlation between the variables. Table 4’s bold values demonstrate that all components have both strong and weak correlations with one another. The bold values in the cross-loadings table are compared row-wise with other factors in the test discriminant validity. The AU(SHS) values are more significant than other factors row-wise and show strong discriminant validity. The bold numbers in Table 4 are smaller than the other left and correct values.Table 3 Fornell-Larcker analysis
AU(SHS) EES(SHS) LCE(SHS) PSSI QSHS
AU(SHS) 0.732
EES(SHS) 0.394 0.831
LCE(SHS) 0.335 0.385 0.824
PSSI 0.325 0.507 0.371 0.832
QSHS 0.726 0.470 0.384 0.408 0.841
N = 357; AU, awareness and understanding of SHS; EES, enhance energy supply through SHS; LCE, low-cost energy; PSSI, performance of small-scale industry; QSHS, quality of SHS
Fig. 2 Measurement model assessment
Table 4 Cross-loadings
AU(SHS) EES(SHS) LCE(SHS) PSSI QSHS
AU(SHS)1 0.885 0.458 0.372 0.378 0.828
AU(SHS)2 0.670 0.134 0.120 0.090 0.267
AU(SHS)3 0.677 0.081 0.109 0.099 0.264
AU(SHS)4 0.675 0.113 0.118 0.071 0.243
EES(SHS)1 0.331 0.912 0.340 0.447 0.391
EES(SHS)2 0.362 0.823 0.328 0.465 0.434
EES(SHS)3 0.356 0.905 0.345 0.446 0.445
EES(SHS)4 0.352 0.904 0.339 0.453 0.440
EES(SHS)5 0.327 0.912 0.336 0.434 0.381
EES(SHS)6 0.361 0.846 0.351 0.438 0.400
LCE(SHS)1 0.313 0.307 0.880 0.341 0.322
LCE(SHS)2 0.276 0.351 0.879 0.339 0.350
LCE(SHS)3 0.289 0.340 0.888 0.327 0.330
LCE(SHS)4 0.308 0.369 0.896 0.305 0.359
PSSI1 0.333 0.473 0.295 0.808 0.378
PSSI2 0.288 0.455 0.306 0.818 0.350
PSSI3 0.242 0.369 0.285 0.854 0.296
PSSI4 0.214 0.392 0.335 0.817 0.330
PSSI5 0.261 0.402 0.321 0.863 0.330
QSHS1 0.713 0.448 0.338 0.399 0.958
QSHS2 0.705 0.444 0.400 0.360 0.943
QSHS3 0.718 0.448 0.386 0.391 0.952
QSHS4 0.709 0.453 0.342 0.401 0.959
Bold data are square root of AVEs
Numerous scholars have questioned the Fornell-Larcker criterion. Therefore, the heterotrait-monotrait ratio of correlations (HTMT) is considered suitable for measuring discriminant validity (Wang et al. 2020). Less than 0.85 is the confirmed value of discriminant validity or 0.90 (Urbach and Ahlemann 2010). The total values in Table 5 are less than 0.90. The findings section also highlights how the factors of discriminant validity are related to one another. The values are highlighted in Table 5.Table 5 Discriminant validity heterotrait-monotrait (HTMT)
Factors AU(SHS) EES(SHS) LCE(SHS) PSSI QSHS
AU(SHS)
EES(SHS) 0.266
LCE(SHS) 0.249 0.416
PSSI 0.220 0.548 0.411
QSHS 0.537 0.492 0.419 0.436
N = 357; AU, awareness and understanding of SHS; EES, enhance energy supply through SHS; LCE, low-cost energy; PSSI, performance of small-scale industry; QSHS, quality of SHS
Structural assessment model
We measured the measurement assessment model; in the second round, the link between exogenous and endogenous factors is examined using the structural assessment model. The evaluation of the structural model is based on various statistical metrics, such as f2 shows effect size, Q2 indicating predictive relevance, R2 expressing the coefficient of determination, t values, and β values displaying path coefficient. The PLS-SEM literature includes evaluation criteria for hypotheses and path coefficient significance estimates (Wang et al. 2020). The bootstrapping procedure was active through 5000 sub-samples to test the significance level of the hypotheses, based on a 5% one-tailed significance level [81]. All hypotheses are accepted, which is discussed in the results. Low-cost energy (β = 0.326, t = 5.105 > 1.64, p < 0.05), low-cost energy association (moderator) (β = 0.095, t = 1.599 > 1.64, p < 0.05), enhance energy supply (β = − 0.239, t = 3.681 > 1.64, p < 0.05), enhance energy association (moderator) (β = 0.237, t = 2.604 > 1.64, p < 0.05), quality of SHS (β = 0.201, t = 3.514 > 1.64, p < 0.05), and quality of SHS association (moderator) (β = 0.189, t = 2.681 > 1.64, p < 0.05) indicated a substantial progressive influence on the performance of the small-scale industry.
The model has a considerable descriptive influence on enhancing the income level of the small-scale industry in Pakistan due to an R2 value of 0.462 for low-cost energy through SHS. However, the value of R2 has some difficulties in supporting the model, and we consider it ineffective and unsuitable based on the R2 value (Brislin 1970). Consequently, the model’s Q2 predicted relevance measurement is the most accurate. The latent exogenous standards are highly predictive and imply that the value of Q2 is more complex than zero. The value of Q2 is 0.242, indicating that the model has considerable predictive validity and that SHS can improve the performance of the small-scale sector. The f2 has typical values, including 0.02, 0.15, and 0.35, indicating modest, medium, and significant effects, respectively (Wang et al. 2020).
Consequently, the value of f2 assumes that effect size differs between moderate and significant (see Table 6). Table 6 is a listing of numerous statistical methods. Figure 3 depicts the structural assessment model, which suggests a significant association between the variables because the T values are greater than zero (1.64). Except for H4, all hypotheses are accepted. All the values of the moderated variables are positive, indicating that the structural evaluation model for solar projects in Pakistan reveals a significant link.Table 6 Hypotheses testing results
Hypothesis Relationship Beta S.D T-statistics p values Supported R2 Q2 F2
H1 LCE(SHS) → PSSI 0.326 0.064 3.514 0.000 Yes 0.462 0.242 0.097
H2 EES(SHS) PSSI 0.239 0.065 5.105 0.000 Yes 0.237 0.041
H3 QSHS → PSSI 0.201 0.057 2.604 0.009 Yes 0.116
H4 LCE(SHS)* AU(SHS) → PSSI 0.095 0.056 1.599 0.056 No 0.019
H5 EES(SHS)* AU(SHS) → PSSI 0.237 0.091 3.681 0.000 Yes 0.015
H6 QSHS* AU(SHS) → PSSI 0.189 0.067 2.681 0.004 Yes 0.483 0.021
N = 357; AU, awareness and understanding of SHS; EES, enhance energy supply through SHS; LCE, low-cost energy; PSSI, performance of small-scale industry; QSHS, quality of SHS. The moderating relationship has indicated by the asterisk among the variables
Fig. 3 Structural assessment model
The moderation results are indicated in Fig. 4, so a positive or negative relationship is indicated among moderated variables through the green line in the figures. It will be considered a positive relationship if the red and blue lines are below the green line. The moderate variable shows a negative relationship in Fig. 4 because the red and blue lines are above the green. Conversely, the moderate variables show a positive relationship in Fig. 5 because the red and blue lines are below the green line. There are progressive associations amid the moderate variables in Fig. 6 because the red and blue lines are below the green.Fig. 4 ESS(SHS)* AU(SHS)
Fig. 5 LCR(SHS)* AU(SHS)
Fig. 6 QSHS* AU(SHS)
Discussion
The research outcomes have revealed that the low-cost energy of an SHS has a positive relationship with the performance of the small-scale industry. The studies imply that the energy harnessed through solar panels is renewable and less costly. This sort of energy enables small-scale industries with limited financial sources to save the money spent to pay electricity bills. The money is saved and used in business operations and production procedures to meet the market requirements and improve performance. The past study (Amado and Poggi 2014) addresses the SHS as a significant energy source and profitable to small-scale industries. It implies that the solar panel system does not require much maintenance after installation at the beginning. The industries do not pay electricity bills because it is a personal energy source. Thus, the small-scale industries’ total cost and per-unit cost are low, and profitability is high. The results have also indicated that the enhanced energy supply through SHS positively affects the performance of small-scale industries. The results support the past study of Sarker et al. (2020), which shows that the application of the solar system for generating power has enhanced the energy supply, and the solar energy storage system is an excellent complement to it.
The improved energy produced from installing solar panels can use running technology even when electricity is unavailable to meet the power requirements. It creates fluency in the activities of small-scale enterprises and improves their performance. The results have shown that the SHS quality positively links with the performance of small-scale industries. The high quality of SHS reduces the cost and improves the functioning of small-scale enterprises. These results are in line with past studies (Urban and Heydenrych 2015), which show that the SHS had good quality, such as durability, lack of maintenance requirements, high generation of energy, and the attachment of solar energy storage system, which is helpful in small-scale industries to achieve higher performance. The study results have indicated that the awareness and understanding of SHSs play a significant moderating role between the low-cost energy of SHSs and the small-scale industry’s performance. These results align with the scholarly article of Urban and Heydenrych (2015), which shows that the influences of low-cost energy of SHSs on the performance of small-scale industries become more robust in the case of high awareness and understanding of SHS. The research outcomes have similarly indicated that the awareness and understanding of SHSs is a substantial arbitrator amid enhanced energy supply through SHSs and the performance of small-scale industries. These outcomes align with the previous study (Brislin 1970), which designates that the awareness and understanding of SHSs strengthen the influences of enhanced supply through SHSs on the performance of small-scale industries.
This work has both empirical and theoretical consequences. It is a substantial fictional work and participates in literature economically. The study contributes to the literature with the influence of three new factors such as LCE, EES, PSSI, and QSHS, to increase the performance of the small-scale industry. The present research work is a guideline for private investors, household families, and government officials in alternative energy to embrace SHSs for better output and financial performance for the small-scale industry. The study has prodigious importance to the investors of Pakistan and the developing world because the study suggestions can play an active part in enhancing the concert of small-scale industries by adopting low-cost energy sources (SHS). The easy adoption of SHS can decrease production costs and improve the financial performance of the small-scale industry in Pakistan. Hence, adopting SHS can increase low-cost energy, decrease the drive of disaster, and progress Pakistan’s small-scale industry’s economic performance.
Consequently, the study indicates that solar home systems have a positive and significant correlation with the increasing performance of the small-scale industry in Pakistan. The study also investigates how the energy cost of the small-scale industry can reduce through SHSs in Pakistan. The existing research has shown that SHSs are substantial for low energy cost, increasing off-grid energy levels, overcoming the energy shortfall, positive change in lifestyle, and better financial performance of the small-scale industry in the country. The R2 value for low-cost energy through SHSs, as reported in Table 6, is 0.462, indicating that the model has strong explanatory power for enhancing the performance of small-scale manufacturing in Pakistan. The model has substantial predictive relevance since the results indicate that the value of Q2 is 0.242, which shows that the performance of the small-scale sector can be improved using SHSs. Figure 5 depicts the structural assessment model, which suggests a significant association between the variables because the T values are greater than zero (1.64). Except for H4, all hypotheses are accepted. All moderated variable values are positive and show a meaningful link in the structural evaluation model for solar energy projects in Pakistan. The moderated variables indicated positive signs and showed an entirely substantial relationship in the structural assessment model for the small-scale industry in Pakistan. Finally, the present study expressed that SHS and its adopted moderation have a significant and positive impact on the financial performance of the small-scale industry in Pakistan.
Moreover, the results indicate that the awareness and understanding of SHS are significant moderators between SHS quality and the small-scale industry’s performance. These results align with the past study (Brislin 1970), which shows that the awareness and understanding of SHS improve the quality of SHSs applied and the performance of small-scale industries and their mutual association between the quality of SHSs and the small-scale industry’s performance. The present study suggested to the investors that low-cost energy through SHSs is advantageous for the development and high performance of the small-scale industry. The SHS is an easy and low-cost energy source and has significant and positive links to increasing the financial performance of the small-scale industry in Pakistan.
Conclusions and policy implications
The study examines the role of SHS in the economy, especially in small-scale industries. It deals with the influences of three factors: the low-cost energy of SHS, the enhanced energy supply through SHS, and the quality of SHS on the performance of small-scale industries. The study examines that SHS is a cheap way of generating energy. It also helps the small-scale industries control other energy costs; thus, the per-unit cost of the small-scale industries is low, and profitability is high. The study states that, in the modern age, where the energy requirements have increased, the solar energy system is the cheap but sole way to obtain the energy to meet the excessive requirements. Due to the application of the SHSs, the need for power can be completed quickly, and it creates fluency in the operational and production processes. The quality of an SHS affects the operational and financial performance of small-scale industries. The study shows that an SHS has good quality, including high durability, fewer maintenance requirements, a large amount of energy generated from it, and the availability of an excellent solar energy storage system. It proves to be helpful to small-scale industries to achieve higher performance.
The current study examines the influences of the awareness and understanding of the personnel of small-scale enterprises about the nature and installation of SHS, making it possible for them to apply, install, and maintain SHSs. This awareness and understanding of SHSs enable the three factors, such as the low-cost energy of SHS, the enhanced energy supply through SHS, and the quality of SHS, to improve the performance of small-scale industries. The study results have confirmed that low-cost energy through SHS, enhanced energy supply through SHS, and quality of SHS have positive nexus with the small-scale industry of Pakistan. Hypotheses H1, H2, and H3 are accepted. The outcomes similarly revealed that awareness and understanding of SHS suggestively moderated amid the links enhanced energy supply through SHS, quality of SHS, and performance of small-scale industry of Pakistan and accepted H5 and H6. However, the findings also exposed that awareness and understanding of SHS insignificantly moderated the energy of the low-cost link through SHS and the performance of the small-scale industry of Pakistan and rejected H4.
The results of this study displayed that the enhanced energy supply through SHS is positively associated with the output and economic performance of the small-scale industry. The quality of SHS is an effective instrument that facilitates investors and improves the performance of the small-scale industry in Pakistan. The findings of this study offer valuable guidelines for regulators and policymakers to adopt the solar home system to develop the small-scale industry. The relevant renewable energy institutions and competent authorities need to consider low-cost and low CO2 emission energy sources such as SHSs to develop small-scale industries in Pakistan. This study adds literature to introduce the awareness and understanding of SHS as a moderator between the low-cost energy of SHS, enhanced power through SHS, the quality of SHS, and the performance of small-scale industries. The study suggests that the performance of the small-scale industries can enhance with the low-cost energy from SHS, enhanced power through SHS, and increase profit with better quality of SHS. This study also implies how the influences of these variables mentioned above on small-scale industries’ performance improve if the individuals have great awareness and understanding of SHS.
Though the study has a practical implementation, it also has several limitations, which scholars must explore in the future. First, the study has addressed only four factors, such as the low-cost energy of SHS, enhanced power through SHS, and quality of SHS, and checks their influences on the performance of small-scale industries. Many factors other than these also substantially affect small-scale industries’ performance. Scholars must also analyze other factors except for the four elements above with the performance of small-scale industries. The data collected through a single source supporting this study do not give comprehensive data per the requirements. For better validity, in the future, scholars recommend applying more than one source to collect data while replicating the current study.
Appendix
Table Table A1 Different questions and behavior of respondents was mentioned during the semi-structured interview
Part A: Demographic characteristics of the participating respondents
Range Options Frequency Percentage
Gender Male 271 75.9
Female 86 24.0
Age Less than 25 111 31.09
26–32 51 14.28
33–42 47 13.16
43–57 57 15.96
58–65 48 13.44
66 and above 43 11.48
Income of small-scale enterprises owners Less than 250,000 81 22.68
250,001–275,000 79 22.12
275,001–300,000 85 23.80
300,001–355,000 56 15.68
355,001 and above 56 15.68
Selected small-scale industries Furniture manufacturing 61 17.08
Food bakeries 83 23.24
Book making sector 78 21.84
Fast food making sector 84 23.52
Hotel industry 51 14.28
Owners of solar panels 2–3 145 40.61
4–5 107 29.97
6–7 71 19.88
8–9 63 17.64
10 and above 34 0.095
Solar panel brand name adopted by the owners Jinko Eagle 72 H.M. G2 113 31.65
Trina solar TSM 103 28.85
Sun power × 22 64 17.92
Ja solar MR series 51 14.28
Hanwha Q cell Speak Duo 43 12.04
Part B: Influencing factors of solar home systems
Variables Items Questions Percentage
Low-cost energy through SHS LCE1 Solar home systems are the low-cost energy source for small-scale industry 29.3
LCE2 Performance of small-scale industry can increase adopting solar home systems due to low-cost energy 26.4
LCE3 Small-scale industry by adopting solar home systems can support the development of solar energy 27.1
LCE4 There is a need to adopt solar home systems for the low cost of the products produced by the small-scale industry 17.2
Enhance energy supply through SHS EES1 Adopting solar home systems can enhance the energy supply for local people and a small-scale industry 26.1
EES2 Solar home systems can eliminate the energy crisis by increasing the level of energy supply 23.2
EES3 There is a need to adopt solar home systems to increase energy supply at a low cost for local industry 13.1
EES4 Energy supply through solar home systems can support to minimize the conventional energy expense 18.3
EE5 Solar home systems are the reliable energy source with low maintenance cost 11.3
EE6 Small-scale industry is required to adopt solar home systems for saving utility bills and earning money 8.0
Quality of the solar home system QSHS1 There is needed a set of quality standards using certification procedure for the performance of the solar home system 33.7
QSHS2 Solar home systems are required to ensure high quality with durability and reliability for small-scale industry 34.9
QSHS3 Government should monitor private and government-based solar organizations for quality assurance of the solar home systems 20.3
QSHS4 Solar organizations are required to set a specific quality assurance for the components of solar home systems 11.3
Performance of small-scale industry PSSI1 Small-scale industry can significantly contribute to economic development through energy supply of solar home system 27.4
PSSI2 Small-scale industry can also be an essential source of foreign exchange earnings by adopting solar home systems 20.2
PSSI3 Promotion of small-scale industry can help in reducing the problem of widespread unemployment using energy produced by solar home systems 18.1
PSSI4 There is a need to increase exports and reduce imports of goods through the high performance of small-scale industry by adopting solar home systems 23.1
PSSI5 Small-scale industry can play an essential role in gross domestic products with the energy source of solar home systems 11.2
Awareness and understanding of the solar home system AUSHS1 Awareness and understanding of solar home systems is a crucial factor for solar companies but suitable for users 26.3
AUSHS2 There is a need to encourage solar home systems by providing awareness and promoting material 18.3
AUSHS3 There is a need to educate rural people about the installation and maintenance of solar home systems for the improvement of small-scale industry 27.2
AUSHS4 Awareness and understanding are most effective spread through word of mouth for solar home systems 28.2
7.
Author contribution
S.A.: conceptualization, writing—original draft, formal analysis, data handling, variable construction, and methodology. Q.Y.: supervision, funding acquisition. A.D.: writing—review and editing. M.I.: conceptualization, writing—review and editing. S.F.: writing—review and editing. All authors have read and agreed to the published version of the manuscript.
Data availability
The data supporting to findings of this study are available from the first author upon reasonable request.
Declarations
Ethical approval
This research study was conducted according to the Declaration of Helsinki guidelines. The Institutional Review Board of North China Electric Power University has approved the study (protocol code 110826–3).
Consent to participate
Informed consent was obtained from all respondents belonging to this research study.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Abdelhady S Performance and cost evaluation of solar dish power plant: sensitivity analysis of levelized cost of electricity (LCOE) and net present value (NPV) Renew Energy 2021 168 332 342 10.1016/j.renene.2020.12.074
Abdullah Zhou D Shah T Jebran K Ali S Ali A Ali A Acceptance and willingness to pay for solar home system: survey evidence from northern area of Pakistan Energy Rep 2017 3 54 60 10.1016/j.egyr.2017.03.002
Aberilla JM Gallego-schmid A Stamford L Azapagic A Design and environmental sustainability assessment of small-scale off-grid energy systems for remote rural communities Appl Energy 2019 258 114 125 10.1016/j.apenergy.2019.114004
Adetona ZA, Ogunyemi J, Bitrus I (2020) Investigation into the maintenance management awareness on solar PV renewable energy system in Nigeria. In: NSE Ilaro Branch, 1st National Conference, 2–3 November, 2020, pp 52–60. http://eprints.federalpolyilaro.edu.ng/id/eprint/1634
Adwek G Boxiong S Ndolo PO Siagi ZO The solar energy access in Kenya: a review focusing on Pay-As-You-Go solar home system Environ Dev Sustain 2020 22 5 3897 3938 10.1007/s10668-019-00372-x
Ahmad I Balkhyour MA Occupational exposure and respiratory health of workers at small scale industries Saudi J Biol Sci 2020 27 985 990 10.1016/j.sjbs.2020.01.019 32140043
Akbar A Ali S Ahmad MA Akbar M Danish M Understanding the antecedents of organic food consumption in Pakistan: moderating role of food neophobia Int J Environ Res Public Health 2019 16 4043 10.3390/ijerph16204043 31652520
Ali S Rashid H Khan MA The role of small and medium enterprises and poverty in Pakistan: an empirical analysis Theor Appl Econ 2014 18 67 80
Ali MY Hassanb M Rahmana MA Kafyc A-A Ara I Javede A Rahman MR Life cycle energy and cost analysis of small scale biogas plant and the 15th international symposium on district heating and cooling assessing the feasibility of using the heat demand-outdoor solar PV system in rural areas of Bangladesh Energy Procedia 2019 160 277 284 10.1016/j.egypro.2019.02.147
Ali S Yan Q Hussain MS Irfan M Ahmad M Razzaq A Dagar V Işık C Evaluating green technology strategies for the sustainable development of solar power projects: evidence from Pakistan Sustainability 2021 13 12997 10.3390/su132312997
Ali S Yan Q Irfan M Ameer W Atchike DW Green investment for sustainable business development: the influence of policy instruments on solar technology adoption Front Energy Res 2022 10 874824 10.3389/fenrg.2022.874824
Ali S Yan Q Irfan M Chen Z Evaluating barriers on biogas technology adoption in China: the moderating role of awareness and technology understanding Front Environ Sci - Environ Econ Manag 2022 10.3389/fenvs.2022.887084
Ali S, Yan Q, Razzaq A, Khan I, Irfan M (2022c) Modeling factors of biogas technology adoption: a roadmap towards environmental sustainability and green revolution. Environ Sci Pollut Res 1–23. 10.1007/s11356-022-22894-0
Amado M Poggi F Solar energy integration in urban planning: GUUD model Energy Procedia 2014 50 277 284 10.1016/j.egypro.2014.06.034
Awan U, Knight I (2020) Domestic sector energy demand and prediction models for Punjab Pakistan. J Build Eng 32:101790. 10.1016/j.jobe.2020.101790
Azimoh CL Klintenberg P Wallin F Karlsson B Illuminated but not electrified: an assessment of the impact of solar home system on rural households in South Africa Appl Energy 2015 155 354 364 10.1016/j.apenergy.2015.05.120
Baig IA, Irfan M, Salam MA, Işik C (2022) Addressing the effect of meteorological factors and agricultural subsidy on agricultural productivity in India: a roadmap toward environmental sustainability. Environ Sci Pollut Res 1–18. 10.1007/s11356-022-23210-6
Balkhyour MA Ahmad I Rehan M Assessment of personal protective equipment use and occupational exposures in small industries in Jeddah: health implications for workers Saudi J Biol Sci 2018 26 4 653 659 10.1016/j.sjbs.2018.06.011 31048988
Barman M Mahapatra S Palit D Chaudhury MK Energy for sustainable development performance and impact evaluation of solar home lighting systems on the rural livelihood in Assam, India Energy Sustain Dev 2017 38 10 20 10.1016/j.esd.2017.02.004
Bisaga I Parikh P Mulugetta Y Hailu Y The potential of performance targets (imihigo) as drivers of energy planning and extending access to off-grid energy in rural Rwanda Energy Environ 2019 8 310 335 10.1002/wene.310
Brislin RW Back-translation for cross-cultural research J Cross Cult Psychol 1970 3 185 216 10.1177/135910457000100301
Charles RG Davies ML Douglas P Hallin IL Mabbett I Sustainable energy storage for solar home systems in rural Sub-Saharan Africa – a comparative examination of lifecycle aspects of battery technologies for circular economy, with emphasis on the South African context Energy 2018 166 1207 1215 10.1016/j.energy.2018.10.053
Chowdhury SA Mourshed M Off-grid electrification with solar home systems: an appraisal of the quality of components Renew Energy 2016 97 585 598 10.1016/j.renene.2016.06.017
Deng QS Alvarado R Cuesta L Tillaguango B Murshed M Rehman A Işık C López-Sánchez M Asymmetric impacts of foreign direct investment inflows, financial development, and social globalization on environmental pollution Econ Anal Policy 2022 76 236 251 10.1016/j.eap.2022.08.008
Diouf B Tontine: self-help financing for solar home systems Renew Energy 2016 90 166 174 10.1016/j.renene.2015.12.050
Diouf B Avis C The potential of Li-ion batteries in ECOWAS solar home systems J Energy Storage 2019 22 295 301 10.1016/j.est.2019.02.021
Dogan E Ulucak R Kocak E Isik C The use of ecological footprint in estimating the environmental Kuznets curve hypothesis for BRICST by considering cross-section dependence and heterogeneity Sci Total Environ 2020 723 138063 10.1016/j.scitotenv.2020.138063 32217396
Dong K, Ren X, Zhao J (2021) How does low-carbon energy transition alleviate energy poverty in China? A nonparametric panel causality analysis. Energy Econ 103:105620. 10.1016/j.eneco.2021.105620
Duan K Ren X Shi Y Mishra T Yan C The marginal impacts of energy prices on carbon price variations: evidence from a quantile-on-quantile approach Energy Econ 2021 95 105131 10.1016/j.eneco.2021.105131
Elavarasan RM Leoponraj S Dheeraj A Irfan M Gangaram Sundar G Mahesh GK PV-diesel-hydrogen fuel cell based grid connected configurations for an institutional building using BWM framework and cost optimization algorithm Sustain Energy Technol Assess 2021 43 100934 10.1016/j.seta.2020.100934
Gandini D Almeida ATD Direct current microgrids based on solar power systems and storage optimization, as a tool for cost-effective rural electrification Renew Energy 2017 111 275 283 10.1016/j.renene.2017.04.009
Ghafoor A Rehman TU Munir A Ahmad M Iqbal M Current status and overview of renewable energy potential in Pakistan for continuous energy sustainability Renew Sustain Energy Rev 2016 60 1332 1342 10.1016/J.RSER.2016.03.020
Global solar council (2021) Global market outlook for solar power [WWW Document]
GoP. government of Pakistan (2021) Alternative Energy Development Board (AEDB) [WWW Document]
Gray L Boyle A Francks E Yu V The power of small-scale solar: gender, energy poverty, and entrepreneurship in Tanzania entrepreneurship in Tanzania Dev Pract 2018 29 29 39 10.1080/09614524.2018.1526257
Groenewoudt AC Romijn HA Alkemade F From fake solar to full service: an empirical analysis of the solar home systems market in Uganda Energy Sustain Dev 2020 58 100 111 10.1016/j.esd.2020.07.004
Halder PK Potential and economic feasibility of solar home systems implementation in Bangladesh Renew Sustain Energy Rev 2016 65 568 576 10.1016/j.rser.2016.07.062
Hao X Li Y Ren S Wu H Hao Y The role of digitalization on green economic growth: does industrial structure optimization and green innovation matter? J Environ Manage 2023 325 116504 10.1016/j.jenvman.2022.116504 36272290
Hoque SMN Das BK Analysis of cost, energy and CO2 emission of solar home systems in Bangladesh Int J Renew Energy Res 2013 3 347 352
Hossain CA, Chowdhury N, Longo M, Yaïci W (2019) System and cost analysis of stand-alone solar home system applied to a developing country. Sustainability 11(5):1403. 10.3390/su11051403
Iqbal T, Dong CQ, Lu Q, Ali Z, Khan I, Hussain Z, Abbas A (2018) Sketching Pakistan’s energy dynamics: prospects of biomass energy. J Renew Sustain Energy 10(2):023101. 10.1063/1.5010393
IRENA (2020) Statistical Review of World Energy International Renewable Energy Agency (IRENA); Renewable capacity statistics report. Availa-bleonline:https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2020/Mar/IRENA_RE_Capacity_Highlights_2020.pdf 2020. [WWW Document]
Irfan M Salem S Ahmad M Acevedo-Duque Á Abbasi KR Ahmad F Razzaq A Işik C Interventions for the current COVID-19 pandemic: frontline workers’ intention to use personal protective equipment Front Public Health 2022 9 793642 10.3389/fpubh.2021.793642 35186871
Isik C Ongan S Ozdemir D Ahmad M Irfan M Alvarado R Ongan A The increases and decreases of the environment Kuznets curve (EKC) for 8 OECD countries Environ Sci Pollut Res 2021 28 28535 28543 10.1007/s11356-021-12637-y
Jan A Xin-gang Z Ahmad M Irfan M Ali S Do economic openness and electricity consumption matter for environmental deterioration: silver bullet or a stake? Environ Sci Pollut Res 2021 10.1007/s11356-021-14562-6
Kamran M Current status and future success of renewable energy in Pakistan Renew Sustain Energy Rev 2018 82 609 617 10.1016/j.rser.2017.09.049
Khan I Impacts of energy decentralization viewed through the lens of the energy cultures framework: solar home systems in the developing economies Renew Sustain Energy Rev 2019 119 109 134 10.1016/j.rser.2019.109576
Khan DHA, Qurashi PDMM, Hussain ET, Hayee MI (2005) Commission on Science and Technology for Sustainable Development in the South COMSATS
Khan MI, Khan IA, Chang YC (2020) An overview of global renewable energy trends and current practices in Pakistan - a perspective of policy implications. J Renew Sustain Energy 12(5):056301. 10.1063/5.0005906
Khattak N, Hassnain SR, Shah SW, Mutlib A (2006) Identification and removal of barriers for renewable energy technologies in Pakistan, in: IEEE--ICET 2006 2nd international conference on emerging technologies Peshawar, Pakistan, 13–14 November 2006. pp. 397–402. 10.1109/ICET.2006.335984
Komatsu S Kaneko S Pratim P Morinaga A Determinants of user satisfaction with solar home systems in rural Energy 2013 61 52 58 10.1016/j.energy.2013.04.022
Lemaire X (2018) Solar home systems and solar lanterns in rural areas of the Global South: what impact? Wiley Interdiscip. Rev Energy Environ 7(5):e301. 10.1002/wene.301
López-vargas A Fuentes M Vivar M Current challenges for the advanced mass scale monitoring of solar home systems: a review Renew Energy 2021 163 2098 2114 10.1016/j.renene.2020.09.111
Mandal S Das BK Hoque N Optimum sizing of a stand-alone hybrid energy system for rural electrification in Bangladesh J Clean Prod 2018 200 12 27 10.1016/j.jclepro.2018.07.257
Mondal AH Klein D Impacts of solar home systems on social development in rural Bangladesh Energy Sustain Dev 2011 15 17 20 10.1016/j.esd.2010.11.004
Mujaddad HG Ahmad HK Measuring efficiency of manufacturing industries in Pakistan: an application of DEA double bootstrap technique Pak Econ Soc Rev 2016 54 363–384 26616713
Narayan N Papakosta T Vega-garita V Qin Z Popovic-gerber J Bauer P Zeman M Estimating battery lifetimes in solar home system design using a practical modelling methodology Appl Energy 2018 228 1629 1639 10.1016/j.apenergy.2018.06.152
NEPRA (2018) NEPRA, “State of industry report 2018,” National Electric and Power Regulatory Authority (NEPRA), Government of Pakistan
Newcombe A Ko E Energy for sustainable development sustainable solar home systems model: applying lessons from Bangladesh to Myanmar’s rural poor Energy Sustain Dev 2017 38 21 33 10.1016/j.esd.2017.03.002
Nicholas S, Buckley T (2018) Pakistan’s power future: renewable energy provides a more diverse, secure and cost-effective alternative, Institute for Energy Economics and Financial Analysis (IEEFA), Islamabad
Ongan S Işık C Bulut U Karakaya S Alvarado R Irfan M Ahmad M Rehman A Hussain I Retesting the EKC hypothesis through transmission of the ARMEY curve model: an alternative composite model approach with theory and policy implications for NAFTA countries Environ Sci Pollut Res 2022 29 46587 46599 10.1007/s11356-022-19106-0
Opiyo NN Energy research & social science impacts of neighbourhood influence on social acceptance of small solar home systems in rural western Kenya Energy Res Soc Sci 2019 52 91 98 10.1016/j.erss.2019.01.013
Pakistan economic survey (2021) Pakistan Council of Renewable Energy Technologies (PCRET), solar energy technology - experience of PCRET. Report, Islamabad. Available online: http://www.pcret.gov.pk/. (Accessed 11 May 2021)
Perpiña C Batista F Lavalle C An assessment of the regional potential for solar power generation in EU-28 Energy Policy 2020 88 86 99 10.1016/j.enpol.2015.10.004
Rafique MM Rehman S National energy scenario of Pakistan – current status, future alternatives, and institutional infrastructure: an overview Renew Sustain Energy Rev 2017 69 156 167 10.1016/j.rser.2016.11.057
Rafique MM Rehman S Alhems LM Assessment of solar energy potential and its deployment for cleaner production in Pakistan J Mech Sci Technol 2020 34 3437 3443 10.1007/s12206-020-0736-9
Ren S Hao Y Wu H Government corruption, market segmentation and renewable energy technology innovation: evidence from China J Environ Manage 2021 300 113686 10.1016/j.jenvman.2021.113686 34521008
Rena X Zudi Lua B Chengc C Shid Y Shena J On dynamic linkages of the state natural gas markets in the USA: evidence from an empirical spatio-temporal network quantile analysis Energy Econ 2019 80 234 252 10.1016/j.eneco.2019.01.001
SAARC (2020) USAID. South Asian regional initiative for energy integration: developing power system integration & cross border electricity trade in SAARC region: current status & future outlook
Sarker SA, Wang S, Adnan KMM, Anser MK, Ayoub Z, Ho TH, Tama Z, Trunina A, Hoque MM (2020) Economic viability and socio-environmental impacts of solar home systems for off-grid rural electrification in Bangladesh. Energies 13(3):679. 10.3390/en13030679
Shaughnessy EO Nemet GF Pless J Margolis R Addressing the soft cost challenge in U.S. small-scale solar PV system pricing Energy Policy 2019 134 110956 10.1016/j.enpol.2019.110956
Sovacool BK Success and failure in the political economy of solar electrification: lessons from World Bank Solar Home System (SHS) projects in Sri Lanka and Indonesia Energy Policy 2018 123 482 493 10.1016/j.enpol.2018.09.024
Stojanovski O Thurber M Wolak F Energy for sustainable development rural energy access through solar home systems: use patterns and opportunities for improvement Energy Sustain Dev 2017 37 33 50 10.1016/j.esd.2016.11.003
Ullah I Rehman A Khan FU Shah MH Khan F Nexus between trade, CO2 emissions, renewable energy, and health expenditure in Pakistan Int J Health Plann Manag 2020 35 818 831 10.1002/hpm.2912
Urbach N Ahlemann F Structural equation modeling in information systems research using partial least squares J Inf Technol Theory Appl 2010 11 5 40
Urban B Heydenrych J Technology orientation and effectuation – links to firm performance in the renewable energy sector of South Africa South African J Ind Eng 2015 26 125 136
Wang Z Ali S Akbar A Rasool F Determining the influencing factors of biogas technology adoption intention in Pakistan: the moderating role of social media Int J Environ Res Public Health 2020 17 2311 10.3390/ijerph17072311 32235442
Wu H Hao Y Ren S Yang X Xie G Does internet development improve green total factor energy efficiency? Evidence China Energy Policy 2021 153 112247 10.1016/j.enpol.2021.112247
Wu X, Sadiq M, Chien F, Ngo Q, Trinh T (2021b) Testing role of green financing on climate change mitigation: evidences from G7 and E7 countries
Yadav P Heynen AP Palit D Energy for sustainable development Pay-As-You-Go financing: a model for viable and widespread deployment of solar home systems in rural India Energy Sustain Dev 2019 48 139 153 10.1016/j.esd.2018.12.005
Yadav P Davies PJ Khan S Breaking into the photovoltaic energy transition for rural and remote communities: challenging the impact of awareness norms and subsidy schemes Clean Technol Environ Policy 2020 22 817 834 10.1007/s10098-020-01823-0
Yang X, Wang W, Wu H, Wang J, Ran Q, Ren S (2021) The impact of the new energy demonstration city policy on the green total factor productivity of resource-based cities: empirical evidence from a quasi-natural experiment in China. J Environ Plan Manag 1–34. 10.1080/09640568.2021.1988529
Yang X Su X Ran Q Ren S Chen B Wang W Wang J Assessing the impact of energy internet and energy misallocation on carbon emissions: new insights from China Environ Sci Pollut Res 2022 29 23436 23460 10.1007/s11356-021-17217-8
Zubi G Spertino F Carvalho M Adhikari RS Khatib T Development and assessment of a solar home system to cover cooking and lighting needs in developing regions as a better alternative for existing practices Sol Energy 2017 155 7 17 10.1016/j.solener.2017.05.077
| 36441331 | PMC9707279 | NO-CC CODE | 2022-12-01 23:20:22 | no | Environ Sci Pollut Res Int. 2022 Nov 28;:1-20 | utf-8 | Environ Sci Pollut Res Int | 2,022 | 10.1007/s11356-022-24387-6 | oa_other |
==== Front
Brittonia
Brittonia
Brittonia
0007-196X
1938-436X
Springer US New York
9728
10.1007/s12228-022-09728-6
Original Research
Nomenclatural notes on Menthinae (Mentheae — Lamiaceae) native to Brazil
Oliveira Aron Barbosa [email protected]
12
http://orcid.org/0000-0001-8109-4544
Antar Guilherme Medeiros [email protected]
3
Trovó Marcelo [email protected]
2
1 grid.8536.8 0000 0001 2294 473X Programa de Pós-graduação em Ciências Biológicas (Botânica), Museu Nacional, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Rio de Janeiro Brazil
2 grid.8536.8 0000 0001 2294 473X Laboratório Integrado de Sistemática Vegetal, Departamento de Botânica, Instituto de Biologia, Universidade Federal do Rio de Janeiro, CCS, Bloco A1, Cidade Universitária, Rio de Janeiro, Rio de Janeiro 21941-590 Brazil
3 grid.412371.2 0000 0001 2167 4168 Departamento de Ciências Agrárias e Biológicas, Universidade Federal do Espírito Santo, Centro Universitário Norte do Espírito Santo, Rodovia BR-101 Norte, Km 60, Litorâneo, 29934-900 São Mateus, Espírito Santo Brasil
28 11 2022
123
17 5 2022
5 9 2022
13 10 2022
© The New York Botanical Garden 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Menthinae, one of the five subtribes of Mentheae (Lamiaceae), with ca. 735 species, is represented in Brazil by a wide range of herbs and shrubs. During an investigation of the taxonomic history of the group for the Flora e Funga do Brasil project, several nomenclatural issues with typifications in the group have been identified. This study attempts to resolve them, making suggestions of new types and re-evaluation of synonyms. Thirteen lectotypifications and one neotypification are being proposed, along with the recognition of Thymus selloi Spreng. ex J.A.Schmidt as a nomen nudum. Most of the suggested changes are due to the loss of many of Sellow’s types at B during WWII, lack of specific herbaria references and collector numbers in the protologues, or to existence of designated syntypes in the referred collections. Our study is a contribution to the nomenclatural and taxonomic knowledge of Menthinae, providing a stable foundation for future study of its taxonomy, biogeography, and evolution.
Keywords
Brazilian flora
Labiatae
lectotypification
Nepetoideae
neotypification
taxonomy
http://dx.doi.org/10.13039/501100002322 Coordenação de Aperfeiçoamento de Pessoal de Nível Superior 001 http://dx.doi.org/10.13039/501100004586 Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro E-26/202.708/2019 http://dx.doi.org/10.13039/501100003593 Conselho Nacional de Desenvolvimento Científico e Tecnológico 132032/2020-1 306758/2019-9 Oliveira Aron Barbosa http://dx.doi.org/10.13039/100010402 American Society of Plant Taxonomists Graduate Student Research Grant
==== Body
pmcMentheae, the largest tribe of Lamiaceae, together with tribe Ocimeae, encompasses most of the diversity of Lamiaceae in the Neotropics (Drew & Sytsma, 2012). Among the five Mentheae subtribes, Menthinae stands out as a group comprising ca. 40 genera and 735 species spread throughout the temperate, subtropical, and tropical regions of all continents, except for Antarctica, and is mostly associated with Europe, Asia, and the Americas. The species of the subtribe are frequently found in high altitude vegetation formations, occupying mostly open phytophysiognomies, such as grasslands, riverine forests, bogs, rocky montane areas, forest glades or margins and, rarely, deserts (Harley et al. 2004; Bräuchler et al. 2010).
Menthinae are usually aromatic herbs or shrubs, with ascending or divergent stamens bearing often divaricate thecae, a symmetrical (or ± symmetrical) nectariferous disk, usually strongly 2-lipped corollas, 11- to 15-nerved calyx, and an areolate abscission scar on the nutlet (Harley et al. 2004). The group is monophyletic and well supported, as demonstrated in recent phylogenetic studies based on molecular data (Drew & Sytsma, 2012; Zhao et al. 2021). However, as shown by Drew & Sytsma (2012), Bräuchler et al. (2010), and Drew et al. (2017), several of its Neotropical genera require further investigation regarding their circumscription.
Menthinae is represented in Brazil by 42 native species in 8 genera and distributed mostly in open vegetation types, such as the campos de altitude (herbaceous to shrubby vegetation associated with igneous or metamorphic outcrops, composed mainly of granites, gneisses, or nepheline syenites, that had been uplifted during the Tertiary, and starting at altitudes above 1500 m.a.s.l.), the campos rupestres (herbaceous to shrubby vegetation on quartzite, arenite, and ironstone soils and outcrops of the Pre-Cambrian origin, mostly at elevations above 900 m.a.s.l.), both with high levels of endemism, the campos limpos (grasslands at lower altitudes dominated by herbs, mainly by members of Poales, among a diversity of other herbaceous taxa, and slender shrubs), or along forest margins of the Atlantic Forest, the Cerrado and the Pampa domains (Rizzini, 1997; Vasconcelos, 2011; Antar et al. 2022).
The taxonomy of the group had been initially presented by Sprengel (1827), who did not specify herbaria or collectors for the newly described species, followed by that of Bentham (1834, 1848), who on several occasions had referred to more than one herbarium and/or collector — often without collector’s numbers — for a single species, and that of Briquet (1896, 1887). Epling (1936) and Epling & Stewart (1939) made significant contributions to the advancement of taxonomic study of the group, including several lectotypifications (sometimes inadvertent — see Prado et al. 2015), and the proposal of several new taxa. Thereafter, new species or combinations have been proposed, but the taxonomy of most of the group remained largely stable (Brade, 1943; Epling, 1951; Epling & Mathias, 1957; Epling & Játiva, 1963; Irving, 1980).
With the completion of the Menthinae genera project for the online Flora e Funga do Brasil effort (Antar et al. 2022), there remained several unresolved nomenclatural issues for the native Brazilian species of Cunila D.Royen, Glechon Spreng., Hesperozygis Epling, Hoehnea Epling, and Rhabdocaulon (Benth.) Epling. These genera are represented in the Atlantic Forest domain, with Hoehnea restricted to it and the others having it as their main site of occurrence in Brazil (Epling, 1936; Epling & Stewart, 1939; Antar et al. 2022). Cunila, Glechon and mainly Rhabdocaulon are also represented in the Cerrado domain, while the first of those and Hesperozygis are found in the Pampa domain (Epling & Stewart, 1939; Antar et al. 2022).
The objective of our study is to establish correct typifications for the Menthinae species in Brazil, incorporating new nomenclatural decisions and summarizing the ones already made (Epling, 1936, 1939; Xifreda & Malo, 2004, 2006), thus establishing a stable foundation for future study of their taxonomy, biogeography, and evolution.
Material and methods
Field collecting trips have been conducted in South and Southeast Brazil since 2015. Digital images available from JSTOR Global Plants (https://plants.jstor.org/), JABOT (http://rb.jbrj.gov.br/), REFLORA (http://reflora.jbrj.gov.br/), and SpeciesLink (https://specieslink.net/) databases have been examined, as well as specimens deposited at B, BHCB, BM, BR, C, CEN, CTBS, E, F, HAL, HUEFS, IBGE, LA, G, GH, K, M, MBM, MO, NY, P, PACA, R, RB, RFA, SAMES, SP, SPF, UB, UC, US, and W herbaria (acronyms follow Thiers, continuously updated). Historic photos of type specimens from F and UC were also consulted. Our taxonomic decisions follow the rules and recommendations of the International Code of Nomenclature for algae, fungi, and plants (Turland et al. 2018).
Results and discussion
Most of the required lectotypes selections are due to the loss of Friedrich Sellow’s specimens housed at B during WWII (BGBM, 2021). They had been initially designated as syntypes by George Bentham (1834, 1848), but later inadvertently lectotypified or proposed as types of new taxa by Carl C. Epling (1936), who provided for them originally absent collection numbers, as can be seen in the photos of the types in the Field Museum Negatives collection. They represent Epling’s interpretations of them based on specimens with collection numbers from other herbaria, such as P, and should be treated with caution.
Possible duplicates of the specimens lost at B may be found at several herbaria in Europe and the Americas, sometimes even carrying the original label from B and thus indicating direct links to Sellow’s collections analyzed by Bentham. It is worth mentioning that B still holds many type specimens for these taxa, but they represent a later transfer, likely from HAL, of the Rudolf Baschant Herbarium (Hiepko, 1980), and as such cannot be original holotypes or lectotypes. Similar cases of problematic lectotypifications of some of the Brazilian taxa of Menthinae have been referred to by Xifreda & Mallo (2004, 2006).
Problematic cases with the designated syntypes are also common, as is the case for many species of Rhabdocaulon, where second step lectotypifications are required to stabilize prioritization among possible lectotypes. Details of remaining, historic, syntypes, disregarded as lectotypes by Epling’s decisions (1936), are also provided. We also found rare cases of the complete lack of herbarium citations, as for example for Hesperozygis kleinii Epling & Játiva.
We present a nomenclatural treatment for the following taxa of Menthinae. The conclusions reflect our nomenclatural decisions after numerous analyses of type specimens housed in different herbaria, together with the protologues and other references, leading to the proposition of thirteen lectotypifications, one neotypification, and one nomen nudum:
1. Cunila fasciculata Benth., Labiat. Gen. Spec. 363 (1834). Hedyosmos fasciculatus (Benth.) Kuntze, Revis. Gen. Pl. 2: 520 (1891).—Type: Brazil: [without specific locality or date], F. Sellow s.n. (holotype: B [destroyed]; lectotype, here designated: K barcode K000248135 [image!]; possible isolectotypes: B barcode B_10_0248417 [image!], HAL 3X barcode HAL0064927 [image!], barcode HAL0064928 [image!], barcode HAL0064929 [image!], UC accession 61939 [image!], UC accession 61940 [image!], S accession 08-15360 [image!], W barcode W0027334 [image!]).
Gardoquia capitata Spreng., Syst. Veg., ed. 16 [Sprengel] 4(2, Cur. Post.): 225 (1827).—Type: Brazil, Rio Grande [do Sul]: [without specific locality or date], F. Sellow s.n. (holotype: HBG barcode HBG-518225 [image!]).
Thymus selloi Spreng. ex J.A.Schmidt, Fl. Bras. (Martius) 8(1): 168 (1858), nom. nud.
Notes.—Bentham (1834) described Cunila fasciculata based on a specimen Sellow s.n., housed at B. Epling (1936) cited the collection number 1524 for the Sellow specimen at B. However, the specimens at B had not been numbered, as can be seen in the Field Museum Negatives N° 28974. Nevertheless, most of them had been seen and annotated by Bentham or Epling and confirmed as types. The specimen at B was lost during WWII, but one of Sellow’s types that comes from the Rudolf Baschant Herbarium is still housed at B (Hiepko, 1980).
Bentham (1834) referred to southern Brazil (“in Brasilia merid.”) as the general geographic region for the type. W0027334 is the only specimen matching that general location, that is more precisely restricted to the state of Rio Grande do Sul, part of the Brazilian South (Antar et al. 2022). There cannot be much certainty that all the specimens identified as Sellow s.n. are in fact duplicates. Nevertheless, a combination of Bentham’s annotation, the stamp of the Herbarium Benthamianum, and the designation by Epling (1936) as an isotype at K offers enough confidence to select this material (Fig. 1) as the lectotype. Fig. 1 Lectotype of Cunila fasciculata Benth. (F. Sellow s.n., K000248135). ©The Board of Trustees of the Royal Botanic Gardens, Kew, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
Schmidt (1858) considered Thymus selloi to be a synonym, reportedly in an annotation on the holotype of Gardoquia capitata. Since no description for that name had been made, and in accordance with Art. 38, Ex.1, Rec. 50B of the Shenzhen Code (Turland et al. 2018), it is here considered a nomen nudum.
2. Cunila galioides Benth., Labiat. Gen. Spec. 363 (1834). Hedyosmos galioides (Benth.) Kuntze, Revis. Gen. Pl. 2: 520 (1891).—Type: Brazil: [without specific locality or date], F. Sellow s.n. (lectotype, designated by Epling [1936: 142]: B [destroyed]; replacement lectotype, here designated: K barcode K000248136 [image!]; possible isolectotypes: B barcode B_10_0311251 [image!], HAL barcode HAL0064932 [image!], K 4X barcode K000640689 [image!], barcode K000640690 [image!], barcode K000640691 [image!], barcode K000640692 [image!], M barcode M0096112 [image!], P 3X barcode P00737717 [image!], barcode P00737718 [image!], barcode P00737720 [image!], TUB barcode TUB007556 [image!], UC accession 61941 [image!], US accession 616883!, W 3X barcode W0027331 [image!], barcode W0027332 [image!], barcode W0027333 [image!]).
Cunila galioides var. hispida J.A.Schmidt, Fl. Bras. (Martius) 8(1): 168 (1858).—Type: Brazil: [without specific locality or date], F. Sellow s.n. (holotype: B [destroyed]; Brazil, Rio de Janeiro: Itatiaia, no campo, 5 Jun 1871, A.F.M. Glaziou 4893 (neotype, here designated: P barcode P02970010 [image!]; isoneotype: R barcode R000011485 [!]).
Hedeoma glaziovii Briq., Annuaire Conserv. Jard. Bot. Genève 2: 185 (1898). —Type: Brazil: prope Rio de Janeiro, 1876, A.F.M. Glaziou 6651 (holotype: G accession 8423-7 [image!]; isotypes: BR barcode BR0000005113934 [image!], C 3X barcode C10012996 [image!], barcode C10012997 [image!], barcode C10012998 [image!], G accession 8423-8 [image!]; P 2X barcode P00737727 [image!], barcode P00737728 [image!]).
Notes.—Bentham (1834) described Cunila galioides based on a specimen Sellow s.n., housed at B and on the material A.F.C.P Saint-Hilaire s.n., housed at P. Epling (1936) cited the specimen housed at B as the type, an action interpreted in Prado et al. (2015) as an inadvertent lectotype selection according to the ICN (Turland et al. 2018). Epling (1936) also cited the collection number 4400 for the Sellow specimen at B. However, the specimens at B had not been originally numbered, as can be seen in the Field Museum Negative N° 28975. The only numbered collection is that of one specimen at P, but the calligraphy on its label corresponds to that of Bentham, while the original Sellow’s collection at B does not have a collection number. The specimens at B, however, were lost during WWII, and are now represented solely by a photograph attached to the UC 61941 material, as confirmed by an annotation left by Epling who defined it as the type. Nevertheless, B still houses one of Sellow’s types, the one coming from the Rudolf Baschant Herbarium (Hiepko, 1980).
Bentham (1834) cited southern Brazil (“in Brasilia merid.”) as the general geographic reference for the type. The only specimens matching that information are K000640692 and W0027331, even though the species is widely distributed throughout the South, Southeast and parts of the Center-West of Brazil (Antar et al. 2022). There is not much certainty that all specimens identified as Sellow s.n. are in fact duplicates, as Sellow could have collected the species more than once. Nevertheless, taking into consideration Bentham’s annotations, its origin from the Herbarium Benthamianum, and Epling’s annotation stating its status as an isotype, we hereby propose the specimen from K (Fig. 2) as the replacement lectotype. The Saint-Hilaire unnumbered specimens, housed at P and marked as types, are thus recognized as the remaining syntypes. Fig. 2 Replacement lectotype of Cunila galioides Benth. (F. Sellow s.n., K000248136). ©The Board of Trustees of the Royal Botanic Gardens, Kew, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
The synonym Cunila galioides var. hispida was proposed by Schmidt (1858) using one of Sellow’s types at B, with no references to an herbarium or collection number. Later on, Epling (1936) inadvertently proposed Sellow 4524 as the lectotype. That material was lost during WWII, leaving it without an existing type. A neotype (Fig. 3) is hereby proposed based on the quality of preserved morphology of the specimen matching the description in the protologue. Fig. 3 Neotype of Cunila galioides var. hispida J.A.Schmidt (A.F.M. Glaziou 4893, P02970010). ©MNHN — Muséum national d'Histoire naturelle, made openly available through SpeciesLink (https://specieslink.net/). Reproduced with the consent of the MNHN — Muséum national d'Histoire naturelle.
3. Cunila incisa Benth., Labiat. Gen. Spec. 361 (1834). Hedyosmos incisus (Benth.) Kuntze, Revis. Gen. Pl. 2: 520 (1891). —Type: Brazil: [without specific locality or date], F. Sellow s.n. (holotype: B [destroyed]; lectotype, here designated: K barcode K000247819 [image!]; possible isolectotypes: B barcode B_10_0248416 [image!]; G barcode G00680943 [image!], HAL barcode HAL0064935 [image!], UC accession 61953 [image!], UC accession 61955 [image!], W barcode W0027327 [image!]).
Notes.—Bentham (1834) described Cunila incisa based on a specimen Sellow s.n., housed at B. Epling (1936) inadvertently designated the collection number 3161 as the type, which is not found among examined types except for UC 61953 in its label annotations written at a later date, presumably by the same author. The specimen at B was lost during WWII and is now represented solely by a photograph attached to the UC 61955 material, as confirmed by a note left by Epling defining it as the type. B, within the Rudolf Baschant Herbarium, still houses one of Sellow’s types (Hiepko, 1980).
Bentham (1834) had southern Brazil (“in Brasilia merid.”) as the general geographic reference for the type. The only available specimen that matches that information is W0027327, with a distribution restricted to the states of Santa Catarina and Rio Grande do Sul, parts of the Brazilian South (Antar et al. 2022). There is not much certainty that all specimens identified as Sellow s.n. are in fact duplicates, but a combination of Bentham’s annotations, the stamp of the Herbarium Benthamianum, and the designation as an isotype in K by Epling (1936) gives us enough confidence to select the referred material (Fig. 4) as the lectotype. Fig. 4 Lectotype of Cunila incisa Benth. (F. Sellow s.n., K000247819). ©The Board of Trustees of the Royal Botanic Gardens, Kew, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
4. Cunila menthoides Benth., Labiat. Gen. Spec. 363 (1834).—Type: Brazil: [without specific locality or date], F. Sellow s.n. (holotype: B [destroyed]; lectotype, here designated: K barcode K000248137 [image!]; possible isolectotypes: K barcode K000640693 [image!], G barcode G00681160 [image!], HAL barcode HAL0104706 [image!], UC accession 61960 [image!], UC accession 61961 [image!]).
Notes.—Bentham (1834) described Cunila menthoides based on a specimen Sellow s.n., housed at B. Epling (1936) inadvertently designated the number 3990 as the type, which is presently found only in UC 61960 in the annotations written at a later date on its label, presumably by the same author, and in a photograph of the material attached to UC 61961. No other specimen has a collector’s number, although most of them are marked as seen by Bentham or Epling. That specimen at B, however, was lost during WWII. No other available duplicates have collectors’ numbers.
Bentham (1834) cited southern Brazil (“in Brasilia merid.”) as the general geographic reference for the type. The only specimen matching that information is HAL0104706, more precisely that of the states of Santa Catarina and Rio Grande do Sul (Antar et al. 2022). The specimens at G and HAL, though representative of the species, are identified as Cunila pulegioides Benth., a name that was never published —highlighted by Dr. Christian Bräuchler in a note on the second specimen — and that should not be confused with Cunila pulegioides L., the basionym of Hedeoma pulegioides (L.) Pers.
There is no certainty that all specimens identified as Sellow s.n. are in fact duplicates, but a combination of Bentham’s annotations, the stamp of the Herbarium Benthamianum, its designation as an isotype in K by Epling (1936), and an Epling’s note treating it as an isotype should be enough to confidently select the referred material (Fig. 5) as the lectotype. Fig. 5 Lectotype of Cunila menthoides Benth. (F. Sellow s.n., K000640693). ©The Board of Trustees of the Royal Botanic Gardens, Kew, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
5. Glechon discolor Epling, Repert. Spec. Nov. Regni Veg. Beih. 85: 145 (1936).—Type: Brasilia [Brazil]: [without specific locality or date], F. Sellow 4210 (holotype: B [destroyed]; lectotype, here designated: UC accession 55173 [image!]; isolectotypes: K 2X barcode K000485443 [image!], barcode K000485444 [image!]).
Notes.—Epling (1936) described Glechon discolor based on the specimen Sellow 4210, housed at B. The specimen at B, however, was lost during WWII. There is no certainty that the specimens at K identified as Sellow s.n. are in fact duplicates but, considering Epling’s annotations on them and his decision to treat them as isotypes at K, they are here treated as such. Taking into consideration the herbarium where Epling was working (Stafleu & Cowan, 1976), and the designation of an isotype at LA, later transferred to UC, we treat the material chosen by Epling (Fig. 6) as the most adequate option for the lectotype. Fig. 6 Lectotype of Glechon discolor Epling (F. Sellow 4210, UC accession 55173). Photo taken by Dr. Paul Fine, curator of Neotropical Flora at UC. Reproduced with the consent of University and Jepson Herbaria, Berkeley.
It is noteworthy to mention that even though the province of São Paulo was defined by Epling (1936) in the protologue as the area of occurrence for the type, no collections matching that information have been found. The only geographic reference more specific than the country (“in Brasilia merid.”, southern Brazil) was found in a note written by Bentham on the sheet of the K000485444 specimen. This geographic term is fairly general and does include São Paulo, but that state is not part of the known distribution of the species, restricted to the states of Santa Catarina and Rio Grande do Sul (Epling & Stewart, 1939; Antar et al. 2022).
6. Hesperozygis kleinii Epling & Játiva, Brittonia 15: 369 (1963).—Type: Brazil, Santa Catarina: Bom Jardim [da Serra], São Joaquim, Serra do Oratório, Aparados da Serra, 18 Set 1958, R. Reitz & R.M. Klein 7172 (lectotype, here designated: UC accession 49799 [image!]; isolectotypes: HBR accession 54909 [image!], RFA accession 30375 [!]).
Notes.—When describing Hesperozygis kleinii based on Reitz & Klein 7172, Epling & Játiva (1963) did not define where the type materials were housed. Upon investigation, three specimens were uncovered. Among those, we consider the specimen at UC (Fig. 7), previously housed at LA, the most suitable candidate for lectotypification, since it was housed at LA where Epling was mainly working (Stafleu & Cowan 1976) and therefore likely the main specimen on which the description of the new species was based. Fig. 7 Lectotype of Hesperozygis kleinii Epling & Játiva (R. Reitz & R.M. Klein 7172, UC accession 49799). Photo taken by Dr. Paul Fine, curator of Neotropical Flora at UC. Reproduced with the consent of the University and Jepson Herbaria, Berkeley.
7. Hesperozygis nitida (Benth.) Epling, Repert. Spec. Nov. Regni Veg. Beih. 85: 133 (1936). Hedeoma nitida Benth., Nat. Pflanzenfam. [Engler & Prantl] iv. III A. 295 ≡ Keithia nitida Benth. Prodr. [A. P. de Candolle] 12: 247. 1848.—Type: Brazil, Rio de Janeiro, in Serra dos Órgãos, 1841, G. Gardner 5835 (lectotype, designated by Epling [1936: 133]: K; second step lectotype, here designated: K barcode K000485461 [image!]; isolectotypes: K barcode K000485460 [image!]; UC accession 1943458 [image!]).
Notes.—Bentham (1848) described Keithia nitida without defining an herbarium for the specimen, which led to a lectotypification by Epling (1936), who designated the type specimen as housed in K, and an isotype in LA (later transferred to UC). The appointed collection for the type, however, holds two specimens, which brought the necessity for a second step lectotypification.
Although Epling stated on its label that the specimen K000485460 should be recognized as the lectotype, it presents the contradictory information that it was collected in Minas Gerais state on a printed note, which is not in agreement with the species protologue. Furthermore, the only specimen clearly mentioning Serra dos Órgãos Mountains is UC1943458, but the note concerning that location was written later, as confirmed by its distinct calligraphy. Finally, the specimen K000485461, suggested as an isotype on a note made by Epling, has its location appointed as “Brazil” with a handwritten note stating “Near summit (…)” followed by words which are almost illegible, but were interpreted, as seen in the UC1943458 isotype, as “(…) Organ Mts.”, the same locality presented in the protologue. Therefore, we hereby choose the last K specimen (Fig. 8) as the second step lectotype due to the collection where it is housed, the notes left by Bentham, and its available geographical information. Fig. 8 Second step lectotype of Hesperozygis nitida (Benth.) Epling (G. Gardner 5835; K000485461). ©The Board of Trustees of the Royal Botanic Gardens, Kew, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
8. Hesperozygis ringens (Benth.) Epling, Repert. Spec. Nov. Regni Veg. Beih. 85: 134 (1936). Glechon ringens Benth., Labiat. Gen. Spec. 409. 1834.—Type: Brazil: [without specific locality or date], F. Sellow s.n. (holotype: B [destroyed]; lectotype, here designated: K barcode K000485458 [image!]; possible isolectotypes: B barcode B_10_0004117 [image!], BR barcode BR0000006913878 [image!], G barcode G00437837 [image!], HAL barcode HAL0114389 [barcode!], K barcode K000485459 [image!], UC accession 70231 [image!], W 2X barcode W0027343 [image!], barcode W0031664 [image!]).
Notes.—Bentham (1834) described Glechon ringens based on a specimen Sellow s.n., housed at B. Later, Epling (1936) proposed a new combination Hesperozygis ringens for the new genus Hesperozygis, and inadvertently designated the collection number 3305 as its type, which can only be seen in the photograph of the duplicate described by him as the type and attached to UC 70231 and at a later date annotated in Epling’s notes attached to the selected lectotype. No other specimen, including the one at B seen in the Field Museum Negative N° 28968, displays a collector’s number, although most of them are marked as seen by Bentham or Epling. The types at B have been lost during WWII, even though B still houses one of the Sellow’s type, that from the Rudolf Baschant Herbarium (Hiepko, 1980).
Bentham (1834) designated “in Brasilia merid.” as the general geographic reference for the type. No similar information was found among the examined specimens, even though the species is restricted to the state of Rio Grande do Sul in the extreme South of Brazil (Antar et al. 2022). There is no certainty that all specimens identified as Sellow s.n. are in fact duplicates, but a combination of Bentham’s annotations, the stamp of the Herbarium Benthamianum, the designation of an isotype at K by Epling (1936) and his note on the material treating it as an isotype should be sufficient to confidently select the referred material (Fig. 9) as the lectotype. Fig. 9 Lectotype of Hesperozygis ringens (Benth.) Epling. (F. Sellow s.n., K000485458). ©The Board of Trustees of the Royal Botanic Gardens, Kew, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
9. Hoehnea minima (J.A.Schmidt) Epling, Repert. Spec. Nov. Regni Veg. Beih. 115: 9 (1939). Keithia minima J.A.Schmidt, Fl. Bras. (Martius) 8(1): 173. 1858.—Type: Brasilia [Brazil]: [without specific locality or date], F. Sellow 4441 (lectotype, designated by Epling [1936: 132]: B [destroyed]; replacement lectotype, here designated: UC accession 64016 [image!]).
Notes.—Schmidt (1858) described Keithia minima using a specimen Sellow s.n., housed at B. Epling (1936) assigned to it the collection number 4441, what is here understood as an inadvertent lectotypification. Later on, Epling & Stewart (1939) proposed Hoehnea minima, based on Epling’s new genus Hoehnea. The lectotype at B was lost during WWII and is only preserved as a negative photograph in the UC 64015 material. The replacement lectotype (Fig. 10), marked as an isotype in Epling’s annotation, represents a complete specimen adequate for detailed morphological observation, and therefore most suitable to be treated as such. Fig. 10 Replacement lectotype of Hoehnea minima (J.A.Schmidt) Epling. (F. Sellow 4441, UC accession 64016). Photo digitalized by Dr. Paul Fine, curator of Neotropical Flora at UC. Reproduced with the consent of the University and Jepson Herbaria, Berkeley.
10. Rhabdocaulon denudatum (Benth.) Epling, Repert. Spec. Nov. Regni Veg. Beih. 85: 136 (1936). Hedeoma denudata (Benth.) Briq., Nat. Pflanzenfam. [Engler & Prantl] iv. III A. 295 (1887). Keithia denudata Benth., Labiat. Gen. Spec. 412. 1834.—Type: Brazil: [without a specific locality or date], F. Sellow s.n. (lectotype, designated by Epling [1936: 136]: B [destroyed]; replacement lectotype, here designated: K barcode K000485596 [image!]; possible isolectotypes: E 2X barcode E00208046 [image!], barcode E00208047 [image!], HAL barcode HAL0114392 [image!], G barcode G00435270 [image!], K barcode K000485595 [image!]).
Notes.—Bentham (1834) described Keithia denudata based on a specimen Sellow s.n., housed at B, and specimens A.F.C.P. Saint-Hilaire s.n., housed at P and M. Epling (1936) later on proposed the name Rhabdocaulon denudatum, following the recombination of significant parts of Keithia Spreng. into the new genus Rhabdocaulon, and inadvertently lectotypified the specimens by citing the Sellow s.n. specimen at B as the type. The specimen at B, however, was lost during WWII and can be seen only in a photograph of a specimen in the Field Museum Negative N° 28957, without, however, certainty that it represents the specimen selected by Epling.
Though Bentham (1834) described the occurrence of the species in southern Brazil (“in Brasilia merid.”), and Epling (1936) cited Minas Gerais as the province of its occurrence, such geographic references have not been found among the analyzed specimens. Other geographic references, found on the label of the HAL0114392, were “Brasil aequin.” (Which may be interpreted as a reference to Minas Gerais), and, in a note to the K000485595 isolectotype, St. Paulo, part of southern Brazil. Both latter regions are in fact parts of the actual distribution for the species (Antar et al. 2022).
A combination of Bentham’s annotations, the stamp of the Herbarium Benthamianum, the designation of an isotype at K by Epling (1936), and his note treating it as a “cotype” — or, rather, an isotype — makes the designated specimen (Fig. 11) the most suitable replacement as the lectotype. It is also worth mentioning that both materials at K are part of the same sheet, though with different barcodes and associated notes. Fig. 11 Replacement lectotype of Rhabdocaulon denudatum (Benth.) Epling. (F. Sellow s.n., K000485596). ©The Board of Trustees of the Royal Botanic Gardens, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
11. Rhabdocaulon gracile (Benth.) Epling, Repert. Spec. Nov. Regni Veg. Beih. 85: 136 (1936). Hedeoma gracilis (Benth.) Briq Nat. Pflanzenfam. [Engler & Prantl] iv. III A. 295. 1887. Keithia gracilis Benth., Labiat. Gen. Spec. 412. 1834.—Type: Brazil: São Paulo, Campos près Carapatos dans les Campos Geraes, 1816–1821, A.F.C.P. Saint-Hilaire C2 1572 (first step lectotype, designated by Epling [1936: 136]: P; second step lectotype, here designated: P barcode P00714595 [image!]; isolectotype: P barcode P00714596 [image!]).
Notes.—Bentham (1834) described Keithia gracilis based on the specimen A.F.C.P. Saint-Hilaire s.n., housed in P. Later on, Epling (1936) proposed the name Rhabdocaulon gracile based on his recombination of a significant part of Keithia Spreng. into his new genus Rhabdocaulon and made an inadvertent lectotypification by assigning the collection number 1572 to the cited type specimen.
Upon investigation, we found that there are in fact two specimens with the same collector’s number in the collection, which necessitates a second-step lectotypification. Considering prioritizing between the two specimens, we decided on the one (Fig. 12) with the higher quality of preserved morphology and more closely matching the description. A third specimen, from Saint-Hilaire’s collection and numbered C2-2791-10, is designated as a syntype. Considering the decisions made, this last specimen should be recognized as a remaining syntype. Fig. 12 Second step lectotype of Rhabdocaulon gracile (Benth.) Epling (A.F.C.P. Saint-Hilaire C2 1572, P00714595). ©MNHN — Muséum national d'Histoire naturelle, made openly available through SpeciesLink (https://specieslink.net/). Reproduced with the consent of the MNHN — Muséum national d'Histoire naturelle.
12. Rhabdocaulon stenodontum (Briq.) Epling, Repert. Spec. Nov. Regni Veg. Beih. 85: 136 (1936). Hedeoma stenodonta Briq., Bull. Herb. Boissier Ser. II. vii. 612. 1907.—Type: Paraguay, Ypejhú: Sierra de Maracayú, Nov. 1885–1902, E. Hassler 5253 (lectotype, here designated: G barcode G00307329 [image!]; isolectotypes: B [destroyed], F barcode F0BN017707 [image!], G barcode G00307330 [image!], GH GH00001191 [image!], K K000485594 [image!], NY NY00000578 [image!], P 4X barcode P00714584 [image!], barcode P00714585 [image!], barcode P00753762 [image!], barcode P00753763 [image!], S accession 04-2713 [image!], UC accession 944369 [image!]).
Notes.—Briquet (1907) described Hedeoma stenodonta based on the gathering numbered 5253 in his work discussing Hassler’s collection. Epling (1936) later on proposed the combination Rhabdocaulon stenodontum based on the new genus Rhabdocaulon. Searching for the type specimens, we found that two of these specimens are housed at G, where the Delessert collection had been transferred (Stafleu & Cowan 1976), and thus requiring lectotypification. Considering Briquet’s determination on the specimen label, and the fact that it had a priority of 23 years in the transfer from the Delessert collection to G, we hereby designate one of the available specimens (Fig. 13) — effectively syntypes — as the lectotype. It is noteworthy to mention that, although the details of locality and collection date are not completely available for these specimens, they can be confirmed in Briquet (1907). Fig. 13 Lectotype of Rhabdocaulon stenodontum (Briq.) Epling (E. Hassler 5253, G00307329). Openly available on the Catalogue des herbiers de Genève (CHG) (http://www.ville-ge.ch/musinfo/bd/cjb/chg/). Reproduced with the consent of the Conservatoire et Jardin botaniques de Genève.
13. Rhabdocaulon strictum (Benth.) Epling, Repert. Spec. Nov. Regni Veg. Beih. 85: 137 (1936). Cunila stricta Benth., Labiat. Gen. Spec. 362. 1834. Hedyosmos strictus Kuntze, Revis. Gen. Pl. 2: 520 (1891).—Type: Brazil: [without specific locality or date], F. Sellow s.n. (lectotype, designated by Epling [1936: 137]: B [destroyed]; replacement lectotype, here designated: K barcode K000485589 [image!]; possible isolectotypes: G barcode G00681158 [image!], HAL barcode HAL0064945 [image!], UC accession 1943464 [image!], W barcode W0013599 [image!]).
Notes.—Bentham (1834) described Cunila stricta using a specimen Sellow s.n., housed at B, and the material A.F.C.P. Saint-Hilaire, housed in the collector’s collection. Later on, Epling (1936) proposed the name Rhabdocaulon strictum as a combination in the newly created genus Rhabdocaulon, and made an inadvertent lectotypification by designating the Sellow material as the type. The material at B, however, was lost during WWII.
Bentham (1834) used “in Brasilia merid.” as a general geographic reference for the type. The only specimen in accordance with such reference is W0013599, even though the species is only found in the states of Santa Catarina and Rio Grande do Sul, parts of the Brazilian South (Antar et al. 2022). A combination of Bentham’s annotations, the stamp of the Herbarium Benthamianum, and the designation as an isotype at K by Epling (1936) gives us enough confidence to select the referred material (Fig. 14) as the replacement lectotype. The specimens at P and US and marked as types are part of Saint-Hilaire’s collection and should be recognized as remaining syntypes. Fig. 14 Replacement lectotype of Rhabdocaulon strictum (Benth.) Epling. (F. Sellow s.n., K000485589). ©The Board of Trustees of the Royal Botanic Gardens, Kew, made openly available through Reflora Virtual Herbarium (http://reflora.jbrj.gov.br/). Reproduced with the consent of the Royal Botanic Gardens, Kew.
The authors are grateful to the personnel of several herbaria who provided inestimable help under extremely difficult conditions of the COVID-19 pandemic, making this work possible. Especially among others, Dr. Paul Fine, curator of Neotropical Flora at UC; Dr. Robert Vogt, Senior Curator at B; Dr. Freud Stauffer, Head Curator of Phanerogams, and his assistant secretary, Laurence Loze, at G; María del Rosário García-Peña, responsible for the Program of Loaning and Scanning of type specimens at MEXU; Dr. Nilber Gonçalves da Silva, and the technician Renata Santos, at R; Dr. Rosana Lopes and the technician Carla G. Manão, at RFA; Dr. Sylvia Orli, IT and Digitization Manager at US; and James C. Solomon, Curator Emeritus at MO. This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior — Brasil (CAPES) — Finance Code 001. The first author would like to thank CNPq and CAPES for the financial support throughout his research projects. GMA thanks CAPES, ASPT, the Bentham Moxon Trust, IdeaWild, and the Smithsonian Institution for financial support. MT thanks Alexander von Humboldt Foundation (proc. BRA/1139098), FAPERJ (procs. E26/203.269/2016 and E-26/202.708/2019—JCNE), and CNPq (proc. 470349/2013-1—Universal, proc. 306758/2019-9—Pq2).
Funding
This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) through grants awarded to ABO (grant# 132032/2020-1) and MT (grant# 306758/2019-9); Fundação de Amparo a Pesquisa do Rio de Janeiro (grant# E-26/202.708/2019) awarded to MT; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) — Finance Code 001 awarded to GMA; The American Society of Plant Taxonomists through a Graduate Student Research Grant for GMA; and Idea Wild grant awarded to GMA.
Declarations
Conflicts of interests/Competing interests
All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.
==== Refs
Literature cited
Antar, G.M., R.M. Harley, A.B. Oliveira, M.G. Buchoski, F. França, M.T. Faria, A.S. Soares, M.C.A. Mota, M.A. Schliewe & J.F.B. Pastore. 2022. Lamiaceae. In: Flora e Funga do Brasil. Jardim Botânico do Rio de Janeiro, Rio de Janeiro. http://floradobrasil.jbrj.gov.br/reflora/floradobrasil/FB142 (Accessed: 01 September 2022).
Bentham G. 1834. Labiatarum Genera et Species. Pp. 360–412. Ridgeway & Sons, London.
Bentham G de Candole ALPP Labiatae Prodomus systematis naturalis regni vegetabilis 1848 Paris Fortin. Masson & Sociorum 83 143
BGBM — Botanischer Garten und Botanisches Museum Berlin-Dahlem. 2021. 1913 to March 1, 1943. https://www.bgbm.org/en/history-collections-botanical-museum-berlin-dahlem-b-1. (Accessed: 13 December 2021).
Brade AC Labiadas Novas do Brasil Rodriguésia 1943 7 16 23 34
Bräuchler C Meimberg H Heubl G Molecular phylogeny of Menthinae (Lamiaceae, Nepetoideae, Mentheae) — Taxonomy, biogeography, and conflicts Molecular Phylogenetics and Evolution 2010 55 501 523 10.1016/j.ympev.2010.01.016 20152913
Briquet J Engler A Prantl K Labiatae Die Natürlichen Pflanzenfamilien nebst ihren Gattungen und wichtigeren Arten, insbesondere den Nutzpflanzen, unter Mitwirkung zahlreicher hervorragender Fachgelehrten begründet IV(3a) 1887 Engelmann, Leipzig Verlag von Willhelm 183 375
Briquet J Fragmenta monographiae Labiatarum IV — Labiatae americanae Kuntzeanae Bull. Herb. Boissier 1896 4 676 878
Briquet J. 1907. Bull. Herb. Boissier Ser. II, VII: 611–612.
Drew BT Sytsma KJ Phylogenetics, biogeography, and staminal evolution in the tribe Mentheae (Lamiaceae) American Journal of Botany 2012 99 933 953 10.3732/ajb.1100549 22539517
Drew BT Liu S Bonifacino JM Sytsma KJ Amphitropical disjunctions in New World Menthinae: three Pliocene dispersals to South America following late Miocene dispersal to North America from the Old World American Journal of Botany 2017 104 1695 1707 10.3732/ajb.1700225 29158343
Epling C Synopsis of the South American Labiatae Repertorium Specierum Novarum Regni Vegetabilis Beihefte 1936 85 97 192
Epling C Supplementary notes on American Labiatae V Brittonia 1951 7 3 129 142 10.2307/2804702
Epling C Játiva CD Supplementary notes on American Labiatae VIII Brittonia 1963 15 4 366 376 10.2307/2805381
Epling C Mathias ME Supplementary notes on American Labiatae VI Brittonia 1957 8 4 297 313 10.2307/2804980
Epling C Stewart WS A revision of Hedeoma, with a review of allied genera Repertorium Specierum Novarum Regni Vegetabilis Beihefte 1939 115 1 49
Harley RM Atkins S Budantsev AL Cantino PD Conn BJ Grayer R Harley MM De Kok R Krestovskaja T Morales R Paton AJ Ryding O Upson T Kadereit JW Labiatae The families and genera of vascular plants. VII. Flowering plants. Dicotyledons (Acanthaceae including Avicenniaceae) 2004 Berlin Springer 167 275
Hiepko P. 1980. Das Herbar von Rudolf Baschant. [The herbarium of Rudolf Baschant.] — Willdenowia 10: 83–86.
Irving RS The systematics of Hedeoma (Labiatae) Sida 1980 8 3 218 295
Prado J Hirai RY Moran RC Proposals concerning inadvertent lectotypifications (and neotypifications) Taxon 2015 64 3 651 10.12705/643.29
Rizzini C.T. 1997. Tratado de fitogeografia do Brasil: aspectos ecológicos, sociológicos and florísticos. Rio de Janeiro. Âmbito Cultural Edições Ltda. 2.ed., 747p.
Schmidt J.A. 1858. Labiatae (Lamiaceae). In: Martius, CFP. Flora Brasiliensis Frid. 8(1): 65–206, 215–226, tab: 14–38. Fleischer, Leipzig.
Sprengel CPJ Cryptogamia, Sect. II: Jungermannia Syst. Veg 1827 4 2 227
Stafleu F.A. & R.S. Cowan. 1976. Taxonomic literature 1(2): 508, A–G. Utrecht and Bohn: Scheltema and Holkema.
Thiers B. continuously updated. Index herbariorum: a global directory of public herbaria and associated staff. New York Botanical Garden’s Virtual Herbarium. http://sweetgum.nybg.org/science/ih/ (Accessed: 13 December 2021).
Turland N.J., J.H. Wiersema, F.R. Barrie, W. Greuter, D.L. Hawksworth, P.S. Herendeen, S. Knapp, W-H. Kusber, D-Z. Li, K. Marhold, T.W. May, J. McNeill, A.M. Monro, J. Prado, M.J. Price & G.F. Smith (eds.). 2018. International Code of Nomenclature for algae, fungi, and plants (Shenzhen Code) adopted by the Nineteenth International Botanical Congress Shenzhen, China, July 2017. Regnum Vegetabile 159. Koeltz Botanical Books, Glashütten. 10.12705/Code.2018.
Vasconcelos MF O que são campos rupestres e campos de altitude nos topos de montanha do leste do Brasil? Revista Brasileira de Botânica 2011 34 2 241 246 10.1590/S0100-84042011000200012
Xifreda C.C. & A.C. Malo A.C. 2004. Las especies argentinas de Glechon (Lamiaceae, Menthinae). Darwiniana 42(1–4): 333–346. 2004.
Xifreda CC Malo A.C AC El género Cunila (Lamiaceae, Mentheae) en Argentina Darwiniana 2006 44 1 298 308
Zhao F Chen YP Salmaki Y Drew BT Wilson TC Scheen A Celep F Bräuchler C Bendiksby M Wang Q Min D Peng H Olmstead RG Li B Xiang C An updated tribal classification of Lamiaceae based on plastome phylogenomics BMC Biology 2021 19 2 1 27 10.1186/s12915-020-00931-z 33407428
| 36466788 | PMC9707280 | NO-CC CODE | 2022-12-01 23:20:22 | no | Brittonia. 2022 Nov 28;:1-23 | utf-8 | Brittonia | 2,022 | 10.1007/s12228-022-09728-6 | oa_other |
==== Front
SN Comput Sci
SN Comput Sci
Sn Computer Science
2662-995X
2661-8907
Springer Nature Singapore Singapore
1457
10.1007/s42979-022-01457-7
Original Research
MERN Stack Web-Based Education Management Information Systems for Pacific Island Countries
http://orcid.org/0000-0002-7033-5506
Badru Lawal Olarotimi [email protected]
1
Vasudevan Vani [email protected]
2
Lingam Govinda Ishwar [email protected]
3
Khan M. G. M. [email protected]
1
1 grid.33998.38 0000 0001 2171 4027 The School of Information Technology, Engineering, Mathematics and Physics, The University of South Pacific, Suva, Fiji
2 grid.444321.4 0000 0004 0501 2828 Department of Computer Science and Engineering, Nitte Meenakshi Institute of Technology, Bengaluru, Karnataka India
3 grid.33998.38 0000 0001 2171 4027 The School of Pacific Arts, Communication and Education, The University of South Pacific, Suva, Fiji
29 11 2022
2023
4 1 702 3 2022
13 10 2022
© The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Generation of quality data to aid planning, assessing the education performance, monitoring programs implementation and learning outcomes are the basic functions of the Education Management Information System (EMIS). This study examines the existing EMIS in the Pacific Island Countries PICs and proffers solution. A SWOT analysis on the selected PICs EMIS through the published technical reports and policy documents from government and donors’ between years 2000 and 2021, revealed that EMIS in PICs have not progressed beyond the stages of collecting demographic data and generating basic indicators. Fiji EMIS which has prospect of leading other PICs EMIS could only generate few indicators manually, and the findings indicate that these indicators are not being considered in decision making. To solve these defects, we proposed data-driven microservices architecture developed with MERN (MongoDB, ExpressJS, ReactJS, and NodeJS) stack on 13 NoSQL collections, tested with pseudonymised data from Fiji Ministry of Education (consisting of 98.6% Learners, 100.2% Schools and 99.5% Teachers in post). Our solution renders dynamic visualized production-ready education data and 28 UNESCO standard indicators to guide decision making and this may serve as a model for PICs EMIS.
Keywords
Education indicators
Web-based application
School management solution
Microservices architecture
Fiji
issue-copyright-statement© Springer Nature Singapore Pte Ltd 2023
==== Body
pmcIntroduction
Education sectors have received increase attention in recent years on Artificial Intelligence (AI) in the areas such as profiling and prediction, intelligent tutoring systems, assessment and evaluation, and adaptive systems and personalization [1]. Learning Analytics on Learning Management System (LMS) in tertiary institutions have also started receiving researchers’ attention because of its importance in improving students' academic performances and aiding decision in policymaking [2, 3]. This advancement has uniquely separated plethora of information management systems from one another. Intelligent Information Systems (IIS) which combines AI advancement [4], database technologies and knowledge to exhibit intelligent behavior [5] in assisting users and other systems to retrieve or manipulate data [6, 7] becomes a game changer in allowing non-data analytics experts to make evidence-based decisions in complex situations [8]. Cruz-Domínguez and Santos-Mayorga [9] opined that IIS must possess the ability to efficiently store and retrieve large amounts of data needed to solve problems or make decisions. In spite of new contributions in IIS, there has been a growing demand for valid and reliable historical education data for planning, management of educational services and providing comparative information on pre and post-primary education performances in the developing countries [10].
The design specifications and operations of Education Management Information System (EMIS) are within the Ministries of Education country’s objectives and priorities, yet the efficacy is still largely low [11]. EMIS is described as a smart system embedded with functions of sensing, self-inducing and controlling in analyzing education situation and decisions on available data in a predictive or adaptive manner [12, 13]. However, a survey conducted by Asian Development Bank (ADB) in 2018 revealed that no EMIS in the Pacific Island Countries (PICs) appears to have progressed beyond the stages of collecting demographic data and reporting on sector indicators [14]. Those issues are still at the front burner in recent times, many countries have performed poorly in rendition of annual education indicators into the UNESCO education database.1 The situation is the same in other regions, as reproducing 5 years old data without data discrepancy [10] remains a major challenge, since education data at the ministry is managed manually which oftentimes are characterized with data overlap and data gap. These problems are responsible for delay in publishing annual education digest report and affect utilization of education data in planning and decision making by relevant stakeholders (i.e., Ministry of Education, Government, Schools, Parents and donors etc.). [15] observed that developing an EMIS is complex and takes several years to arrive at a working system to accommodate all stakeholders’ needs. EMIS implementation design models has received less attention and scholarly articles are limited in the public domain. One of the PICs EMIS reference most is Fiji EMIS (FEMIS). FEMIS became popular in 2017 following the world bank recognition for its evolving affordable school-centered integrated web solution [16]. However, Abdul-Hamid [17], Saraogi et al. [18] identified some weaknesses in FEMIS and these include:i. Architectural design deficiency (FEMIS is being used to manage the allocation of school grants and provision of vouchers to students against the generation of education statistic data which cater for education performances),
ii. Ineffective use in decision processes such as policy making and evaluation,
iii. Manual generation of a few education indicators,
iv. Lack of built-in validation mechanism which reduces the productivity of quality data analysts on data preparation and
v. Login security vulnerabilities among other weaknesses.
To solve these defects, this paper proposes a modern web-based EMIS termed “Intelligent EMIS (IEMIS)” for PICs through the lens of FEMIS’ weaknesses. The proposed solution is built on microservices architecture framework using MongoDB, ExpressJS, ReactJS, and NodeJS (MERN) stack. These technologies are free, open-source, cross-platform, JavaScript and implemented on 13 NoSQL collections of pseudonymised data2 consisting of 98.6% Learners, 100.2% Schools and 99.5% Teachers in post for the Year 2020 Academic Session. The NodeJS and ExpressJS bind web backend together with MongoDB (NoSQL database) which has an expressive query language, indexing for fast lookup and aggregation that provides powerful ways to access and analyze data and ReactJS handles the frontend that the users see and interact with [19]. The rest of the paper is organized as follows: “Evolution of EMIS in PICs” Section presents an overview of the related works on the evolution of Pacific EMIS and MERN stack underlying concept. “MERN Stack Concept” Section presents design methodology and implementation. Followed by result discussion and conclusion in Section “Design Methodology and Implementation” and “Discussion”, respectively.
Evolution of EMIS in PICs
Desktop Solution Era
Samoa EMIS (SEMIS) was built on the Microsoft Access system in 2001 with a data warehouse put in place to create various modules of data capturing in line with the Ministry of Education needs and demands [20] and there has not been any upgrade till now. Pacific Island Nation Evaluation Analysis Policy and Planning Leveraging Education Statistics (PINEAPPLES) built on Microsoft SQL Server and compatible with Microsoft applications became the Solomon Islands EMIS (SIEMIS) in 2004 [21]. Between 2004 and 2005 Microsoft Access System and Excel were used to manage education data in Papua New Guinea Education Management Information System (PNG EMIS) [22]. Fiji had three main standalone databases built on MySQL, namely SIMS (School Information System), FESA (Education Staffing Information System) and LANA (Literacy and Numeracy Assessment System) [18]. This desktop phase was also characterized by non-communication among databases, data were manually collected, collated and fed into the database with many data issues and data cleansing becomes important before analysis could be done for data integrity which invariably wasted most of data quality analysts’ productivity time, thereby delay result in the production of education data for planning, decision making and respond to an international survey request from donors – International Development Partner (IDP).
Unfortunately, many developing countries’ EMIS are yet to advance from this stage to a web-based solution, in a survey conducted in the Q3 of 2020 by the UNESCO Institute for Statistics (UIS), 25% of pacific countries still uses spreadsheet packages, 88% who claimed to have an EMIS in place recorded 25% with online interface [10] for collecting data. Tables 1 and 2 show data management and collection mode respectively.StatEduc is a built-in EMIS software developed with UIS involvement on system development.
Global ED*ASSIST or more commonly referred to as EdAssist is an integrated built-in EMIS software.
Table 1 Data management platform by region (% of countries)
Regions Own developed Stateduc Ed assist Openemis Others (excel)
Sub-saharan Africa 63 47 6 6 0
Arab states 85 8 0 15 7
South and West Asia 78 0 0 22 0
Central Asia 100 0 0 33 0
East Asia 100 0 0 0 0
Pacific 88 0 0 13 25
Latin America and caribbean 83 8 0 21 8
Central and eastern Europe 100 13 0 0 0
Average 80 18 2 13 5
Source: EMIS typology survey UIS (2020)
Table 2 Mode of data collection
Regions Paper (% of countries) Standalone electronic mode (% of countries) Online interface (% of countries)
Sub-saharan Africa 81 31 19
Arab states 39 31 46
South and West Asia 33 22 78
Central Asia 33 33 67
East Asia 67 67 67
Pacific 75 75 25
Latin America and Caribbean 42 38 71
Central and Eastern Europe 0 13 100
Average 53 36 51
Source: EMIS typology survey UIS (2020)
In a separate survey conducted by Global Partnership for Education (GPE) [23], showed 43% of developing country partners (DCPs) reported 10 out of 12 core education indicators to UIS and only 23% of countries worldwide currently report on the most important Sustainable Development Goals (SDG) global indicators with COVID-19 ravaging across the continents, the huge data gap is inevitable.
Integration of Web Presence in Pacific EMIS
FEMIS databases were upgraded to allow intranet for transmission of data into three (3) databases and later merged into two (2) physical databases vis-à-vis LANA and FESA, while information on SIMS database was migrated into FESA with a corresponding web application, FESA application maintains staffing data and learners’ detail and the LANA application maintains assessment data in the LANA database. Schools’ class lists are being administered in the LANA database using FEMIS and hardware design consists of a webserver to serve web pages and reports over the open Internet became available in 2017 [17, 18]. Vanuatu Government introduced Open VEMIS3 developed on FEMIS source code in the same year (2017) using the cloud hosting platform, while PNG adopted Oracle Application Express and old EMIS data was migrated into new web-based software development environment. Due to non-support for Discoverer by Oracle in 2015, PNG switched to Business Intelligence (BI) for querying, reporting, data analysis and web publishing. Solomon Islands adopted OpenEMIS4 and is in the process of recruiting consultant-project manager for managing its OpenEMIS.5 Figures 1 and 2 show web interface login page of OpenEMIS and own developed EMIS.Fig. 1 Maldives -OpenEMIS login page
Fig. 2 Fiji -FEMIS school login page
Challenges with Early Web Solution and the Modern Web App
Potency to manage data/information effectively gave rise to application of information technology in educational management [24]. Thus, early web solutions received low user satisfaction from inability to receive immediate feedback, a clear perception contrast and experience from desktop applications. Wang and Yang [25] attributed the synchronous interaction to traditional web application development model of request-wait-request, where browser renders blank pages until the server returns data pages which made user’s experience with desktop application different from early web solution. The emergence of Ajax (Asynchronous JavaScript and XML)6 in public in 2005 enhances the page rendering performances and the web2.0 application based on Ajax became a powerful competitor to Java-based desktop application.7 A new era from static to dynamic web page became popular, companies such as Google, IBM, Oracle, Yahoo and other famous companies that started the Open Ajax projects promote the adoption of Ajax development in web application. The built-in XMLHttpRequest "fetch()" function within JavaScript introduced around 2017 allowed the execution of Ajax on webpages to load content onto the screen without refreshing the page.8
With the explosion of web-enabled devices, Almeida and Monteiro [26] observed that users' web browsers and devices determine how web pages are rendered, this development forced the web application solutions to migrate to responsive web design that can adapt to the users' technology platforms such as user personal preferences, screen resolutions and internet connection speed. Nowadays, web-based application development has undergone a transformation with many cross-platforms, high-performance and open-source framework such as ASP.NET Core versions (.NET 5 released in November 20209), Ruby on Rails, Cake PHP, Laravel, Aftertaste, Spring Web MVC,10 Django, LAMP stack (Linux, Apache, MySQL or MongoDB, PHP) and MERN stack etc.
Modern Web Application with MERN Stack
Web apps are developed by widgets for quicker loading, interactive, mobile-responsive and Single Page Application (SPA). Researchers’ contributions in modern web development showed that MEAN (MongoDB, Express.js, AngularJS, and Node.js) and MERN (MongoDB, Express.js, ReactJS, and Node.js) stack framework are simple to learn, flexible, faster and more suitable for web-based application than LAMP stack framework [27, 28]. To support these claims, [27] designed and implemented real-time digital signage system (DSS), which supports handling real-time tasks like urgent/instant messaging, system status monitoring and conventional digital signage Content Management System (CMS) service tasks with both MEARN (ME(A + R)N) stack which performed better and render fetch data faster than DSS implemented with LAMP stack.
Aggarwal and Verma [29] compared two of the most popular and extremely powerful stacks used in the development of modern web applications MEAN and MERN stacks JavaScript framework, noticeable difference is AngularJS (supported by Google Company and uses TypeScript coding and Model View Controller (MVC)) and React.js (a Facebook initiative written in JavaScript and based on Virtual DOM (Document Object Model)). A comparison between both stacks showed that MEAN stack is a better option for large-scale applications with better productivity in design and operations because it is a complete framework. However, MEAN data flow is bidirectional which connotes a change in the user interface (UI) automatically changes the model state.11 In contrast, MERN stack is preferred for fast development of small-scale applications and suitable for cases that are JSON-heavy, cloud-native, dynamic web interfaces.12 Unlike MEAN, it has unidirectional data binding which provides better data overview in managing large projects. React.js is not a framework but a JavaScript library, this lowers its productivity as it is being used in conjunction with 3rd party libraries and remains the most popularly used stack by developers [30, 31].
MERN is being adopted in many of the recent modern web application research works. Bagade, et al. [32] developed the Farmer Portal System with MERN Stack, a modern web application that helps farmers to directly connect with the seller and also deploys a chatbot to clear some of customers doubts. Real-Time BPMN Website web application developed with MERN Stack by Handoyo, et al. [33], helps several users to make BPMN diagrams collaboratively. Deepika, et al. [31] implemented a Virtual Assistant AI-powered RASA framework with MERN stack, a web solution that was demonstrated with virtual C laboratory, and is accessible by the students anywhere with capacity for simulated laboratory assistants to help students solve problems faced during an experiment, which in turn reduces laboratory assistants spent time for solving students repetitive mistakes. Mishra, et al. [34] implemented Management Information System with MERN stack, the solution was specially designed for non-governmental organizations (NGOs) and the Government of India in reducing a lot of paper work by integrating database of registered NGOs to manage medium and small enterprises who want to apply for loan from the government. Analysis on search engine optimization techniques for a MERN stack web applications conducted by Shivani, et al. [35] revealed that Google page rank algorithm works best and yielded positive results which improve the ranking of application on search engines. The popularity of MERN in most recent web-based applications showed that it is more suitable when intensive interactive UI and faster development are required and thus, its adoption in the development of our proposed “IEMIS”.
MERN Stack Concept
The MERN architecture allows 3-tier architecture constructs (frontend, backend, database) entirely using JavaScript and JSON, Fig. 3 shows the interaction among the 3-tier architectures.Fig. 3 Full stack MERN stack architecture. Source: Clusterdb.com
Front End (React.js)
React.js is a client-side application that uses a declarative and versatile JavaScript library for building UI. Building complex UI is possible through Components (small, isolated and reusable pieces of code) which is then connected to backend server data, to render UI as HTML. React is suitable for handling stateful, data-driven interfaces with minimal code and support modern web framework and forms error handling.
Server Tier (Express.js and Node.js)
Backend (server-side application) is the backbone of any web app, this tier protects APIs (Application Programming Interfaces) against external attacks, authenticating users, enabling seamless interaction with databases, and handling user requests to collect and present required information. Express.js server-side framework runs inside a Node.js server. Express.js has powerful models for URL routing (matching an incoming URL with a server function) and handling HTTP requests and responses [19].
Database Tier (MongoDB)
MongoDB is a NoSQL database. It is document-oriented using collections and documents instead of traditional SQL databases tables and rows, where similar types of documents are grouped to form collections. MongoDB stores the data in binary JSON format (BSON-provides additional data types such as date and binary which is not supported in JSON) that allows the fast exchange of data between client and server. MongoDB works extremely well with Node.js [19], and makes storing, manipulating, and representing JSON data at every tier of application which makes it perfect for our proposed solution that allows individual records management with thousands of routine data manipulation operations.
Design Methodology and Implementation
Requirement Analysis
Basic EMIS requirements should house data for schools, learners, teachers and infrastructures, while a complete EMIS ideally should also include data on enrollments, attendance, completion rates, learning assessments, student health, finance, teacher characteristics, and some administrative statistics [36].
UIS [10] posited that among the four primary data sources used in calculating the SDG4 indicators such as household-based survey and assessment data; Annual School census data; school-based survey and assessment data; and EMIS or administrative data. The school/individual-based administrative data has the capacity to produce around half of the 43 thematic indicators. However, most PICs EMIS are not tailored towards this direction and this has been responsible for the low/no generation of indicators as evident in internationally comparable data in the UIS database.13
In this paper, we proposed 13 NoSQL collections and implemented 11 of these collections as depicted in the relationship diagram, while the description on all 13 collections on Fig. 4 and Table 3 respectively, Table 4 presents comparison data on number of records used to populate the IEMIS and Fiji Ministry of Education, Heritage and Arts (MoEHA) published 2020 Academic Session data. (Fig. 5).Fig. 4 IEMIS Collections Relationship Diagram
Table 3 IEMIS Collections description
Collection Description
Admin This collection contains details on Ministry of Education Administrative Management staff, access level and login credentials
District District details, login credentials and Education District Head officer name
Staff Staff details and login credential of all dedicated staff from MoEHA or District who handles assigned district administrative work on students, teacher, school and parents according to their role
Schools School details and login credentials for all schools
Teachers All Teachers’ information and login details with their privileges
Students All students’ details and other related school records
Parents Parent model captures information and login credentials about parent/guardian as the case maybe and allow monitoring of student progress in school from their dashboard
Classes All classes to ease class management of students and allow uniformity
Assessment captures student academic score per term by session, subject teacher can populate the model from teacher or school dashboard
Attendance All schools attendance details
Sessions Session details to track activities in a session and allow uniformity
Subjects Subject details provides efficient subject allocation to teacher and students taken the subject, available on school dashboard
Population estimate Population details on district projection population data by gender by single age
Table 4 Data comparison of 2020 academic session data published by MoEHA and IEMIS
Education Level Learners Schools Teachers
MoEHA IEMIS Variation MoEHA IEMIS Variation MoEHA IEMIS Variation
Pre-School 17,912 17,469 443 874 871 3 1426 1411 15
Primary 154,206 153,757 449 736 739 − 3* 6232 6222 10
Secondary 67,547 67,610 − 63* 172 175 − 3* 5665 5612 53
Vocational 806 1041 − 235* N/A 2 − 2* N/A 10 − 10*
Technical 2741
Total 243,212 239,877 594 1,782 1,787 − 5* 13,323 13,255 68
Fig. 5 IEMIS Model
The above table shows that pseudonymised data used were really close to the 2020 Academic Session data with deficit of 594 learners’ record, excess of 5 schools and 68 teachers shortfall. Due to data privacy policy in Fiji, only records that did not identify the record owners were released and used.
IEMIS Model
The activities diagram below captures, IEMIS users’ activities in addressing FEMIS’ weaknesses and other PICs EMIS’ challenges.
The IEMIS Model component are briefly discussed as follows;
Users
Seven (7) out of the proposed thirteen (13) collections are dedicated to IEMIS users for seamless data management, accessibility, data integrity and buy-in from stakeholders for effective use in planning and decision making, these users include Admins from the Ministry, District Heads, School Portal Admin, Teacher, Staff (non-teaching and other staff), Student and Parent. Figures 6 and 7 show all the IEMIS users and Admin login page with features on password reset and email notification.Fig. 6 Users in IEMIS
Fig. 7 Admin Login Pages
Management Dashboard
Any interaction with IEMIS is through the users’ dashboard which is accessible after successful login with proper authentication and authorization. Figures 8 below depict Admin dashboard. Anyone with Admin role can manage all users and generate interactive education indicators or statistics report at School, District and Ministry Level for decision making.Fig. 8 Admin Dashboard
IEMIS CRUD Operations
IEMIS is built on the FEMIS schemas, this enable users to perform CRUD operations through a protected API route. Few attributes were carefully structured to assist in the generation of education indicators for example CohortA combines “Admission Year” and “Entry Class” the code snippets below.
Implementation on the middlewares that must be passed before the API endpoint to delete school can be executed.
Delete School controller implementation.
Figure 9 shows Students’ CRUD operations available on MoEHA Admin dashboard.Fig. 9 CRUD Operations
Validation checker (frontend and backend) protects the integrity of data to be stored or prevent missing required data to be sent to the database by ensuring the schema types match with mongoose schema design. The following code snippets perform a validation check and ensure data consistency in the database.i. Backend Validator for creating IEMIS MoE Admin
ii. Validator for creating IEMIS MoE Admin from Frontend
IEMIS NoSQL Database
After a successful CRUD operations, MongoDB removes or stores data in a closer way to the applications which makes it highly scalable and providing superior performance even on low-resource devices [19]. Embedding and referencing design patterns are implemented in the IEMIS schemas, this improves fast retrieval of records, generation of statistics data and indicators with visualization to aid policy making. Figure 10 depicts a school record stores as JSON in MongoDB. (Fig. 11).Fig. 10 School record in School collection
Fig. 11 IEMIS Education Statistics and FEMIS dashboard
Education Statistics summary
Depending on the users’ request, the phase allows generation of aggregated data on basic education statistics in a granular decision making form. Each of the attributes in the collection were carefully selected to produce every day statistic data needed for generating education indicators for day to day school routine activities and decision making.
Education Indicators:
According to Mendes [37], a good education indicator system is expected to provide accurate and precise information to illuminate the condition of education and contribute to its improvement. Thus, indicators are expected to assist policymakers in the formulation of school goals and translate those goals into actions.
Panapa14 of PICs submitted that low response rates to international surveys reflected the lack of effective EMIS to produce data and indicators in line with international standards in the region. IEMIS implements 28 education indicators across;i. Access & Participation indicators used to analyze education coverage e.g. gross intake, net intake, adjusted net intake, gross enrollment, net enrollment etc.
ii. Internal Efficiency to measure the number of years it takes a child to complete a particular cycle or level of education e.g. repetition rates, promotion rate, survival by grade, school life expectancy etc.
iii. Resources Utilization indicators generate learners-teacher ratio.
Our IEMIS mathematical formula representation followed UIS Education Indicators Technical guidelines of November 2009, to produce real-time granular disaggregated education indicators in a dynamic visualized forms as shown on Fig. 12 and the FEMIS Score card on Indicators is depicted on Fig. 13.Fig. 12 Education Indicator
Fig. 13 Fiji Records with UNESCO on 2 indicators. Source: http://uis.unesco.org/en/country/fj. (Last accessed 14/10/2021)
Knowledge Based System
Applying formula correctly and interpreting result remains another challenge to many countries due to lack of technical expertise, this is why an integration of Knowledge Based System (KBS) is necessary to intelligently guide policy makers in making decision. Proposed KBS is a form of AI that captures the knowledge and analytical skills of experts in a specific problem domain to support decision making [38]. Baierle, et al. [39] observed that the development of a KBS aided by rules can support the decision-making processes for sales department of a company, the authors illustrated the application with a simplified Management Information System model on Fig. 14.Fig. 14 Knowledge Management Chain
We proposed IEMIS KBS to enhance the intelligence and will be implemented in the second phase of this research work. The KBS model would allow interpretation of generated indicators and implication to the country’s education policy. The report could be on current year or previous years and Predictive Analytics API would be integrated into KBS for an AI enhanced EMIS.
The proposed model is captured on Fig. 15 and a brief explanation on 3 out 28 education indicators are provided on Table 5 below:Fig. 15 Proposed IEMIS KBS Business Model
Table 5 IEMIS rule-based model
Explanation facilities Rules
Access & Participation Formula Explanation of indicator Purpose of the Indicator Action1 Action2 (year comparison) Action3 (complementary indicators information)
Gross Intake Ratio (GIR) GIRt=NtPat∗100
Where
GIRtGIRinschoolyeart
NtNumberofnewentrant
grade1inschoolyeart
PatPopulationofofficial
primaryschoolentranceageainschoolyeart
A high GIR indicates a high degree of access to primary education. As this result includes all new entrants to first grade (regardless of age), the ratio can exceed 100%, due to over aged and under-aged children entering primary school for the first time To indicate the general level of access to primary education IF $GIR ≥ 100
THEN “there are more over aged and under-aged children entering primary school for the first time. Hence, there is need embark on community sensitization exercise for enrollment into grade1 at official school age to reduce number of out of school children”
ELSEIF $GIR > $NIR THEN $GIR—$NIR “are percentage of over aged and under-aged school learners accommodated in grade1, a concerted effort should be made to improve admission drive at official age at grade1”
ELSE flag error_1
IF YearN $GIR < YearN-1 $GIR
AND YearN-1 $GIR > 0 THEN “A decline trend depicts reduction in over-age gap and a positive outcome in the adherence to official age entry”
ELSEIF YearN $GIR > = YearN-1 $GIR AND YearN-1 $GIR > 0 THEN “There is a significant number of children who did not enter primary education at the official age in the previous year”
ELSE flag error_2
IF YearN $GIR < YearN-1 $GIR AND YearN $NIR > YearN-1 $NIR THEN “The growth is healthy and should be sustained”
ELSE “This situation is unhealthy and the root cause should be investigated”
NET INTAKE RATE (NIR) NIRt=NatPat∗100
Where
NIRtNIRinschoolyeart
NatNumberofchildrenof
officialgrade1Pryschool
entranceageainschool
yeart
PatPopulationofofficial
primary s choolentranceagea
inschoolyeart
A high NIR indicates a high degree of access to primary education for the official primary school-entrance age children. NIR of 100% is a necessary condition for the policy goal of universal primary education To precisely measure access to primary education by the eligible population of primary school-entrance age IF $NIR > 100
THEN “Incorrect population data or Date of Birth of grade1 data, please cross check”
ELSEIF $ANIR > $NIR
THEN “This is an indication that a proportion pupils enrolled before official school age in the previous year, those pupils should be monitored in their academic performance” ELSE “The low NIR indicates a low degree of access to primary education of the official primary school-entrance age children”
IF YearN $NIR > YearN-1 $NIR AND YearN-1 $NIR > 0 THEN
An increasing trend reflects efficient policy goal of universal primary education, A positive indicator for reduced number out of school children ELSEIF YearN $NIR < = YearN-1 $NIR AND YearN-1 $NIR > 0 THEN
There is a significant number of children who did not enter primary education at the official age in the current year in review
ELSE flag error_2
ADJUSTED NEW INTAKE RATE (ANIR) ANIRt=Na,ktPat∗100
Where
ANIRtANIRinschoolyeart
NaktNumberofchildrenof
officialgrade1Pryschool
entranceageainanyclassk
ofschoolyeart
PatPopulationofofficialpry
schoolentranceageain
schoolyeart
A value of 100% indicates theoretically that the system has reached all children of the official primary school-entrance age This is used to measure the actual level of access to primary education of the population of the official primary school-entrance age IF $ANIR = 100
THEN “there is no out of school children of official school age in the country”
ELSE $ANIR—$NIR “is the actual percentage of official school age that had been enrolled in grade1 before their official age”
IF YearN $ANIR > YearN-1 $ANIR AND YearN-1 $ANIR > 0 THEN
An increasing trend early reflects early entry In grade1
ELSEIF YearN $ANIR < YearN-1 $ANIR AND YearN-1 $ANIR > 0 THEN A decline trend depicts reduction in underage gap
ELSE flag error_2
Cloud Services architecture
To leverage on cloud services’ scalability and availability, IEMIS MongoDB Atlas (database-as-a-service) clusters is paired with Heroku,15 one of the most popular Platform as a Service (PaaS) solutions for developers to build and deploy fully managed cloud applications services for both backend and frontend [40].
Results and Discussion
EMIS design specifications and operations are within ministry of education country’s objectives and priorities [36]. This flexibility has resulted into plethora of systems and deciding what system to use is a difficult and challenging task. A microservices architecture adopted in the development of proposed IEMIS allows data interoperability as no EMIS is optimally designed to produce all the datasets needed to meet the requirements of the education sector, by allowing the proposed system to separate portions of apps into logical sections that can be maintained, scaled, or updated independently. IEMIS provides a richer useful data require for decision making to improve the standard of education in the PICs through automated visual education indicators. FEMIS limitation on statistics data generation is enhanced with granular disaggregated with corresponding dynamic visualization charts. Decision-makers from schools to ministry can easily interpret their education data in real-time without the need for technical assistant or further analysis to understand the data. For instance, it is evident from Fig. 16 that no admission took place in the Year 2014 across the level of education while the trends showed Year 2020 recorded highest admission rate in the country. (Fig. 17).Fig.16 Student by year of Admission
Fig. 17 Granular Education data report
Authors have identified unavailable of data [41], lack of data integrity [42] and data not presented in a useful way16 as top reasons why data are not being used in decision making. EMIS in PICs is characterized by those reasons. IEMIS not only provides education data but also ensure granular disaggregated data with modern intelligent and dynamic visualization.
Automation on 28 education indicators makes IEMIS timely to PICs challenges in generating education indicators, the region has shown great concern and the data gap observed in the UIS database corroborated those challenges. However, since the population projected data falls outside the purview of the Ministry of Education. IEMIS implementation considers this constraint and allows indicator generation without population, EMIS officers can divide the result with corresponding population or historical data as they become available.
The security issue was also recognized as one of the weaknesses of FEMIS, with the use of UUID4 in IEMIS for salting and password hashing algorithm implementation the user credential is safe from attacker. A stateless authentication mechanism JSON Web Token (JWT) is also used for Auth implementation to protect API endpoints from unauthorized users.
Conclusion
IEMIS offers a modern fast non-blocking data-driven web-based application, the implementation prioritizes the availability of education data to facilitate planning & budgeting, policy formulation, decision making and routine management at the national, district and school level for PICs. It is imperative to note that pseudonymised data released from MoEHA is misleading and does not reflect the situation in the country. This level of distrust to share pseudonymised data (which does not compromise data privacy policy) would discourage researchers and EMIS developers to contribute towards the advancement of EMIS development if such practices continues.
3 out of 7 EMIS users’ dashboards (i.e., Ministry, District and School) have been implemented covering 11 collections (i.e., admin, district, school, staff, teacher, student, assessment, class, session population and parent) out of 13 proposed collections.
Cloud services (platform-as-a-service) such as servers and database-as-a-service remains cost-effective for web-based EMIS deployment because they offer high availability, scalability, elasticity and disaster recovery infrastructures. Migration to the cloud would relieve the government from the responsibility of managing infrastructures’ security and updates without impacting services availability and the problems of aftermath damage on physical ICT infrastructures witnessed during and after the impact of seasonal climate change disasters in the scattered PICs.
Further improvement on state management with Redux or its variants is required to prevent event listeners from Page reloads on screen resize, scroll and keyboard pop up on mobile devices.
Declarations
Conflict of Interest
My coauthors and I do not have any interests that might be interpreted as influencing the research “Conflicts of interest: none”.
Ethical Approval
This article does not contain any studies with human participants or animals performed by any of the authors.
1 http://uis.unesco.org/en/country/fj (Last accessed 14/10/2021).
2 Pseudonymisation of data means replacing any sensitive attributes which could be used to identify an individual with a pseudonym.
3 A cloud based information system for collecting, storing and disseminating basic data, statistics and indicators to support education planning and management in Vanuatu.
4 OpenEMIS is a built-in open-source EMIS software developed by Community System Foundation.
5 https://www.pacifictenders.com/tenders/recruitment-consultant-project-manager-open-solomon-islands-education-management-information.
6 Microsoft ASP.NET AJAX, lunched by Microsoft Company, is the package of Ajax technology and visible Ajax application framework.
7 Jerome Sonnet, Charles Savage. OGC Web Service Soap Experiment Report 0.8 Document#03–014, Jan 2003.
8 https://en.wikipedia.org/wiki/Ajax_(programming).
9 https://devblogs.microsoft.com/dotnet/announcing-net-5-0/.
10 https://opensource.com/article/20/3/frameworks.
11 https://medium.com/@devathon_/mean-vs-mern-stack-development-5ba3e517bc68.
12 https://www.mongodb.com/mern-stack.
13 http://data.uis.unesco.org/.
14 http://uis.unesco.org/en/blog/across-pacific-building-effective-education-management-information-systems.
15 https://blog.back4app.com/netlify-vs-heroku/.
16 https://bi-survey.com/decision-making-no-information.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Zawacki-Richter O Marín VI Bond M Gouverneur F Systematic review of research on artificial intelligence applications in higher education–where are the educators? Int J Educ Technol High Educ 2019 16 39 10.1186/s41239-019-0171-0
2. Lemay DJ Baek C Doleck T Comparison of learning analytics and educational data mining: A topic modeling approach Comput Edu: Artif Intell 2021 2 100016
3. Jacob J, Jha K, Kotak P, Puthran S. Educational data mining techniques and their applications. In 2015 International Conference on Green Computing and Internet of Things (ICGCIoT), 2015;1344–1348.
4. Shabbir J, Anwer T. Artificial intelligence and its role in near future. arXiv preprint arXiv:1804.01396 (2018).
5. Ras Z Advances in intelligent information systems 2018 NY Springer
6. Wiederhold G The roles of artificial intelligence in information systems J Intell Info Syst 1992 1 35 55 10.1007/BF01006413
7. Collins C Dennehy D Conboy K Mikalef P Artificial intelligence in information systems research: A systematic literature review and research agenda Int J Inf Manage 2021 60 102383 10.1016/j.ijinfomgt.2021.102383
8. Ahmed M Ahmad MS Yusoff MZM A collaborative framework for multiagent systems Int J Agent Technol Syst (IJATS) 2011 3 1 18 10.4018/jats.2011100101
9. Cruz-Domínguez O Santos-Mayorga R Artificial intelligence applied to assigned merchandise location in retail sales systems S Afr J Ind Eng 2016 27 112 124
10. UIS, "Data Innovation for Producing SDG 4 Indicators: An EMIS Metadata Global Analytical Report," UNESCO Institute for Statistics, P.O. Box 6128, Succursale Centre-Ville Montreal, Quebec H3C 3J7 Canada UIS/2020/LO/IP/65, September 2020 2020.
11. GPE. (2019). What can we learn from EMIS diagnostics? Available: https://www.globalpartnership.org/blog/what-can-we-learn-emis-diagnostics
12. Abdul-Hamid H. Data for Learning: Building a Smart Education Data System. World Bank Publications. 2017.
13. World-Bank. SABER annual report 2016 : smarter education systems for brighter futures (English). World Bank Group, 2017;114293 2017
14. Pillay HK, Velasco-Rosenheim RP, Thonden C. ICT for Better Education in the Pacific. Asian Development Bank, 2018.
15. Durnali M. Managing the Utilization of Technologies in Adult Education, Training, and Administration: The Case Study of Turkish MoNE. in Utilizing Technology, Knowledge, and Smart Systems in Educational Administration and Leadership. IGI Global 2020; 1–19.
16. Abdul-Hamid H, Data for Learning: Building a Smart Education Data System. World Bank Publications, 2017.
17. Abdul-Hamid H. Developing an Affordable and School-Centered EMIS: The Case of Fiji. ed, 2017.
18. Saraogi N, Mayrhofer DK, Abdul-Hamid H. Fiji Education Management Information Systems: SABER Country Report 2017. World Bank, 2017.
19. Hoque S Full-stack React Projects 2020 Birmingham, UK Packt Publishing
20. Mintz SM, Saraogi N. Samoa-Education management information systems. The World Bank 2015.
21. WorldBank G. Solomon Islands Education Management Information Systems: SABER Country Report 2015. World Bank, 2015.
22. World-Bank-Group. 2015, Papua New Guinea Education Management Information System : SABER Country Report 2015.
23. GPE. Meeting the Data Challenge in Education. Global Partnership for Education254/04/2021 2019.
24. Shah M Impact of management information systems (MIS) on school administration: what the literature says Procedia Soc Behav Sci 2014 116 2799 2804 10.1016/j.sbspro.2014.01.659
25. Wang H, Yang J. Research and application of web development based on ASP. NET 2.0+ Ajax. in 2008 3rd IEEE Conference on Industrial Electronics and Applications, 2008;857–860.
26. Almeida F, Monteiro J. The Role of Responsive Design in Web Development. Webology 14, (2017)
27. Khue TD Nguyen TB Jang U Kim C Chung S-T Design and Implementation of MEARN Stack-based Real-time Digital Signage System J Korea Multimedia Soc 2017 20 808 826 10.9717/kmms.2017.20.5.808
28. Porter P, Yang S, Xi X. The Design and Implementation of a RESTful IoT Service Using the MERN Stack. in 2019 IEEE 16th International Conference on Mobile Ad Hoc and Sensor Systems Workshops (MASSW). 2019, pp. 140–145.
29. Aggarwal S, Verma J. Comparative analysis of MEAN stack and MERN stack 2018.
30. Shetty J, Dash D, Joish AK, Guruprasad C. Review Paper on Web Frameworks, Databases and Web Stacks. Intern Res J Eng Technol (IRJET), 07: 5, 2020.
31. Deepika N, Bala MM, Kumar R. Design and implementation of intelligent virtual laboratory using RASA framework. Materials Today: Proceedings 2021.
32. Bagade S, Bhatt M, Gondhali S, Jadhav V. Farmer Portal System using MERN Stack 2021.
33. Handoyo R Santoso LW Setiawan A Real-time BPMN website menggunakan teknologi MERN stack Jurnal Infra 2019 7 75 80
34. Mishra SK Kumar A Kaundal A Joshi U Management Information System Int J Res Eng Sci Manag 2021 4 144 146
35. Shivani G, Mutreja P, Nagrath P. Analysing search engine optimization techniques for a full stack web application. in IOP Conference Series: Materials Science and Engineering. 2021;012097.
36. Wyk CV, Crouch L. Efficiency and Effectiveness in Choosing and Using an EMIS. Available: http://emis.uis.unesco.org/wp-content/uploads/sites/5/2020/09/EMIS-Buyers-Guide-EN-fin-WEB.pdf 2020.
37. Mendes S. Progress and result indicators and their relevance for educational policy analysis. 2003.
38. Tripathi K A review on knowledge-based expert system: concept and architecture IJCA Special Issue Artif Intell Technique-Novel Approach & Practical Appl 2011 4 19 23
39. Baierle IC Sellitto MA Frozza R Schaefer JL Habekost AF An artificial intelligence and knowledge-based system to support the decision-making process in sales S Afr J Ind Eng 2019 30 17 25
40. Ansaharju J. Improving Software Development with Platform-as-a-Service Product–Using Heroku in Web Application Project. 2016.
41. Svensson RB, Feldt R, Torkar R. The unfulfilled potential of data-driven decision making in agile software development. in International Conference on Agile Software Development, 2019;69–85.
42. Beck B Seeliger D Kriegl JM The impact of data integrity on decision making in early lead discovery J Comput Aided Mol Des 2015 29 911 921 10.1007/s10822-015-9871-2 26409840
| 36467856 | PMC9707281 | NO-CC CODE | 2022-12-01 23:20:22 | no | SN Comput Sci. 2023 Nov 29; 4(1):70 | utf-8 | SN Comput Sci | 2,022 | 10.1007/s42979-022-01457-7 | oa_other |
==== Front
Found Comut Math
Found Comut Math
Foundations of Computational Mathematics (New York, N.y.)
1615-3375
1615-3383
Springer US New York
9599
10.1007/s10208-022-09599-z
Article
Counting Real Roots in Polynomial-Time via Diophantine Approximation
Rojas J. Maurice [email protected]
grid.264756.4 0000 0004 4687 2082 Department of Mathematics, Texas A & M University, TAMU 3368, College Station, TX 77843-3368 USA
Communicated by Peter Bürgisser.
28 11 2022
143
29 12 2020
17 9 2022
19 9 2022
© SFoCM 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
Suppose A={a1,…,an+2}⊂Zn has cardinality n+2, with all the coordinates of the aj having absolute value at most d, and the aj do not all lie in the same affine hyperplane. Suppose F=(f1,…,fn) is an n×n polynomial system with generic integer coefficients at most H in absolute value, and A the union of the sets of exponent vectors of the fi. We give the first algorithm that, for any fixed n, counts exactly the number of real roots of F in time polynomial in log(dH). We also discuss a number-theoretic hypothesis that would imply a further speed-up to time polynomial in n as well.
Keywords
Sparse polynomial system
Real root
Positive root
Circuit
Baker–Wustholtz theorem
Descartes’ rule
Rolle’s theorem
Mahler’s theorem
Gale dual
Mathematics Subject Classification
Primary: 14P99
Secondary: 11J86
14Q20
65Y20
==== Body
pmcIntroduction
Solving sparse polynomial systems remains a challenging problem, even 40 years after the dawn of fewnomial theory [61, 62]. More recently, connections have emerged between fewnomial theory over finite fields, cryptography, and number theory [26, 32, 37], and sparse polynomial systems over the real numbers continue to form the foundation of applications including computational biology and biochemistry [18, 19, 42, 43] and circuit complexity [64]. However, efficiently counting the number of real roots and even just finding a reasonably tight upper bound on the number of real roots are still open problems. Here, we focus on the problem of exactly counting real roots and roots in any given orthant. In what follows, all O-constants and o-constants are absolute (and can be made explicit), time will refer to the number of (deterministic) bit operations in the classical Turing model of computation, and we will use #S for the cardinality of a set S.
Assuming A={a1,…,at}⊂Zn, xaj:=x1a1,j⋯xnan,j, and f(x)=∑j=1tcjxaj∈Zx1±1,…,xn±1, we define the support of f to be Supp(f):={aj|cj≠0}. We then call a system of the form F:=(f1,…,fn)∈Zx1±1,…,xn±1n, with fi(x):=∑j=1tci,jxaj for all i and #⋃i=1nSupp(fi)=t, a t-nomial n×n system (over Z) supported on A. We denote the positive orthant by R+n, R∗:=R\{0} and call a root of F in R+n a positive root.
If the aj do not all lie in the same affine hyperplane, then we clearly have t≥n+1. It is natural to assume that the exponent vectors are non-coplanar in this sense, and we will do so, for otherwise one could use a monomial change of variables to reduce F to a system in fewer variables: See Remark 2.6 from Sect. 2. Our main theorem gives a dramatic new speed-up for counting the exact number of real roots of F in the special case t=n+2.
Theorem 1.1
Following the notation above, assume further that t≤n+2, A⊂{-d,…,d}n, and the coefficient matrix [ci,j] lies in {-H,…,H}n×(n+2) and is generic. Then, in time (n2log(dH))(2+o(1))n, we can determine the number of roots of F in Rn, (R∗)n, and R+n. Furthermore, if t=n+1, then we can do the same in time n3.373log1+o(1)(dH).
We prove Theorem 1.1 in Sect. 5.1, based mainly on Algorithms 4.1 and 4.3 from Sect. 4. A key new ingredient is diophantine approximation over number fields.
Remark 1.2
We count roots without multiplicity. In particular, degenerate1 isolated roots are not a problem and are counted correctly by our algorithms. ⋄
A sufficient genericity condition for our faster counts in (R∗)n and R+n is the non-singularity of the n×n sub-matrices of [ci,j]. This condition is checkable in time On4.373log1+o(1)(nH): See Lemma 2.15 and Corollary 2.18 of Sect. 2.3. In particular, the fraction of coefficient matrices failing to satisfy this genericity condition is no greater than n(n+2)(n+1)4H+2. A sufficient genericity condition for faster counting in Rn is detailed in Sect. 5 and also holds practically often: Probability 1-ε when H has Ωnlog(nd)+log1ε bits (see Remark 5.8 of Sect. 5.1).
Root counting without genericity assumptions is rather non-trivial: Deciding finiteness for the number of real (or positive) roots in time (nlog(dH))O(n) (i.e., roughly as fast as our main algorithms), when t=n+2 and f2=⋯=fn identically, is still an open problem [15, 22]. Furthermore, for any fixed ε>0, deciding whether the non-generic system F=(f1,…,f1) has any real (or positive) roots is NP-hard already for t=n+nε [22]: With our underlying measure of size being nt(log(H)+nlog(d)) (which has asymptotically the same order as the number of bits needed to write down all the monomial terms of F), the existence of an algorithm with complexity (nlog(dH))O(1), for just one ε>0, would imply P=NP.
Other than an algorithm for the very special case (n,t)=(1,3) from [22], the best previous deterministic complexity bound for t=n+2 appears to have been (nnlogndn)O(log2n) arithmetic operations [9], via an algorithm solving the harder problem of computing roadmaps (see, e.g., [8, Ch. 15]) for arbitrary real algebraic sets. One can also speed up to a (dlogH)O(n) arithmetic complexity bound via [85] if one assumes the complex zero set is finite. (All of these works build upon the seminal works [12, 33, 38, 89]. See also [95] for more recent speed-ups via randomization.) There have also been important recent advances from the point of view of numerical conditioning (e.g., [40, 41]), even enabling practical computation of homology of real projective sets, but work in this direction has not yet focused on speed-ups like Theorem 1.1: With few exceptions, earlier work on solving polynomial systems over the real numbers focused on coarser complexity bounds that ignored the finer monomial term structure.
Example 1.3
Consider the 7-nomial 5×5 system F=(f1,…,f5) defined by
Then, Algorithm 4.3 from Sect. 4 (simulated in a few lines of Maple code2) tells us in under a second that F has exactly 2, 6, 6, 2, 2, or 0 positive roots, respectively, when c is 120731, 120730, 114392, 114391, 113059, or 113058. (All roots in (R∗)5 of these F happen to lie in R+5.) One can also easily check that each these F has infinitely many roots in R5, since they each vanish identically on the 3-plane {x1=x4=0}. We will return to this family in Sect. 2.3, and see another example there as well. It is interesting to observe that Maple’s Solve command (which employs Gröbner bases) gives no useful information about any of these systems, even after 3 hours.2 Bertini (a state-of-the-art homotopy solver, Version 1.4 [11]), on each of the preceding systems, immediately returns a message stating“ERROR: The system is numerically zero 0! Please input a non- degenerate system. Bertini will now exit due to this error.”
This is partially because each such F has3 over 245 million roots in (C∗)5, and older polynomial system solving techniques have complexity super-linear, or worse, in the number of complex roots. ⋄
The main intent of Theorem 1.1 is to set the stage (building on the framework of [48, 49, 83]) for more practical improvements in real-solving such as complexity sub-exponential in n, in the average-case/smoothed analysis setting, for sparse systems. In particular, just as binomial systems are a building block for polyhedral homotopy algorithms for arbitrary n×n systems [58, 69, 102], (n+2)-nomial n×n systems are a building block behind recent optimization techniques such as SONC/SAGE-optimization (see, e.g., [35, 44, 84]). While tackling the remaining exceptional cases (e.g., infinitely many real roots in (R∗)n) is important, such cases are provably rare for random coefficients.
Connection to Fewnomial Bounds Over R
There has been growing interest in generalizing Descartes’ Rule of Signs (see, e.g., [53, 100]) from univariate polynomials to n×n polynomial systems. This began with Khovanski’s seminal Theorem on Real Fewnomials [62] which, in our notation, asserted an upper bound of 2t2(n+1)t for the number of non-degenerate positive roots of any t-nomial n×n system. It was then shown in [71] that Khovanski’s bounds could be greatly improved for various structured systems, e.g., the correct tight upper bound on the number of isolated4 positive roots for 2×2 systems of trinomials is 5 — far less than the best previous bound of 248832. Sharper upper bounds for new families of systems, including a tight upper bound of n+1 (resp. (n+1)2n) non-degenerate roots in R+n (resp. (R∗)n) for the case t=n+2 were then derived in [14]. Explicit families of systems attaining these bounds for each n were then given in [86] (see also [15, 21]). Khovanski’s general upper bound was vastly improved to e2+342t-n-12nt-n-1 positive roots in [23], and a remarkable (sometimes much sharper) bound for curve intersections was derived later in [63]. More recently, an elegant and near-optimal average-case upper bound of 12n-1·t!n!(t-n)! for the number of positive roots was proved in [31], using independent real Gaussians for the coefficients.
Fewnomial bounds so far have not made significant use of the signs of the coefficients (much less their values) when n≥2, and such refined bounds remain elusive: see, e.g., [10, Thm. 2.1] and [16, 17, 20]. The latter works, particularly [17], culminated in a refined characterization of the maximal number of positive roots — incorporating the signs of n×n sub-determinants of the coefficient matrix [ci,j] and the matroidal structure of A —in the case t=n+2. Nevertheless, no algorithm for exactly counting the real or positive roots, faster than combining more general results on rational univariate reduction (see, e.g., [66, 90, 94]) with the computation of real dimension (see, e.g., [7]) or real root isolation (see, e.g., [96]), appears to have been known before Theorem 1.1.
Exactly counting the real or positive roots of F, and even formulating a reasonable generalization of Descartes’ Rule, appears to be much harder for t≥n+3. This is why there is much recent attention on the case t=n+2 to develop further intuition. An even harder open question is the complexity of actually approximating the real roots of such F and we hope to address this in future work. For instance, finding real approximate roots (in the sense of Smale [24]) in deterministic time (log(dH))O(1), just for the special case (n,t)=(1,3), is still an open problem [27, 59, 92].
Our main tools are reduction to a canonical form (a special case of Gale Dual Form from [23]) and a careful application of diophantine approximation to the critical values of this reduction. In particular, the locus of F with degenerate roots forms a discriminant variety which partitions the coefficient space into connected open regions we call chambers (see, e.g., [52, Ch. 11]). Classical topological results, such as Hardt’s Triviality Theorem [55], tell us that counting the real roots of F is tantamount to identifying the chamber in which F lies. Such a calculation is challenging, since the theory of A-discriminants [52] does not directly provide a tractable description of our chambers. However, applying Rolle’s Theorem to the resulting Gale Dual Form allows one to replace chamber identification by the determination of signs of the critical values and poles of a single univariate rational function.
A new obstacle is that the usual univariate root-finding algorithms, combined with classical height bounds on polynomial roots, do not yield a useful complexity bound. Indeed, the degree of the resulting univariate reduction can be so high that a naive use of real root isolation would lead to complexity super-linear in nn/2dn. So we leverage the special structure of the derivative of our univariate reduction to apply a powerful theorem from diophantine approximation: A refinement of an estimate of Baker and Wustholtz on linear forms in logarithms of algebraic numbers (see, e.g., [4, 6, 30, 77]).
Linking Diophantine Approximation and Algorithmic Complexity
A nice warm-up to the Baker–Wustholtz Theorem is the following problem:
Rational Binomial Sign Problem (RBSP). Given positive rational numbers α1,…,αm with numerators and denominators no greater than A, and integers b1,…,bm with absolute value at most B, what is the sign (+, −, or 0) of the binomial ∏i=1mαibi-1?
While the RBSP is a fundamental real algebraic question, there is a still an exponential gap between its best current lower and upper complexity bounds. [22] applied the theory of A-discriminants [52] to show that determining the isotopy type of the real zero set of a single n-variate (n+2)-nomial over Z is (under mild assumptions) equivalent to the RBSP. A consequence of this connection between the RBSP and real zero sets was then the first polynomial-time algorithm to decide non-emptiness for the real zero sets of (single) n-variate (n+2)-nomials with n fixed [22].
Note that the RBSP is clearly equivalent to deciding the sign of the linear combination of logarithms Λ(α,b):=∑i=1mbilogαi. Indeed, since simply evaluating the binomial can lead to numbers with huge bit-length, it makes sense to solve the RBSP by instead approximating the linear combination of logarithms Λ(α,b) to sufficient accuracy. However, this is feasible only if we know a sufficiently good (and explicit!) lower bound on the minimum of |Λ(α,b)| over all such αi and bi. Alan Baker won a Fields medal in 1970, due in large part to finding such a bound—over arbitrary number fields—and deriving numerous landmark results in number theory as a consequence [4]: The most recent refinements of his bound [77, 80], in the special case of Q, can be coarsely summarized as follows.1 Λ(α,b)≠0⟹log|Λ(α,b)|>-[O(logA)]mlogB(CoarseBaker′sBound)
Combined with classical results on quickly approximating logarithms [13, 28, 29, 97], Bound (1) readily implies that the RBSP can be solved in time [O(logA)]mlogB [22], and thus polynomial-time if m is fixed.
Remark 1.4
It is perhaps surprising that efficiently distinguishing between Λ(α,b)=0 and Λ(α,b)≠0 (neglecting positivity or negativity) has already been known for decades: This simplification of the RBSP can be solved in time (mlog(AB))O(1) via gcd-free bases (see, e.g., [23, Ch. 4] and [22]). ⋄
Sadly, little is known about the true asymptotics of log|Λ(α,b)|. However, around 1978, Lang and Waldschmidt conjectured that a much sharper bound of the shape2 Λ(α,b)≠0⟹log|Λ(α,b)|>-O(mlog(AB))(Lang-WaldschmidtConjecture)
should hold [68, Pg. 213] and this conjecture remains open. Later, Baker proved [5] that a bound of the form3 Λ(α,b)≠0⟹log|Λ(α,b)|>-O(mlog(A)logB)(ConsequenceofRefinedabc)
follows from a refined version [5] of the famous Masser-Oesterle abc-Conjecture [76, 81, 82]. Unfortunately, the latter refinement also appears out of reach.
Such conjectures are important not just in diophantine geometry but also in real algebraic geometry:
Theorem 1.5
[50]5 The truth of either of the Lang–Waldschmidt Conjecture or Baker’s refined abc-Conjecture implies that we can decide non-emptiness for the positive zero set of any f∈Zx1±1,…,xn±1—with coefficients in {-H,…,H}, and support of cardinality n+2 lying in {-d,…,d}n not contained in any affine hyperplane—in time (nlog(dH))O(1). □
Our framework here implies an even deeper conditional speed-up: Sufficiently sharp lower bounds for linear forms in logarithms of real algebraic numbers imply that root counting in Rn, for the (n+2)-nomial n×n systems from Theorem 1.1, can be sped up to time polynomial in n as well. We formalize this through a conjecture, and a theorem, below:
Real Algebraic Log Conjecture (RALC)
Suppose K is a real, degree d algebraic extension of Q, α1,…,αm∈K, b1,…,bm∈Z\{0}, logA is the maximum of the logarithmic heights6 of the αi, and B:=maxi|bi|. Then, there are constants x0,C>0 such that Λ(α,b)≠0 ⟹log|Λ(α,b)|>-(dmlog(AB))C for all d,m,A,B≥x0.
Note that the latest refinement of the Baker–Wustholtz Theorem by Matveev [77, Cor. 2.3] (stated as Theorem 2.25 in Sect. 2.4) can be coarsely summarized asΛ(α,b)≠0⟹log|Λ(α,b)|>-d2log(d)[O(logA)]mlogB.
Note also that the special case d=1 of the RALC (with C≤2) is implied by the truth of either of the Lang–Waldschmidt Conjecture or Baker’s Refined abc-Conjecture.
The special case d=1 case of the RALC (with any C>0) is an even weaker hypothesis that still implies the speed-up for fewnomial hypersurfaces from Theorem 1.5 above. If we have the RALC true in full, then we also obtain the following speed-up for fewnomial systems:
Theorem 1.6
If the RALC is true then the complexity bound from Theorem 1.1, for counting roots in Rn, (R∗)n, and R+n, in the hardest case t=n+2, can be replaced by (nlog(dH))O(1).
We prove Theorem 1.6 in Sect. 6. It is tantalizing to speculate that some kind of converse may hold, i.e., sufficiently fast real root counting for sparse polynomial systems may imply a strengthenings of Baker’s Theorem on linear forms in logarithms. Such an implication is still not clear, and it seems natural that information like the underlying real root-spacing is also needed. It would indeed be worthwhile to prove new root-spacing bounds for (n+2)-nomial n×n systems, in order to apply them to improving Baker’s Theorem. However, the current state of the art goes in the opposite direction: [65, 93] prove new root-spacing bounds in the case n=1 over C (and even Cp), but heavily use Baker’s Theorem (and its p-adic cousin, Yu’s Theorem [103]).
What is clearer (and true) is the contrapositive of Theorem 1.6: If counting real roots for generic (n+2)-nomial n×n systems (as in Theorem 1.1) is not doable in time (nlog(dH))O(1), then the RALC is false. Similarly, if deciding non-emptiness for real zero sets of n-variate (n+2)-nomials (as in Theorem 1.5) is sufficiently hard, then we can falsify both the Lang–Waldschmidt Conjecture and Baker’s Refined abc-Conjecture. These are curious examples of how proving complexity lower bounds in real algebraic geometry is as hard as falsifying conjectures from diophantine approximation.
The hardness of proving new diophantine approximation bounds is one reason that new average-case speed-ups, using geometric numerical conditioning techniques (e.g., [48, 49]) instead of diophantine approximation, may arrive sooner than new worst-case speed-ups.
Background
The Complexity of Linear Algebra Over Z
Let ω denote the well-known matrix multiplication exponent, i.e., the infimum over all ω such that there exists an algorithm that can multiply an arbitrary pair of n×n matrices, in any field K, using O(nω) field operations in K. The best current upper bound is ω<2.3728596 [2, 70]. Recall the notions of reduced row echelon form and leading entries of a matrix, from basic linear algebra (see, e.g., [87]). For any nonzero rational number pq with p,q∈Z and gcd(p,q)=1, its (absolute) logarithmic height is h(p/q):=max{log|p|,log|q|}. (We set h(0):=0.) We will first need a result on the bit complexity of row reduction for matrices:
Lemma 2.1
[101, Pg. 17 & Cor. 2.12]7 Suppose M∈Zn×t has rank r and all the entries of M have absolute value at most H. Then, in deterministic timeOntr[rω-2loglog(rlog(rH))+log(r)log2(rlog(rH))]log(rH)
we can find the reduced row echelon form R∈Qn×t of M. Furthermore, every nonzero entry of R has logarithmic height O(rlog(rH)). □
An illuminating alternative discussion of the bit complexity of linear algebra can be found in [24, Ch. 15, Sec. 15.5].
Via Cramer’s Rule and Hadamard’s classical inequality on the absolute values of determinants [78, Thm. 1], we can easily obtain the following related bound:
Lemma 2.2
If A∈Zn×(n+1) has rank n and all the entries of the ith row of A have absolute value at most di, then any generator (b1,…,bn+1)⊤∈Z(n+1)×1 of the right-null space of A, with gcd(b1,…,bn+1)=1, satisfies |bj|≤nn/2∏i=1ndi for all j. □
Definition 2.3
[57, 98] We call a matrix U∈Zn×n with determinant ±1 unimodular. Given any matrix M∈Zn×t, we then call any identity of the form UM=R, with U∈Zn×n unimodular and R upper-triangular with all leading entries positive, a Hermite factorization. Finally, we call any identity of the form UMV=S, with U∈Zn×n and V∈Zt×t both unimodular, and S with diagonal entries s1,…,sn satisfying s1|s2,…,sn-1|sn, a Smith factorization of M. ⋄
We will also need the following complexity bound on Smith factorization:
Theorem 2.4
[101, Ch. 8, Prop. 8.10] Suppose M∈Zn×t has rank r and all its entries having absolute value at most d. Then, a Smith factorization UMV=S for M can be found in deterministic timeOntr[rω-2loglog(rlog(rd))+log2(rlog(rd))]log(nt)log(rd)
with all the entries of U, V, S having logarithmic height O(rlog(rd)). □
The bound above also assumes that we use an O(hlogh)-time algorithm for h-bit integer multiplication, e.g., [56].
Binomial and (n+1)-Nomial Systems Over (R∗)n
A simple, folkloric algebraic/analytic fact we will need is the following:
Proposition 2.5
Suppose A,B∈Zn×n and x=(x1,…,xn) is a vector of indeterminates. Let us define xA to be the vector of monomials x1a1,1⋯xnan,1,…,x1a1,n⋯xnan,n, where A=[ai,j]. Then, (xA)B=xAB and, if A is unimodular, the function defined by x↦xA defines an analytic group automorphism of (C∗)n that restricts to an analytic group automorphism of R+n. □
Remark 2.6
A simple consequence of Proposition 2.5 is that if f∈Rx1±1,…,xn±1 is an n-variate t-nomial with support A, and d is the dimension of the smallest affine subspace containing A, then there is a monomial change of variables x=yU (with U unimodular), and a monomial yb∈Ry1±1,…,yd±1, such that g(y):=ybfyU∈Ry1±1,…,yd±1 is a d-variate t-nomial, and the zero set of f in (R∗)n is analytically isomorphic to the Cartesian product of (R∗)n-d and the zero set of g in (R∗)d. So A in an affine hyperplane implies that the zero set of f in (R∗)n can be easily characterized by the zero set of another t-nomial in fewer variables. ⋄
Another consequence of Lemma 2.1 is that we can almost trivially count the positive roots of binomial systems, provided the exponent vectors are in general position.
Proposition 2.7
Suppose c=(c1,…,cn)∈(R∗)n, a1,…,an∈Zn, A is the n×n matrix with jth column aj for all j, UAV=S is a Smith factorization of A, and c′:=c1′,…,cn′:=cV. Let sj be the (j, j) entry of S. Then, G:=(xa1-c1,…,xan-cn) and y1s1-c1′,…,ynsn-cn′ have the same number of roots in R+n (resp. (R∗)n, (C∗)n). In particular, G has exactly 0, 1, or infinitely many roots in R+n under the following respective conditions: : Some ci is negative or [Rank(A)=j<n and ci′≠1 for some i∈{j+1,…,n}].
: c∈R+n and detA≠0.
: c∈R+n, Rank(A)=j<n, and cj+1′=⋯=cn′=1. □
Proposition 2.7 follows directly from Proposition 2.5. Both facts are folkloric in the toric geometry/Lie group literature (see, e.g., [58] and [25, Ch. 3]). A more in-depth discussion of binomial systems can be found in [34, 36, 83].
Counting roots in (R∗)n is slightly more complicated but still admits efficient formulae.
Proposition 2.8
Following the notation of Proposition 2.7, assume the exponent vectors a1,…,an are linearly independent. Let r denote the rank, over the field F2, of the mod 2 reduction of A. Then, the map m:(R∗)n⟶(R∗)n defined by m(x):=xA is 2n-r-to-1, and the ith coordinate of the range of mV is R∗ (resp. R+) if and only if i≤r (resp. i≥r+1). In particular, F has exactly 0 (resp. 2n-r) roots in (R∗)n if and only if ci′<0 for some (resp. no) i≥r+1.
Proof
First note that by the definition of Smith factorization, we have that the diagonal entries si of S are such that s1,…,sr are odd and sr+1,…,sn are even. By Proposition 2.5, exponentiating by U or V induces a permutation of the open orthants of Rn. In particular, we see that the range of x↦xS is exactly (R∗)r×R+n-r. So the pre-image assertion on m is proved.
Now, note that the range of m must be (R∗)r×R+n-rV-1 thanks to Proposition 2.5. So now we know the range of m.
The final remaining assertion follows from our earlier definition c′:=cV and our earlier assumption that c∈(R∗)n. □
We can now state more explicitly how we deal with positive root counting for t-nomial systems in the case t=n+1.
Lemma 2.9
If F=(f1,…,fn)∈Zx1±1,…,xn±1n is an (n+1)-nomial n×n system, with union of supports A={a1,…,an+1} not lying in an affine hyperplane, and the coefficient matrix of F has rank n, then the number of positive roots of F is either 0 or 1. Furthermore, if all the coefficients of all the fi have absolute value at most H, then we can determine the number of positive roots of F in time n3.373log1+o(1)(nH).
Remark 2.10
The reader disturbed by the complexity bound being independent of A may be reassured to know that (a) checking the hyperplane condition takes time dependent on A and (b) the analogue of our lemma for counting roots in (R∗)n (Corollary 2.11) has complexity depending on A. ⋄
Proof of Lemma 2.9:
By our assumption on the coefficient matrix, we may reorder monomials so that the left-most n×n minor of the coefficient matrix has nonzero determinant. So we may divide every fi by xan+1 without changing the roots of F in (C∗)n, and assume a1,…,an are linearly independent and an+1=O.
From Lemma 2.1 (and the fact that ω<2.373 [2]), it is then clear that we can reduce the coefficient matrix of F, [ci,j]∈Zn×(n+1), to a reduced row echelon form in Qn×(n+1), in time n3.373log1+o(1)(nH). The underlying linear combinations of rows can then be applied to the equations fi=0 so that F=O can be reduced to a binomial system of the form xA=γ where γ=(γ1,…,γn), A∈Zn×n, and the solutions of xA=γ in (C∗)n are the same as the roots of F in (C∗)n.
Clearly then, γi≤0 for any i implies that F has no positive roots. In which case, we simply report that F has 0 positive roots and conclude, having taken time n3.373log1+o(1)(nH).
Otherwise, γ∈R+n implies that F has exactly 1 positive root by Proposition 2.7, and we are done. □
A simple consequence of our development so far is a method to efficiently count roots in (R∗)n for generic (n+1)-nomial systems.
Corollary 2.11
Following the notation and assumptions of Lemma 2.9 and its proof, the number of roots of F in (R∗)n is either 0 or 2n-r, where r is the rank, over the field F2, of the mod 2 reduction of A. In particular, we can determine the number of roots of F in (R∗)n in time n3.373log1+o(1)(ndH), where d is the maximum absolute value of any entry of A.
Proof
Continuing from the proof of Lemma 2.9 (and having already reduced our input (n+1)-nomial n×n system to a binomial system), it is clear that Proposition 2.8 tells us that we can easily count the roots of F in (R∗)n: We merely need to check the signs of γr+1′,…,γn′ where γ′:=γV and UAV=S is a Smith factorization of A. More precisely, instead of computing γV, we compute sign(γ)(Vmod2). Computing the mod 2 reduction of V takes time O(n2) and then computing the resulting vector of signs clearly takes time just O(n2). So the only remaining work (after applying Lemma 2.1 to the coefficient matrix of F) is extracting the Smith factorization of A via, say, Theorem 2.4. So our final complexity bound is n3.373log1+o(1)(nH)+n3.373log1+o(1)(nd), which is no greater than our stated bound. □
Circuits, (n+2)-Nomial Systems, and Gale Dual Form with Heights
We now show how to reduce root counting in (R∗)n for F to root counting in certain sub-intervals of R for a linear combination of logarithms in one variable. This reduction dates back to [23], if not earlier, but our statement here includes height and computational complexity bounds that appear to be new. Before proving our reduction, however, let us recall the combinatorial/geometric notion of a circuit8:
Definition 2.12
Given any subset A={a1,…,am+2}⊂Qn with #A=m+2, we define A^∈Z(n+1)×(m+2) to be the unique matrix with jth column 1aj for all j. We then call A a circuit if and only if A^ has right nullspace of dimension one. In which case, we call any generator b∈Z(m+2)×1\{O} for the right nullspace of A^, with 1 for its gcd of coordinates, a (minimal) circuit relation of A. We also call A a degenerate circuit if and only if b has at least one zero coordinate. ⋄
Note that m≤n if A as above is a circuit, since m≥n+1 would imply A^ has a right nullspace of dimension at least 2. Note also that all circuit relations for a fixed circuit (other than the trivial relation O) have zero entries occuring at the same set of coordinates. More precisely, the following proposition is elementary.
Proposition 2.13
Any circuit A={a1,…,am+2}⊂Zn has a unique subset Σ={ai1,…,aiℓ+2} with Σ a non-degenerate circuit of cardinality ℓ+2. In particular, {i1,…,iℓ+2} is exactly the set of indices of the nonzero coordinates of any (non-trivial) circuit relation for A. Furthermore, if J⊆{i1,…,iℓ+2} and ∑j∈Jaj=O, then J={i1,…,iℓ+2}. □
We call Σ the unique non-degenerate sub-circuit of A. Note that any A={a1,…,an+2}⊂Zn with cardinality n+2, and A not lying in any affine hyperplane, is a circuit.
Example 2.14
It is easily checked that A= 0000,1000,2000,0100,0010,0001⊂R4 is a degenerate circuit, and that letting Σ consist of the first 3 points of A yields the unique non-degenerate sub-circuit of A. In particular, Σ has the same minimal circuit relation (up to sign) as the non-degenerate circuit {0,1,2} in R1. ⋄
Lemma 2.15
Suppose F=(f1,…,fn)∈Zx1±1,…,xn±1n is an (n+2)-nomial n×n system supported on a circuit A={a1,…,an+2}⊂{-d,…,d}n for some d∈N. Suppose also that F has coefficient matrix [ci,j]∈{-H,…,H}n×(n+2) with all of its n×n sub-matrices non-singular. Then in time n3.373log1+o(1)(ndH), we can give either a true declaration that F has no positive roots, or find γ1,0,γ1,1,…,γn+1,0,γn+1,1∈Q and b1,…,bn+1∈Z such that: The number of roots of the function L(u):=∑i=1n+1bilog|γi,1u+γi,0| in the open interval I:=u∈R|γi,1u+γi,0>0for alli∈{1,…,n+1} is finite and exactly the number of positive roots of F.
I is non-empty and, for each i∈{1,…,n+1}, we have max{γi,1,γi,0}>0.
L is a non-constant real analytic function on I.
We have height bounds h(bi)=O(nlog(nd)) and h(γi,j)=O(nlog(nH)) for all i and j.
Example 2.16
Returning to Example 1.3, one can easily apply Gauss-Jordan elimination to the underlying linear combinations of monomials, and then divide every equation by the last monomial x3166x468x5343, to reduce F=O to the following system having the same roots in (R∗)5:x136x2194x3-116x414x5-283=16384cx158x2194x3-142x4-32x5-318+14x176x2240x3-166x4-27x5-342=4096cx158x2194x3-142x4-32x5-318+1x174x2179x3-141x4-68x5-286=256cx158x2194x3-142x4-32x5-318+1x125x2203x3-122x4-67x5-343=16cx158x2194x3-142x4-32x5-318+1x120x2167x3-102x4-56x5-275=cx158x2194x3-142x4-32x5-318+1
Note that this new system reveals why all the roots of F in (R∗)5 (for our earlier chosen values of c) must in fact lie in R+5: The right-hand sides are all positive on (R∗)5. The underlying circuit relation for the exponent vectors above is the same as the circuit relation for the exponent vectors of F: b=(-2,2,-2,2,-2,1,1)⊤. Part of the proof of Lemma 2.15, applied to our example here, will imply that the resulting linear combination of logarithms L(u) can be easily read from b and the right-hand sides of our reduced system:-2log16384cu+14+2log4096cu+1-2log256cu+1+2log16cu+1-2log|cu+1|+log|u|.
In particular, for any c>0, the number of roots of L in I=R+ is the same as the number of roots of F in R+5. Our family of examples here is in fact an obfuscated version of a family derived in [86], thus accounting for the nice coefficients and high number of positive roots (6) for c∈120730,114392. A more realistic example of coefficient growth can be found in Example 2.21 (see also Example 3.2 from Sect. 3). ⋄
Example 2.17
Even if Σ=A (so that m=n), we still need enough non-singular minors to guarantee that F has just finitely many roots in R+n. For instance, the 4-nomial 2×2 systemx1-1=0x1x2-x2=0,
has c1,1c1,2c2,1c2,2 and c1,3c1,4c2,3c2,4 non-singular (among other 2×2 sub-matrices), but has infinitely many positive solutions: (1, t) for all t>0. In particular, we are missing the non-singularity of c1,2c1,3c2,2c2,3. ⋄
Proof of Lemma 2.15:
First let b∈Z(n+2)×1 be the unique (up to sign) minimal circuit relation of A. By identifying the nonzero entries of b, we can then reorder the ai so that the unique non-degenerate circuit in A is Σ={a1,…,am,an+1,an+2}. (So Σ has cardinality m+2 and bm+1,…,bn are the only 0 entries of b.) Note that the coordinates of b are of logarithmic height O(nlog(nd)), and the computation of b takes time n3.373log1+o(1)(nd), thanks to Lemmata 2.1 and 2.2 (and the fact that ω<2.373 [2]).
We can then divide f1,…,fn by xan+2 without affecting the positive roots of F. So we may assume further that an+2=O and, since this at worst doubles our original d, our O-estimates will be unaffected. We can then apply Lemma 2.1, thanks to our assumption on the n×n sub-matrices of [ci,j], to reduce F=O to a system of equations of the form G=O, having the same solutions in Rn as F, where G:=(g1,…,gn),gi(x)=xai-γi,1xan+1-γi,0for alli,
and the γi,j are rational with logarithmic height O(nlog(nH)). This reduction takes time just n3.373log1+o(1)(nH), by Lemma 2.1 (and our earlier observations on ω). To complete our notation, let us also set γn+1,1:=1, γn+1,0:=0, γn+2,1:=0, and γn+2,0:=1.
Clearly, if there is an i such that both γi,0 and γi,1 are non-positive, then G (and thus F) has no positive roots, and we can simply stop, having spent time just n3.373log1+o(1)(nH). So we may assume the following:4 For eachi∈{1,…,n+1}we havemax{γi,1,γi,0}>0.
We can easily check whether I is non-empty after sorting the (possibly infinite) numbers -γi,0/γi,1, using just O(nlogn) comparisons of integers with O(nlog(nH)) bits (via, say, merge sort [39]). If I is empty, then we can conclude that F has no positive roots and stop (having spent time just n3.373(log1+o(1)(nd)+log1+o(1)(nH))). So we may also assume the following:5 Iis non-empty.
We now establish Assertions (1)–(4) via G and Σ: Observe that any root ζ∈(R∗)n of G must satisfy6 1=(ζa1)b1⋯(ζam)bm(ζan+1)bn+1=(γ1,1ζan+1+γ1,0)b1⋯(γm,1ζan+1+γm,0)bm(ζan+1)bn+11bn+2,
So let P(u):=(γ1,1u+γ1,0)b1⋯(γm,1u+γm,0)bmubn+1-1. Note that n=1⟹(γ1,1,γ1,0)=(-c2/c1,-c3/c1)∈(R∗)2 and thus P is a non-constant real rational function when n=1. So L(u) is a non-constant real analytic function on I when n=1. Let us then assume n≥2. By Cramer’s Rule, and our assumption on the n×n sub-matrices of [ci,j], we have γi,0≠0 and γi,1≠0 for all i. P is then a non-constant real rational function since bn+1≠0 (thanks to Σ being a non-degenerate circuit and Proposition 2.13), and there is thus no way to cancel the ubn+1 factor in the product term of P. So L(u) is a non-constant real analytic function on I.
Now observe that any root ζ∈R+n of F yields ζan+1∈I as a root of P by Equation (6). Moreover, by Proposition 2.5, any root ζ′∈R+n of F with (ζ′)an+1=ζan+1 must satisfy ζ′=ζ, since G reduces to a binomial system with a unique positive root once the value of xan+1 is fixed. (This is because the vectors a1,…,an are linearly independent, thanks to {an+1,O}⊂Σ and A being a circuit.) So P has at least as many roots in I as F has in R+n.
Conversely, Proposition 2.5 tells us that any root u∈I of P yields a unique ζ∈R+n satisfying ζa1,…,ζan=(γ1,1u+γ1,0,…,γn,1u+γn,0). Recall that bn+1≠0. So we also obtain ζan+1=(γ1,1u+γ1,0)-b1⋯(γm,1u+γm,0)-bm1/bn+1=ubn+1/bn+1=u by the definition of P. So ζ is in fact a root of G. Similarly, a root u′∈I of P with u′≠u would yield a positive root of (ζ′)a1,…,(ζ′)an=(γ1,1u′+γ1,0,…,γn,1u′+γn,0) with (ζ′)an+1≠ζan+1 and thus a root ζ′≠ζ of F. So F has at least as many roots in R+n as P has in I.
Observing that P(u)=0⟺L(u)=0 (for u∈I), and recalling Assumptions (4) and (5), we thus obtain Assertions (1)–(4). Noting that m≤n, we are done. □
Our sub-matrix condition from Lemma 2.15 in fact holds for a large fraction of integer coefficients:
Corollary 2.18
The fraction of matrices [ci,j]∈{-H,…,H}n×(n+2) with all n×n sub-matrices of [ci,j] non-singular is at least 1-n(n+2)(n+1)4H+2. Also, the fraction of matrices [ci,j]∈{-H,…,H}n×(n+1) with leftmost n×n sub-matrix of [ci,j] non-singular is at least 1-n2H+1.
Proof
The DeMillo–Lipton–Schwartz–Zippel (DLSZ) Lemma [73, 99, 104] is a classic result that tells us that if f∈C[z1,…,zn] has degree d and S⊂C is a set of finite cardinality N, then f vanishes at no more than dNn-1 points of Sn. The condition stated in our corollary is then equivalent to the non-vanishing of a product of n+22 many n×n sub-determinants of [ci,j]. The resulting polynomial clearly has degree n(n+2)(n-1)2. Taking S={-H,…,H} (which has cardinality 2H+1) and applying the DLSZ Lemma, we obtain our first bound. Our second bound follows almost identically, just considering one determinant instead. □
Recall that a critical point of a function L:R⟶R is a root of the derivative L′.
Proposition 2.19
Following the notation and assumptions of Lemma 2.15, let u0:=infI, uk:=supI, and suppose u1<⋯<uk-1 are the critical points of L in I (k=1 implying no critical points). Then, the number of positive roots of F is exactly the number of i∈{0,…,k-1} such that limu→ui+L(u)limu→ui+1-L(u)<0, plus the number of degenerate roots of L in I.
Proof
It is clear that L is strictly monotonic on any open sub-interval (ui,ui+1) of I. So the image of (ui,ui+1) under L isLi:=minlimu→ui+L(u),limu→ui+1-L(u),maxlimu→ui+L(u),limu→ui+1-L(u),
and we see by the Intermediate Value Theorem that Li does not contain 0⟺[limu→ui+L(u) and limu→ui+1-L(u) are both non-positive or both non-negative]. So by Lemma 2.15, we are done. □
We can now state analogues of Lemma 2.15 and Proposition 2.19 for roots in (R∗)n.
Lemma 2.20
Following the notation and assumptions of Lemma 2.15, assume further that bn+1 is odd, an+2=O, and let A:=[a1,…,an]. Let UAV=S be a Smith factorization of A, and r the rank, over the field F2, of the mod 2 reduction of A. Also, for any u∈R, let εi:=sign(γi,1u+γi,0), Λ(u):=∏i=1mεibimod2, and (Γ1′(u),…,Γn′(u)):=(ε1,…,εn)Vmod2. Then, the number of roots of F in (R∗)n is exactly 2n-r times the number of roots u∈R of L satisfying both Λ(u)=sign(u) and Γr+1′(u),…,Γn′(u)>0.
Example 2.21
Consider the 6-nomial 4×4 system F=(f1,…,f4) defined by-12x18x218x416-5x14x2x33x48+17x111x219x3x417-4x111x29x314+2x218x313x417+3x15x314x416,-9x18x218x416+14x14x2x33x48-8x111x219x3x417+3x111x29x314+12x218x313x417-x15x314x416,5x18x218x416+4x14x2x33x48+11x111x219x3x417-16x111x29x314+18x218x313x417-19x15x314x416,-x18x218x416+2x14x2x33x48+11x111x219x3x417-17x111x29x314-14x218x313x417-6x15x314x416.
Proceeding as in Lemmata 2.15 and 2.20, we see that Gauss–Jordan elimination on the coefficient matrix, and computing the circuit relation underlying the exponent vectors, yields the following linear combination of logarithms:L(u)=54667log3989827281-8455627281u-16978log4721027281-12568027281u-43727log4213927281-12675427281u+5123log2084527281-11429627281u-10129log|u|.
In particular, L is an analytic function onR\{0,0.182377...,0.332447...,0.375636...,0.471852...}
whose roots encode the roots of F in (R∗)4: Observing thatΛ(u)=sign39898-84556usign42139-126754usign20845-114296u,
we see that the only open intervals containing u satisfying Λ(u)=sign(u) are(0,0.182377...),(0.332447...,0.375636...),(0.375636...,0.471852...).
(The mod 2 reduction in our A here has full rank r=4 and thus the condition involving the Γi′(u) becomes vacuously true.) It is then easily checked that L is strictly decreasing, with range R, on the first and third intervals; and L is positive on the second interval. (See also Corollary 2.22.) So L has exactly 2 roots in R∗ satisfying the sign conditions9 of Lemmata 2.20, and thus F has exactly 2 roots in (R∗)4: The roots in (R∗)4 respectively lie in the ++++ and +-++ orthants. (It is also easily checked that F has infinitely many roots in R4, since F vanishes on the entire subspace defined by x1=x2=0.) PHCpack (a state-of-the-art polyhedral homotopy solver [102]) confirms our root count for this F in about 15 minutes, along with a count of 70834 for the total number of roots in (C∗)4, as well as approximations of all these roots to 14 decimal places. Our Maple code counts the roots of F in (R∗)4 and R+4 in under one second. ⋄
Proof of Lemma 2.20:
Continuing the notation of the proof of Lemma 2.15, we need to revisit the properties of the rational function P defined earlier. In particular, whereas before we had a natural bijection between the roots of F in R+n and the roots of P in a particular interval I, we now need to consider roots of F with negative coordinates and roots of P throughout R. In particular, a key difference from our earlier lemma is the following simple equivalence, valid for all u∈R: P(u)=0⟺[L(u)=0 and Λ(u)=sign(u)]. (Indeed, we could encounter u with P(u)=-2 without the condition involving Λ(u).) Note also that by construction, P(0) is either -1 or undefined.
So let ζ∈(R∗)n be a root of F. By Relation (6), ζan+1 must be a nonzero real root of P and, by the definition of G and the γi,j (and Proposition 2.8), we must have Γr+1′ζan+1,…, Γn′ζan+1>0. By Proposition 2.8, there must also be exactly 2n-r many roots ζ′∈(R∗)n of F with (ζ′)an+1=ζan+1, because G reduces to a binomial system once the value of ζan+1 is fixed. So F has no more than 2n-r times as many roots in (R∗)n as P has in R∗.
Conversely, if u∈R∗ is a root of P, then Proposition 2.8 tells us that Γr+1′(u),…,Γn′(u)>0 implies that there are exactly 2n-r many ζ∈(R∗)n satisfying(ζa1,…,ζan)=(γ1,1u+γ1,0,…,γn,1u+γn,0).
(Note also that γi,1u+γi,0≠0 for all i since P(u)≠0 when γi,1u+γi,0=0.) We then also obtain ζbn+1an+1=(γ1,1u+γ1,0)-b1⋯(γm,1u+γm,0)-bm=ubn+1 by Relation (6). Since bn+1 is odd, all our resulting ζ must satisfy ζan+1=u and therefore be roots of G (and thus of F). Similarly, a real root u′ of P with u′≠u would yield a collection of 2n-r many ζ′∈(R∗)n that are roots of F but with (ζ′)an+1≠ζan+1, since bn+1 is odd and u∈R∗. So the number of roots of F in (R∗)n is at least 2n-r times the number of roots of P in R∗.
Our stated root count for F in (R∗)n is thus correct. □
The following variant of Proposition 2.19 can be proved almost the same as Proposition 2.19, simply using Lemma 2.20 instead of Lemma 2.15:
Corollary 2.22
Following the notation and assumptions of Lemma 2.20, let w1<⋯<wℓ-1 be the critical points and poles of L in R, and set w0:=-∞ and wℓ:=+∞. Let N be the number of i∈{0,…,ℓ-1} such that Λ(u)=sign(u) and Γr+1′(u),…,Γn′(u)>0 for all u∈(wi,wi+1), and limu→wi+L(u)limu→wi+1-L(u)<0. Then, the number of roots of F in (R∗)n is exactly N plus the number of degenerate roots of L in R. □
Heights of Algebraic Numbers and Linear Forms in Logarithms
Recall that if β is in the algebraic closure Q¯ of Q, with minimal polynomial m(x1):=c0+⋯+cdx1d∈Z[x1] satisfying gcd(c0,…,cd)=1, then we may define the (absolute) logarithmic height of β to beh(β):=1dlog|cd|+∑i=1dlogmax{|βi|,1},
where β1,…,βd (among them, β) are all the roots of m. This definition in fact agrees with our earlier definition for rational numbers. Since m must be irreducible we have #{β1,…,βd}=d.
Proposition 2.23
(See, e.g., [25, Prop. 1.5.15, pg. 18].) If α1,…,αk∈Q¯, then h∑i=1kαi is no greater than log(k)+∑i=1kh(αi). Also, h∏i=1kαi≤∑i=1kh(αi). □
Letting c0+c1x1+⋯+cdx1d2:=∑i=0d|ci|2, we recall the following classical inequality:
Landau’s Inequality
[78] If β∈Q¯ has minimal polynomial g∈Z[x1] with relatively prime coefficients then h(β)≤log|g|2degg. □
It will also be useful to have a mildly refined version of Liouville’s classic bound [72] on the separation between rational numbers and irrational algebraic numbers.
Theorem 2.24
Suppose β∈Q¯, with minimal polynomial m∈Z[x1] of degree d≥2. Then,β-pq<1⟹β-pq≥|m′(β)|+m′′(β)2!+⋯+m(d)(β)d!-1qd
for all p,q∈Z with q>0.
Proof
First note that the parenthesized expression in the numerator of the large fraction above is positive since m is the minimal polynomial of β and thus m′(β)≠0.
Via Taylor expansion we then obtain the following:|m(p/q)|=m(β)+m′(β)pq-β+m′′(β)2!pq-β2+⋯+m(d)(β)d!pq-βd=β-pq0+m′(β)+m′′(β)2!pq-β+⋯+m(d)(β)d!pq-βd-1≤β-pq|m′(β)|+m′′(β)2!pq-β+⋯+m(d)(β)d!pq-βd-1<β-pq|m′(β)|+m′′(β)2!+⋯+m(d)(β)d!.
Since m is irreducible and of degree ≥2, m has no rational roots, and thus qdm(p/q) must be a nonzero integer. So we obtain qd|m(p/q)|≥1 and thus |m(p/q)|≥1/qd. Combined with our last Taylor series inequalities, we are done. □
Finally, we recall the following paraphrase of a bound of Matveev [77, Cor. 2.3], considerably strengthening earlier bounds of Baker and Wustholtz [6]. (See also [30, Thm. 9.4].)
Theorem 2.25
Suppose K is a degree d real algebraic extension of Q, α1,…,αm∈K\{0}, and b1,…,bm∈Z\{0}. Let B:=max{|b1|,…,|bm|} and logAi:=max{dh(αi),|logαi|,0.16} for all i. Then, ∑i=1mbilogαi≠0 implies thatlog∑i=1mbilogαi>-1.4·m4.530m+3d2(1+logd)(1+logB)∏i=1mlogAi.
□
Bounds on Coefficients, Roots, and Derivatives of Univariate Polynomials
Letting c0+c1x1+⋯+cdx1d∞:=maxi|ci|, recall the following classic bounds on the size and minimal spacing of roots of polynomials:
Proposition 2.26
(See, e.g., [88, Thm. 8.1.4 & Thm. 8.1.7, (i), (8.1.3)].) If f∈Z[x1] satisfies |f|∞≤H and ζ∈C is a nonzero root of f, then 11+H<|ζ|<1+H. □
Mahler’s Theorem
[74] Suppose f∈Z[x1] is square-free, has degree d, and |f|∞≤H. Then, any two distinct complex roots ζ1,ζ2 of f satisfy|ζ1-ζ2|>3(d+1)-(2d+1)/2H-(d-1).
In particular, |log|ζ1-ζ2||=O(dlog(dH)). □
Letting c0+c1x1+⋯+cdx1d1:=∑i=0d|ci|, recall also the following nice bound on the coefficients of divisors of polynomials:
Lemma 2.27
[78, Thm. 4] Suppose f,g∈C[x1] have respective leading coefficients c and γ, and g|f. Then, |g|1≤2deggγc|f|2. □
Recall that the content of a polynomial r(x):=c0+⋯+cdxd∈Z[x1] is con(r):=gcd(c0,…,cd), and that we call such an r primitive if and only if con(r)=1. We will need the following consequence of the classical Gauss’ Lemma for polynomials (see, e.g., [46, Ex. 3.4, pg. 109]):
Lemma 2.28
If f,h∈Z[x1] with h primitive, g∈Q[x1], and f=gh identically, then g∈Z[x1]. □
We can now prove the following extension of Mahler’s bound to the case of polynomials with degenerate roots.
Corollary 2.29
Suppose f∈Z[x1] has degree d and |f|∞≤H. Then, any two distinct complex roots ζ1,ζ2 of f satisfy |log|ζ1-ζ2||=O(d2+dlogH).
Proof
Let g:=gcd(f,f′), where we compute the gcd of two polynomials in Q[x1] via the Euclidean Algorithm. (So g∈Q[x1] and g is monic.) In particular, since γg is primitive for some minimal γ∈N, Lemma 2.28 tells us that f/(γg)∈Z[x1]. Moreover, since g is monic, γ must divide the leading coefficient of f and thus γ≤H.
Recall that the square-free part of f is p:=f/gcd(f,f′). It is then elementary that p has the same roots in C as f, p is square-free, and p∈Q[x1]. In particular, from the last paragraph, we see that p∈Z[x1], with the same leading coefficient as that of f since g is monic. Lemma 2.27 then tells us that |p|1≤2d|f|2. So then, |p|∞≤2dd|f|∞≤dH2d. Applying Mahler’s Theorem, we see that log|ζ1-ζ2|=O(dlog(ddH2d))=O(d[log(d)+d+logH])=O(d2+dlogH). □
We will also need the following bound on the coefficients of products of polynomials:
Lemma 2.30
If f1,…,fk∈Z[x1] then ∏i=1kfi∞≤∏i=1k(1+degfi)|fi|∞.
Proof
Via direct expansion (and the Triangle Inequality) it is clear that the 1-norm for polynomials is sub-multiplicative, i.e., ∏i=1kfi1≤∏i=1k|fi|1. It is also immediate that |f|∞≤|f|1 and |f|1≤(1+degf)|f|∞ for any polynomial. So we obtain∏i=1kfi∞≤∏i=1kfi1≤∏i=1k|fi|1≤∏i=1k(1+degfi)|fi|∞.
□
Finally, we will need the following bound on higher derivatives of polynomials, dating back to work of Duffin and Schaeffer [45], based on a classic bound of A. A. Markov [75]:
Corollary 2.31
Suppose f∈C[x1] has degree d and t>0. Then,max-t≤x1≤tf(j)(x1)≤d2(d2-12)⋯(d2-(j-1)2)1·3⋯(2j-1)·max-t≤x1≤t|f(x1)|tj.
□
After rescaling the variable so it ranges over [-1,1], the statement above follows immediately from [88, Thm. 15.2.6 & Cor. 15.2.7, Sec. 15.2]. The latter results in fact include conditions for equality in the bound above.
Critical Values of Linear Forms in Logarithms and Their Signs
We are now ready to prove two key lemmata 3.1 and 3.4 that enable our new complexity bounds.
Lemma 3.1
Suppose m≥2, bi∈Z\{0} and γi,1,γi,0∈Q with h(γi,j)≤logH (for some integer H≥3) for all i∈{1,…,m}, B:=maxi|bi|, and L(u):=∑i=1mbilog|γi,1u+γi,0| is non-constant and differentiable on some non-empty open interval. Then, the critical points of L in R are exactly the real roots of a polynomial g∈Z[u] of degree at most m-1 with |g|∞≤m2m-1BH2m. In particular, log|g|∞=O(log(B)+mlogH), L has at most m roots in any open interval I not containing a pole of L, and L has at most 2m real roots.
Example 3.2
Example 2.21 is more representative (than Example 2.16) of the coefficient growth one encounters when converting F into a univariate linear combination of logarithms L: There we saw an input 6-nomial 4×4 system F with coefficients and exponents having at most 2 digits, resulting in an L with coefficients having 6 or fewer digits. In particular, the polynomial encoding the critical points of L isg(u):=-85015812446550320118784u4+160578806134338659719072u3-78932164016242868100268u2+13833463598904597755876u-837930167824219163155,
which has coefficients with at most 24 digits, and 2 real roots, neither of which lies in the sub-intervals of R contributing to the root count of F in (R∗)4. So in Example 2.21, it is the signs of the poles of L, instead of the signs of the critical values that determine the number of roots of F in (R∗)4. ⋄
Example 3.3
Returning to Example 2.16, which had L(u) being-2log16384cu+14+2log4096cu+1-2log256cu+1+2log16cu+1-2log|cu+1|+log|u|,
it is easily checked via Maple that this L has exactly 5 critical values, alternating in sign, and the underlying critical points interlace the 6 positive roots of L. ⋄
Proof of Lemma 3.1:
First observe that L′(u)=∑i=1mbiγi,1γi,1u+γi,0. Thanks to our non-constancy assumption, L′ has at most m distinct poles. Letting νi be the least common multiple of the denominators of γi,1 and γi,0, and settinggi(u):=biγi,1νi∏j=1m(γj,1u+γj,0)νj/((γi,1u+γi,0)νi),
let us define g(u):=∑i=1mgi(u). Clearly, gi∈Z[u] for all i, g∈Z[u], and g(u) is nothing more than L′(u)∏j=1m(γj,1u+γj,0)νj. So we clearly obtain the statement on the real critical points of L being the real roots of g, and it is clear that degg≤m-1. Lemma 2.30 implies that |gi|∞≤BH22m-1H2(m-1). Clearly then, |g|∞≤m2m-1BH2m. That L has at most m roots in I is immediate from Rolle’s Theorem, since degg≤m-1. We similarly obtain at most 2m roots in R since L has no more than m poles (as well as no more than m-1 critical points). □
Recall that a critical value of a function L:R⟶R is the value of L at a critical point of L.
Lemma 3.4
Following the notation and assumptions of Lemma 3.1, let I be any open interval defined by consecutive real poles of L, let ε denote any nonzero critical value of L, and let δ (resp. η) be the minimum of |ζ1-ζ2| over all distinct roots ζ1,ζ2∈I of L (resp. the derivative L′). Finally, let Δ denote the minimum of |ζ-μ| as ζ (resp. μ) ranges over the critical points (resp. poles) of L. Then: logη>-O(mlog(B)+m2logH).
log|ε|>-O61mlogm+1BH2m-1.
logΔ>-Omlog(B)+m2logH.
logδ>-O61mlogm+1BH2m-1.
Proof
If L has no critical points then, by Rolle’s Theorem, L has at most 1 root in I and Assertions (1)–(3) are vacuously true. So let us assume L has exactly k-1 critical points (with k≥2) in the open interval I, u0:=infI, uk:=supI, and suppose u1<⋯<uk-1 are the critical points of L in I. Also let g denote the polynomial from Lemma 3.1. Below, we illustrate a coarse approximation of what the graph of L can look like, along with some of our notation:
Assertion (1) then follows immediately by applying Corollary 2.29 to g, thanks to Lemma 3.1. In particular, we get |logη|=O((m-1)2+(m-1)log(m2m-1BH2m))=O(m2+m[log(m)+m+log(B)+mlogH]),
which clearly reduces to the stated bound.
Assertion (2) then follows routinely from Theorem 2.25 upon observing that |ε| is nothing more than the absolute value of a linear combination of logarithms of real algebraic numbers. In particular, the arguments of the logarithms constituting L(uj) at a critical point uj∈I (for some j∈{1,…,k-1}) all lie in the same real algebraic extension: Q(uj). Noting that the minimal polynomial, p, of uj has degree ≤m-1, Lemmata 2.27 and 3.1 then tell us that |p|∞≤2m-1γpγg|g|2 (since p|g), where γp and γg are, respectively, the leading coefficients of p and g. Moreover, since p,g∈Z[u], we have that γp|γg and thus7 |p|∞≤2m-1·1·(m(mB2m-1H2m))=m3/24m-1BH2m.
So log|p|2≤logm·m3/24m-1BH2m, and thus Landau’s Inequality tells us that8 deg(p)h(uj)≤logm24m-1BH2m.
Proposition 2.26 and Lemma 3.1 then tell us that |uj|<1+|g|∞≤1+m2m-1BH2m. Also, h(γi,1uj+γi,0)≤log(2)+(h(γi,1)+h(uj))+h(γi,0), thanks to Proposition 2.23. So then, by Inequality (8),deg(p)h(γi,1uj+γi,0)≤(m-1)log(2)+(m-1)log(H)+logm24m-1BH2m+(m-1)logH≤logm28m-1BH4m-2.
Theorem 2.25 then tells us that9 log|ε|>-1.4·30m+3m4.5(m-1)2(1+log(m-1))(1+logB)logm28m-1BH4m-2m>-1.4·303·30mm6.5(1+logm)(1+logB)logmm28m-1BH4m-2=-O30mm6.5log(m)log(B)logmm28m-1BH4m-2.
Now observe that m6.5logm=O(1+θ1)m for any θ1>0, and m28m-1≤(BH4m-2)θ2 when H≥8.0814θ2, m≥1467, and θ2>0. So then, if we pick θ1=θ2=30.5/30-1=0.00829... we obtain: log|ε|>-O(30m(1+θ1)mlog(B)logmBH4m-21+θ2)=-O(30(1+θ1)(1+θ2))mlogm+1BH4m-2=-O30.5m2m+1logm+1BH2m-1=-O61mlogm+1BH2m-1,
thus proving Assertion (2).
To prove Assertion (3), observe that Δ=min{u1-u0,uk-uk-1}, i.e., Δ=|uj′-uℓ| for some j′∈{1,k-1} and ℓ∈{0,k}, by our earlier definitions. If Δ=∞, then there is nothing to prove, so let us assume Δ<∞. If uj′∈Q, then Assertion (3) follows easily from Proposition 2.23, since uℓ has logarithmic height no greater than 2logH and uj′ must have logarithmic height no greater than O(log(B)+mlogH). So we may assume that uj′ is algebraic of degree at least 2 over Q.
We can then apply Theorem 2.24 and Lemma 3.1 to obtain that Δ must be bounded from below by10 min|p′(uj′)|+p′′(uj′)2!+⋯+p(m-1)(uj′)(m-1)!-1H2(m-1),1.
We know that |uj′|<1+m2m-1BH2m by Proposition 2.26, so it is enough to minimize the preceding sum of derivative norms over the interval J:=-1-m2m-1BH2m,1+m2m-1BH2m.
Noting the easy inequality max-t≤x1≤tf(x1)≤|f|1max1,|t|d for any f∈R[x1] of degree d and t∈R, we then have:11 maxx1∈J|g(x1)|≤|g|1max1,(1+m2m-1BH2m)m-1≤m|g|∞1+m2m-1BH2mm-1,≤m·m2m-1BH2m1+m2m-1BH2mm-1,
and Corollary 2.31 then implies|p(r)(uj′)|≤maxx1∈Jp(r)(x1)≤(m-1)2((m-1)2-12)⋯((m-1)2-(r-1)2)1·3⋯(2r-1)·|p|1·1+m2m-1BH2mm-1,
since p has degree ≤m-1. Since |p|1≤m|p|∞, Inequality (7) tells us that p(r)(uj′) is bounded from above by(m-1)2((m-1)2-12)⋯((m-1)2-(r-1)2)1·3⋯(2r-1)·m·m3/24m-1BH2m1+m2m-1BH2mm-1≤(m-1)2rm·m3/24m-1BH2m1+m2m-1BH2mm-1≤(m-1)2errm5/24m-1BH2m1+m2m-1BH2mm-1,
where the last inequality follows easily from Stirling’s classical estimate for the factorial function. So then, |p′(uj′)|+p′′(uj′)2!+⋯+p(m-1)(uj′)(m-1)! is strictly less than(m-1)2e+((m-1)2e/2)22!+⋯+((m-1)2e/(m-1))m-1(m-1)!m5/24m-1BH2m1+m2m-1BH2mm-1.
Now, by the Maclaurin series for ex, the bracketed factor above is strictly less than e(m-1)2e. So then, |p′(uj′)|+p′′(uj′)2!+⋯+p(m-1)(uj′)(m-1)! is strictly less thane(m-1)2em5/24m-1BH2m1+m2m-1BH2mm-1≤em(m-1)em5/24m-1BH2m1+m2m-1BH2mm-1=Oee+θ3m(m-1)BH2m2(1+θ4)(m-1)BH2mm=O21+θ4ee+θ3m(m-1)Bm+1H2m2+2m,
for any θ3,θ4>0. Observing that 2ee<30.31, we can then clearly pick θ3 and θ4 to obtain|p′(uj′)|+p′′(uj′)2!+⋯+p(m-1)(uj′)(m-1)!=O31m(m-1)Bm+1H2m2+2m,
and thus, combining with Inequality (10), 1Δ=O31m(m-1)Bm+1H2m2+4m-2, and we obtain Assertion (3) by taking logarithms.
To prove Assertion (4), we merely use the mean value theorem. First, let δ′ be the minimum distance between any critical point uj of L (with nonzero critical value) and ζ′, where ζ′ is any root of L. Clearly, δ≥2δ′>0 (thanks to Rolle’s Theorem), so it is enough to prove a sufficiently good lower bound on δ′. Note in particular that if δ′>Δ/2, then we are done, thanks to Assertion (3). So let us assume δ′≤Δ/2.
Recall that from the proof of Lemma 3.1, we have L′(u)=g(u)/∏i=1m(γi,1u+γi,0)νi where νi is the least common multiple of the denominators of γi,1 and γi,0.
Clearly, if γi,1≠0, then |(γi,1u+γi,0)νi|=u--γi,0γi,1|γi,1νi|>|γi,1νi|Δ/2≥Δ/2 for all u∈uj-δ′,uj+δ′, since ζ′∈I, -γi,0γi,1 is a pole of L, and γi,1νi is a nonzero integer. On the other hand, if γi,1=0, then |(γi,1u+γi,0)νi|=|γi,0νi|≥1 since γi,0νi is a nonzero integer. So then,12 ∏i=1m(γi,1u+γi,0)νi>Δm/2mfor allu∈uj-δ′,uj+δ′.
By the mean value theorem, we must have |L′(ξ)|=εδ′ for some ξ∈uj-δ′,uj+δ′. So then, thanks to Inequalities (11) and (12), we obtain13 |L′(ξ)|=g(ξ)/∏i=1m(γi,1ξ+γi,0)νi<m22m-1BH2m1+m2m-1BH2mm-12mΔm≤m222m-1BH2m1+m2m-1BH2mm-1O31m(m-1)Bm+1H2m2+4m-2m
Since δ′=|ε/L′(ξ)| we thus obtain that logδ′=log|ε|-log|L′(ξ)|, which is then bounded frombelow by-O61mlogm+1BH2m-1-Olog(m)+m+log(B)+mlog(H)+mlogm2m-1BH2m-mOm2log(31)+mlog(B)+m2logH,
which reduces to -O61mlogm+1BH2m-1. □
The Complexity of Approximating Logarithms and Real Roots of Polynomials
Any real number can be expressed in binary. Since 2log2x≤x≤21+log2x for any x∈R+, it is easy to check that 1+log2x is the number of bits for the integer part of x. It then makes sense to call the binary expansion of 2ℓ-1-log2xx the ℓ most significant bits of an x∈R+. Clearly, knowing the ℓ most significant bits of x means that one knows x within a multiple of (1+2-ℓ)±1.
Let us recall the following classical fact on approximating logarithms via arithmetic–geometric Iteration:
Theorem 3.5
[13, Sec. 5] Given any positive x∈Q of logarithmic height h, and ℓ∈N with ℓ≥h, we can compute log2max{1,log|x|} and the ℓ most significant bits of logx in time O(ℓlog2ℓ). □
The underlying technique dates back to Gauss and was refined for computer use in the 1970s by many researchers (see, e.g., [28, 29, 97]). We note that in the complexity bound above, we are applying the recent O(nlogn) algorithm of Harvey and van der Hoeven for multiplying two n-bit integers [56]. Should we use a more practical (but asymptotically slower) integer multiplication algorithm, then the time can still be kept at Oℓ1.585 or lower.
Recall that bisection is the ancient technique of approximating a root of a continuous function f:[r1,r2]⟶R by the following trick: If sign(f(r1)f(r2))<0, then f must have a root in the open interval (r1,r2), and this root lies in the left half-interval r1,r1+r22 if and only if signf(r1)fr1+r22<0. Bisection thus allows one to extract an extra bit of precision for a root of f at the cost of one more evaluation of f. Put another way, bisection allows one to halve the size of an isolating interval at the cost of one more evaluation of f.
We will also need the following result on the bit complexity of approximating the real roots of a polynomial in Z[x1] by rational numbers.
Lemma 3.6
Suppose f∈Z[x1] has degree d, |f|∞≤H, and ℓ∈N with ℓ≥2. Let δ(f) denote the minimum of |ζ1-ζ2| over all distinct real roots ζ1,ζ2 of f. Then, in timeOd4[log2(H)+ℓ(ℓ+d2+dlogH)log(dℓlogH)],
we can find a collection of disjoint non-empty open intervals {Ji}i=1k with the following properties: k is the number of real roots of f.
Each Ji contains exactly one root of f.
The endpoints of all the Ji are rational numbers with logarithmic height O(ℓ+d2+dlogH).
All the Ji have width no greater than 2-ℓδ(f).
Proof
The case ℓ=0 is well-known in the computational algebra community and is elegantly described in [96]. (In fact, [96] even allows polynomials with real coefficients known only up to a given tolerance.) In particular, we merely apply the real root isolation algorithm from Theorem 24 of [96] to the square-free part, p:=f/gcd(f,f′), of f: From the proof of Corollary 2.29, we know that p∈Z[x1]. Also, by [51, Cor. 11.20] and fast integer multiplication [56] (and the development of [51, Ch. 6]), gcd(f,f′) can be computed within time O(d2log2(d)log(dH)log2(dlog(dH))), and this dominates the complexity of the division needed to compute f/gcd(f,f′). Moreover, the coefficients of p have logarithmic height logH′=O(d+logH), thanks to Lemma 2.27. So an overall complexity bound of O(d4(d+logH)2) holds, via the O(d4log2H′) bit-complexity bound (in our notation) from [96, Thm. 24].
The case of arbitrary ℓ≥2 can be derived simply by applying bisection after using the ℓ=0 case to start with isolating intervals that are likely larger than desired, but correct in number, for the real roots of f: One first observes that if α∈Q has logarithmic height L, then Proposition 2.23 implies that f(α) has height O(dlog(H)+d2L). So we can correctly find the sign of f(α) by, say, Horner’s Rule [51], using O(dlog(H)+d2L) bits of accuracy in all intermediate calculations. Since there are at most d roots, and each application of Horner’s Rule takes O(d) multiplications and additions, we see that the complexity of one step of bisection, applied to all of our initial isolating intervals (to halve the size of each interval), is dominated by O(d2) many multiplications of integers of height O(dlog(H)+d2L). Assuming we use the fast multiplication algorithm of [56], this will take time14 O(d2·(dlog(H)+d2L)·log(dlog(H)+d2L))=O(d3(log(H)+dL)·log(dlog(H)+d2L)).
Corollary 2.29 then tells us that |logδ(f)|=O(d2+dlogH). This means that our bisection must start with at least L=O(d2+dlogH) bits of accuracy, and this accuracy will successively increase to L+ℓ bits when we finish. So then, by Equality (14), getting ℓ additional bits of accuracy beyond the minimum root separation will require time15 ∑i=0ℓOd3(log(H)+d(L+i))log(dlogH+d2(L+i)).
Note that log(H)+d(L+i)=O(d3+d2log(H)+di)=O(d3(log(H)+i)) and thus log(d(log(H)+d(L+i)))=O(log(idlogH)). So each term of the sum (15) admits an upper bound ofO([d6+d5log(H)+d4ℓ]log(ℓdlogH)).
So our final bound is O(d4(d+logH)2)+ℓ·O(d4(d2+dlog(H)+ℓ)log(dℓlogH)). Cancelling dominant terms, we get our stated bound. □
Remark 3.7
We have opted for a streamlined proof at the expense of a larger complexity estimate. In particular, the exponent of d in our bound can likely be lowered slightly if one uses more sophisticated techniques, some of which are discussed further in [96] and the references therein. ⋄
Our Main Algorithms and Their Complexity
Our central algorithm for counting roots in (R∗)n is conceptually simple but ultimately somewhat laborious: Reduce to computing the signs of a linear combination of m logarithms, evaluated at its critical points and poles. To compute the signs at the critical points, we approximate the input to each logarithm, and each resulting summand, to extremely high accuracy. The devil is in the level of accuracy, but thanks to our earlier development, the required accuracy can be estimated explicitly, and the resulting complexity bound is quadratic in (n2log(dH))(1+o(1))n. We will see an even better conditional speed-up in Sect. 6, but let us first explore what is provable with current technology.
Algorithm 4.1
Input Integers b1,…,bm, rational numbers γ1,1,γ1,0,…,γm,1,γm,0,u0,u∞, with m≥2, γi,1u+γi,0>0 for all u∈(u0,u∞) and i∈{1,…,m}, and L(u)=∑i=1mbilogγi,1u+γi,0 non-constant and differentiable on some non-empty open interval.
Output The signs of L at all its critical points in (u0,u∞).
Description 0. Let B:=maxi|bi|, logH:=max1,maxi,jhγi,j, A:=m28m-1BH4m-2, E:=1.4·m6.530m+3(1+logm)(1+logB)logmA,D:=m2e+(m+2)log8+m2m+2BH2m,
and ρ:=1.443(D+log(12m)+E).
1. Compute the polynomial g(u):=∑i=1mbiγi,1νi∏j=1m(γj,1u+γj,0)νj/((γi,1u+γi,0)νi),where νi denotes the least common multiple of the denominators of γi,1 and γi,0.
2. Via Lemma 3.6, find respective isolating intervals J1,…,Jk-1 to the roots u1<⋯<uk-1 of g in (u0,u∞) such that each Ji has width no greater than 2-ρ.
3. For all i∈{1,…,k-1} do:
4. Let u¯i:=supJi+infJi2.
5. For all j∈{1,…,m} do:
6. Compute, via Theorem 3.5, a rational number Lj agreeing with log|γj,1ui¯+γj,0| in its first 1.443E+log2(6m) most significant bits.
7. End For
8. Let Li:=∑j=1mbjLj andθi:=sign(Li).
9. If |Li|>122-1.443E then
10. Output “The sign of L at ui is θi.”
11. Else
12. Output “L(ui)=0.”
13. End If
14. End For
Lemma 4.2
Algorithm 4.1 is correct and runs in timeO901m(log(B)+mlogH)2mlog2(B)log2(log(B)+mlogH).
Proof
The correctness of our algorithm follows directly from Theorem 2.25 and Lemmata 3.1 and 3.4. First note that the classical inequality 1-1x≤logx≤x-1 (for all x>0), yields sv+s≤log(v+s)-logv≤sv (for all v>0 and s>-v), upon setting x=v+sv. Setting v=γj,1u¯j+γj,0 and s=γj,1(uj-u¯j), and assuming γj,1uj+γj,0,γj,1u¯j+γj,0>0, we then obtain16 (uj-u¯j)γj,1γj,1uj+γj,0≤log(γj,1uj+γj,0)-log(γj,1u¯j+γj,0)≤(uj-u¯j)γj,1γj,1u¯j+γj,0.
The proof of Assertion (4) of Lemma 3.4 tells us that |γj,1uj+γj,0||γj,1|≥Δ/2. Since 1/log2<1.443, we have Δ>2-1.443D, thanks to the definition of D. So the definition of s tells us that |uj-u¯j|≤122-1.443D is sufficient to guarantee that |γj,1u¯j+γj,0||γj,1|≥Δ/2. So, by Inequality (16), we obtain that |uj-u¯j|≤2-ρ guarantees |log(γi,1ui+γi,0)-log(γi,1u¯i+γi,0)|≤16m2-1.443E. Should γi,1u+γi,0<0 we can repeat our preceding argument, with a sign flip, to obtain that |uj-u¯j|≤2-ρ guarantees |log|γi,1ui+γi,0|-log|γi,1u¯i+γi,0||≤16m2-1.443E. So then, thanks to Step 6 and the Triangle Inequality, we see that our algorithm computes, for each i∈{1,…,k-1}, a rational Li such that |L(ui)-Li|≤132-1.443E.
Theorem 2.25 then tells us that |L(ui)| is either 0 or strictly greater than 2-1.443E. So the threshold on |Li| from Step 9 indeed correctly distinguishes between L(ui) being nonzero or zero, and the signs of L(ui) and Li also match when L(ui)≠0 thanks to our chosen accuracy. In other words, our algorithm is correct.
We now analyze the complexity of our algorithm. First note that H, A, E, D, and ρ need not be computed exactly: it is sufficient to work with the ceilings of these quantities, or even the smallest powers of 2 respectively greater than these quantities. In particular, these parameters can easily be computed via standard efficient methods for computing the exponential function [1] (along with Theorem 3.5) and thus the complexity of Step 0 is negligible, and in fact asymptotically dominated by Steps 2 and beyond.
Likewise, Step 1 is easily seen to take time within O(m2(log2(B)+mlog2H)), by combining the fast polynomial multiplication method from, say, [51, Sec. 8.4] with the fast integer multiplication method of Harvey and van der Hoeven [56].
Lemma 3.1 tells us that the complexity of Step 2 can be estimated by replacing (d,H,ℓ) in the statement of Lemma 3.6 by (m-1,m2m-1BH2m,ρ). Noting that ρ=O(E) and mr,logrB,logrH=O(E) for any m≥r>0, Lemma 3.6 then tells us that Step 2 takes time17 O(m4[log2(m2m-1BH2m)+ρ(ρ+m2+mlog(m22m-1BH2m))log(ρmlog(m·m2m-1B2m))])=O(m4E2log2E).
Thanks to Theorem 3.5, a simple over-estimate for the complexity of Step 6 is O(Elog2E), so then the time spent in (each run of) Steps 4–7 in total is O(mElog2E). Since k-1≤m-1, Steps 3–14 then take time no greater than O(m2Elog2E).
We thus see, from comparison to Estimate (17), that Step 2 in fact dominates the asymptotic complexity of our entire algorithm. Since mr=O((1+ε)m) for any fixed r,ε>0, we have that E=O(log(B)(30+ε)m(log(B)+mlogH)m). Since logE=O(mlog(log(B)+mlogH)), we see similarly that the complexity of our algorithm isO(900+ε)m(log(B)+mlogH)2mlog2(B)log2(log(B)+mlogH)
which is dominated by our stated bound. □
We can now state our algorithm for counting the positive roots of circuit systems:
Algorithm 4.3
Input Polynomials f1,…,fn∈Zx1±1,…,xn±1 with A:=⋃iSupp(fi) a circuit and #A=n+2.
Output The number of roots of F=(f1,…,fn) in R+n.
Description 0. Find the unique (up to sign) minimal circuit relation b∈Z(n+2)×1 of A, and re-index the points of b and A so that bm+1=⋯=bn=0 and the unique non-degenerate sub-circuit Σ of A is Σ={a1,…,am,an+1,an+2}. Then, translate a1,…,an+1 by -an+2, and set an+2:=O.
1. Letting [ci,j] be the coefficient matrix of F, check whether all the n×n sub-matrices of [ci,j] are non-singular. If not, then output
“Your system might have infinitely many roots but I’m not sure: Please check if there are any updates to this algorithm, addressing the cases of vanishing minors for the coefficient matrix.”
and STOP. 2. Reduce F=O to a system of equations of the form G=O, where G:=(g1,…,gn) and gi(x):=xai-γi,1xan+1-γi,0 for all i.
3. Let L(u):=bn+1log|u|+∑i=1mbilog|γi,1u+γi,0| and I:=u∈R+|γi,1u+γi,0>0for alli∈{1,…,n}.
4. Via Algorithm 4.1 and Proposition 2.19, compute the number, N, of roots of L in I, and output N.
Example 4.4
Depending on the coefficient matrix of F, the number of poles of L can certainly be smaller than m+1: For instance, one can directly build a 4×4 circuit system F∈(Z[x1,x2,x3,x4])4 yielding, say,L(u)=2log|u|+log|u+1|+2log|2u+2|+3log|9u+9|-5log|7u+7|
and b=[1,2,3,-5,2,-3]⊤. This L clearly has just 2 poles: 0 and -1. ⋄
Lemma 4.5
Algorithm 4.3 is correct and runs in timeO(31n2log(ndH))2n+2(log(nd)log(nlog(dH)))2,
where d is the largest absolute value of an entry of A and H:=maxi|fi|∞.
Proof
First note that Step 1 tests a natural genericity condition mentioned earlier, and the algorithm proceeds to Step 2 if and only if the genericity condition holds. So let us assume we have proceeded to Step 2.
Observe then that the sum L from Step 3 is non-constant and differentiable on a non-empty open sub-interval J⊆R+: By the sub-matrix non-singularity assumption of Step 1, the γi,j must all be nonzero, and thus any cancellation between terms of L can not affect the term bn+1log|u|. So then J=R+ if γi,j>0 for all i and j, and J=0,minγi,1γi,0<0{-γi,0/γi,1} otherwise.
Letting u0:=infI, u∞:=uk:=supI, and letting u1<⋯<uk-1 be the critical points of L in I as before, note that the sign of limu→u0+L(u) is either -sign(bn+1) (if u0=0) or -sign∑u0=-γi,0/γi,1bi. Similarly, the sign of limu→uk-L(u) is simply -sign∑uk=-γi,0/γi,1bi (resp. -sign(bn+2)) if uk<∞ (resp. uk=+∞). So the use of Proposition 2.19 is clear.
The correctness of Algorithm 4.3 then follows directly from Lemmata 2.15, Proposition 2.19, and Lemma 4.2. So we now analyze the complexity of our algorithm.
Thanks to Lemmata 2.1 and 2.2, it is clear that Steps 0–3 are doable in time n3.373log1+o(1)(nd)+n4.373log1+o(1)(nH). This will not be the dominant part of the algorithm: Observing that h(γi,j)=O(nlog(nH)) and h(bi)=O(nlog(nd)) for all i, j (simply by Hadamard’s Inequality and Cramer’s Rule), the proof of Lemma 4.2 tells us that applying Algorithm 4.1 and Proposition 2.19 (with m≤n+1) takes time18
for any fixed ε>0. Note in particular that the underlined expressions are clearly bounded from above by:n2+n2log(d)+n2log(nH)≤n2(1+log(ndH)),
provided n,d,H≥1. So the O-estimate from (18) is bounded from above byO((30+ε′)(n2log(ndH))2n+2log2(nd)log2(n2log(ndH))
for any ε′>ε. Since n2log(ndH)=n2log(n)+n2log(dH)≤n3+n3log3(dH)=O((nlog(dH))3), we are done. □
We are now ready to state the analogue of Lemma 4.5 for counting roots in (R∗)n:
Lemma 4.6
Given any (n+2)-nomial n×n system F=(f1,…,fn)∈Zx1±1,…,xn±1n supported on a circuit A with cardinality n+2, we can count exactly the number of roots of F in (R∗)n in time O(31n2log(ndH))2n+2(log(nd)log(nlog(dH)))2, where d is the largest absolute value of an entry of A and H:=maxi|fi|∞.
Proof
The proof is almost identical to that of Lemma 4.5, save that we apply Corollary 2.22 instead of Proposition 2.19, Lemma 2.20 instead of Lemma 2.15, and that we use a modified version of Algorithm 4.3.
In particular, the modifications to Algorithm 4.3 are that (a) the output is now the number of roots in (R∗)n, (b) we re-index so that bn+1 is odd and (c) we replace Step 4 by Step 4’ stated below: □
Note in particular that b must have an odd coordinate since minimal circuit relations are assumed to have relatively prime coordinates. Also, the left or right-handed limits of L at a real (possibly infinite) pole are easy to compute via the sign of a suitable sum of bi (if the pole is finite) or the sign of -bn+2 (if the pole is ±∞), as in the proof of Lemma 4.5. The correctness of our modified algorithm is then immediate.
The complexity analysis for our modified algorithm is almost identical to that of Algorithm 4.3, save that there is extra work taking time O(n·n2) to compute the signs of Λ(ui) and the Γj′(ui). This is negligible compared to the other steps, so our final asymptotic complexity bound remains the same.
Remark 4.7
Reducing counting the roots of F in (R∗)n to counting the roots of L in ≤n+2 sub-intervals of R (as in our proof above) is much more efficient than naively applying Algorithm 4.3 to each of the 2n orthants of (R∗)n. In particular, our proof helps enable the conditional speed-up to time (nlog(dH))O(1) from Theorem 1.6. ⋄
Affine Roots and Proving Theorem 1.1
Before finally proving our main theorem, we will need to establish some simple facts on roots of over-determined systems on coordinate subspaces. Our first observation is immediate from basic convexity (see, e.g., [54, Ch. 3–4]).
Proposition 5.1
If A⊂Zn is a circuit (resp. the vertex set of a simplex), and X is any coordinate subspace of Rn, then A∩X is either empty, the vertex set of a simplex, or a circuit (resp. either empty or the vertex set of a simplex). Furthermore, for any given circuit A, there is at most one coordinate subspace X with A∩X a non-degenerate circuit. □
For any I⊆{1,…,n} let CI:={(x1,…,xn)∈Cn|i∈{1,…,n}\I⟹xi=0} and CI∗:={(x1,…,xn)∈Cn|i∈{1,…,n}\I⟺xi=0}. Note that C∅=C∅∗={O}, CI is a coordinate subspace of dimension #I, and CI∗ is a dense open subset of CI that we will call a sub-orbit. We also define RI:=CI∩Rn and RI∗:=CI∗∩Rn. Note that if I⊆J, then CI⊆CJ and CJ is the disjoint union of CI∗ over all I⊆J. The corresponding containments hold for RI, RJ, and RI∗ as well. In particular, R{i}∗ is the real xi-axis sans the origin.
Lemma 5.2
Suppose f∈Cx1±1,…,xn±1, I⊆{1,…,n}, and f is well-defined on CI∗. Then, f vanishes on all of CI∗⟺A∩RI=∅.
Proof
By symmetry, we can simply permute coordinates so that I={1,…,r}. By the Ideal-Variety Correspondence in the coordinate ring R:=C[x1±1,…,xr±1,xr+1,…,xn] (see, e.g., [60, Thm. 1.23]), f vanishing on all of CI∗ is equivalent to f lying in the ideal of R generated by xr+1,…,xn. Equivalently, for each monomial xs:=x1s1⋯xnsn of f there must be an i∈{r+1,…,n} with xi|xs. But then this is equivalent to each monomial xs of f having si>0 for some i∈{r+1,…,n}. In other words, (s1,…,sn) can not lie in RI. □
Lemma 5.3
[52, Ch. 8] Suppose A={a1,…,at}⊂Zn has cardinality t and does not lie in any affine hyperplane, and F=(f1,…,fn+1) with fi(x)=∑j=1tci,jxaj for all i and the ci,j indeterminates. Then, there is an irreducible polynomialRA∈Zci,j|(i,j)∈{1,…,n}×{1,…,t}\{0},
such that [ci,j]∈Cn×t and RA(…,ci,j,…)≠0⟹F has no roots in (C∗)n. In particular, we can pick RA so that degRA≤(n+1)·n!V (where V is the volume of the convex hull of A, normalized so that the unit n-cube has volume 1) and, if t=n+1, then we can use RA=det[ci,j]. □
Example 5.4
When t=n+2 the polynomial RA can already be far more unwieldy than an (n+1)×(n+1) determinant. For instance, with n=1, A={0,1,d}, and d≥2, the corresponding over-determined circuit system yields RA being a 2d×2d determinant (a special case of the Sylvester resultant) having degree 2d. One can also check via any reasonable computer algebra system that such an RA has exactly 3d+1 monomial terms, at least for d∈{2,…,100}. ⋄
The polynomial RA above is an example of a sparse resultant, and is one of many ways to formulate the fact that (n+1)-tuples of n-variate polynomials generically have no roots in (C∗)n. The same of course holds for (n+k)-tuples of n-variate polynomials for k≥2, but then the sufficient condition need not be determined by an irreducible polynomial that is unique up to sign.
Example 5.5
Consider F=(c1+c2x1,c3+c4x1,c5+c6x1). Then, the non-vanishing of either of detc1c2c3c4 or detc3c4c5c6 suffices to make F have no roots in C∗. Put another way, the non-vanishing of the resultant of some sub-pair of the original triple of polynomials suffices to obstruct roots in C for the triple. ⋄
From Lemma 5.3 (and using resultants of suitable sub-(d+1)-tuples of F restricted to d-dimensional coordinate subspaces), it is easy to see that if A∩X is non-empty for every coordinate subspace X⊂Rn, then a generic n×n system F supported on A can have no roots on the union of coordinate hyperplanes in Cn. We will need a more explicit refinement of this fact.
Lemma 5.6
Suppose A={a1,…,at}⊂Zn has cardinality t≤n+2 and does not lie in any affine hyperplane, I:={i∈{1,…,n}|min(a1,…,an)∈Aai<0}, and F=(f1,…,fn) with fi(x)=∑j=1tci,jxaj∈Rx1±1,…,xn±1 for all i∈{1,…,n}. Suppose further that every square sub-matrix of [ci,j] is non-singular and, if A∩Y contains a non-degenerate sub-circuit for some proper coordinate subspace Y⫋Rn, assume further that RA∩Y(G)≠0 for some sub-tuple G:=(fℓ|ℓ∈M) of F with #M=1+dimY. Then, for all J⊇I we have that F is well-defined on CJ∗. Furthermore, for I⊆J≠{1,…,n}, we have that F either vanishes on all of RJ∗ or has no roots in CJ∗, with the latter occuring if and only if A∩RJ≠∅.
Example 5.7
The additional condition involving A∩Y is necessary when t=n+2: For instance, taking n=2 and I={1}, the system F=(1+x2+2x1-3x14,1+2x2+3x1-4x14) has all square sub-matrices of its coefficient matrix non-singular, but F has a unique root in C1∗: (1, 0). The missing condition is in fact the non-vanishing of the resultant of c1,1+c1,3x1+c1,4x14 and c2,1+c2,3x1+c2,4x14. ⋄
Proof of Lemma 5.6:
The statement on F being well-defined on CJ∗ is immediate since the only coordinate hyperplanes possibly containing poles for the fℓ are {xi=0} with i∈I.
Let us now assume that there is no subspace Y as stated. Then, the intersection of A with each coordinate subspace of Rn is the vertex set of a simplex, and Lemma 5.3 combined with our sub-matrix assumption implies that F has no roots in CJ∗ if A∩RJ≠∅. If A∩RJ=∅ then Lemma 5.2 implies that F vanishes on all of CJ∗, and thus on all of RJ∗.
Should there instead be a subspace Y as stated, then Lemma 5.3 combined with our augmented genericity assumption implies that F has no roots in CJ∗ if A∩RJ≠∅. If A∩RJ=∅, then Lemma 5.2 implies that F vanishes on all of CJ∗, and thus on all of RJ∗. □
The Proof of Theorem 1.1
For convenience, let us first name the genericity assumptions we defined above:G+∗: Every n×n sub-matrix of [ci,j] is non-singular. ⋄
Gaff: Every square sub-matrix of [ci,j] is non-singular and, if A∩Y contains a non-degenerate sub-circuit for some proper coordinate subspace Y⫋Rn, assume further that RA∩Y(G)≠0 for some sub-tuple G:=(fℓ|ℓ∈M) of F with #M=1+dimY. ⋄
The cases of Theorem 1.1 for root counting in R+n and (R∗)n then follow respectively from Lemmata 4.5 and 4.6 when t=n+2, under the genericity assumption G+∗. For t=n+1, we simply use Lemma 2.9 and Corollary 2.11 instead. Note in particular that a consequence of Corollaries 2.11 and 2.22 is that F generically has only finitely many roots in (R∗)n.
Let Z∗ denote the zero set of F in (R∗)n. To count the roots of F in Rn, let us switch our genericity assumption to Gaff, and count the roots of F in (R∗)n as in the last paragraph, save for one small change: We assume our input is generic and skip Step 1 when applying Algorithm 4.3, i.e., we no longer check if our system is generic (since there are exponentially many determinants underlying Gaff). Observe now that Lemma 5.6 implies that the real zero set of F in Rn will either be (a) Z∗∪{O}, (b) the union of Z∗ with a real positive-dimensional sub-orbit, or (c) Z∗. So now we merely need to distinguish these possibilities efficiently.
This will reduce to indexing, with complexity negligible compared to Algorithm 4.3 and its variants. First, observe that μj:=min(a1,…,an)∈Aaj>0 for any j implies that xj|fi for all i, which in turn implies F has infinitely many roots in Rn. Computing all these minima takes time O(n). So we may assume μj≤0 for all j, and let I:={j|μj<0}.
If I=∅ then F is well-defined on all of Cn, and thus F vanishes on all of RJ—for some J⊆{1,…,n} with #J≥1—if only if F vanishes on all of R{j}∗ for some j∈{1,…,n}. Lemma 5.6 then tells us this happens if and only if A∩{j}=∅. The last condition is decided easily by checking if, for each ℓ∈{1,…,t}, aℓ has a positive ith_ coordinate for some i∈{1,…,n}\{j}. This can clearly be done in time O(n2), so let us now assume I≠∅.
Lemma 5.6 then implies that F has infinitely many roots in Rn if and only if there is a J⊇I with J≠{1,…,n} and A∩RJ=∅. The last condition holds if and only if A∩RI=∅ (since I⊆J⟹RI⊆RJ). Checking A∩RI=?∅ can be done in time O(n3) simply by checking if, for each ℓ∈{1,…,t}, aℓ has a positive ith_ coordinate for some i∈{1,…,n}\I. So we are done, and we see that the complexity of counting the roots of F on the union of real coordinate hyperplanes is well-dominated by the complexity of counting the roots of F in (R∗)n, thanks to our genericity assumptions. □
Remark 5.8
Note that condition Gaff involves the non-vanishing of each entry of [ci,j], as well as the determinants of each k×k sub-matrix of [ci,j] for k∈{2,…,n}. The product of all these determinants clearly has degree D:=n(n+2)+2n2n+22+⋯+nnnn+2n. Since ∑k=0nnk2=2nn, it is then easy to see that D<n2n+4n+2=2O(n). The resultant underlying condition Gaff has degree (nd)O(n), thanks to Lemma 5.3 and Hadamard’s Inequality. So then, the DLSZ Lemma implies that H≥2Ω(n)+(nd)Ω(n)ε is enough to guarantee that at worst an ε-fraction of the [ci,j]∈{-H,…,H}n×(n+2) fail condition Gaff. This implies our earlier statement on the number of bits needed for H to have Gaff hold with probability 1-ε. ⋄
Remark 5.9
Unlike our setting here—where we restrict the union of supports to be a circuit of cardinality n+2—we conjecture that counting real affine roots for arbitrary binomial systems is #P-hard, parallelling the complex case studied in [34]. Indeed, n×n binomial systems can have union of supports with cardinality up to 2n, and this complicates counting real roots on coordinate sub-spaces. ⋄
Conditionally Speeding Up to Time (nlog(dH))O(1): Proving Theorem 1.6
The key to proving our conditional speed-up will be to modify (a) two key bounds from Lemma 3.4 and (b) the initial step of a key algorithm (Algorithm 4.1), by incorporating the Diophantine improvements granted by the RALC should it be true.
Lemma 6.1
Following the notation of Lemma 3.4, if the RALC is true, then we can respectively replace the bounds from Assertions (2) and (4) by:2′.log|ε|>-O(m2(log(B)+mlogH))C.4′.logδ>-O(m2(log(B)+mlogH))C+m2(log(B)+mlogH).
Proof
The critical juncture is Bound (9), from our proof of Assertion (2) of Lemma 3.4: Replacing the use of the Baker–Wustholtz theorem there with the bound from the RALC immediately yields Bound (2’).
Bound (4’) then follows easily from Bound (13) (near the end of the proof of Lemma 3.4), and the line following that bound, where see that logδ≥log(2)+log|ε|-log|L′(ξ)| (in the notation of the proof of Lemma 3.4). So the last term (fully expanded in the proof of Lemma 3.4) accounts for the final term in Bound (4’). □
Lemma 6.2
Suppose the RALC is true, b1,…,bm∈Z, γ1,1,γ1,0,…,γm,1,γm,0,u0,u∞∈Q, with m≥2, γi,1u+γi,0>0 for all u∈(u0,u∞) and i∈{1,…,m}, and L(u)=∑i=1mbilogγi,1u+γi,0 is non-constant and differentiable on some non-empty open interval. Also let B:=maxi|bi| and logH:=max1,maxi,jhγi,j. Then, we can compute the signs of L at all its critical points in (u0,u∞) in time O(m4+2max{1,C}(log(B)+mlogH)2max{1,C}log(mlog(BH))).
Proof
We simply use Algorithm 4.1with one crucial change: We replace the definition of E in Step 0 by O(m2(log(B)+mlogH))C. We then proceed as in the proof of Lemma 4.2. In particular, Lemma 6.1 tells us that this modified version of Algorithm 4.1 is correct. As for complexity, instead of ρ=O(E) with the old value of E, the new value of E yieldsρ=Om[log(B)+mlogH]+(m2(log(B)+mlogH))C.
Bound (17) from the proof of Lemma 4.2, combined with our new values of E and ρ, then easily yield complexity O(m4ρ2(log(m)+loglog(B)+loglogH)). This reduces to our stated bound. □
We can now give our final remaining proof.
Proof of Theorem 1.6:
To speed up root counting in R+n, we use a modified version of Algorithm 4.3: We use Lemma 6.2 in place of Algorithm 4.1 in Step 4. We then proceed as in the proof of Lemma 4.5. In particular, correctness is immediate. As for complexity, Step 4 of our modified version of Algorithm 4.3 is the dominant part. Our final bound then amounts to substituting (n+1,nlog(nd),nlog(nH)) for (m,logB,logH) into the bound from Lemma 6.2, easily yielding a complexity bound ofOn4+4max{1,C}(log(nd)+nlog(nH))2max{1,C}log(nlog(ndH)),
which is clearly (nlog(dH))O(1).
To speed up root counting in (Rn)∗, we further modify Algorithm 4.3: Lemma 4.6 gives a modification to Step 4 (called Step 4’) that enables counting in (R∗)n instead of R+n. We make one more modification: We replace the use of Algorithm 4.1 in Step 4’ with an application of Lemma 6.2. Continuing as in the proof of Lemma 4.6 then gives us correctness. The complexity analysis is almost identical, save for an extra step involving the quantities Λ(ui) and Γj′(ui) (with complexity still negligible compared to the dominant steps). So our bound remains of the same asymptotic order.
To speed up root counting in Rn, the last three paragraphs of the proof of Theorem 1.1 from Sect. 5.1 tell us that we can count roots on the union of real coordinate hyperplanes in Rn in time O(n3) simply by checking the intersection of the support A against ≤n coordinate subspaces, thanks to Lemma 5.6 and our genericity condition Gaff. So counting roots in (R∗)n dominates our complexity, and we are done. □
Farewell to a Friend
Tien-Yien Li passed away a few months into the COVID-19 pandemic. TY (as he was known to his friends) was an immensely kind and generous man, and a dear friend, in addition to being a great mathematician. Through hours-long grilling sessions in October 1993, at the Centre de Recerca Matematica in Barcelona, he taught me lessons on perseverance, curiosity, scholarship, and generosity that I would always remember. It was there that I also got to know TY and his unique sense of humor. He always faced the greatest difficulties with a smile. I admired him both as a person and a mathematician. I truly miss him.
Acknowledgements
I thank Dan Bates and Jon Hauenstein for answering my questions on how Bertini handles polynomial systems of extremely high degree. I also thank Jan Verschelde for answering my questions on fine-tuning the options in PHCpack. Special thanks to Timo de Wolff for pointing out reference [35] and Alexander Barvinok for pointing out reference [85]. I also thank Weixun Deng, Alperen Ergür, and Grigoris Paouris for good company and inspirational conversations. I am also indebted to the referees for their detailed suggestions which greatly improved this paper. In particular, they encouraged me to include Theorem 1.6, and they pointed out Lemma 2.30 (which gives a simpler and sharper bound than I was using for the same purpose earlier), as well as a simplification of an earlier genericity condition for Theorem 1.1.
1 Roots yielding a Jacobian with less than full rank.
2 Using Maple 2019 on a Dell XPS 13 laptop with an Intel core i7-5500u microprocessor, 8 Gb RAM, and a 256Gb solid state hard-drive, running Ubuntu 19.10. Maple code available on request.
3 Via Kushnirenko’s Theorem [67, 91], Ioannis Emiris’ MixVol code [47], and a simple check that the underlying facial systems have no roots in (C∗)5.
4 Even allowing degenerate isolated roots.
5 [50] in fact proves a stronger theorem by using a weaker hypothesis that we will clarify below.
6 See Sect. 2.4 for the definition of heights for algebraic numbers.
7 Our stated bound assumes that we use an O(hlogh)-time algorithm for h-bit integer multiplication, e.g., [56].
8 ...not to be confused with the circuits from complexity theory (which are layered directed graphs with specially labeled nodes having additional structure).
9 L also happens to be increasing, with range R, on (-∞,0) and (0.182377..., 0.332447...), and thus L has 2 more roots in R∗ that do not satisfy the necessary sign conditions.
In memory of Tien-Yien Li, June 28, 1945–June 25, 2020.
Partially supported by NSF Grant CCF-1900881.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
1. Timm Ahrendt, “Fast computations of the exponential function,” in proceedings of STACS ’99 (16th annual conference on Theoretical aspects of computer science), pp. 302–312, Springer-Verlag Berlin, 1999.
2. Josh Alman and Virginia Vassilevska, “A Refined Laser Method and Faster Matrix Multiplication,” in Proceedings of the 2021 ACM-SIAM Symposium on Discrete Algorithms (SODA 2021), pp. 522–539, SIAM, 2021.
3. Eric Bach and Jeff Shallit, Algorithmic Number Theory, Vol. I: Efficient Algorithms, MIT Press, Cambridge, MA, 1996.
4. Alan Baker, “The Theory of Linear Forms in Logarithms,” in Transcendence Theory: Advances and Applications: proceedings of a conference held at the University of Cambridge, Cambridge, Jan.–Feb., 1976, Academic Press, London, 1977.
5. Alan Baker, “Logarithmic forms and the abc-conjecture,” Number theory (Eger, 1996), pp. 37–44, de Gruyter, Berlin, 1998.
6. Baker Alan Wustholtz Gisbert Logarithmic forms and group varieties J. Reine Angew. Math. 1993 442 19 62
7. Saugata Basu; Richard Pollack; Marie-Françoise Roy, “Computing the dimension of a semi-algebraic set,” reprinted in J. Math. Sci. (N.Y.) 134 (2006), no. 5, pp. 2346–2353.
8. Saugata Basu; Richard Pollack; and Marie-Françoise Roy Algorithms in real algebraic geometry, 2nd edition, Algorithms and Computation in Mathematics, 10 2006 Berlin Springer-Verlag
9. Basu Saugata Roy Marie-Françoise Divide and conquer roadmap for algebraic sets Discrete and Computational Geometry 2014 52 278 343 10.1007/s00454-014-9610-9
10. Daniel J. Bates; Jonathan D. Hauenstein; Matthew E. Niemerg; Frank Sottile, “Software for the Gale transform of fewnomial systems and a Descartes rule for fewnomials,” Numer. Algorithms 73 (2016), no. 1, pp. 281–304.
11. Daniel J. Bates, Jon D. Hauenstein, Andrew J. Sommese, and Charles W. Wampler, Numerically solving polynomial systems with Bertini, Software, Environments, and Tools 25, SIAM, 2013.
12. Ben-Or Michael Kozen Dexter Reif John “The Complexity of Elementary Algebra and Geometry,” J. Computer and System Sciences 1986 32 251 264 10.1016/0022-0000(86)90029-2
13. Daniel J. Bernstein, “Computing Logarithm Intervals with the Arithmetic-Geometric Mean Iterations,” available from http://cr.yp.to/papers.html .
14. Benoit Bertrand; Frédéric Bihan; and Frank Sottile Polynomial Systems with Few Real Zeroes Mathematisches Zeitschrift 2006 253 2 361 385 10.1007/s00209-005-0912-8
15. Frédéric Bihan, Topologie des variétés creuses, Habilitation thesis, Université de Savoie, France, 2011.
16. Bihan Frédéric Dickenstein Alicia Descartes’ Rule of Signs for Polynomial Systems Supported on Circuits International Mathematics Research Notices 2017 2017 22 6867 6893
17. Frédéric Bihan; Alicia Dickenstein; Jens Forsgård, “Optimal Descartes’ rule of signs for systems supported on circuits,” Mathematische Annalen 381 (2021), pp. 1283-1307.
18. Frédéric Bihan; Alicia Dickenstein; and Magalí Giaroli, “Regions of multistationarity in cascades of Goldbeter-Koshland loops,” J. Math. Biol. (2019) Vol. 78(4), pp. 1115–1145.
19. Frédéric Bihan; Alicia Dickenstein; and Magalí Giaroli, “Lower bounds for positive roots and regions of multistationarity in chemical reaction networks,” J. Algebra (2020), Vol. 542, pp. 367-411.
20. Frédéric Bihan; Alicia Dickenstein; and Magalí Giaroli, “Sign conditions for the existence of at least one positive solution of a sparse polynomial system,” Advances in Mathematics, 375, 2 2020.
21. Frédéric Bihan, J. Maurice Rojas, and Frank Sottile, “On the Sharpness of Fewnomial Bounds and the Number of Components of Fewnomial Hypersurfaces,” Algorithms in Algebraic Geometry (Alicia Dickenstein, Frank-Olaf Schreyer, and Andrew J. Sommese, eds.), IMA Volumes in Mathematics and its Applications, Vol. 146, pp. 15–20, Springer-Verlag, 2007.
22. Frédéric Bihan; J. Maurice Rojas; Casey E. Stella, “Faster Real Feasibility via Circuit Discriminants,” proceedings of International Symposium on Symbolic and Algebraic Computation (ISSAC 2009, July 28–31, Seoul, Korea), pp. 39–46, ACM Press, 2009.
23. Bihan Frédéric Sottile Frank New Fewnomial Upper Bounds from Gale Dual Polynomial Systems Moscow Mathematical Journal 2007 7 3 387 407 10.17323/1609-4514-2007-7-3-387-407
24. Lenore Blum; Felipe Cucker; Mike Shub; and Steve Smale, Complexity and Real Computation, Springer-Verlag, 1998.
25. Enrico Bombieri and Walter Gubler, Heights in Diophantine Geometry, new mathematical monographs: 4, Cambridge University Press, 2006.
26. Enrico Bombieri; Jean Bourgain; and Sergei Konyagin, “Roots of polynomials in subgroups of Fp∗ and applications to congruences,” Int. Math. Res. Not. IMRN 2009, no. 5, pp. 802–834.
27. Erick Boniface, Weixun Deng, and J. Maurice Rojas, “Trinomials and Complexity Limits Over the Reals,” in progress, 2022.
28. John M. Borwein and Peter B. Borwein; “On the Complexity of Familiar Functions and Numbers,” SIAM Review, Vol. 30, No. 4, (Dec., 1988), pp. 589–601.
29. Brent Richard P Fast Multiple-Precision Evaluation of Elementary Functions Journal of the Association for Computing Machinery 1976 23 2 242 251 10.1145/321941.321944
30. Yan Bugeaud; Maurice Mignotte; and Samir Siksek, “Classical and modular approaches to exponential Diophantine equations, I, Fibonacci and Lucas perfect powers,” Ann. of Math. (2) 163 (2006), pp. 969–1018.
31. Bürgisser Peter Ergür Alperen A Tonelli-Cueto Josué On the Number of Real Zeros of Random Fewnomials SIAM Journal on Applied Algebra and Geometry 2019 3 4 721 732 10.1137/18M1228682
32. Ran Canetti; John B. Friedlander; Sergey Konyagin; Michael Larsen; Daniel Lieman; and Igor E. Shparlinski, “On the statistical properties of Diffie-Hellman distributions,” Israel J. Math. 120 (2000), pp. 23–46.
33. John F. Canny, “Some Algebraic and Geometric Computations in PSPACE,” Proc. 20th_ ACM Symp. Theory of Computing, Chicago (1988), ACM Press.
34. Cattani Eduardo Dickenstein Alicia Counting solutions to binomial complete intersections Journal of Complexity 2007 23 82 107 10.1016/j.jco.2006.04.004
35. Chandrasekaran Venkat Shah Parikshit Relative Entropy Relaxations for Signomial Optimization SIAM J. Optim. 2016 26 2 1147 1173 10.1137/140988978
36. Chen Tianran Li Tien-Yien Solutions to Systems of Binomial Equations Annales Mathematicae Silesianae 2014 28 7 34
37. Qi Cheng; Shuhong Gao; J. Maurice Rojas; and Daqing Wan, “Sparse Univariate Polynomials with Many Roots Over a Finite Field,” Finite Fields and their Applications, Vol. 46, July 2017, pp. 235–246.
38. Alexander L. Chistov and Dima Yu Grigoriev, “Complexity of Quantifier Elimination in the Theory of Algebraically Closed Fields,” Lect. Notes Comp. Sci. 176, Springer-Verlag (1984).
39. Thomas H. Cormen; Charles E. Leiserson; Ronald L. Rivest; and Clifford Stein, Introduction to Algorithms, 3rd edition, MIT Press, 2009.
40. Cucker Felipe Krick Teresa Malajovich Gregorio Wschebor Mario A numerical algorithm for zero counting. I: Complexity and Accuracy J. Complexity 2008 24 582 605 10.1016/j.jco.2008.03.001
41. Cucker Felipe Krick Teresa Shub Michael Computing the homology of real projective sets Found. Comput. Math. 2018 18 929 970 10.1007/s10208-017-9358-8
42. Alicia Dickenstein; Magalí Giaroli; Rick Rischter; Mercedes Pérez Millán, “Parameter regions that give rise to 2[n/2]+1 positive steady states in the n-site phosphorylation system,” Mathematical Biosciences and Engineering, 2019, 16(6):7589–7615.
43. Alicia Dickenstein; Mercedes Pérez Millán; Anne Shiu; and Xiaoxian Tang, “Multistationarity in Structured Reaction Networks,” Bulletin of Mathematical Biology (2019) 81(5), 1527-1581.
44. Mareike Dressler, Adam Kurpisz, and Timo de Wolff, “Optimization over the boolean hypercube via sums of nonnnegative circuit polynomials,” Foundations of Computational Mathmematics, 2021, https://doi.org/10.1007/s10208-021-09496-x .
45. Duffin RJ Schaeffer AC A refinement of an inequality of the brothers Markoff Transactions of the American Mathematical Society 1941 50 517 528 10.1090/S0002-9947-1941-0005942-4
46. David Eisenbud, Commutative Algebra with a View Toward Algebraic Geometry, Graduate Texts in Mathematics 150, Springer-Verlag, 1995.
47. Ioannis Z. Emiris, MixedVolume-SparseResultants software package, https://github.com/iemiris/MixedVolume-SparseResultants .
48. Alperen A. Ergür, Grigoris Paouris, and J. Maurice Rojas, “Probabilistic Condition Number Estimates for Real Polynomial Systems I: A Broader Family of Distributions,” Foundations of Computational Mathematics, Feb. 2019, Vol. 19, No. 1, pp. 131–157.
49. Alperen A. Ergür, Grigoris Paouris, and J. Maurice Rojas, “Smoothed analysis for the condition number of structured real polynomial systems,” Math. Comp. 90 (2021), pp. 2161-2184.
50. Alperen A. Ergür, Grigoris Paouris, and J. Maurice Rojas, “Randomized Baker’s Theorem and Real Algebraic Geometry,” preprint, Texas A &M University, 2022.
51. Joachim von zur Gathen and Jürgen Gerhard, Modern Computer Algebra, 3rd ed., Cambridge University Press, 2013.
52. Gel’fand Israel M Kapranov Mikhail M Zelevinsky Andrei V Discriminants 1994 Birkhäuser, Boston Resultants and Multidimensional Determinants
53. David J. Grabiner, “Descartes’ Rule of Signs: Another Construction,” The American Mathematical Monthly, Vol. 106, No. 9 ( 1999), pp. 854–856.
54. Branko Grünbaum, Convex Polytopes, 2nd ed. (prepared by Volker Kaibel, Victor Klee, and Günter Ziegler), Graduate Texts in Mathematics, vol. 221, Springer-Verlag, 2003.
55. Robert M. Hardt, “Semi-Algebraic Local-Triviality in Semi-Algebraic Mappings,” American Journal of Mathematics, Vol. 102, No. 2 (Apr., 1980), pp. 291–302.
56. David Harvey and Joris van der Hoeven, “Integer multiplication in time O(nlogn),” Annals of Mathematics, Vol. 193, No. 2 (March 2021), pp. 563–617.
57. Hermite Charles Sur l’introduction des variables continues dans la théorie des nombres J. Reine Angew. Math. 1851 41 191 216
58. Huber Birkett Sturmfels Bernd A Polyhedral Method for Solving Sparse Polynomial Systems Math. Comp. 1995 64 212 1541 1555 10.1090/S0025-5718-1995-1297471-4
59. Gorav Jindal and Mikael Sagraloff, “Efficiently computing real roots of sparse polynomials,” in: Proceedings of the 2017 ACM ISSAC (International Symposium on Symbolic and Algebraic Computation), ACM, New York. pp. 229–236. 10.1145/3087604.3087652 .
60. Kemper Gregor A Course in Commutative Algebra, Graduate Texts in Mathematics 256 2011 Berlin Heidelberg Springer-Verlag
61. Khovanskiĭ Askold G A class of systems of transcendental equations Dokl. Akad. Nauk SSSR 1980 255 4 804 807
62. Askold G Khovanskiĭ 1991 Fewnomials AMS Press, Providence, Rhode Island
63. Pascal Koiran; Natacha Portier; and Sebastian Tavenas, “On the intersection of a sparse curve and a low-degree curve: A polynomial version of the lost theorem,” Discrete and Computational Geometry, 53(1):48-63, 2015.
64. Pascal Koiran; Natacha Portier; and Sebastian Tavenas, “A Wronskian approach to the real tau-conjecture,” Journal of Symbolic Computation, 68(2):195-214, 2015.
65. Koiran Pascal Root separation for trinomials J. Symbolic Comput. 2019 95 151 161 10.1016/j.jsc.2019.02.004
66. Kronecker Leopold Werke 1895 Leipzig Teubner
67. Anatoly Georgievich Kushnirenko, “Newton Polytopes and the Bézout Theorem," Functional Analysis and its Applications (translated from Russian), vol. 10, no. 3, July–September (1977), pp. 233–235.
68. Serge Lang, Elliptic Curves: Diophantine Analysis, Springer, 1978.
69. Tsung-Lin Lee and Tien-Yien Li, “Mixed volume computation in solving polynomial systems,” in Randomization, Relaxation, and Complexity in Polynomial Equation Solving, Contemporary Mathematics, vol. 556, pp. 97–112, AMS Press, 2011.
70. François Legall, “Powers of tensors and fast matrix multiplication,” Proceedings of ISSAC (International Symposium on Symbolic and Algebraic Computation) 2014, ACM Press, pp. 296–303, 2014.
71. Tien-Yien Li; J. Maurice Rojas; and Xiaoshen Wang, “Counting Real Connected Components of Trinomial Curves Intersections and m-nomial Hypersurfaces,” Discrete and Computational Geometry, 30:379–414 (2003).
72. Liouville Joseph Sur des classes très étendues de quantités dont la valeur n’est ni algébrique, ni même réductible á des irrationnelles algébriques Journal Math. Pures et Appl. 1851 16 133 142
73. Richard A. Lipton and Richard J. Lipton, “A probabilistic remark on algebraic program testing,” Information Processing Letters 7 (4): 193–195.
74. Mahler Kurt An inequality for the discriminant of a polynomial The Michigan Mathematical Journal 1964 11 3 257 262 10.1307/mmj/1028999140
75. A. A. Markov, “On a certain problem of D. I. Mendeleiff,” (in Russian) Utcheniya Zapiski Imperatorskoi Akademii Nauk, 62, pp. 1–24, 1889.
76. David W. Masser, “Open Problems,” Prod. Symp. Analytic Number Theory (ed. by W. W. L. Chen), Imperial Coll. London, 1985.
77. E. M. Matveev, “An explicit lower bound for a homogeneous rational linear form in logarithms of algebraic numbers, II”, Izv. Ross. Akad. Nauk Ser. Mat. 64 (2000), pp. 125–180; English transl. in Izv.l Math. 64 (2000), pp. 1217–1269.
78. Maurice Mignotte, “Some Useful Bounds,” Computing, Suppl. 4, pp. 259–263 (1982), Springer Verlag.
79. Pinaki Mondal, How many zeroes? Counting the number of solutions of systems of polynomials via geometry at infinity, Springer Verlag, 2021.
80. Yuri Nesterenko, “Linear forms in logarithms of rational numbers,” Diophantine approximation (Cetraro, 2000), pp. 53–106, Lecture Notes in Math., 1819, Springer, Berlin, 2003.
81. Abderrahmane Nitaj, The abc Conjecture Home Page, https://nitaj.users.lmno.cnrs.fr/abc.html
82. Oesterlé Joseph Nouvelles approches du ‘Théorème’ de Fermat Astérisque 1988 161–2 165 186
83. Grigoris Paouris, Kaitlyn Phillipson, and J. Maurice Rojas, “A Faster Solution to Smale’s 17th Problem I: Real Binomial Systems,” in proceedings of ISSAC 2019 (July 15-18, 2019, Beihang University, Beijing, China), ACM Press, 2019.
84. Philippe Pébay; J. Maurice Rojas; and David C. Thompson, “Optimization and NPR -completeness of certain fewnomials,” proceedings of SNC 2009 (August 3–5, 2009, Kyoto, Japan), pp. 133–142, ACM Press, 2009.
85. Paul Pedersen, Marie-Françoise Roy, and Aviva Szpirglas, “Counting real zeros in the multivariate case,” in proceedings of Computational algebraic geometry (Nice, 1992), pp. 203–224, Progr. Math., 109, Birkhäuser Boston, Boston, MA, 1993.
86. Kaitlyn Phillipson and J. Maurice Rojas, “Fewnomial Systems with Many Roots, and an Adelic Tau Conjecture,” in proceedings of Bellairs workshop on tropical and non-Archimedean geometry (May 6–13, 2011, Barbados), Contemporary Mathematics, vol. 605, pp. 45–71, AMS Press, 2013.
87. Victor V. Prasolov, Problems and Theorems in Linear Algebra, translations of mathematical monographs, vol. 134, AMS Press, 2004.
88. Qazi Ibadur Rahman and Gerhard Schmeisser, Analytic Theory of Polynomials, London Mathematical Society Monographs 26, Oxford Science Publications, 2002.
89. Jim Renegar, “ On the Computational Complexity and Geometry of the First-Order Theory of the Reals, I–III,” J. Symbolic Comput. 13 (1992), no. 3, pp. 255–352.
90. J. Maurice Rojas, “Solving degenerate sparse polynomial systems faster,” J. Symbolic Comput. 28 (1999), no. 1-2, pp. 155–186.
91. J. Maurice Rojas, “Why Polyhedra Matter in Non-Linear Equation Solving,” Contemporary Mathematics, vol. 334, pp. 293–320, AMS Press, 2003.
92. J. Maurice Rojas and Yinyu Ye, “On Solving Sparse Polynomials in Logarithmic Time,” Journal of Complexity, special issue for the 2002 Foundations of Computation Mathematics (FOCM) meeting, February 2005, pp. 87–110.
93. J. Maurice Rojas and Yuyu Zhu, “Root Repulsion and Faster Solving for Very Sparse Polynomials Over p -adic Fields,” Journal of Number Theory, Vol. 241, Dec. 2022, pp. 655–699.
94. Rouillier Fabrice Solving zero-dimensional systems through the rational univariate representation Appl. Algebra Engrg. Comm. Comput. 1999 9 5 433 461 10.1007/s002000050114
95. Mohab Safey El Din and Éric Schost, “A nearly optimal algorithm for deciding connectivity queries in smooth and bounded real algebraic sets,” Journal of the ACM, Vol. 63, No. 6, pp. February 2017.
96. Michael Sagraloff, “A General Approach to Isolating Roots of a Bit-stream Polynomial,” Mathematics in Computer Science 4, 481 (2010), Springer-Verlag.
97. Eugene Salamin, “Computation of π using arithmetic-geometric mean,” Math. Comput., 30 (1976), pp. 565–570
98. Alexander Schrijver, Theory of Linear and Integer Programming, John Wiley & Sons, 1986.
99. Jacob T. Schwartz, “Fast Probabilistic Algorithms for Verification of Polynomial Identities,” J. of the ACM 27, 701–717, 1980.
100. David Eugene Smith and Marcia L. Latham, The Geometry of René Descartes, translated from the French and Latin (with a facsimile of Descartes’ 1637 French edition), Dover Publications Inc., New York (1954).
101. Storjohann Arne "Algorithms for Matrix Canonical Forms," doctoral dissertation 2000 Zurich Swiss Federal Institute of Technology
102. Verschelde Jan Polynomial Homotopy Continuation with PHCpack ACM Communications in Computer Algebra 2010 44 4 217 220
103. Kunrui Yu, “p -adic logarithmic forms and group varieties III,” Form Math., Vol. 19, No. 2, pp. 187–280, 2007.
104. Richard Zippel, “Probabilistic algorithms for sparse polynomials,” Symbolic and Algebraic Computation, Lecture Notes in Computer Science 72, pp. 216–226, Springer 1979.
| 36465263 | PMC9707291 | NO-CC CODE | 2022-12-01 23:20:22 | no | Found Comut Math. 2022 Nov 28;:1-43 | utf-8 | Found Comut Math | 2,022 | 10.1007/s10208-022-09599-z | oa_other |
==== Front
IFAC-PapersOnLine
2405-8963
2405-8963
, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd.
S2405-8963(22)02916-0
10.1016/j.ifacol.2022.11.273
Article
A Physics-Based Data-Driven Approach for Finite Time Estimation of Pandemic Growth⁎
Uppaluru Harshvardhan *
Rastgoftar Hossein ⁎⁎
⁎ Scalable Move and Resilient Traversability (SMART) Lab, Aerospace and Mechanical Engineering Department, University of Arizona, Tucson, AZ 85721, USA
⁎⁎ Scalable Move and Resilient Traversability (SMART) Lab, Aerospace and Mechanical Engineering Department, University of Arizona, Tucson, AZ 85721, USA
29 11 2022
2022
29 11 2022
55 37 758763
© 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
COVID-19 is a global health crisis that has had unprecedented, widespread impact on households across the United States and has been declared a global pandemic on March 11, 2020 by World Health Organization (WHO). According to Centers for Disease Control and Prevention (CDC), the spread of COVID-19 occurs through person-to-person transmission i.e. close contact with infected people through contaminated surfaces and respiratory fluids carrying infectious virus. This paper presents a data-driven physics-based approach to analyze and predict the rapid growth and spread dynamics of the pandemic. Temporal and spatial conservation laws are used to model the evolution of the COVID-19 pandemic. We integrate quadratic programming and neural networks to learn the parameters and estimate the pandemic growth. The proposed prediction model is validated through finite-time estimation of the pandemic growth using the total number of cases, deaths and recoveries in the United States recorded from March 12, 2020 until October 1, 2021.
Keywords
COVID-19
Discrete Event Dynamic Systems
Estimation
Modeling and Validation
Pandemic Growth Analysis
==== Body
pmc⁎ This work has been supported by the National Science Foundation under Award Nos. 2133690 and 1914581.
==== Refs
REFERENCES
(2020). Worldometer. https://www.worldometers.info/coronavirus/country/us/.
Chimmula V.K.R. Zhang L. Time series forecasting of COVID-19 transmission in Canada using LSTM networks Chaos, Solitons & Fractals 135 2020 109864
Chinazzi M. Davis J.T. Ajelli M. Gioannini C. Litvinova M. Merler S. y Piontti A.P. Mu K. Rossi L. Sun K. The effect of travel restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak Science 2020
Fanelli D. Piazza F. Analysis and forecast of COVID-19 spreading in China, Italy and France Chaos, Solitons & Fractals 134 2020 109761
Grenfell B. Harwood J. (Meta) population dynamics of infectious diseases Trends in ecology & evolution 12 10 1997 395 399 21238122
Hanski I. Metapopulation dynamics Nature 396 6706 1998 41 49
Keeling M.J. Bjørnstad O.N. Grenfell B.T. Metapopulation dynamics of infectious diseases Ecology, genetics and evolution of metapopulations 2004 Elsevier 415 445
Kermack W.O. McKendrick A.G. A contribution to the mathematical theory of epidemics Proceedings of the royal society of london. Series A, Containing papers of a mathematical and physical character 115 772 1927 700 721
Kermack W.O. McKendrick A.G. Contributions to the mathematical theory of epidemics. ii.—the problem of endemicity Proceedings of the Royal Society of London. Series A, containing papers of a mathematical and physical character 138 834 1932 55 83
Kermack W.O. McKendrick A.G. Contributions to the mathematical theory of epidemics. iii.—further studies of the problem of endemicity Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character 141 843 1933 94 122
Kırbas¸ İ. Sözen A. Tuncer A.D. Kazancıoğlu F.S¸ Comparative analysis and forecasting of COVID-19 cases in various european countries with ARIMA, NARNN and LSTM approaches Chaos, Solitons & Fractals 138 2020 110015
Liao Z. Lan P. Liao Z. Zhang Y. Liu S. TW-SIR: time-window based sir for COVID-19 forecasts Scientific reports 10 1 2020 1 15 31913322
Nabi K.N. Tahmid M.T. Rafi A. Kader M.E. Haider M.A. Forecasting COVID-19 cases: A comparative analysis between recurrent and convolutional neural networks Results in Physics 24 2021 104137
Rastgoftar H. Atkins E. A mass-conservation model for stability analysis and finite-time estimation of spread of COVID-19 IEEE Transactions on Computational Social Systems 2021
Singh P. Gupta A. Generalized sir (GSIR) epidemic model: An improved framework for the predictive monitoring of covid-19 pandemic ISA transactions 2021
van Doremalen N. Bushmaker T. Morris D.H. Holbrook M.G. Gamble A. Williamson B.N. Tamin A. Harcourt J.L. Thornburg N.J. Gerber S.I. Aerosol and surface stability of SARS-CoV-2 as compared with SARS-CoV-1 New England Journal of Medicine 2020
| 0 | PMC9707306 | NO-CC CODE | 2022-12-06 23:15:57 | no | 2022 Nov 29; 55(37):758-763 | utf-8 | null | null | null | oa_other |
==== Front
Int J Hosp Manag
Int J Hosp Manag
International Journal of Hospitality Management
0278-4319
1873-4693
Elsevier Ltd.
S0278-4319(22)00272-9
10.1016/j.ijhm.2022.103406
103406
Article
The resilience of the lodging industry during the pandemic: Hotels vs. Airbnb
Dogru Tarik a
Hanks Lydia a⁎
Suess Courtney b
Line Nathan a
Mody Makarand c
a Florida State University, Dedman College of Hospitality, Tallahassee, FL, USA
b Texas A&M University, Department of Recreation, Park and Tourism Sciences, College Station, TX, USA
c Boston University, School of Hospitality Administration, Boston, MA, USA
⁎ Corresponding author.
29 11 2022
2 2023
29 11 2022
109 103406103406
16 2 2022
15 11 2022
25 11 2022
© 2022 Elsevier Ltd. All rights reserved.
2022
Elsevier Ltd
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
The adverse impact of the recent pandemic on the lodging industry has largely been based on anecdotal evidence. The extent to which different parts of this broad industry were individually affected by the COVID-19 pandemic also remains unclear. The purpose of this study is to investigate the effects of the COVID-19 pandemic in the various sectors of the lodging industry to identify patterns that may not be consistent with the idea that the entire hospitality industry was negatively affected by the pandemic. The results show that while the COVID-19 pandemic did have a generally negative effect on lodging demand, hotel room and peer-to-peer accommodation property bookings were not affected equally. Importantly, it appears that these variations were attributable, at least in part, to state-level variations in policy that made travel and hospitality services relatively more (or less) difficult for consumers to obtain. Theoretical and managerial implications are extensively discussed.
Keywords
COVID-19
Pandemic
Lodging industry
Hotels
Airbnb
Resilience
==== Body
pmc1 Introduction
Anecdotal evidence abounds regarding the negative effect of the COVID-19 pandemic on the lodging industry. Headlines such as “North American inbound tourism spend declined by 74.1% in 2020″ (Johnson, 2021) and “Coronavirus Pandemic Sets Hotel Industry Back 10 Years, Report Finds” (Lardieri, 2021) have led to both an academic and a practical understanding that, in a general sense, the COVID-19 pandemic negatively affected consumers’ booking overnight accommodations. However, the accommodations industry is quite complex. For example, the lodging industry comprises both traditional hotels and resorts as well as individuals’ home properties listed through companies such as Airbnb, Vrbo, HomeAway, etc. (peer-to-peer accommodations). Additionally, there are lodging sub-types within this stratification. Within the traditional hotel sector, there are six distinct lodging classes ranging from economy to luxury (STR, 2021). Likewise, amongst the peer-to-peer accommodation listings by companies such as Airbnb, available property types include entire homes, private rooms within an owner-occupied home, and shared-rooms with other guests. Thus, while the accommodations industry is often considered in aggregate when it comes to reporting purchasing decline from COVID-19, it is possible that not all of the various operating sectors were affected equally.
Many different factors exist that could have caused variation between (and within) bookings in the hotel and the peer-to-peer accommodation sectors. On one hand, consumer sentiment about safety could have potentially shifted demand away from hotels and toward the peer-to-peer accommodations due to the perception that private accommodations are more conducive to social distancing than hotels. On the other hand, consumers may have been willing to forego the relative isolation inherent to private or even shared homes in favor of hotels that typically provide a standardized and more professional approach to health and sanitation (Verleye et al., 2021). In addition, it is possible that factors outside of consumer sentiment, such as state-level policy, travel restrictions, and limitations on hospitality operations, could have caused differential effects on various booking patterns throughout the lodging industry. While many states implemented relatively strict mandates in an effort to mitigate spread of COVID-19 (e.g., stay at home orders, curfews, limited business operating hours, business closures, visitor restrictions, etc.) other states imposed relatively few restrictions (McCann, 2021). As such, it is likely that inter-state disparities in COVID-19-based restrictions contributed, at least in part, to variations in demand effects between and within the various operating sectors of the lodging industry.
Overall, it is evident that occupancy rates were disrupted at the industry level during the COVID-19 pandemic. However, it remains unclear how various sectors of the broad industry were affected. Accordingly, the purpose of this study is to investigate the effects of the COVID-19 pandemic in the various sectors of the lodging industry to identify patterns that may not be consistent with the idea that the entire hospitality industry was negatively affected by the pandemic. Additionally, this research seeks to identify the effects of state-level policy on the demand across various operating sectors of the lodging industry. The results from this study are expected to shed light on several issues ranging from the resilience from pandemic-related demand shocks to the effect of public-health regulations on guest booking hotel rooms and peer-to-peer accommodations.
2 Literature review
2.1 The impact of COVID-19
With the onset of the COVID-19 pandemic, travel restrictions and consumer concerns have had a significant impact on all facets of the travel and tourism industry. In an effort to quantify these effects, understand their precursors, and develop mitigating strategies, researchers have conducted a number of studies in various contexts, beginning very shortly after the onset of the pandemic. For example, Ugir & Akbiyik (2020) analyzed data from a large travel forum and found that prior to the worldwide proliferation of the virus, even while the pandemic was still localized, travelers were changing, adjusting, and canceling travel plans. Likewise, Li et al., (2020) found that as the pandemic began to spread globally, travelers were delaying travel plans and booking shorter stays than originally planned. Indeed, in 2020, a flurry of papers were published, analyzing the impact of the pandemic on various aspects of the tourism industry, such as trip type (Wen et al., 2020), travel likelihood (Yang et al., 2020), tourism consumer behavior (Zenker & Kock, 2020; Suess et al., 2022), and homesharing (Dolnicar and Zare, 2020). Research into the impact of the pandemic on the generalized hospitality and tourism industry continued during the following two years, demonstrating the adverse effects on tourism asset management (Crespi-Cladera et al., 2021), tourism company valuations (Kocak et al., 2022), and various subsectors of the tourism and hospitality industry. Cepni et al. (2022) found that the hotel subsector of the tourism industry was hit particularly hard by the pandemic.
Several studies have been conducted over the past two years investigating the impact of COVID-19 on the hotel industry. For example, Guillet & Chu (2021) conducted a qualitative study, speaking with revenue managers at hotels in Asia, and found that hotel demand was down, regardless of revenue management strategies. Likewise, Ozdemir et al. (2021) found that demand was down across the entire hotel industry, although economy-scale hotels were more resilient than other segments. Ozdemir et al. (2022) found that demand was weaker in states where travel and gathering restrictions were the most stringent, and that this was not uniform across all hotel segments, with economy hotels being more resistant than luxury segment hotels. These studies would suggest that there is much to be learned about the ways in which the COVID-19 pandemic impacts various types of hotels differentially. To fill this gap, the present study aims to investigate how the pandemic impacted the various hotel classes, as well as comparing these results to homesharing platforms.
2.2 The effect of COVID-19 on the hotel industry
There is no doubt that COVID-19 has significantly impacted the lodging industry. In the face of the pandemic, visiting friends and family has been canceled, business trips have shifted into online videoconferencing, and leisure and vacation travel have been severely restricted. These changes have been a major disruptor, seriously affecting the traditional hotel business models (Schulz, 2021). However, as the course of the pandemic has ebbed and flowed, it is difficult to discern exactly how the hotel industry has been affected. Impacts to hotel occupancy have been relatively fluid, with some phases of the pandemic producing drastic drops in travel and associated room bookings during times of high-risk and peak infection rates, followed by months of recovery fueled by demand from those pent-up in their residences, seeking to travel (Eggleston and Lee, 2021). Consumer perceptions of safety, the ability and willingness to travel, and anxiety about the pandemic have all resulted in demand-side fluctuations in hotel reservations. In addition, government mandated closures and stay-at-home orders have had a profound effect. Given the number and the complexity of factors affecting hotel reservations, it is difficult to discern how COVID-19 has affected overall hotel occupancy. Thus, the following is proposed:
Research Question 1:RQ1 To what extent does the COVID-19 pandemic affect hotel occupancy?
To further illuminate the ways in which COVID-19 has impacted the hotel industry, it is important to examine whether the effects of the pandemic on hotel occupancy vary across hotel classes. For classification and comparison purposes, hotels are often divided into categories (i.e., economy, midscale, upper midscale, upscale, upper upscale, and luxury) (STR Chain Scales, 2019). While there are many ways to classify hotels, these categories are widely accepted due to their use by Smith Travel Research and the STAR report, thus creating a taxonomy that is commonly understood by hoteliers and researchers alike (Miller, 2016). Ozdemir et al. (2021) have suggested that the pandemic has affected various hotel classes differently, but aside from this study, there is little understanding of how each class of hotel has been specifically impacted. Thus, research question 2 is proposed as follows:RQ2 To what extent does the COVID-19 pandemic affect the occupancy levels of various hotel classes (i.e., economy, midscale, upper midscale, upscale, upper upscale, and luxury)?
2.3 The effect of COVID-19 on Airbnb occupancy
In addition to affecting hotel occupancy, it is possible that the restrictions on travel have affected the peer-to-peer accommodations sectors as well. While there are many home-sharing platforms, for the purposes if this study, the effects of the pandemic on Airbnb property occupancy levels is most germane. Airbnb is the largest among the peer-to-peer accommodation platforms, and a primary competitor to the hotel industry, providing travelers with the opportunity to book a stay in an individual’s private home, either with or without them occupying it as a host. Moreover, data on the properties listed on Airbnbs’ occupancy levels is readily available. While COVID-19 restrictions, lockdowns, and canceled travel plans have certainly affected occupancy among Airbnb properties, research has suggested that the impact has not been entirely negative (Dolnicar and Zare, 2020). As safety-conscious travelers opt out of staying in hotels where they must share space such as elevators, lobbies, and restaurants with other patrons, they have looked toward renting the private properties listed on Airbnb as an alternative, where they will not encounter other guests and extensive staff, thus decreasing the risk of catching the virus (Carville and Bloomberg, 2021). Additionally, the shift at many companies to working at home has given rise to a new group of customers who choose to rent a property through Airbnb on a long-term basis so that they can live in and explore new and different cities and countries while working remotely (Hines, 2021). Such digital nomads have generated demand for peer-to-peer accommodations in a way that could not be foreseen prior to the pandemic (Rodriguez, 2020). Given that travel restrictions and health concerns significantly decreased travel overall during the pandemic, yet health-conscious guests and digital nomads generated new demand, it remains unclear how COVID-19 has impacted occupancy of Airbnb listings overall. Thus, research question 3 is proposed:RQ3 To what extent does the COVID-19 pandemic affect Airbnb occupancy?
One unique characteristic of Airbnb listings is that guests have a number of property options. First, the guest can choose to book an entire home with no host residing at the premises during their stay. Second, guests can book a private bedroom or suite within a home in which the host is living during their stay. Third, guests can also book a shared room (i.e., the room is shared with another guest) within a home (Airbnb, 2021). Because each of these property types carries different implications in terms of contact with people during COVID-19, it is important to examine each one, separately. Prior research has demonstrated that these three peer-to-peer accommodation property types vary in such areas as pricing (Voltes-Dorta and Sánchez-Medina, 2020) and occupancy (Bresciani et al., 2021).
It is important to note that shared rooms and shared homes involve direct contact and shared common areas with others, a situation that is contrary to social distancing practices. Alternatively, entire-home rentals, with no host occupying the residence, provide the opportunity to isolate and social distance. Indeed, Bresciani et al. (2021) found that during the pandemic, travelers were reluctant to book homes shared with other guests or occupied by hosts during the pandemic, and the preference for whole home rentals was predicated on the desire for social distancing. Based on this idea, it seems logical to investigate whether the effect of COVID-19 varies across property types in Airbnb rentals. Thus, research question 4 is proposed:RQ4 Is the effect of COVID-19 on occupancy levels homogenous across Airbnb property types?
2.4 The effect of COVID-19 state restrictions
While consumer concerns about cleanliness, social distancing, and isolation while traveling certainly had an impact on the lodging industry’s occupancy levels, these are not the only factors to consider. State-level restrictions on hospitality businesses also had a significant impact on lodging demand, throughout the pandemic. Many states enforced social-distancing and ordered restrictions on business openings, hours of operation, number of patrons allowed in any given business, and private gatherings -in some cases any gathering over six people (Multistate, 2021). However, there has been vast deviation with regards to the number and type of restrictions each state has imposed during the pandemic. For example, California, New York, and Massachusetts enforced the strictest measures, maintaining them throughout periods when infection rates were very low, and vaccination was prevalent (McCann, 2021). Meanwhile, states such as Florida and Texas had minimal restrictions, and only during brief periods when the infection cases were reported to be at their highest and when all other states were enforcing the public safety measures under federal pressure (Hunnicutt & Mason, 2021).
Such restrictions obviously had a significant impact on hotel occupancy rates, as they not only set a precedent of discouraging people from going out in public, but also limited the ability of those who chose to stay mobile to travel and book a hotel (Gursoy and Chi, 2020, Lai and Wong, 2020). Hotels were further affected as they became vulnerable to government restrictions on gathering size, impacting conferences, events, and group business significantly. Thus, it is important to understand how state-level restrictions impacted hotel occupancy, and Research Question 5 is thus proposed:RQ5 How has the level of state restrictions regarding COVID-19 impacted hotel occupancy levels?
As mentioned above, peer-to-peer accommodations listed on Airbnb differ from traditional hotels in terms of their guest capacity and host occupancy. Both of these factors are important in determining the impact of state COVID-19 restrictions on demand for Airbnb properties. First, because listings on Airbnb do not typically include property types that are dependent on large group business (as opposed to hotels), state restrictions regarding group size, for the most part, would not impact bookings on Airbnb. Second, because properties listed are mostly individual-owned private homes, as opposed to businesses, state mandates to close did not apply to the properties. Third, while state restrictions and social distancing reduced the number of guests permitted on hotel premises and amenity operations, these mandates, in general, did not apply to properties listed on Airbnb or significantly impact host operations for guests in their homes. Namely, almost none of the state restrictions on gatherings applied to entire-home listings, often booked by individuals or family and small groups. In light of the differential social distancing standards among hotels and peer-to-peer accommodations, it is important to investigate Research Question 6:RQ6 How has the level of state restrictions regarding COVID-19 impacted Airbnb property occupancy levels?
3 Methodology
3.1 Sample and data
The sample of this study comprises 49 out of the 50 U.S. states and the District of Columbia for the period between October 2014 and December 2020. The study covers the time period from the first availability of Airbnb property data (e.g., occupancy, supply, demand, etc.) to its latest availability (December 2020). Delaware was omitted from the study sample because there was no data available on this state. Accordingly, the study’s sample contains 3850 state-month observations.
The study’s dependent variable is the monthly occupancy rate, which is the most commonly accepted industry-metric reported on a monthly basis for both hotels and Airbnb peer-to-peer accommodation listings. Specifically, hotel room occupancy rates and Airbnb properties occupancy rates were specified as the main dependent variables of the study to examine the effects of COVID-19 pandemic on hotels and Airbnb property demand. The hotel occupancy data were provided by Smith Travel Research (STR), while data on Airbnb property occupancy were obtained from AirDNA, a well-recognized research company that collects detailed data on Airbnb listings through web crawling. Furthermore, monthly hotel occupancy rate for hotel categories, including economy class, midscale class, upper midscale class, upscale class, upper upscale class, and luxury class hotels were used to analyze the extent to which the effect of COVID-19 pandemic on hotels varies based on diverse hotel categories. Likewise, listings on Airbnb include entire-home, private-room, and shared-room property types. Therefore, monthly Airbnb occupancy rates for entire homes, private rooms, and shared room Airbnb properties were specified as additional dependent variables to investigate the effect of COVID-19 pandemic across alternative Airbnb property types. Fig. 1 presents the occupancy levels of the accommodations during the study period. The dramatic fall in occupancy levels across different accommodation properties is evident during the pandemic period.Fig. 1 Occupancy Levels of Hotels and Airbnb Properties.
Fig. 1
The total monthly number of confirmed COVID-19 cases was specified as the main independent variable of the study, to investigate the effect of COVID-19 pandemic on the US accommodation sector. In addition, the total monthly number of confirmed COVID-19 deaths, monthly new-confirmed COVID-19 cases, and monthly new-confirmed COVID-19 deaths as alternative measures. The COVID-19 new-confirmed cases and death data were obtained from the Centers for Disease Control and Prevention (CDC). Fig. 2 presents the severity of the total COVID-19 cases across the US states during the study period in the form of heat maps.Fig. 2 Total Number of COVID-19 Cases in the US.
Fig. 2
A number of control variables were included to account for macroeconomic factors and travel demand that might affect hotel and Airbnb property occupancy rates regardless of the COVID-19 pandemic. Based on extant studies analyzing hotel supply and demand dynamics (e.g., Canina & Carvell, 2005; Dogru et al., 2020; Lee & Jang; 2012; Lei & Lam, 2015; Tsai, Kang, Yeh, & Suh, 2006), five control variables were included in the models to account for the effects of macroeconomic factors and travel demand on the accommodation sector. Specifically, airport passenger arrivals, hotel room supply, Airbnb property supply, income, and unemployment rate were controlled for. The first control variable in the empirical models, which accounts for general tourism and travel demand dynamics, was the number of airport passenger arrivals to all state airports. Airport passenger arrivals data were obtained from the Bureau of Transportation Statistics. Because a change in hotel room supply and Airbnb properties supply is likely to have an effect on the occupancy rates (Dogru et al., 2020), the effect of hotel room supply and Airbnb properties supply was also controlled for. Furthermore, the coincident index, a variable developed by the Federal Reserve Bank of Philadelphia to measure the monthly gross domestic product (GDP) for each state, was included in the empirical models to account for the effects of general macroeconomic dynamics on hotel and Airbnb property occupancy rates (Crone & Clayton-Matthews, 2005; Lee & Jang, 2012). The coincident index is a commonly used proxy for income when analysis consists of monthly level data frequency (see e.g., Dogru et al., 2019; Lee & Jang, 2012). To further account for overall macroeconomic conditions, unemployment rate was included in the empirical models.
Unemployment rate can help capture overall macroeconomic conditions, which are likely to affect the accommodation sector (Chen et al., 2011). Table 1 presents the summary statistics of the study’s dependent, independent and control variables.Table 1 Summary Statistics.
Table 1Study Variables Mean Median Std. Dev. Min. Max.
Occupancy
All Hotels 0.58 0.60 0.13 0.08 0.89
Economy Class 0.52 0.53 0.15 0.00 0.89
Midscale Class 0.54 0.55 0.14 0.00 0.94
Upper Midscale Class 0.61 0.63 0.14 0.14 0.91
Upscale Class 0.64 0.67 0.15 0.06 0.92
Upper Upscale Class 0.60 0.66 0.20 0.00 0.94
Luxury Class 0.60 0.65 0.18 0.03 0.96
All Airbnb 0.33 0.32 0.12 0.02 0.76
Entire Home Airbnb 0.34 0.33 0.13 0.01 0.79
Private Room Airbnb 0.32 0.32 0.12 0.01 0.72
Shared Room Airbnb 0.22 0.21 0.12 0.00 1.00
Log Total COVID-19 Cases 1.69 0.00 4.01 0.00 15.07
Log Airport Arrivals 12.66 12.56 1.68 6.09 15.80
Log Hotel Supply 14.57 14.64 0.86 12.56 16.64
Log Airbnb Supply 11.49 11.55 1.36 6.51 15.19
Log Coincident Index (Income) 4.81 4.81 0.09 3.93 5.03
Unemployment 0.05 0.04 0.02 0.02 0.29
Although the overall examination of the effects of COVID-19 pandemic on the supply of hotel rooms and Airbnb properties is useful, individual states in the U.S. have had varying regulations in place regarding the COVID-19 pandemic. These regulations, in turn, have resulted in differential effects on the supplies across different states. To account for these potential differences, the data were sorted into two groups based on the state and the extent of the regulations, respectively. Specifically, each state was categorized into either a more restrictive or less restrictive group, depending on the degree of regulations as measured by the WalletHub index of “States with the Fewest COVID-19 Restrictions”. For a detailed description of the index and its methodology, readers are directed to the following reference: WalletHub’s States with the Fewest Coronavirus Restrictions at https://wallethub.com/edu/states-coronavirus-restrictions/73818/. The resulting groups were fairly even, consisting of 25 states in each.
3.2 Empirical methodology
The COVID-19 pandemic is treated as a variable intervention in time against hotel room occupancy and Airbnb properties occupancy rates in 50 states, which allows for the investigation of both before and after effects of the COVID-19 pandemic on hotel room occupancy and Airbnb properties occupancy rates. The dataset covered the pre-pandemic period between October 2014 and December 2019, where total number of COVID-19 cases take the value of zero based on the Centers for Disease Control and Prevention agency, and the post-COVID period between January 2020 and December 2020, where total number of COVID-19 cases take the value of the number of cumulative COVID-19 cases recorded every month in each state. This empirical approach better captures the effect of COVID-19 pandemic on the US accommodation sectors because the COVID-19 pandemic is treated as a variable intervention in time against the hotel room occupancy and Airbnb properties occupancy rates in 50 states. The empirical models are specified as follows:(1) HotelRoomOCCit=αi+ß1logCOVIDit+∑ßkXit+εit
(2) EconomyClassHotelRoomOCCit=αi+ß2logCOVIDit+∑ßkXit+εit
(3) MidscaleClassHotelRoomOCCit=αi+ß3logCOVIDit+∑ßkXit+εit
(4) UpperMidscaleClassHotelRoomOCCit=αi+ß4logCOVIDit+∑ßkXit+εit
(5) UpscalseClassHotelRoomOCCit=αi+ß5logCOVIDit+∑ßkXit+εit
(6) UpperUpscaleClassHotelRoomOCCit=αi+ß6logCOVIDit+∑ßkXit+εit
(7) LuxuryClassHotelRoomOCCit=αi+ß7logCOVIDit+∑ßkXit+εit
(8) AirbnbPropetiesOCCit=αi+ß8logCOVIDit+∑ßkXit+εit
(9) EntireRoomAirbnbOCCit=αi+ß9logCOVIDit+∑ßkXit+εit
(10) PrivateRoomAirbnbOCCit=αi+ß10logCOVIDit+∑ßkXit+εit
(11) SharedRoomAirbnbOCCit=αi+ß11logCOVIDit+∑ßkXit+εit
The dependent variables are hotel room occupancy (OCC), which consists of all hotels, and economy, midscale, upper midscale, upscale, upper upscale and luxury class occupancy of hotel rooms, and Airbnb properties occupancy, entire home Airbnb occupancy, private room Airbnb occupancy and shared room Airbnb occupancy in state i at time t. The independent variable is the total number of COVID-19 cases (i.e., logCOVID) in state i at time t. X represents a set of control variables of the state i at time t that includes log airport arrivals, log hotel room supply, log Airbnb properties supply, log coincident index (income) and unemployment rate. The variable e is the error term, and ß1−11, ßk are the model parameters. All models include year and month effects to control for seasonality and time-specific economic and other conditions over time, and state-fixed effects to control for state-specific characteristics and dynamics. A logarithmic transformation was utilized to account for data-skewness in the study variables, with the exception of occupancy rates and unemployment rates. The central focus of the empirical specifications detailed above is the coefficient of logCOVID (ß1−11).
The panel data fixed effect regression model was employed to estimate the effect of COVID-19 on hotel and Airbnb. While the fixed effect approach would be recommended from a theoretical perspective due to the nature of data and analysis, a Hausman test to examine whether fixed effect or random effect models was conducted, and more appropriate from an empirical methodology perspective. The results from the Hausman test suggests that the panel data fixed effect regression model is a better fit for the analysis (Chi2: 223.88 Probability> Chi2 =0.00). Therefore, the panel data fixed effect regression model was utilized throughout the analyses. The errors were further clustered based on the states to account for potential autocorrelation and heteroskedasticity problems that may exist in the model estimates. Thus, empirical estimations yielded robust estimates based on robust standard errors (Hoechle, 2007).
4 Empirical results
4.1 The effect of COVID-19 pandemic on hotels
This section presents the results from the investigation of the effect of COVID-19 pandemic on the US hotel industry. The analyses are carried out utilizing the panel data fixed effect regression analysis. Table 2 presents the results from the analysis of the effects of COVID-19 pandemic on all hotels and for each hotel class separately (i.e., economy, midscale, upper midscale, upscale, upper upscale, and luxury) controlling for macroeconomic factors, such as income and hotel industry specific factors, such as hotel room supply.Table 2 The Effect of COVID-19 on Hotel Occupancy.
Table 2 (1) (2) (3) (4) (5) (6) (7)
Study Variables All Hotels Economy Class Midscale Class Upper Midscale Class Upscale Class Upper Upscale Class Luxury Class
Log Total COVID-19 Cases -0.007a
(−6.34) -0.004c
(−1.85) 0.001
(0.37) -0.005a
(−5.08) -0.010a
(−7.97) -0.018a
(−10.60) -0.017a
(−6.19)
Log Airport Arrivals 0.093a
(9.28) 0.058a
(6.78) 0.089a
(4.68) 0.113a
(10.20) 0.125a
(11.09) 0.164a
(7.86) 0.171a
(4.88)
Log Hotel Supply 0.460a
(6.43) 0.166b
(2.05) 0.415a
(3.46) 0.336a
(6.31) 0.279a
(5.01) 0.366a
(5.65) 0.484a
(5.31)
Log Airbnb Supply -0.025b
(−2.59) -0.039b
(−2.65) 0.006
(0.19) -0.035a
(−3.84) -0.029a
(−2.99) -0.023b
(−2.20) -0.060a
(−3.91)
Log Coincident Index (Income) -0.104
(−1.36) -0.001
(−0.01) 0.017
(0.21) -0.087
(−1.25) -0.223b
(−2.56) -0.284b
(−2.33) -0.482
(−1.68)
Unemployment -0.055
(−0.19) 0.184
(0.57) 0.067
(0.19) -0.141
(−0.50) -0.439
(−1.35) 0.298
(0.42) 0.391
(0.41)
Constant -6.646a
(−6.74) -2.364
(−1.67) -6.870a
(−4.35) -5.051a
(−6.45) -3.726a
(−4.51) -5.304a
(−4.56) -6.001a
(−3.29)
R-Squared
Within groups 0.83 0.57 0.64 0.85 0.86 0.85 0.78
Between groups 0.28 0.13 0.21 0.20 0.19 0.19 0.17
Overall 0.14 0.20 0.10 0.14 0.16 0.20 0.11
F-Test 309.21a 224.85a 333.62a 417.73a 390.94a 173.44a 146.26a
Number of Observations 3750 3750 3750 3750 3750 3750 1710
Number of Groups (States) 50 50 50 50 50 50 39
Notes: Occupancy rate is the dependent variable. Robust t-statistics are in parentheses. a, b, and c denote 1%, 5% and 10% statistical significance levels, respectively.
The results show that hotel room occupancy is adversely affected by the COVID-19 pandemic. Specifically, a 1% increase in total number of COVID-19 cases decreases hotel occupancy rate by almost 0.01%. This result is both economically and statistically significant (β=−0.007,p<0.01). While the COVID-19 pandemic is expected to have an adverse impact on the hotel industry, the magnitude of this impact was not clear. Furthermore, the effect of the COVID-19 pandemic may not be uniform across demand associated with different hotel categories. Therefore, a further analysis was conducted on the effect of the COVID-19 pandemic including all hotel class categories.
The results showed that demand for all hotel classes was adversely affected by the COVID-19 pandemic at varying degrees with the exception of the midscale class hotels. The effect of the COVID-19 pandemic on midscale class hotels was not statistically significant at conventional statistical significance levels. The magnitude of the effect of the COVID-19 pandemic was larger in higher class hotels. Specifically, a 1% increase in total number of COVID-19 cases decreases hotel occupancy rate by between 0.01% and 0.02% in upscale class hotels (β=−0.01,p<0.01), upper upscale class hotels (β=−0.018,p<0.01) and luxury class hotels (β=−0.017,p<0.01). The results further show that economy class hotels and upper midscale class hotels showed higher resilience against the COVID-19 pandemic. That is, a 1% increase in total number of COVID-19 cases decreases hotel occupancy rate by about 0.005% in economy class hotels (β=−0.004,p<0.10) and upper midscale class hotels (β=−0.005,p<0.01). These outcomes collectively suggest that the COVID-19 pandemic has had a significantly negative effect on demand in the hotel sector, in general. However, the magnitude of the effect is not uniform across hotel class categories and that the midscale class hotels are the most resilient against the COVID-19 pandemic, followed by economy scale and upper midscale hotels, while the highest impact was observed in the upper upscale and luxury class hotels. The results further showed that the differences in magnitude were also statistically significant for all hotel classes except for upscale hotels compared to all hotels, as the COVID-19 pandemic has had a similar effect on upscale hotels compared to all hotels. The significance difference tests of the coefficients are presented as follows: all hotels vs. economy hotels (t = 2.54, p < 0.05); all hotels vs. midscale hotels (t = 2.75, p < 0.01); all hotels vs. upper midscale hotels (t = 1.90, p < 0.10); all hotels vs. upscale hotels (t = 1.62, n.s.); all hotels vs. upper upscale hotels (t = 7.28, p < 0.01); and all hotels vs. luxury hotels (t = 4.15, p < 0.01).
4.2 The effect of the COVID-19 pandemic on Airbnb
Although results of the analyses showed that the COVID-19 pandemic had an adverse impact on the hotel industry’s demand, and that the magnitude of this impact varies across hotel class categories, the accommodation sector is not limited to hotels only. Peer-to-peer accommodations listed through the company Airbnb have become a major player in the lodging industry, and the effect of the COVID-19 pandemic on booking demand for properties listed by Airbnb might not be similar to the that of the hotel industry. Therefore, we further analyzed the effects of COVID-19 pandemic on Airbnb property demand utilizing a panel data fixed-effect regression analysis.
Table 3 presents the results from the analysis of the effects of COVID-19 pandemic on both the demand for Airbnb properties, overall, and for each Airbnb property type separately (i.e., entire home, private room, and shared room) controlling for macroeconomic factors, such as income and hotel industry specific factors, such as Airbnb properties supply. Results from the analyses show that when the effect of the COVID-19 pandemic on demand for all Airbnb properties was examined, the COVID-19 pandemic did not have a statistically significant impact. Results are presented in Column 1 of Table 3. This outcome is surprising when it is compared with that of hotel occupancy. However, the effect of the COVID-19 pandemic does not seem to be uniform across Airbnb property types. While the demand for entire home Airbnb properties does not seem to be adversely affected by the COVID-19 pandemic, the demand for private room and shared room Airbnb properties is negative. Specifically, a 1% increase in total number of COVID-19 cases decreases private room and shared room Airbnb occupancy rates by 0.011% (p < 0.01) and 0.012% (p < 0.01), respectively. The results further showed that the differences in magnitude were also statistically significant for all Airbnb properties except for entire room Airbnb properties compared to Airbnb properties, as the COVID-19 pandemic did not have any adverse impact on all Airbnb properties in general and entire room Airbnb properties in particular. The significance difference tests of the coefficients are presented as follows: all Airbnb vs. entire room Airbnb (t = 1.34, n.s.); all Airbnb vs. private room Airbnb (t = 7.07, p < 0.01); and all Airbnb vs. shared room Airbnb (t = 7.77, p < 0.01).Table 3 The Effect of COVID-19 on Airbnb Occupancy.
Table 3 (1) (2) (3) (4)
Study Variables All Airbnb Entire Home Airbnb Private Room Airbnb Shared Room Airbnb
Log Total COVID-19 Cases 0.001
(0.02) 0.002
(1.43) -0.011a
(−8.68) -0.012a
(−6.60)
Log Airport Arrivals 0.045a
(4.12) 0.047a
(4.06) 0.026a
(2.84) 0.016
(1.51)
Log Hotel Supply 0.441a
(8.46) 0.422a
(7.16) 0.476a
(10.69) 0.330a
(4.64)
Log Airbnb Supply -0.062a
(−3.45) -0.069a
(−3.59) -0.039a
(−2.74) -0.024
(−1.45)
Log Coincident Index (Income) -0.134
(−0.99) -0.159
(−1.15) -0.008
(−0.07) -0.066
(−0.40)
Unemployment 0.586
(1.50) 0.505
(1.23) 0.935a
(2.80) 0.660
(1.60)
Constant -5.670a
(−6.27) -5.228a
(−5.34) -6.807a
(−8.36) -4.415a
(−4.21)
R-Squared
Within groups 0.74 0.72 0.74 0.29
Between groups 0.21 0.21 0.17 0.16
Overall 0.14 0.14 0.12 0.08
F-Test 177.02a 168.14a 173.54a 34.58a
Number of Observations 3750 3750 3750 3750
Number of Groups (States) 50 50 50 50
Notes: Occupancy rate is the dependent variable. Robust t-statistics are in parentheses. a, b, and c denote 1%, 5% and 10% statistical significance levels, respectively.
These results indicate that guests continued to book stays in entire home properties listed on Airbnb during the pandemic, while private rooms and shared rooms experienced significant decline in bookings from guests. It stands to reason that guests might feel less comfortable staying in a private room or shared room during the pandemic due to an increased likelihood of infection during their stay in properties with other people, whether it be hosts or other travelers. Therefore, guests appear to have preferred entire home Airbnb properties during pandemic to avoid social contact and reduce the risk of infection.
The analyses of the effect of COVID-19 pandemic on hotels and properties listed on Airbnb showed that the COVID-19 pandemic does not have equally adverse effects on the accommodation sector. Certainly, midscale class hotels are more resilient than the other classes of hotels, operating throughout the pandemic with little or no change to their demand. Likewise, entire home Airbnb property bookings were not affected by the pandemic. Moreover, comparisons of the effects the COVID-19 pandemic had on demand between hotels and properties listed on Airbnb revealed that peer-to-peer accommodations are more resilient to the pandemic, in general. In sum, while the effect of the COVID-19 pandemic is not uniform across the demand for different accommodation sectors, there was little or no impact overall to Airbnb properties. When the effect of the COVID-19 pandemic was examined across different property types, it appears that entire home Airbnb properties and midscale class hotels are the most resilient against the COVID-19 pandemic. Luxury class and upper upscale class hotels, followed by the private room and shared room Airbnb properties, have experienced the largest impact of COVID-19 pandemic in terms of magnitude.
4.3 The effect of COVID-19 pandemic on Hotels vs. Airbnb: the role of Restrictions
Main analyses showed that the hotel room occupancy is adversely affected by the COVID-19 pandemic, while the effect of COVID-19 pandemic on Airbnb properties occupancy was not statistically significant. However, these analyses do not take into consideration the policy responses of the government authorities across the states regarding the COVID-19 pandemic. That is, during the pandemic, state authorities have reacted differently to COVID-19 pandemic. Some states have taken strict measures, such as stay-at-home orders, curfews, business closures, and travel restrictions to cope with spread of the COVID-19 infections, while some states did not impose any restrictions in this context. Accordingly, the effect of the COVID-19 pandemic is likely to be more prevalent in states with higher restrictions than in states with little or no restrictions in place. Therefore, we further examined the effect of COVID-19 pandemic on hotel and Airbnb occupancies based on the degree of restrictions in 50 US states. Specifically, we categorized each state into more restrictive and less restrictive state groups based on their degree of regulations as measured by the WalletHub index of “States with the Fewest COVID-19 Restrictions”. Table 4 presents these results.Table 4 The Effect of COVID-19 on Hotels and Airbnb Occupancy: The Role of State Restrictions.
Table 4 Hotels Airbnb
(1) (2) (3) (4)
Study Variables More Restrictive States Less Restrictive States More Restrictive States Less Restrictive States
Log Total COVID-19 Cases -0.011a
(−6.97) -0.005a
(−3.13) -0.003
(−1.38) -0.001
(−0.13)
Log Airport Arrivals 0.083a
(6.03) 0.095a
(6.02) 0.031c
(1.96) 0.046a
(3.74)
Log Hotel Supply 0.443a
(5.57) 0.455a
(3.21) 0.428a
(7.47) 0.439a
(4.58)
Log Airbnb Supply -0.040a
(−3.06) -0.028c
(−1.89) -0.102a
(−4.17) -0.084a
(−3.39)
Log Coincident Index (Income) -0.042
(−0.27) -0.171
(−1.39) -0.027
(−0.12) -0.252c
(−2.01)
Unemployment 0.147
(0.28) -0.283
(−0.71) 0.827
(1.12) 0.388
(0.97)
Constant -6.503a
(−5.13) -6.136a
(−3.45) -5.400a
(−4.42) -4.885a
(−3.44)
R-Squared
Within groups 0.87 0.80 0.74 0.80
Between groups 0.10 0.42 0.09 0.27
Overall 0.12 0.15 0.11 0.16
F-Test 1083.55a 694.42a 423.96a 484.18a
Number of Observations 1875 1875 1875 1875
Number of Groups (States) 25 25 25 25
Notes: Occupancy rate is the dependent variable. Robust t-statistics are in parentheses. a, b, and c denote 1%, 5% and 10% statistical significance levels, respectively.
The results from Columns 1 and 2 of Table 4 show that the COVID-19 pandemic has had a significantly adverse impact on hotel occupancies in states with both more restrictive and less restrictive regulations. Further, the magnitude of the results demonstrate that the COVID-19 pandemic has much higher impact on hotel occupancies in more restrictive states (β=−0.011,p<0.01) than less restrictive states (β=−0.005,p<0.01). The difference in the magnitude of the impact is statistically significant (t=4.24,p<0.01). These findings suggest that pandemic related restrictions have had a significantly higher adverse impact on the hotel industry.
The results from the examination of the effect of COVID-19 pandemic on Airbnb occupancy yields results similar to that of the main findings. That is, the results from Columns 3 and 4 of Table 4 show that the COVID-19 pandemic does not have any adverse effects on Airbnb property occupancies in more restrictive or less restrictive states.
These results suggest that Airbnb properties continued to operate regularly regardless of the restrictions imposed to contain the COVID-19 pandemic by more restrictive states. These findings may indicate that guests preferred to stay in Airbnb properties while traveling during COVID-19 pandemic. Alternatively, privately owned Airbnb properties may not have been held to the same restrictions imposed to commercially operated hotel business. Regardless, peer-to-peer accommodations, at least those listed on Airbnb, appear to be more resilient to the COVID-19 pandemic in both more restrictive and less restrictive states.
We further investigated the effect of the COVID-19 pandemic on hotels and Airbnb properties using alternative measures to test the robustness of our initial findings. More specifically, we analyzed one-month, two-month, and three-month lagged effects of the COVID-19 pandemic on the accommodation sector, because the effect of the COVID-19 pandemic might last for few months. These results collectively confirmed the initial findings that Airbnb properties are more resilient against the COVID-19 pandemic than hotels. Due space limitation, we present these findings in detail in the Appendix.
5 Discussion and conclusion
The COVID-19 pandemic has had major adverse impacts on the overall economy, and many, if not all, industries and businesses have experienced significant challenges resulting from social distancing and public health regulations. While the COVID-19 pandemic continues to be a challenge around the world, the adverse effect of the COVID-19 pandemic on the lodging industry has largely been based on anecdotal evidence. That is, the effect of the COVID-19 pandemic on the lodging industry has not yet been quantified. Furthermore, although the COVID-19 pandemic has had an impact on the overall economy and the lodging industry, its effect might not be uniform across various segments (e.g., economy-scale, midscale, luxury-scale). That is, some hotel segments could be more resilient than others to the external shock.
In this study, the effects of the COVID-19 pandemic on all hotels and for each hotel class separately (i.e., economy, midscale, upper midscale, upscale, upper upscale, and luxury) were analyzed. Specifically, the COVID-19 pandemic was specified as a variable intervention in time against hotel room occupancy, which allows for investigation of the before and after effect of the COVID-19 pandemic on hotel room occupancy in 50 states. The results show that the COVID-19 pandemic has had a significantly negative effect on the hotel sector in general. However, the magnitude of the effect is not uniform across hotel class categories, and the midscale class hotels are the most resilient against the COVID-19 pandemic, followed by economy scale and upper midscale hotels, while the highest impact was observed in the upper upscale and luxury class hotels. The resilience of the midscale hotel can be attributed to the pricing points of these properties compared to upper scale or luxury hotel properties (Ozdemir et al., 2021). During the pandemic, some people continued to travel; however, these travels were mostly business related and hence midscale hotels were preferred. Also, some midscale hotels were utilized by first responders and that provided a resilience in occupancy in this category of hotels. Furthermore, the quality of amenities compared to that of economy scale hotels may offer a better value for travelers and hence may have been preferred during the pandemic. Maintaining the cost of operation with low occupancy level can still be manageable in midscale and economy scale hotels because of the low level of employment needed compared to that of upper scale and luxury scale hotel properties (Dogru et al., 2020). Therefore, midscale hotels showed higher resilience followed by economy scale hotels, while upscale and luxury scale hotels are the most affected by the pandemic.
In addition to alternative hotel segments, the lodging sector consists of peer-to-peer accommodations property listing platforms, such as Airbnb. Therefore, we further examined the effects of COVID-19 pandemic on all Airbnb properties and for each Airbnb property type separately (i.e., entire home, private room, and shared room). The results showed that the COVID-19 pandemic did not have a statistically significant impact on Airbnb properties, in general. However, the effect of the COVID-19 pandemic is not uniform across Airbnb property types. While entire home Airbnb property bookings were not adversely affected by the COVID-19 pandemic, there were significant declines of private room and shared room Airbnb property bookings. These outcomes suggest that travelers do not feel comfortable staying in a private room or shared room Airbnb properties due to likely risk of infection and interaction with other guests, however Airbnb guests continued to book in entire home Airbnb properties during the pandemic which facilitated social distancing.
Moreover, while some states pursued a relatively strict approach to public health regulations in an effort to combat the spread of the COVID-19 virus, including stay-at-home orders, curfews, business closures, and travel restrictions, other states imposed little or no restrictions (McCann, 2021). Therefore, we investigated the effect of the COVID-19 pandemic on hotel and Airbnb property occupancies based on the degree of restrictions by state.
The results showed that the COVID-19 pandemic had a significantly adverse impact on hotel occupancies in both more restrictive and less restrictive states. However, the magnitude of these effects was not uniform. Specifically, the COVID-19 pandemic had a much higher impact on hotel occupancies in more restrictive states than less restrictive states, which suggests that pandemic related restrictions further intensified the adverse impacts of the COVID-19 pandemic on the hotel industry’s demand. However, the COVID-19 pandemic did not have any adverse effects on Airbnb property occupancies, regardless of the degree of restrictions across states. These results suggest that peer-to-peer accommodation properties listed on Airbnb seem immune to the state-level COVID-19 policy and restrictions either because the pandemic-related restrictions did not apply to privately-owned Airbnb properties, or because travelers preferred staying at Airbnb properties for social-distancing reasons.
5.1 Research implications
This study makes several key contributions to the limited empirical literature on the effects of the COVID-19 pandemic on the lodging industry. First, much of the extant studies in this context are descriptive in nature, and the impact of the COVID-19 pandemic on the lodging industry has largely been based on anecdotal evidence (Lardieri, 2021). In a recent study, Ozdemir et al. (2021) showed that US hotels have experienced significant drops in demand during the early stages of the pandemic; however, the extent to which COVID-19 pandemic affects the lodging industry was not empirically investigated. In the present study, we examined and quantified the effect of the COVID-19 pandemic on the lodging industry across states. Second, while the COVID-19 pandemic might appear to have an adverse effect on the entire industry, the hotel industry is quite complex, with six distinct categories ranging from economy to luxury. As such, the COVID-19 pandemic might not have a uniform effect on these various hotel classes. However, the extent to which the COVID-19 pandemic affects the occupancy levels of various hotel classes has not been previously investigated. Thus, the effect of the COVID-19 pandemic on the economy, midscale, upper midscale, upscale, upper upscale, and luxury scale hotel classes was analyzed.
Next, the lodging industry comprises both traditional hotels and peer-to-peer accommodation properties listed for booking by companies such as Airbnb. It is possible that Airbnb property bookings were not equally affected by the COVID-19 pandemic. In a recent study, Dolnicar and Zare (2020) posed the question of whether COVID-19 has disrupted the disruptor (Airbnb). Following this, the extent to which the COVID-19 pandemic affected Airbnb property bookings, including all, entire home, private rooms, and shared room properties in the US was investigated. In so doing, this key question posed by Dolnicar and Zare (2020) was answered, in that results showed that Airbnb demand was not adversely impacted by the COVID-19 pandemic.
Finally, the COVID-19 pandemic is a major external shock to which policymakers in US states have responded differently in terms of regulations. Some states have ordered strict limitations on business openings, number of patrons allowed, hours of operation, large and even small gatherings, and social distancing (Multistate, 2021), while other states imposed almost no rules at all (Hunnicutt & Mason, 2021). This study examined the role of those different policy responses on the effect of the COVID-19 pandemic on the lodging industry. In so doing, analyses showed that while an external shock, such as the COVID-19 pandemic, might have adverse impacts on the lodging industry, policies and strategies implemented to deal with such external shocks have significant impacts on the economic actors of the society.
Accordingly, the present study contributes to the limited formative literature by analyzing and quantifying the impact of the COVID-19 pandemic on the lodging industry. Also, the current study further contributes to the extant literature through its temporal, inferential, and geographical scope. That is, the effect of the COVID-19 pandemic US hotel room and Airbnb property bookings was examined using available data during the most comprehensive timeframe. The research serves as contemporary empirical evidence of the implications of the COVID-19 pandemic on the lodging industry in the US. Results further showed a major external shock (i.e., COVID-19 pandemic) that appears to have adverse impacts to the economy, in general, and the lodging sectors, in particular, has an impact at varying degrees. While some sectors of the lodging industry are more resilient (e.g., peer-to-peer accommodation properties listed on Airbnb and midscale hotels), others are more vulnerable (e.g., upper upscale and luxury hotels). Furthermore, we showed that public health policies and regulations developed to mitigate the COVID-19 virus spread throughout the pandemic have had varying effects on its impacts. Therefore, such policy interventions should always be taken into consideration when analyzing the economic impact of external shocks, such as the COVID-19 pandemic.
5.2 Practical implications
The results of this study have significant managerial implications for the lodging industry and strategic implications for destinations and policymakers. First, although the COVID-19 pandemic is a major external shock that has had severe impacts to the overall economy and hospitality and tourism industry, the effect of the COVID-19 pandemic is not uniform across all platforms. Midscale hotels are the most resilient hotels against the COVID-19 pandemic, followed by economy-scale and upper midscale class hotels. Therefore, hotel corporations should increase midscale, economy-scale, and upper midscale class hotels in their portfolios, as they are the most resilient hotels against major external shocks.
Furthermore, the hotel industry has somewhat recognized that peer-to-peer accommodation networks are major competitors (Dogru et al., 2020), and some hotel chains have even incorporated the home sharing model into their businesses. However, advertising privately-owned properties for guests is often limited to individuals, and commercial hotel companies may not be able to operate within the peer-to-peer accommodation market, in this sense. Therefore, to the extent of the limitations, hotel companies should endeavor to innovate commercially owned properties, as travelers have shifted to staying in more private properties, where they do not interact with other guests and large staff.
In addition, hotels can provide stronger value propositions than peer-to-peer accommodation operations by integrating their large-scale knowledge base, professional cleaning, hygiene standards, and technological infrastructure that enables contactless check-in and entry (Jiang and Wen, 2020, Kaushal and Srivastava, 2021). As hotel corporations have started to acquire properties and operate peer-to-peer accommodations, expanding their business model could also increase their resiliency during major external shocks, such as a pandemic, and it may also provide an opportunity to capitalize on the rapidly growing market. Offering private home and residential style property rentals, such as those advertised by individual owners on the peer-to-peer accommodation networks, would be a winning strategy for hotels, because loyal customers of major hotel corporations might not feel the need to shift their booking preferences to the properties listed on Airbnb. While peer-to-peer accommodation and home sharing business models appear to be ideal for consumers who desire to travel during the pandemic while maintaining social distancing with other travelers, incorporating experiential value propositions, such as localness, serendipity, communitas (Mody et al., 2017) and a sense of “feeling at home” (Song et al., 2021) through such properties would further strengthen the hotel businesses.
Furthermore, hotels could change their business models to decrease their reliance on group businesses, such as conferences, events, corporate meetings, and so on (Hao et al., 2020). Similar to Airbnb’s local travel initiative, Go Near (Jelski, 2020), hotels could attract customers from within or near the cities they operate. Consumer perceptions of safety, the ability and willingness to travel, and anxiety about the pandemic have all resulted in significant changes in consumer travel behavior (Suess et. al, 2021). The highest standards of cleanliness, hygiene, and safety should be in the center of hotels’ marketing strategies, coupled with flexible booking and free cancellation options, and promotions that could alleviate consumers’ anxiety and restore confidence about traveling during the pandemic (Shin et al., 2021). Also, the shift at many companies to working at home has given rise to a new group of customers who choose to rent Airbnb properties on a long-term basis so that they can explore new and different cities and countries while working remotely (Hines, 2021). These digital nomads have generated new demand for the lodging sector, which appears to be captured by Airbnb properties (Rodriguez, 2020). Hotels could offer extended stay options to these new groups of health-conscious digital nomads to create a new stream of revenues during the pandemic and thereafter.
Strict restrictions intensify the adverse impacts of the pandemic. Certainly, public health is of paramount importance during a pandemic. Specifically, mandatory business closures, limited hours of operation, curfews, stay-at-home orders, and other similar policies have had direct and significant effects on businesses. By imposing such strict policies, local and federal governments have lost significant sales, income, and other tax revenues that are essential to funding measures related to maintaining public health. While almost all US states implemented strict policy measures, some states, such as Florida and Texas, relaxed these restrictions significantly and allowed businesses to continue to operate as usual in early 2020. The benefits of these lenient policies have appeared in the results of the present study, showing that the impact of the COVID-19 pandemic to the lodging industry was lower in less-restrictive states. Policymakers in more-strictive states should reconsider their approaches to mitigating the effects of the ongoing pandemic by investigating the methods utilized in less-restrictive states to sustain their tax revenues and attract new investments in local communities from future hotel developers.
The results of the study showed that midscale hotels are the most resilient against the demand decreased related to the COVID-19 pandemic. Prior to the pandemic, Dogru et al. (2020) also showed that midscale hotels are the major hotel classes that contributes to the employment in local economies. Considering the impact that the midscale hotels have on local economies and their resilience to pandemic, local authorities should further incentivize midscale hotel development in their communities. Also, the resilience of the Airbnb peer-to-peer accommodation properties might appear to be a benefit for local communities as they bring in additional tax revenues for cities and local governments from visitor spending, especially during the COVID-19 pandemic. In fact, Airbnb properties, in general, was found to be resilient in both more-restrictive and less-restrictive states, suggesting that more restrictive states only imposed restrictions against traditional hotels or that hosts of Airbnb properties were not mandated by the same restrictions as hotels. Regardless, policymakers should consider regulating Airbnb properties in the same manner as the commercial accommodation counterparts to level the playing field. Otherwise, loose regulations and policies on privately operated properties might drive away major hotel investments to states or destinations that treat hotels and sharing economy properties similarly, where both are allowed to operate with little or no restrictions during a pandemic (Dogru et al., 2019). Also, state and federal governments could initiate tax-incentives to encourage travelers to book hotels and/or Airbnb properties during pandemic to accelerate the recovery (Ozdemir et al., 2021, Salem et al., 2021).
5.3 Limitations and recommendations for future research
Although the present study makes a significant contribution to the extant literature, it has some limitations. The current study examines the effect of the COVID-19 pandemic on various sectors of the lodging industry. However, its sample is limited to the hotel categories and Airbnb property types specifically in the US. It is not clear whether the COVID-19 pandemic affected the lodging industry in other parts of the world, similarly. Future research is necessary to examine the extent to which the COVID-19 pandemic affects the lodging industry in other countries. Also, the classification of hotels in the study is based on class categories (STR). However, hotels in each class might vary in their organizational structures, such as franchised or managed, and the implications of the COVID-19 pandemic might be different. Therefore, future studies are necessary to examine these issues and compare the structures. Similarly, Airbnb properties can be compared based on management by single-unit hosts or multi-unit hosts. The examination of the impact of COVID-19 pandemic to Airbnb properties in this structural context will well serve the extant literature. The present study examined the effect of the COVID-19 pandemic on the demand for hotel and peer-to-peer accommodation properties listed on Airbnb. Finally, the impact of the COVID-19 pandemic to other sectors of the hospitality and tourism industry is not clear. Future studies should investigate the implications of the COVID-19 pandemic on restaurants, museums, and other sectors of the hospitality and tourism industry.
Declarations of interest
None.
Appendix 1 Robustness analysis
Our main analyses provide extensive evidence on the effects of the COVID-19 pandemic on hotels and the peer-to-peer accommodation properties listed on Airbnb. However, we further examined the effects using alternative measures of COVID-19 pandemic. While the COVID-19 pandemic might have immediate effect on the accommodation sectors, its effect could be lagged for several months. Therefore, a further examination of the lagged effects of the COVID-19 pandemic on hotels and Airbnb was warranted. Table 5 presents these findings.Table 5 The Effect of Lagged COVID-19 on Hotels and Airbnb Occupancy.
Table 5 Panel A Panel B
Study Variables (1) (2) (3) (4) (5) (6)
L1. Log Total COVID-19 Cases -0.003b
(−2.61) 0.001
(0.93)
L2. Log Total COVID-19 Cases -0.004a
(−4.22) 0.001
(0.02)
L3. Log Total COVID-19 Cases -0.004a
(−4.14) -0.001
(−0.50)
Log Airport Arrivals 0.104a
(10.36) 0.108a
(11.22) 0.110a
(11.49) 0.047a
(4.64) 0.045a
(4.74) 0.045a
(4.80)
Log Hotel Supply 0.442a
(6.27) 0.442a
(6.28) 0.444a
(6.46) 0.440a
(8.81) 0.446a
(8.81) 0.449a
(8.94)
Log Airbnb Supply -0.021b
(−2.16) -0.024b
(−2.46) -0.025b
(−2.47) -0.062a
(−3.46) -0.067a
(−3.68) -0.062a
(−3.45)
Log Coincident Index (Income) -0.104
(−1.43) -0.123
(−1.61) -0.124
(−1.60) -0.135
(−1.02) -0.148
(−1.14) -0.134
(−0.99)
Unemployment -0.047
(−0.16) -0.070
(−0.23) -0.107
(−0.35) 0.556
(1.44) 0.562
(1.47) 0.586
(1.50)
Constant -6.559a
(−6.74) -6.489a
(−6.68) -6.505a
(−6.83) -5.667a
(−6.40) -5.637a
(−6.34) -5.554a
(−6.26)
R-Squared
Within groups 0.83 0.83 0.83 0.74 0.74 0.73
Between groups 0.30 0.30 0.30 0.21 0.20 0.20
Overall 0.15 0.15 0.15 0.14 0.14 0.13
F-Test 356.16a 341.04a 315.52a 168.49a 179.12a 176.33a
Number of Observations 3700 3650 3600 3700 3650 3600
Number of Groups (States) 50 50 50 50 50 50
Notes: Occupancy rate is the dependent variable. Robust t-statistics are in parentheses. a, b, and c denote 1%, 5% and 10% statistical significance levels, respectively.
The analyses yielded results similar to that of the main findings. That is, the results show that the effect the COVID-19 pandemic on hotels continues to be significant in one-month (β=−0.003,p<0.05), two-month (β=−0.004,p<0.01), and three-month (β=−0.004,p<0.01) lagged total number of COVID-19 cases. While the magnitudes of the effect of the COVID-19 pandemic on hotels are slightly lower, the impact of the COVID-19 pandemic persists for at least three months. However, the effect of COVID-19 pandemic on Airbnb properties remains to be insignificant in a lagged examination of the effect of the COVID-19 pandemic. These results confirm our findings that Airbnb properties are more resilient against the COVID-19 pandemic than hotels.
In addition to the examination of the effect of lagged values of COVID-19 cases, we also analyzed the effect of COVID-19 pandemic on hotels and Airbnb properties utilizing alternative measures of COVID-19 pandemic. Specifically, we utilized total number of COVID-19 deaths, new number of COVID-19 cases, and new number of COVID-19 deaths to analyze the effect of COVID-19 pandemic on hotels and Airbnb. Table 6 presents these results.Table 6 The Effect of COVID-19 on Hotels and Airbnb Occupancy: Alternative COVID-19 Measures.
Table 6 Panel A Panel B
Study Variables (1) (2) (3) (4) (5) (6)
Log Total COVID-19 Death -0.008a
(−4.65) -0.001
(−0.16)
Log New COVID-19 Cases -0.008a
(−6.41) -0.001
(−0.29)
Log New COVID-19 Death -0.008a
(−3.71) -0.001
(−0.62)
Log Airport Arrivals 0.096a
(9.50) 0.093a
(9.07) 0.096a
(9.11) 0.044a
(3.92) 0.044a
(4.12) 0.042a
(3.69)
Log Hotel Supply 0.452a
(6.14) 0.465a
(6.46) 0.454a
(6.20) 0.442a
(8.37) 0.443a
(8.47) 0.446a
(8.27)
Log Airbnb Supply -0.024b
(−2.53) -0.024b
(−2.50) -0.020b
(−2.24) -0.062a
(−3.51) -0.063a
(−3.49) -0.063a
(−3.61)
Log Coincident Index (Income) -0.100
(−1.32) -0.095
(−1.26) -0.082
(−1.14) -0.135
(−0.99) -0.135
(−1.00) -0.133
(−0.97)
Unemployment -0.068
(−0.24) -0.048
(−0.17) -0.051
(−0.18) 0.587
(1.50) 0.586
(1.50) 0.591
(1.50)
Constant -6.584a
(−6.41) -6.762a
(−6.84) -6.741a
(−6.67) -5.679a
(−6.24) -5.675a
(−6.26) -5.714a
(−6.26)
R-Squared
Within groups 0.83 0.83 0.83 0.74 0.74 0.74
Between groups 0.29 0.28 0.29 0.21 0.21 0.21
Overall 0.15 0.14 0.15 0.14 0.14 0.14
F-Test 299.94a 320.38a 307.66a 168.49a 178.51a 174.29a
Number of Observations 3750 3750 3750 3750 3750 3750
Number of Groups (States) 50 50 50 50 50 50
Notes: Occupancy rate is the dependent variable. Robust t-statistics are in parentheses. a, b, and c denote 1%, 5% and 10% statistical significance levels, respectively.
The results from the Panel A of Table 6 validates our main findings that the COVID-19 pandemic has an adverse impact on hotel occupancy rates. The magnitudes of the effect are similar to that our initial findings, suggesting that the effect of the COVID-19 pandemic on hotels does not change based on the measure of the COVID-19 pandemic. That is, regardless of the COVID-19 pandemic variable, the effect of the COVID-19 pandemic on hotels remains negative and statistically significant with similar findings in magnitude when the total number of COVID-19 deaths, new number of COVID-19 cases, or new number of COVID-19 death measures (β=−0.008,p<0.01) were utilized to examine the effect of COVID-19 pandemic on hotels.
These results further confirm the initial analysis findings of the effect that the COVID-19 pandemic has on peer-to-peer accommodation properties listed on Airbnb, which continues to be insignificant. That is, COVID-19 pandemic does not appear to have any adverse impacts on Airbnb properties when alternative measures of COVID-19 pandemic are utilized. These results suggest that our results are robust to alternative measures of the COVID-19 pandemic, and collectively validate our findings that Airbnb property bookings did not experience any adverse impacts from the COVID-19 pandemic and appear to be more resilient than hotels, although hotels experienced economically and statistically significant impacts due to the COVID-19 pandemic.
The anonymous reviewer suggested to examine the effect of COVID-19 pandemic on hotels and Airbnb occupancy by adjusting the total COVID-19 cases by respective state populations. These results are presented in Table 7.Table 7 The Effect of COVID-19 on Hotels and Airbnb Occupancy: Alternative Time Periods and Measures.
Table 7 Panel A Panel B
COVID-19 Period Only (Excluding Pre-Pandemic) Alternative COVID-19 Cases (Adjusted by State Population)
(1) (2) (3) (4)
Study Variables All
Hotels All
Hotels All
Hotels All
Airbnb
Log Total COVID-19 Cases -0.009b
(−2.09) 0.001
(0.02) -0.011a
(−6.34) 0.013
(0.55)
Log Airport Arrivals 0.064a
(5.04) 0.040a
(3.17) 0.094a
(9.67) 0.047a
(4.31)
Log Hotel Supply 0.242a
(4.02) 0.238a
(4.15) 0.456a
(6.52) 0.438a
(8.52)
Log Airbnb Supply -0.190a
(−5.01) -0.342a
(−8.38) -0.025b
(−2.60) -0.061a
(−3.39)
Log Coincident Index (Income) -0.078
(−1.07) -0.084
(−0.90) -0.116
(−1.50) -0.131
(−0.98)
Unemployment -0.561b
(−2.30) 0.276
(1.00) -0.084
(−0.29) 0.587
(1.51)
Constant -1.141
(−1.00) 0.934
(0.80) -6.533a
(−6.84) -5.671a
(−6.32)
R-Squared
Within groups 0.86 0.88 0.83 0.74
Between groups 0.21 0.22 0.28 0.21
Overall 0.18 0.04 0.15 0.14
F-Test 302.88a 730.23a 310.59a 168.07a
Number of Observations 600 600 3750 3750
Number of Groups (States) 50 50 50 50
Notes: Occupancy rate is the dependent variable. Robust t-statistics are in parentheses. a, b, and c denote 1%, 5% and 10% statistical significance levels, respectively.
The results from Panel A of Table 7 show that analyzing the pandemic period alone yields results similar to that of initial analysis. That is the effect of the COVID-19 pandemic on hotels and Airbnb occupancy remains similar regardless of the time period. Furthermore, the findings from the analysis presented in Panel B of Table 7 show coefficients similar to that of initial findings, suggesting that the state population does not have any significant impact on the effect of COVID-19 pandemic on hotels and Airbnb.
Furthermore, in a recent study Deyá-Tortella et al. (2022) argued that COVID-19 pandemic has led to significant changes in tourist behavior and hence there might be structural shifts in this context. Considering the nature of this study’s data period containing the COVID-19 pandemic, we have further tested for the existence structural breaks in the data following the procedures in Ditzen et al. (2021). The results showed that the data contains five structural breaks. Therefore, we re-run the analysis on all hotels and all Airbnb samples considering the existence of the structural breaks following the methodology proposed by the study of Ditzen et al. (2021). Table 8 presents these findings.Table 8 The Effect of COVID-19 on Hotels and Airbnb Occupancy: Testing for Structural Breaks.
Table 8 Hotels Airbnb
Study Variables
Log Total COVID-19 Cases -0.013a
(−8.44) 0.001
(0.42)
Log Airport Arrivals 0.003
(0.31) 0.004
(0.62)
Log Hotel Supply 0.053b
(1.99) 0.026
(1.28)
Log Airbnb Supply -0.048a
(−3.36) -0.0369a
(−2.62)
Log Coincident Index (Income) -0.021
(−0.46) 0.237
(0.63)
Unemployment -1.877a
(−4.97) -0.502c
(1.51)
Constant 0.471a
(5.18) 0.105a
(3.62)
R-Squared
Within groups 0.76 0.68
Between groups 0.36 0.43
Overall 0.63 0.55
F-Test 355.24a 256.48a
Number of Observations 3750 3750
Number of Groups (States) 50 50
Notes: Occupancy rate is the dependent variable. Robust t-statistics are in parentheses. a, b, and c denote 1%, 5% and 10% statistical significance levels, respectively.
The results yield estimates similar to that of the results from the main analysis. While the coefficient estimates changed slightly, the changes are not significant. Therefore, our estimates are valid and robust to alternative specifications. Overall, the results from the robustness tests suggest that our results are robust and collectively validate our findings.
Appendix 2 TableCorrelation Matrix
Occupancy (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)
(1) All Hotels .55 .85 .94 .95 .86 .59 .33 .30 .47 .35 -.55 .47 .18 .12 .15
(2) Economy Class 1 .53 .52 .53 .44 .32 .24 .22 .32 .19 -.19 .27 .30 .17 .16
(3) Midscale Class 0.53 1 .81 .79 .68 .44 .31 .29 .41 .28 -.40 .32 .17 .09 .12
(4) Upper Midscale Class .52 .81 1 .94 .84 .49 .22 .20 .34 .22 -.60 .40 .13 .01 ns .18
(5) Upscale Class .53 .79 .94 1 .86 .53 .18 .15 .33 .24 -.67 .39 .09 .01 ns .14
(6) Upper Upscale .44 .68 .84 .86 1 .57 .16 .14 .29 .20 -.65 .46 .19 .14 .12
(7) Luxury Class .32 .44 .49 .53 .57 1 .30 .29 .38 .34 -.30 .59 .39 .44 .13
(8) All Airbnb .24 .31 .22 .18 .16 .30 1 .99 .90 .61 .30 .21 .21 .45 .27
(9) Entire Home Airbnb .22 .29 .20 .15 .14 .29 .99 1 .86 .57 .33 .20 .21 .43 .25
(10) Private Room Airbnb .32 .41 .34 .33 .29 .38 .90 .86 1 .68 .08 .28 .20 .48 .33
(11) Shared Room Airbnb .19 .28 .22 .24 .30 .34 .61 .57 .68 1 .02 ns .34 .22 .42 .20
(12) Log COVID19Cases -.19 -.40 -.60 -.67 -.65 -.30 .30 .33 .08 02 ns 1 -.22 0.04c .16 -.05
(13) LogAirport Arrivals .27 .32 .40 .39 .46 .59 .21 .20 .28 .34 -.22 1 .07 .57 .37
(14) LogHotel Supply .30 .17 .13 .09 .19 .39 .21 .21 .20 .22 0.04c .07 1 .63 .28
(15) LogAirbnb Supply .17 .09 .01 ns .01 ns .14 .44 .45 .43 .48 .42 .16 .57 .63 1 .43
(16) Log Income .16 .12 .18 .14 .12 .13 .27 .25 .33 .20 -.05 .37 .28 .43 1
(17) Un-employment -.42 -.21 -.32 -.52 -.50 -.50 -.20 .04b .06 -.07 -.01 ns .53 -.24 -.02c -.46
Notes: b and c denote 5% and 10% statistical significance levels, respectively. ns indicates not significant. All other correlations results are significant at 1% statistical significance level.
Acknowledgments
We thank the anonymous reviewers for providing valuable feedback and constructive comments.
==== Refs
References
Airbnb, n.d., Select your home type. Airbnb. 〈https://www.airbnb.com/help/article/317/select-your-home-type〉.
Bresciani S. Ferraris A. Santoro G. Premazzi K. Quaglia R. Yahiaoui D. Viglia G. The seven lives of Airbnb. The role of accommodation types Ann. Tour. Res. 88 2021 103170
Carville, O. & Bloomberg, 2021, February 25, Airbnb revenue beats estimates, showing demand amid COVID surge. Fortune. 〈https://fortune.com/2021/02/25/airbnb-revenue-beats-estimates-amid-covid-surge/〉.
Deyá-Tortella B. Leoni V. Ramos V. COVID-led consumption displacement: A longitudinal analysis of hotel booking patterns Int. J. Hosp. Manag. 107 2022 103343
Ditzen, J., Karavias, Y., & Westerlund, J. (2021). Testing and estimating structural breaks in time series and panel data in stata. arXiv preprint arXiv:2110.14550.
Dogru T. Mody M. Suess C. Adding evidence to the debate: Quantifying Airbnb's disruptive impact on ten key hotel markets Tour. Manag. 72 2019 27 38
Dogru T. McGinley S. Kim W.G. The effect of hotel investments on employment in the tourism, leisure and hospitality industries Int. J. Contemp. Hosp. Manag. 2020
Dolnicar S. Zare S. COVID19 and Airbnb–Disrupting the disruptor Ann. Tour. Res. 83 2020 102961
Eggleston, K. & Lee, M., 2021, August 8, Hospitality industry moves into post-pandemic recovery mode. HotelManagement.Net. 〈https://www.hotelmanagement.net/operate/hospitality-industry-moves-post-pandemic-recovery-mode〉.
Gursoy D. Chi C.G. Effects of COVID-19 pandemic on hospitality industry: Review of the current situations and a research agenda J. Hosp. Mark. Manag. 29 2020 5
Hao F. Xiao Q. Chon K. COVID-19 and China’s hotel industry: Impacts, a disaster management framework, and post-pandemic agenda Int. J. Hosp. Manag. 90 2020 102636
Hines, M., 2021, February 9, As work and school can be done from anywhere, Airbnb sees more monthly stays. USA Today. Retreived from 〈https://www.usatoday.com/story/travel/hotels/2021/02/09/airbnb-rentals-monthly-stays-up-covid-travel-meets-virtual-life/4375982001/〉.
Hoechle D. Robust standard errors for panel regressions with cross-sectional dependence stata J. 7 3 2007 281 312
Jelski, C., 2020, Airbnb increases focus on local travel. Travel Weekly. Retrieved from 〈https://www.travelweekly.com/Travel-News/Hotel-News/Airbnb-increases-focus-on-local-travel〉.
Jiang Y. Wen J. Effects of COVID-19 on hotel marketing and management: a perspective article Int. J. Contemp. Hosp. Manag. 2020
Johnson, H., 2021, North American inbound tourism spend declined by 74.1% in 2020. eTurboNews. 〈https://eturbonews.com/3000768/north-american-inbound-tourism-spend-declined-by-74–1-in-2020/〉 Accessed Aug. 23, 2021.
Kaushal V. Srivastava S. Hospitality and tourism industry amid COVID-19 pandemic: Perspectives on challenges and learnings from India Int. J. Hosp. Manag. 92 2021 102707
Lai I.K.W. Wong J.W.C. Comparing crisis management practices in the hotel industry between initial and pandemic stages of COVID-19 Int. J. Contemp. Hosp. Manag. 2020
Lardieri, A. (2021). Coronavirus Pandemic Sets Hotel Industry Back 10 Years. U.S. News and World Report. 〈https://www.usnews.com/news/national-news/articles/2021–01-27/coronavirus-pandemic-sets-hotel-industry-back-10-years-report-finds〉 Accessed Aug. 23, 2021.
McCann, A. (2021, April 6). States with the fewest Coronavirus restrictions. WalletHub. 〈https://wallethub.com/edu/states-coronavirus-restrictions/73818〉.
Mody M.A. Suess C. Lehto X. The accommodation experiencescape: a comparative assessment of hotels and Airbnb Int. J. Contemp. Hosp. Manag. 2017
Ozdemir O. Dogru T. Kizildag M. Mody M. Suess C. Quantifying the economic impact of COVID-19 on the US hotel industry: Examination of hotel segments and operational structures Tour. Manag. Perspect. 39 2021 100864
Rodriguez, S., 2020, December 12, Airbnb nomads are ditching their apartments in pandemic-ridden cities and living wherever they want. 〈https://www.cnbc.com/2020/12/12/airbnb-nomads-ditch-apartments-during-covid.html〉.
Salem I.E. Elbaz A.M. Elkhwesky Z. Ghazi K.M. The COVID-19 pandemic: The mitigating role of government and hotel support of hotel employees in Egypt Tour. Manag. 85 2021 104305
Schulz, B., 2021, August 14, I just didn’t feel like it was responsible: Growing number of travelers cancel trips as COVID-19 cases surge. USA Today. 〈https://www.pointswithacrew.com/hotels-categorize-difference-upscale-upper-upscale/〉.
Shin H. Sharma A. Nicolau J.L. Kang J. The impact of hotel CSR for strategic philanthropy on booking behavior and hotel performance during the COVID-19 pandemic Tour. Manag. 85 2021 104322
Song S. Suess C. Mody M.A. Dogru T. Comparing the influence of substantive and communicative servicescape on healthcare traveler emotions: the moderating effect of accommodation type and interior design style Int. J. Contemp. Hosp. Manag. 2021
STR Chain Scales, 2019, October 29, STR. Retrieved August 19, 2021, from 〈https://str.com/sites/default/files/2019–10/STR-Chain%20-Scales-20191025_0.pdf〉.
Suess C. Maddock J. Dogru T. Mody M. Lee S. Using the Health Belief Model to examine travelers’ willingness to vaccinate and support for vaccination requirements prior to travel Tour. Manag. 88 2021 104405
Suess C. Maddock J.E. Dogru T. Mody M. Lee S. Using the Health Belief Model to examine travelers’ willingness to vaccinate and support for vaccination requirements prior to travel Tour. Manag. 88 2022 104405
Verleye K. Leroi-Werelds S. Line N. Bove L. Value proposition dynamics in response to external event triggers J. Bus. Res. 2021
Voltes-Dorta A. Sánchez-Medina A. Drivers of Airbnb prices according to property/room type, season and location: A regression approach J. Hosp. Tour. Manag. 45 2020 266 275
| 36467109 | PMC9707363 | NO-CC CODE | 2022-12-01 23:20:23 | no | Int J Hosp Manag. 2023 Feb 29; 109:103406 | utf-8 | Int J Hosp Manag | 2,022 | 10.1016/j.ijhm.2022.103406 | oa_other |
==== Front
Eur Neuropsychopharmacol
Eur Neuropsychopharmacol
European Neuropsychopharmacology
0924-977X
1873-7862
Elsevier B.V. and ECNP.
S0924-977X(22)00892-6
10.1016/j.euroneuro.2022.11.003
Insights
How else can inflammation kill you?
Zalsman Gil ab⁎
a Geha Mental Health Center and Psychiatry Department, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
b Division of Molecular Imaging and Neuropathology, Department of Psychiatry, Columbia University, New York, NY, United States
⁎ Corresponding author at: Chair of Psychiatry, Sackler Faculty of Medicine, Tel Aviv University, Director, Geha Mental Health Center, P.O. Box 102, Petach Tiqwa 49100, Israel.
29 11 2022
2 2023
29 11 2022
67 13
13 10 2022
2 11 2022
4 11 2022
© 2022 Elsevier B.V. and ECNP. All rights reserved.
2022
Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID-19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active.
Keywords
C-reactive protein
Covid-19
Depression
Inflammation
Suicide
Abbreviations
CI, central inflammation
CRP, C-reactive protein
CSF, cerebrospinal fluid
GWAS, genome-wide association study
HC, healthy controls
IL-6, interleukin-6
MDD, major depressive disorder
PET, positron emission tomography
SAs, suicide attempts
SB, suicidal behavior
SNP, single nucleotide polymorphism
TNF-α, tumor necrosis factor α
==== Body
pmcAs data accumulates on the psychiatric effects of COVID-19 and the virus's neuroinvasive characteristics, the connection between the immune system, brain inflammation and neuropsychiatric disorders, including suicidal behavior, seems more relevant than ever. Isolation, social distancing and loneliness that accompany quarantines, may be bad for your brain (Zalsman, 2020).
Inflammation is involved in the pathophysiology of psychiatric disorders irrespective of diagnosis; however, a disorder-specific biomarker has not been established yet. Blood levels of C-reactive protein (CRP), was consistently higher in people with major depression (MDD), bipolar, schizophrenia and anxiety disorders, compared with healthy controls (HC) (Courtet et al., 2016). In patients with MDD increased peripheral inflammation has been reported consistently. Still, few studies explored markers of central inflammation (CI) in patients.
Enanche et al. (2019) reviewed systematically in-vivo and postmortem (PM) markers of CI, including examining cerebrospinal fluid (CSF), positron emission tomography (PET), and postmortem brain tissues, in subjects with MDD compared to controls. In 69 studies they found higher CSF levels of interleukin (IL)−6 and tumor necrosis factor (TNF)-α in patients with MDD, compared to controls. IL-6 levels were also increased in people with suicide attempts (SAs) regardless of their psychiatric diagnosis. Translocator protein, a PET marker of CI, was elevated in the anterior cingulate cortex and temporal cortex of patients with MDD compared to controls.
Of the studies investigating expression of cell-specific markers for astrocytes, thirteen reported a decreased expression of astrocyte-specific markers, two reported increased expression of these markers, and eleven did not detect any differences. Four of six studies reported decreased oligodendrocytes markers in the prefrontal cortex. postmortem TNF-α brain-levels were also found increased in MDD., (Enache et al., 2019).
One may wonder: Can inflammation cause more than depression? Can suicide result from inflammatory processes?
Suicide is a purely human behavior with multifactorial etiologies. Many researchers of suicidal behavior (SB) find it hard to believe that inflammation can lead to such complex behavior. Is there enough evidence for this argument?
It is known that CRP levels of patients with depression and suicidal ideation or SAs tend to be higher, irrespective of severity and duration of the SB, indicating that CRP may be a trait marker for SB. In a systematic review and meta-analysis Courtet et al. found that postmortem studies demonstrated associations between suicide and inflammatory cytokines in the orbitofrontal cortex - a brain region involved in suicidal vulnerability. They furthermore found microgliosis and monocyte-macrophage system activation as markers of suicide neurobiology (Courtet et al., 2016).
Nine years ago, we conducted the first DNA microarray study on brain tissue of suicide victims (Galfalvy et al., 2013), searching for new candidate regions in suicide with and without depression, in a gene-centric, low coverage single nucleotide polymorphism (SNP) pilot genome-wide association study (GWAS). We analyzed and genotyped postmortem, 99 Caucasian subjects, 68 who died by suicide and 31 who died from other causes. GWAS analysis yielded 58 SNPs, 22 of them in or near 19 known genes, with risk allele-associated odds ratios between 2.7 and 6.9. Diagnosis of mood disorders did not explain the associations. Gene expression in the prefrontal and the anterior cingulate cortex for these 19 genes was measured on a separate sample who died by suicide and seven of 19 of their genes showed altered expression, compared to controls. Interestingly, most of those genes were immune system related, emphasizing the role of neuroimmunological effects in SB.
Piras et al. (2022) conducted a meta-analysis of PM brain gene expression in 141 suicide subjects and found reduced expression of four markers connected to the immune system and microglia in suicide.
Meta-analyses of the past five years on the association between immune system markers and suicide, revealed more markers. For example, the association between high IL-6, and SB has been confirmed in blood, CSF and postmortem brains. Gananca et al. (2021) found that polyunsaturated fatty acids and cytokines, both inflammatory biomarkers, have been associated with suicide independent of depression severity.
Recently, Vasupanrajit et al. (2022) conducted a meta-analysis that included studies comparing blood biomarkers in psychiatric patients with (+SA) and without SA (-SA) and HC and combined different biomarkers into immune, inflammatory, and neurotoxic profiles. They compared 4945 patients+SA and 24,148 controls. The patients+SA showed significantly increased immune activation, inflammation, as compared with HC.
Miola et al. (2021) conducted a systematic review and random-effect meta-analysis including observational studies comparing peripheral CRP levels in patients with and without suicidality, affected by any psychiatric disorder with CRP in HC. Observational studies (22) with high median quality, involving 7682 subjects were included and showed CRP to be significantly higher in individuals with high suicide ideation or SA than in non-suicidal patients or HC.
In a meta-analysis of nine studies, Gonzalez-Castro et al. (2021) evaluated IL-6 levels in plasma and CSF to determine increase in individuals with SB compared to HC. They found reduced IL-6 levels in plasma of individuals with SB, but increased levels in serum and CSF.
A meta-analysis of seven studies with 2108 participants by Chen et al. (2020) explored differences in peripheral blood CRP between people with depression and SB (PD+SB) and people with depression and no SB (PD-SB) and between the former and HC. They found that concentrations of peripheral CRP may be significantly increased for PD+SB, compared with both, PD-SB and HC.
Data suggests an increase in IL-6 and TNF-α levels, in CSF and brain parenchyma, possibly associated with increased microglia activity and reduction in astrocyte and oligodendrocyte markers in MDD, which may lead to compromised blood brain barrier integrity with increased monocyte infiltration.
In summary, inflammatory markers are found to be associated with suicidality and this effect may serve as a marker or even have an etiological role. Some data show that patients with immune activation are at higher risk for SAs, possibly due to increased neurotoxicity resulting from inflammation and nitro-oxidative stress. Of all markers, CRP was strongly associated with higher suicidality. Larger studies including longitudinal monitoring of all those markers, are needed and may assist in our effort to prevent suicides.
Funding
No funding.
Declaration of Competing Interest
The author declares no conflict of interests.
==== Refs
References
Chen X. Pu J. Liu Y. Tian L. Chen Y. Gui S. Xu S. Song X. Xie P. Increased C-reactive protein concentrations were associated with suicidal behavior in patients with depressive disorders: a meta-analysis Psychiatry Res. 292 2020 113320
Courtet P. Giner L. Seneque M. Guillaume S. Olie E. Ducasse D. Neuroinflammation in suicide: toward a comprehensive model World J. Biol. Psychiatry 17 2016 564 586 26223957
Enache D. Pariante C.M. Mondelli V. Markers of central inflammation in major depressive disorder: a systematic review and meta-analysis of studies examining cerebrospinal fluid, positron emission tomography and post-mortem brain tissue Brain Behav. Immun. 81 2019 24 40 31195092
Galfalvy H. Zalsman G. Huang Y.Y. Murphy L. Rosoklija G. Dwork A.J. Haghighi F. Arango V. Mann J.J. A pilot genome wide association and gene expression array study of suicide with and without major depression World J. Biol. Psychiatry 14 2013 574 582 22059935
Gananca L. Galfalvy H.C. Cisneros-Trujillo S. Basseda Z. Cooper T.B. Ren X. Figueira M.L. Oquendo M.A. Mann J.J. Sublette M.E. Relationships between inflammatory markers and suicide risk status in major depression J. Psychiatr. Res. 134 2021 192 199 33388702
Gonzalez-Castro T.B. Tovilla-Zarate C.A. Lopez-Narvaez M.L. Genis-Mendoza A.D. Juarez-Rojop I.E. Interleukin-6 Levels in Serum, Plasma, and Cerebral Spinal Fluid in Individuals with Suicide Behavior: systematic Review and Meta-Analysis with Meta-Regression J. Interferon Cytokine Res. 41 2021 258 267 34280025
Miola A. Dal Porto V. Tadmor T. Croatto G. Scocco P. Manchia M. Carvalho A.F. Maes M. Vieta E. Sambataro F. Solmi M. Increased C-reactive protein concentration and suicidal behavior in people with psychiatric disorders: a systematic review and meta-analysis Acta Psychiatr. Scand. 144 2021 537 552 34292580
Piras I.S. Huentelman M.J. Pinna F. Paribello P. Solmi M. Murru A. Carpiniello B. Manchia M. Zai C.C. A review and meta-analysis of gene expression profiles in suicide Eur. Neuropsychopharmacol. 56 2022 39 49 34923210
Vasupanrajit A. Jirakran K. Tunvirachaisakul C. Solmi M. Maes M. Inflammation and nitro-oxidative stress in current suicidal attempts and current suicidal ideation: a systematic review and meta-analysis Mol. Psychiatry 2022
Zalsman G. Neurobiology of suicide in times of social isolation and loneliness Eur. Neuropsychopharmacol. 40 2020 1 3 33161991
| 36459952 | PMC9707409 | NO-CC CODE | 2022-12-01 23:20:23 | no | Eur Neuropsychopharmacol. 2023 Feb 29; 67:1-3 | utf-8 | Eur Neuropsychopharmacol | 2,022 | 10.1016/j.euroneuro.2022.11.003 | oa_other |
==== Front
Manag Rev Q
Management Review Quarterly
2198-1620
2198-1639
Springer International Publishing Cham
308
10.1007/s11301-022-00308-2
Article
Synthesizing research in entrepreneurial bootstrapping and bricolage: a bibliometric mapping and TCCM analysis
http://orcid.org/0000-0001-7517-2892
Singh Mansi [email protected]
1
http://orcid.org/0000-0001-6754-9602
Dhir Sanjay [email protected]
1
Mishra Harsh [email protected]
2
1 grid.417967.a 0000 0004 0558 8755 Present Address: Department of Management Studies, Indian Institute of Technology, Delhi, New Delhi, India
2 Present Address: Quantic School of Business and Technology, Washington, DC USA
29 11 2022
134
22 7 2022
8 11 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022, Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
The concepts of entrepreneurial bootstrapping and bricolage study identical problems; however, there exists a lack of integration between the two. Entrepreneurial bootstrapping and bricolage are resourcefulness behaviors that could potentially be critical dynamic capabilities that help firms accumulate, reconfigure and recombine resources to adapt to the dynamic and resource-constrained environment. This paper aims to focus on the missing integration of the two behaviors by investigating their convergence systematically using published literature. The study involves a bibliometric co-citation, bibliographic coupling, network, and Theory, Context, Characteristics, Methodology analysis to understand the existing literature more methodically to get a comprehensive picture of the behaviors. The results have provided more clarity on the conceptual convergence of these behaviors by analyzing their intellectual structures and research trends. The two entrepreneurial behaviors exhibit a mutual contribution to each other's literature, suggesting scholars to explore a probable integration of the two pieces of literature in future studies.
Keywords
Bootstrapping
Bricolage
Bibliometrics
Entrepreneurship
Co-citation
Bibliographic coupling
JEL Classification
L26 Entrepreneurship
==== Body
pmcIntroduction
Innumerable entrepreneurship and strategy scholars have discussed the precisely planned path entrepreneurs take to reach well-defined business goals (Shane and Venkataraman 2000). In reality, entrepreneurs’ scarcity of resources makes it difficult to follow these paths (Baker and Nelson 2005; Sarasvathy 2001). Despite the dynamic role of small firms in the economy (Audretsch 2002), most startups find it difficult to access or procure the resources required to gain a competitive advantage. However, scholars also note that such resource deficiencies often spur creativity in the enterprise and force the entrepreneur and their team to innovate to find solutions to resource constraints (An et al. 2018; Baker and Nelson 2005). Startups exist in environments characterized by not just resource constraints but also uncertainty and risk (Nelson and Lima 2020). One critical entrepreneurial skill is operating and surviving in these conditions by employing resourcefulness behaviors like entrepreneurial bootstrapping and bricolage. However, more scholarly research in entrepreneurship is needed to study these entrepreneurial behaviors adequately. For years, scholars have emphasized the significance of resources for a new venture (Hoang and Antoncic 2003) and acknowledged that access and arrangement of these resources are critical to the successful establishment of a new firm (Ebben and Johnson 2006). Therefore, it is crucial to study resource mobilization behaviors that help entrepreneurs discover and exploit business opportunities.
The two resourcefulness behaviors considered for this study are entrepreneurial bootstrapping and bricolage. Bootstrapping involves accumulating and acquiring business resources without using formal and traditional sources of finance (Rutherford et al. 2012). Bricolage refers to the extemporized approach of making do with whatever resources entrepreneurs have in hand (Baker et al. 2003). Section 2 explains the two concepts in detail. The concepts of entrepreneurial bootstrapping and bricolage study identical problems; however, their integration in literature seems missing or incomplete. Both entrepreneurial behaviors explain resourcefulness behaviors that are often informal and involve using resources in hand to overcome resource scarcity. While scholars compare bricolage with other related resource processes like effectuation, causation, and improvisation, its relationship with bootstrapping still needs more scholarly attention. The existing conceptualization of bricolage and bootstrapping presents avenues for research in synergy, as bricolage could be a strategic activity for entrepreneurs choosing to bootstrap, as both behaviors trigger actions that help entrepreneurs mobilize resources in unconventional ways without accessing resources from traditional sources. Even though the two behaviors are not entirely synonymous with each other, the strong connections between the two seem blurred in academic research (Rutherford et al. 2022). This lack of integration between the two kinds of literature could possibly lead to ignoring critical insights into the academic understanding of entrepreneurial resource management behaviors. Until this study, existing academic research has not adequately considered systematically studying the two resourcefulness behaviors, aiming to understand their unique similarities and differences. Therefore, research with a synergy of these entrepreneurial concepts presents a promising research gap.
The current study involves a bibliometric and TCCM analysis to help understand the existing literature more methodically to get a comprehensive picture of the behaviors (Grégoire and Cherchem 2020; Singh and Dhir 2019; Zupic and Čater 2015). This analysis will contribute to ascertaining the rational structures of both areas and understanding their complementarities and differences (Habib and Afzal 2019; Rey-Martí et al. 2016). Further, with the increase in the number of publications in both the literature, growing theoretical bases, and diverse methodologies, this bibliometric approach of studying the two behaviors together assists in understanding the evolution of both individually and in harmony. It also helps overcome the idiosyncrasies that might arise in reviewing a large number of works, given the capabilities and biases of a researcher subjectively reviewing them alone (Kessler 1963; Stefano et al. 2010; Verbeek et al. 2002; Vogel and Güttel 2013; Zupic and Čater 2015).
Overall, this study contributes to enhancing our understanding of the dynamics of resource management by entrepreneurs who operate in uncertain environments using flexible and informal approaches. It enhances our understanding of the missing integration and similarities in the two behaviors that could potentially be critical dynamic capabilities that help firms accumulate, reconfigure and recombine resources to adapt to the business environment. This study identified the intellectual structures and research trends of the two approaches using a systematic combined analysis of the two bodies of literature. The findings highlight that bootstrapping helps bolster the resource base, which is further made more flexible and innovative by the creative reinvention of resources using bricolage. Bricolage could also be a strategic action that is triggered when entrepreneurs decide to bootstrap their ventures strategically. These two behaviors complement each other as bootstrapping helps arrange for the finances, material resources, knowledge, and skills that are then creatively applied by the bricoleurs. Therefore, studying the two approaches together is interesting while considering their unique aspects (accumulation vs. deployment).
Conceptual background
Resource theories in management have emphasized the critical role of resources for venture survival and success (Barney 1991; Penrose 1996). Nevertheless, newer ventures find it difficult to access and avail business resources from conventional sources due to lack of credibility, liability of smallness and newness, information asymmetry, and innumerable risks involved (Freeman et al. 1983; Witt 2004). Some entrepreneurs also choose to voluntarily opt out of raising resources with external help (Winborg 2009). Regardless of the motivation (resource scarcity or strategic choice), resource management of entrepreneurial ventures in dynamic business environments without any explicit external help is a promising area of research. This paper focuses on two such resource mobilization behaviors—bootstrapping and bricolage.
The first area of research is entrepreneurial bootstrapping, which refers to using informal methods to meet resource needs without taking external funds as debt or diluting equity (Winborg and Landström 2001). It involves using creative means to acquire financial and non-financial business resources by spending little or no money. Scholars in this area have described it as innovative resource management which aims to minimize the cost of operations while also avoiding external influence on the venture board by investors (Ebben and Johnson 2006). This helps entrepreneurial ventures overcome their resource limitations while also being in complete control of the firms’ strategic decisions. Even without the explicit use of the term ‘bootstrapping’, informal ways of gathering resources like delaying payments, accelerating receivables, reinvesting cash flows, using personal savings, and exploiting social and professional networks have been used by entrepreneurs (Churchill and Thorne 1989). Bhide (1992) then introduced the term ‘bootstrapping’ to explain this innovative resource management method employed by entrepreneurs.
Academics in entrepreneurial research emphasize the unconventional methods used by startup founders to acquire business capital and resources while ensuring freedom of action for themselves in the functioning of their businesses (Grichnik et al. 2014; Winborg 2009). Entrepreneurial bootstrapping typically takes place in two forms- financial bootstrapping for acquiring financial funding from informal sources and resource bootstrapping for gathering all the business resources (non-financial) without taking any external help (Harrison et al. 2004). Scholars have also highlighted the completive advantages to a bootstrapped venture vis a vis other players as bootstrappers tend to work more efficiently with the limited resource base they own or can access (Harrison et al. 2004; Vanacker et al. 2011).
The second resource mobilization behavior considered for this study is entrepreneurial bricolage, which refers to making do by utilizing the existing resource base of the venture. It involves using already owned or easily and freely available resources for unconventional purposes (Baker and Nelson 2005). The concept of bricolage has its roots in anthropology, where Levi-Strauss (1967) introduced it to describe how the actions of an engineer and a bricoleur differ. It was then adapted in entrepreneurship research to explain how founders combine and recombine resources at hand to create new markets and grow (Baker et al. 2003; Baker and Nelson 2005).
Empirical research in this area suggests that resource reconstruction in the bricolage process helps such founders build a new product or market for their customers instead of finding a place for them in the existing markets and industries. These studies in bricolage have suggested that bricoleurs primarily show three specific characteristics. First, bricoleur entrepreneurs do not wait for their inputs, finances, and resources to be available at a covetable level, thereby having a definite bias towards taking action. Second, bricoleurs prefer to take full advantage of their social networks, resources, and skills. Finally, bricoleur entrepreneurs combine and recombine existing resources to come up with new solutions to their entrepreneurial challenges (Baker and Nelson 2005). According to their study, bricoleurs focus on finding a solution to their obstacles instead of wasting valuable time deliberating about the appropriateness of resources at hand. Resources at hand encompass any resources that are cheaply, freely, or easily accessible and available to an entrepreneur (human capital, raw material, know-how, and others). Table 1 explains the similarities and differences between the two resource-focused entrepreneurial behaviors.Table 1 Similarities and differences between entrepreneurial bricolage and bootstrapping
Similarities Entrepreneurial bricolage and bootstrapping Selected References
Resource dependence Focus on the critical role of resources for operations and survival of the business and generally prevalent in resource-scarce environments; resource-led innovation Baker and Nelson (2005), Baker et al. (2003), Bhide (1992), Ebben (2009), Ebben and Johnson (2006), Fisher (2012), Grichnik et al. (2014), Welter et al. (2016)
Focus on action Focus is on taking action to meet business goals by managing with the existing resources without any external help to take actions to control the uncertain environment
Experimentation Unconventional and innovative ways of experimenting with the resources available; "do-it-yourself" approaches
Networks and communities Critical role of the entrepreneurs' social capital and network to access resources and test products
Dependence on entrepreneur (action-taker) Entrepreneur (human capital, social capital, and behavior patterns) is at the center of resource management decision-making in order to exploit business opportunities
Differences Entrepreneurial bricolage Entrepreneurial bootstrapping Selected references
Focus To bolster the ventures' limited resource base without external funds or resources To redefine the use of resources at hand by opting for alternative practices and combinations of existing resources Archer et al. (2009), Baker and Nelson (2005), Bhide (1992), Fisher (2012), Grichnik et al. (2014), Löfqvist (2017), Vanacker et al. (2011), Welter et al. (2016), Winborg (2009)
Resources and uncertainty Generally, in case critical resources required for business operations are not available to the business Generally, resource scarcity forces entrepreneurs to bootstrap; could also be a voluntary and strategic activity
Expertise and experience of entrepreneurs Generally, both nascent and experienced entrepreneurs engage in bricolage Entrepreneurs with more managerial experience tend to be better at bootstrapping their businesses
Opportunity exploitation and resources Focus is not exactly on exploitation of opportunities but on using resources available at hand Focus is on exploiting opportunities despite no external funding
Another related concept of resource management is effectuation, introduced by Sarasvathy (2001). This behavior highlights that in uncertain business environments, entrepreneurs change their focus from the end goals to the means they control. These could be individual characteristics like the entrepreneur’s professional or personal network, knowledge, and skills, or could be the venture’s means like its physical, organizational, or human resources. The focus here is to take action by exploiting the existing means, thereby allowing goals to form during this process of exploration (March 1991). The idea is to identify and exploit opportunities with the given means. Just like entrepreneurial bricolage and bootstrapping, the entrepreneurs’ attention here is on critical resource exploitation by taking action, given the uncertainties of dynamic and resource-constrained business environments. However, unlike bricolage which both new and serial entrepreneurs can undertake, effectuation is more often practiced by experienced entrepreneurs who resort to means-driven actions. Entrepreneurs who apply effectual behavior focus on affordable loss instead of the returns they expect at the end while also focusing on their relationships and partnerships instead of competition. They focus on exploiting contingencies to reach newer outcomes instead of an already set goal (Fisher 2012).
Most of the prominent theories in entrepreneurship suggest that entrepreneurs, who can recognize and capitalize on opportunities supported by a clear long-term vision and predetermined goals, and well-defined plans and strategies while having access to the best resources and talent, are the ones who are able to succeed in this competitive business environment (Shane and Venkataraman 2000; Wiklund and Shepherd 2005). However, recent studies suggest that ventures can sail through in these dynamic ecosystems without having access to all of the aforementioned advantages. Therefore, it is crucial to study behaviors like bootstrapping and bricolage which contribute to the evolution as well as the development of dynamic capabilities of a firm. Such behaviors increase the flexibility of ventures while also honing their abilities to look for alternative resource uses and use recombination of accessible resources to create value for the business.
Method
Bibliometric analysis is a systematic statistical method for analyzing academic research publications (Donthu et al. 2021; Zupic and Čater 2015). It highlights frontiers within research areas to understand their overview, gaps, trends, networks, knowledge producers, and themes (Donthu et al. 2021; Rey-Martí et al. 2016; Zupic and Čater 2015). Compared to traditional literature reviews, bibliometric analysis enables analyzing large quantities of academic and scholarly research data, which is easily accessible due to new technology tools (Hood and Wilson 2001; Merigó et al. 2015). The study also involves a systematic review of bootstrapping and bricolage literature following the TCCM approach (Paul and Rosado-Serrano 2019; Singh and Dhir 2019). This simple but comprehensive framework involves a detailed review of literature by focussing on theory development (T), context (C), characteristics (C), and methodology (M) employed. Citation analysis of bibliometrics helped identify the most cited works in both the literature, which was followed by their content analysis as suggested by Paul and Rosado-Serrano (2019), taking guidance from Keupp and Gassmann (2009), Nicholls-Nixon et al. (2011) and Terjesen et al. (2016). This analysis helped identify the gaps in the existing literature to propose new directions for future research.
Data collection and sampling
The data was collected from the Scopus database, covering a broad range of literature from peer-reviewed journals (“Appendix 1”). For searching relevant literature, the keywords used were “bricolage*”, and “bootstrap*”. The keywords “entrepreneur*”, “startup”, “start-up”, “founder” were also used while searching to ensure filtering out papers that were not directly related to the entrepreneurship literature. The literature was collected for a 30-year period from 1991 to mid-2021 from the Business and Management areas. The search also examined other related areas like psychology and sociology. The first refined search resulted in 565 documents. To reduce the bias in selecting the final studies for this analysis, refining for done to the final sample (Zupic and Čater 2015) to include studies related to entrepreneurial bootstrapping and bricolage, which are a part of the business entrepreneurship literature. For inclusion, the papers were required to discuss the resource management of entrepreneurs or be closely related theoretical concepts along with bootstrapping and bricolage. This refining also helped remove the documents that contained bootstrapping procedures of the PLS-SEM methodology. The final sample excluded papers that did not directly relate to business management, entrepreneurship, strategic resource management, and innovation themes. A similar search in the Web of Science database showed a 93% overlap with the Scopus sample.
Another scholar helped validate the final sample by independently reading the basic details of all the documents chosen, removing papers that did not appear suitable for the analysis by consensus. It included papers that did not necessarily pertain to entrepreneurship or business management literature. The Cohen-Kappa (1960) coefficient was 0.86 for agreement on the documents included in the final sample. The measure was significant at p = 0.01. As per the guidelines by Landis and Koch (1977), a significant kappa statistic of over 0.81 shows an almost perfect agreement between the two raters. In case of disagreements, the papers were discussed with fellow researcher to decide whether to include them in the sample. A final sample of 239 papers, including 167 papers in the bricolage literature and 72 papers in the bootstrapping literature, were considered.
Since research in any scientific field is not conducted in isolation but by a community of people, it is imperative to understand literature from a temporal dimension, as the development and maturity of a scientific discipline involve incremental efforts as it progresses in leaps through different paradigms (Kuhn, 1962). This approach helped analyze the knowledge base of bootstrapping and bricolage literature. Apart from the seminal works explaining the concept, bootstrapping literature consists of multi-disciplinary works on different aspects of entrepreneurial bootstrapping like motives behind opting for it, stages of business development, women entrepreneurship and gender differences, business environment, human and social capital, firm performance and growth, technology orientation as well as the strategic role of the entrepreneur. Similarly, for the bricolage literature, multi-disciplinary works discuss resource constraints, institutional voids, social entrepreneurship, related concepts like effectuation and causation, technical entrepreneurship, intrapreneurship, and impact on firm performance. More recently, scholars have also studied the impact of the COVID-19 pandemic on entrepreneurial resourcefulness behaviors.
Analysis procedures
The first bibliometric analysis procedure applied was co-citation analysis, a commonly used bibliometric procedure that measures how frequently two articles are cited together in the sample (Hjørland 2013; McCain 1990). This helps identify and understand the thematic clusters of the most cited publications (Donthu et al. 2021). Co-citation analysis measures the proximity between works on the sample, assuming that works related in terms of their concepts get cited together (Zupic and Čater 2015). Therefore, the co-citation analysis will help identify the intellectual structure of the bootstrapping and bricolage literature (Rossetto et al. 2018). To make the co-citation results more meaningful, an exploratory factor analysis (EFA), one of the most commonly employed clustering methods in a co-citation bibliometric analysis, was performed (McCain 1990; Zupic and Čater 2015).
For performing the EFA, the co-citation matrix, extracted from Bibexcel (Persson et al. 2009) software, was further converted to a Pearson’s coefficient correlation matrix. The SPSS software helped identify factors by applying the principal component method and scree tests with Varimax rotation and Kaiser normalization (Kaiser 1960; Lin and Cheng 2010; Watkins 2018). This ensured the maximum number of works load to minimum identified factors (Pilkington and Liston-Heyes 1999). The works with a factor loading over 0.4 were considered important for the analysis (Hair et al. 2010; Peterson 2000; Zhao 2006). Exploratory factor analysis is the most suitable method in such a study because it enables loading the cited works to more than one factor, which implies the contributions of these works across different factors and literature (Zupic and Čater 2015). Thus, works loaded to multiple factors were assigned to the factor with the highest loading while also analyzing its role in the other factors (Vogel and Güttel 2013). Works with a factor loading of more than 0.7 are considered the most critical works for the respective factor (McCain 1990; Zupic and Čater 2015). The number of articles considered for co-citation analysis was reduced to 71 most cited works after cleaning the raw data for duplicates (like multiple entries of the same paper in Scopus) from the top cited articles. Bibexcel software assisted in extracting co-citation data. This follows Lotka’s law which states that a smaller proportion of highly cited articles can represent an area of research (Nath and Jackson 1991). A network analysis then helped to increase the robustness of the co-citation analysis (Vogel and Güttel 2013) by using the NETDRAW, UCINET (Borgatti et al. 2002) software, and the co-citation matrix. Network analysis will help visualize the intellectual structure of the literature. The nodes in the network diagram represent the works or the publications, with the ties depicting their proximity.
The next analysis of the bibliometric study is a bibliographic coupling that assumes that works that refer to the same literature are similar in their content and themes (Kessler 1963). This helps to identify similarities in works by considering the overlapping publications referred to in the papers (Vogel and Güttel 2013; Zupic and Čater 2015). A bibliographic coupling analysis assists in establishing the research front and themes in the studied literature (Baker et al. 2021; Vasconcelos Scazziota et al. 2020). While co-citation analysis focuses on how two documents get cited together, a bibliographic coupling analysis focuses on the publications cited in the sample (Donthu et al. 2021). It helps understand how similar works are by analyzing how frequently studies in the selected sample have a common reference.
A co-occurrence matrix was extracted from Bibexcel software to analyze which pairs of documents share the same references for performing the bibliographic coupling analysis. This analysis follows the same process used for co-citation, network analysis, and clustering. Documents with more than five couplings were used for the analysis, thus reducing the number of cited documents in the bibliographic coupling matrix to 58. This helped manage the data better by reducing documents with lesser couplings while also increasing the robustness of the analysis by using references of higher intensity (Vasconcelos Scazziota et al. 2020). As in the co-citation analysis, this analysis generates a network diagram for bibliographic coupling using the same parameters as the co-citation analysis.
A detailed analysis of the works that fall into each cluster for co-citation and bibliographic coupling analysis helped label the different clusters, following a structured reading of the studies in each factor (Vogel and Güttel 2013). This process followed the guidelines from the work of Denyer and Tranfield (2009) for this review. The cross-loadings in the factors analysis help understand the conceptual convergence and overlaps in the two pieces of literature. If an article has a factor loading over 0.40 in not just one but two factors, it implies that it conceptually adds to the literature of the other factor. For instance, as in Table 2, the work of Desa (2012) is a part of factor 1 (0.568 factor loading), but it also contributes to factor 2 and factor 3 as it has a factor loading of greater than 0.4 for those two factors as well.Table 2 Co-citation analysis factor loadings
Paper authors Year Highest factor loading Co-citation factors Paper authors Year Highest factor loading Co-citation factor
Cooper, Gimeno-Gascon and Woo 1994 CC1-0.941 CC1 Wernerfelt 1984 CC2-0.675 CC2, CC1,CC3
Cassar 2004 CC1-0.852 CC1 Barney 1991 CC2-0.663 CC2, CC1,CC3
Bhide 1992 CC1-0.838 CC1 Ciborra 1996 CC2-0.65 CC2, CC1
Berger and Udell 1998 CC1-0.829 CC1 Short, Moss and Limpkin 2009 CC2-0.643 CC2, CC1
Adler and Kwon 2002 CC1-0.821 CC1 Desa and Koch 2014 CC2-0.64 CC2, CC1,CC3
Harrison, Mason and Girling 2004 CC1-0.805 CC1 Linna 2013 CC2-0.629 CC2, CC1,CC3
Winborg and Landstrom 2001 CC1-0.803 CC1, CC2 Zahra et al 2008 CC2-0.624 CC2, CC1
Grichnik et al 2014 CC1-0.775 CC1, CC2 Jacob 1977 CC2-0.614 CC2, CC1,CC3
Ebben and Johnson 2006 CC1-0.774 CC1, CC2 Mair and Marti 2009 CC2-0.596 CC2, CC1,CC3
Baker and Nelson 2005 CC1-0.773 CC1 Sirmon, Hitt and Ireland 2007 CC2-0.567 CC2, CC1,CC3
Baker 2007 CC1-0.766 CC1 Garud and Karnøe 2003 CC2-0.565 CC2,CC1,CC3,CC4
Jones and Jayawarna 2010 CC1-0.755 CC1, CC2 Levi Strauss 1966 CC2-0.564 CC2,CC1,CC3,CC4
Van Auken 2005 CC1-0.749 CC1 Fisher 2012 CC2-0.535 CC2,CC1,CC3,CC4
Winborg 2009 CC1-0.745 CC1, CC2 Shane and Venkataraman 2000 CC2-0.532 CC2,CC1,CC3,CC4
Baker, Miner and Eesley 2003 CC1-0.682 CC1, CC2, CC3 Sarasvathy 2001 CC2-0.529 CC2,CC1,CC3,CC4
Wu, Liu, Zhang 2017 CC1-0.657 CC1, CC3 Alvarez and Busenitz 2001 CC3-0.781 CC3
Ferneley and Bell 2006 CC1-0.629 CC1, CC2, CC3 Hambrick and Mason 1984 CC3-0.734 CC3, CC1
Zahra 2009 CC1-0.617 CC1, CC2, CC3 Lumpkin and Dess 1996 CC3-0.7 CC3, CC4
Halme, Lindeman and Linna 2012 CC1-0.585 CC1, CC2, CC3 Salunke, Weerawardena, Mccoll-Kennedy 2013 CC3-0.624 CC3, CC1,CC2
Miner, Bassof and Moorman 2001 CC1-0.575 CC1, CC2, CC4 Bechky and Okhuysen 2011 CC3-0.6 CC3, CC1,CC2
Di Domenico, Haugh and Tracey 2010 CC1-0.57 CC1, CC2, CC3 Senyard, Baker and Davidsson 2009 CC3-0.599 CC3,CC1,CC4
Duymedjian and Rüling 2010 CC1-0.57 CC1, CC2, CC3 Davidsson, Baker and Senyard 2017 CC3-0.591 CC3, CC1,CC2
Desa 2012 CC1-0.568 CC1, CC2, CC3 Hoang and Antoncic 2003 CC4-0.857 CC4
Phillips and Tracey 2007 CC1-0.567 CC1, CC2,CC3,CC4 Granovetter 1985 CC4-0.811 CC4
Desa and Basu 2013 CC1-0.564 CC1, CC2, CC3 Alvarez and Barney 2007 CC4-0.701 CC4,CC1,CC3
Mair and Marti 2009 CC1-0.549 CC1, CC2, CC3 Perry, Chandler and Markova 2012 CC4-0.699 CC4,CC3
Senyard et al. 2014 CC1-0.549 CC1, CC2, CC3 Read, Song and Smit 2009 CC4-0.695 CC4,CC1,CC3
Stinchfield, Nelson and Wood 2013 CC1-0.538 CC1, CC2,CC3,CC4 Mcmullen and Shepherd 2006 CC4-0.666 CC4,CC3
Meyer and Rowan 1977 CC2-0.81 CC2, CC4 Berends et al 2014 CC4-0.567 CC4,CC1,CC3
Seelos and Mair 2005 CC2-0.701 CC2, CC1 Welter, Mauer and Wuebker 2016 CC4-0.55 CC4,CC1,CC3
Weick 1993 CC2-0.677 CC2, CC1 Shane 2000 CC4-0.546 CC4,CC1,CC2,CC3
Co-citation factors column consists of factors that studies are loaded to (> 0.4 factor loading), the first one depicting the highest factor loading
% of Variance Explained: CC1: 33.2%; CC2: 24.9%; CC3: 21.5%; CC4: 15.5%; % of Variance Accumulated: 95.1%
Results
Co-citation analysis: understanding the intellectual structure
The first analysis procedure applied was co-citation analysis. The factor analysis identified four factors that explain 95% of the variation. Table 2 details the co-citation factors along with the factor loadings. The factor loading analysis process included the exclusion of multiple seminal works dealing with the basics of case study analysis, theory building, statistical methodologies, and qualitative data biases from the articles analyzed for co-citation as they do not contribute directly to the bootstrapping and bricolage literature.
The first factor was labeled entrepreneurial resourcefulness (CC1). The works in this factor primarily discuss small business resourcing behaviors (Cassar 2004; Berger and Udell 1998). The works in bootstrapping consist of seminal work explaining the concept of bootstrapping, the motives behind bootstrapping, changes in this behavior over time, the role of social and human capital, and the impact of bootstrapping on product development and firm performance (Cooper et al. 1994; Ebben and Johnson 2006; Grichnik et al. 2014; Harrison et al. 2004; Jones and Jayawarna 2010; van Auken 2005; Winborg 2009; Winborg and Landström 2001). The bricolage works consist of documents explaining resource recombination and making do with what is in hand (Baker 2007; Baker and Nelson 2005). Here, the factor loadings highlight works in bricolage and their impact on firm performance and innovation (Duymedjian and Rüling 2010; Ferneley and Bell 2006; Miner et al. 2001; Senyard et al. 2014; Stinchfield et al. 2013; Wu et al. 2017). An overlap can be noted here with the other three factors as authors discuss resourcefulness behaviors in social entrepreneurship, impact on entrepreneurial orientation, and subsequent performance (Desa and Koch 2014; Lumpkin and Dess 1996; Seelos and Mair 2005; Senyard et al. 2009; Short et al. 2009). Multiple studies loaded to the other three factors have a contribution to factor CC1 as well (greater than 0.4-factor loading). An overlap between the two pieces of literature is present in this factor as the papers analyzed are cited in works from both behaviors with a focus on getting maximum value out of the resources already owned or easily accessible.
The second factor was labeled resourcing social entrepreneurship (CC2), as it primarily consisted of works that discuss how bricolage and bootstrapping are in the case of a social enterprise (Desa and Koch 2014; Mair and Marti 2009; Seelos and Mair 2005; Short et al. 2009; Zahra et al. 2009). Some of the works here also highlight the role of organizational structure and resilience while discussing how ventures display innovative resourcefulness behaviors to survive in a dynamic resource-deficient environment (Jacob 1977; Linna 2013; Meyer and Rowan 1977; Sirmon et al. 2007; Weick 1993). This factor consists of factor loadings of some classic resource theory work like that of Wernerfelt (1984) and Barney (1991), along with the anthropology paper on bricolage (Honigmann and Levi-Strauss 1967). Further, this factor consists of academic literature on similar behaviors like effectuation, causation, and tinkering (Fisher 2012; Jacob 1977; Sarasvathy 2001). This factor also consists of works that overlap between the two pieces of literature. The works cited by both bricolage and bootstrapping behaviors discuss how social entrepreneurs frequently face resource shortages and resort to innovative ways of resource use by seeking help from their networks to improve firm performance. Factor CC2 also has contributions from works that primarily load to the other three factors.
The third factor was labeled entrepreneurial orientation and firm performance (CC3). It consists of works that discuss how an organization’s resource management reflects its founders and team members (Hambrick and Mason 1984) and how multidimensional entrepreneurial orientation impacts firm performance (Lumpkin and Dess 1996). The works extend the resource-based theory by considering entrepreneurial cognition and the ability to combine resources as a critical component of a firm’s resource management and performance (Alvarez and Busenitz 2001; Senyard et al. 2009). Moreover, it discusses how businesses strategically combine resources to gain a sustained competitive advantage (Salunke et al. 2013). The works loaded on this factor are also cited in both works of literature. Studies loaded to other factors also contribute to factor CC3, as seen in Table 2.
The final factor was labeled social networks and decision-making (CC4). This consisted of works highlighting how business behaviors and decision-making depend on social networks (Granovetter 1985; Hoang and Antoncic 2003). The need to highlight and study the critical role of entrepreneurial action in these related concepts like bricolage, bootstrapping, and effectuation helps to understand these creative aspects of entrepreneurial behavior better (Berends et al. 2014; Perry et al. 2012; Read et al. 2009; Welter et al. 2016). Additionally, it discusses the role of entrepreneurial action and prior knowledge in the opportunity discovery process (Alvarez and Barney 2007; Shane 2000). This factor also consists of works that overlap between the two pieces of literature, as the studies cited by both these behaviors emphasize the importance of sharing resources with their networks for innovatively mobilizing scarce resources. Factor CC4 also has contributions from factors that primarily load to other co-citation factors.
Next, a network analysis helped to increase the robustness of the co-citation analysis. Figure 1 shows the co-citation network, which helped visualize the intellectual structure of the literature. The nodes in the network diagram represent the works or the publications, with the ties depicting their proximity. As in other multidimensional scaling methods, this network layout aims to optimize the node distance. So, in the figure, the path length or the number of edges between two nodes helps measure centrality in the network (Vogel and Güttel 2013). The boxes help to identify the four factors of co-citation analysis. The line thickness connecting the two nodes represents how frequently those documents are co-cited. Since the works loaded on one factor are also connected with the references of the other factors, this network analysis helps understand this overlap, as shown in Fig. 1. This helps understand the relationship between them not just as a network diagram but also by assisting in analyzing their density. Therefore, network analysis enables more than just visualization (Pilkington and Chai 2008; Vogel and Güttel 2013). The density was analyzed to understand how different a factor or subgroup is from the network (Table 3). The density is the ratio of the number of realized ties to all the possible links or ties within the subgroup (Vasconcelos Scazziota et al. 2020; Vogel and Güttel 2013). It represents the extent to which a particular subgroup has a common base. While all factors are dense, CC2 and CC3 have higher densities.Fig. 1 Co-citation network
Table 3 Network Analysis Metrics for Co-citation
Factor Symbol in Fig. 1 Nodes Density Variance explained %
CC1 Square 28 0.50 33.2
CC2 Up Triangle 18 0.81 24.9
CC3 Circle 7 0.67 21.5
CC4 Diamond 9 0.56 15.5
Bibliographic coupling: understanding the research front
The next analysis was a bibliographic coupling. This analysis considered works with more than five couplings. The factor analysis of the bibliographic coupling matrix identified five factors (Table 4).Table 4 Bibliographic coupling analysis factor loadings
Paper authors Year Highest factor loading Bibliographic coupling factors Paper authors Year Highest factor loading Bibliographic coupling factors
An et al. 2020 BC1-0.915 BC1 Weerakoon, Gales and McMurray 2019 BC1-0.915 BC1
An et al. 2018 BC1-0.912 BC1 Wierenga 2019 BC1-0.887 BC1
An et al. 2018 BC1-0.912 BC1 Witell et al 2017 BC1-0.906 BC1
Bacq et al. 2015 BC1-0.901 BC1 Wu, Liu and Zhang 2017 BC1-0.89 BC1
Bellavitis et al. 2017 BC1-0.585 BC1,BC3 Yachin 2017 BC1-0.886 BC1
Bojica, Istanbouli and Fuentes-Fuente 2014 BC1-0.694 BC1,BC4 Yu et al 2019 BC1-0.887 BC1
Bojica et al 2017 BC1-0.829 BC1 Yu et al 2020 BC1-0.894 BC1
Brown, Mawson and Rowe 2019 BC1-0.916 BC1 Zhao et al 2021 BC1-0.9 BC1
Cheung et al. 2019 BC1-0.913 BC1 Horváth 2019 BC2-0.956 BC2
Davidsson, Baker and Senyard 2017 BC1-0.889 BC1 Weigand and Schulte 2020 BC2-0.95 BC2
Desa and Basu 2013 BC1-0.895 BC1 Al Issa 2021 BC2-0.916 BC2
Hooi et al. 2016 BC1-0.902 BC1 Block, Fisch and Hirschmann 2021 BC2-0.769 BC2
Hota, Mitra and Qureshi 2019 BC1-0.908 BC1 Waleczek et al 2018 BC2-0.751 BC2,BC1
Jannsen, Fayolle and Wuilaume 2018 BC1-0.913 BC1 Jayawarna, Jones and Marlow 2015 BC3-0.979 BC3
Korsgaard, Müller and Welter 2021 BC1-0.824 BC1 Jones and Jayawarna 2010 BC3-0.873 BC3
Ladstaetter, Plank and Hemetsberger 2018 BC1-0.912 BC1 Jayawarna, Jones and Macpherson 2014 BC3-0.852 BC3
Olive-Tomas and Harmeling 2020 BC1-0.9 BC1 Vanacker et al 2011 BC3-0.687 BC3, BC5
Salunke, Weerawardena and McColl-Kennedy 2013 BC1-0.873 BC1 Moghaddam et al 2017 BC3-0.568 BC3
Sarkar 2018 BC1-0.918 BC1 Hsiao, Ou and Chen 2014 BC4-0.87 BC4
Schulte-Holthaus 2018 BC1-0.912 BC1 Ritvala, Salmi and Andersson 2014 BC4-0.856 BC4
Servantie and Rispal 2018 BC1-0.92 BC1 Beckett 2016 BC4-0.528 BC4
Sivathanu and Pillai 2019 BC1-0.911 BC1 Beckett 2016 BC4-0.528 BC4
Tasavori, Kwong and Pruthi 2017 BC1-0.911 BC1 Neely and Van Auken 2010 BC5-0.8 BC5
Tindiwensi et al. 2020 BC1-0.809 BC1 Neely and Van Auken 2009 BC5-0.776 BC5
Tsilika et al. 2020 BC1-0.91 BC1 Malmström 2014 BC5-0.647 BC5,BC3
Scazziota et al. 2020 BC1-0.901 BC1 Franquesa and Brandyberry 2009 BC5-0.412 BC5
Bibliographic Coupling factors column consists of factors that studies are loaded to (> 0.4 factor loading), the first one depicting the highest factor loading
% of Variance Explained: BC1: 50.7%; BC2: 9.4%; BC3: 9.4%; BC4: 6.3%; BC5: 6.3%; % of Variance Accumulated: 82.3%
The first factor was labeled resource mobilization in constrained environments (BC1), as it primarily consists of works that discuss how entrepreneurs overcome the challenge of gathering resources that they might not have easy access to (Korsgaard et al. 2021; Wierenga 2020). Some of these works discuss how it is even more difficult for enterprises, especially social ventures, to operate in emerging economies (Cheung et al. 2019; Hota et al. 2019). Academicians also highlight the role of entrepreneurial bricolage in agricultural farming and creative industries (Schulte-Holthaus 2018; Tindiwensi et al. 2021). Others discuss how firms create value through their decision-making logic and resource mobilization while employing resourcefulness behaviors to improve their creativity and innovation performance (An et al. 2018, 2020).
The second factor was labeled venture performance and growth (BC2), which consists of studies that analyze the role of bootstrapping on venture performance. Studies reveal that a firm's financial performance depends on its founders’ business goals and priorities between freedom and maximizing wealth. Findings indicate that the need for capital is driven more by demand instead of supply (Weigand and Schulte 2020). Further, other works highlight a positive relationship between the use of bootstrapping and the venture’s financial performance when moderated with strategic improvisation (Al Issa 2020). Scholars have found that the COVID-19 pandemic has also positively affected bootstrapping, helping ventures manage their liquidity and stay afloat (Block et al. 2021). Further, research also finds that bootstrapping is positively associated with employment growth when positively moderated by the training of entrepreneurs (Horváth 2019). Finally, due to its positive impact on venture performance and growth, bootstrapping is a strategic push driven by the founder(s) and not just a last resort (Waleczek et al. 2018).
The third factor was labeled resourcing through networks (BC3), as it consists of works that analyze the effect of social networks on bootstrapping activities. While bootstrapping has been favorable for most new ventures, studies have found gender differences in utilizing social networks to access resources (Jayawarna et al. 2015). Another study confirms that social networks influence venture sales, growth, and profits, mediated by bootstrapping activities (Jones and Jayawarna 2010). A study highlights that entrepreneurs who have support from their family networks from a young age are better at shaping entrepreneurial ambitions into results (Jayawarna et al. 2011). Contrary to these outcomes, some scholars also highlight that using networks to arrange resources might lead to uncertain outcomes due to a lack of formal arrangement (Vanacker et al. 2011). However, immigrant entrepreneurs prefer a single or few strong network sources over multiple weak sources (Moghaddam et al. 2017).
The fourth factor was labeled resource-led innovation (BC4). The works here discuss how entrepreneurs can innovatively use resources deemed worthless by ventures. Scholars highlight resource transformation, leveraging, and exchange that assists bricoleurs in employing frugal resource combinations (Hsiao et al. 2014). The existing literature has also commented on using bricolage for large companies that might be better off by focusing on more entrepreneurial and pragmatic incremental changes instead of big radical innovations. Moreover, even players in the food industry use entrepreneurial bricolage to innovate based on decision systems to respond and act, focus on iterative learning, and creatively combine resources at hand. (Beckett 2016).
The final factor is labeled slack and strategy (BC5). The works in this cluster highlight strategic and creative resource mobilization. Scholars have recognized three strategic bootstrappers—quick-fix bootstrappers, proactive bootstrappers, and efficient bootstrappers—who adopt strategies like using temporary resources, focusing on operational issues, and even focusing on external activities present in their value chain (Malmström 2014). Scholars find that small firms apply resource invocation using slack resources very differently from larger organizations (Franquesa and Brandyberry 2009). Scholars have noted that factors like founder education, age, and gender impact their strategies to gather resources and funds for their ventures (Neeley and Auken 2009). Another study points out that female business owners are more affected by metrics like firm performance, revenues, and overdraft privileges, due to the inherent gender bias present in capital availability. They tend to strategically improve their revenue-based cash flows that can be re-invested in their businesses (Neeley and van Auken 2010).
Figure 2 shows the network diagram of the bibliographic coupling analysis. It is analyzed as explained in Sect. 4.1. Clusters of bootstrapping papers can be seen on the right side of the network diagram, and we can note a cluster of bricolage literature at the bottom. Table 5 describes the results of the density analysis. The density values of most clusters depict that the factors are very dense (some factors have near-perfect density).Fig. 2 Bibliographic coupling network
Table 5 Network analysis metrics for bibliographic coupling
Factor Symbol in Fig. 2 Nodes Density Variance explained %
BC1 Square 35 0.45 50.78
BC2 Circle 5 1.00 9.44
BC3 Up Triangle 5 0.90 9.43
BC4 Diamond 4 1.00 6.33
BC5 Box 4 0.50 6.30
Integration of the bibliometric results
The harmony between the results of co-citation and bibliographic coupling is essential to understand the synergies between the two pieces of literature (Vasconcelos Scazziota et al. 2020). Figure 3 shows the integration of results based on the content of the studies loaded to each factor. The intellectual structure of the literature comes from the co-citation analysis and the research trends from the bibliographic coupling. In our analysis, factor CC1 (entrepreneurial resourcefulness) finds support by works in factor BC3/CC4 (both discussing resourcing through social networks) that discuss how entrepreneurs use their social networks to acquire limited resources for their firms. Entrepreneurs use their networks not just to gain the information required to acquire resources at the best terms but also to gain the requisite goodwill to access them from their networks. Factor BC1 (resource mobilization in constrained environments) deals with the mobilization of resources in resource-constrained environments to act on available opportunities. Therefore, factor BC1 (resource mobilization in constrained environments) leads entrepreneurs to employ bootstrapping and bricolage (entrepreneurial resourcefulness: CC1) in underdeveloped factor markets with institutional voids. The resource constraints due to the lack of supportive institutions, especially in developing countries, act as a catalyst toward resourcefulness in this situation. Further, the concept of bricolage (entrepreneurial resourcefulness: CC1) often finds a place in the social entrepreneurship literature (resourcing social entrepreneurship: CC2). Thus, factor CC1 (entrepreneurial resourcefulness) helps social ventures (CC2: resourcing social entrepreneurship) scale their businesses by accessing and creatively using resources to impact base-of-the-pyramid communities by finding innovative solutions to customers’ problems while also focusing on both the financial and social impact of their actions. Resource-led innovation (BC4) and the consequent firm performance (CC3 and BC2) find support in resource mobilization using bootstrapping and bricolage. The resource-led hit-and-trial and improvisational ways of resource mobilization, with already owned or cheaply available resources, often balance out with the routine actions to provide exceptional firm performance.Fig. 3 Result integration
Moreover, in terms of innovation, works in factor CC1 (entrepreneurial resourcefulness) explicitly state how entrepreneurs can create unanticipated innovations by creatively employing limited and mostly otherwise worthless resources. This is more critical in resource-constrained environments (BC1) characterized by low-income settings and institutional voids, as the focus is not on using available resources and supporting institutions for conventional purposes but on a more hands-on approach to taking action even without the right resources and support. Finally, the presence of slack (slack and strategy: BC5) resources in the business also support innovation in the firms as it encourages exploration and allows experimentation with the excess resources, thereby promoting initiative. Scholars have suggested that ventures with slack resources tend to strategically experiment and take risks that might benefit their innovativeness (resource-led innovation: BC4) and finally result in better venture performance (CC3/BC2). Different entrepreneurs adopt bootstrapping strategies (slack and strategy: BC5) to achieve their business goals by innovatively transforming their available resource base to create value. Literature suggests that these resourcefulness behaviors are often a planned strategic push from the entrepreneur and not a last resort due to resource shortages.
TCCM analysis
This analysis involves a systematic review of bootstrapping and bricolage literature following the TCCM approach (Paul and Rosado-Serrano 2019; Singh and Dhir 2019). Figure 4 details the TCCM framework which helped identify the various gaps in the existing literature to propose new directions for future research.Fig. 4 TCCM Framework
Theory development
With the increase in academic research on entrepreneurship, scholars have identified multiple theoretical perspectives that help explain the resource mobilization behaviors of entrepreneurs in resource constraints. Due to the differences in the resourcing strategies of big companies and entrepreneurial firms, scholars have used classic theories for studying entrepreneurial behaviors with a modified perspective. Some of the prominent ones are resource dependency theory (Pfeffer and Salancik 2015), resource constraint theory (Loasby and Leibenstein 1976), resource-based view (Barney 1991), dynamic capabilities theory (Teece et al. 1997), pecking order of theory (Myers and Majluf 1984), institutional theory (DiMaggio and Powell 1983; Meyer and Rowan 1977) and traditional finance theory, organizational learning (Argyris and Schön 1997).
The resource-based view and resource dependency theory suggest that a business's success depends on the resources it can acquire, which have an important role to play in achieving competitive advantage (Desa and Basu 2013; Harrison et al. 2004; Jayawarna et al. 2011; Jones and Jayawarna 2010). These theories have been used in this context as the resource dependency theory also holds that firms should reduce their reliance on external parties for resource acquisition. The resource constraint theory highlights how entrepreneurs operate their ventures by combining and recombining available resources under constraints (Baker and Nelson 2005). Scholars have also mentioned the traditional finance theory and Myers pecking order theory which imply that the financing decisions of a business are influenced by the cost of debt or equity while funding initial capital needs using flexible internal sources (Brush et al. 2006; Carter and van Auken 2005; Neeley and van Auken 2010; van Auken 2005; Vanacker et al. 2011). There is also the prevalence of transaction cost economics (Williamson 1979) in analyzing which source of capital is most suitable for newer ventures (Bellavitis et al. 2017). Literature also uses organizational theory to predict the behaviors of entrepreneurs who are bootstrapping their startups (Ebben and Johnson 2006).
Further, institutional theory also finds a place in the studies due to the importance of the external business environment and activities undertaken by startup founders who leverage resources available to them to create new institutions (Bellavitis et al. 2017; Desa 2012; Phillips and Tracey, 2007; Politis et al. 2012). Multiple researchers have chosen social network theory to explain how entrepreneurs gather resources from their networks (Bizri 2017; Carter et al. 2003; Dabić et al. 2020; Evers and O’Gorman 2011; Jonsson and Lindbergh 2013). Even though multiple theories in literature have briefly explained entrepreneurial actions like bricolage and bootstrapping, there is still a need to identify theories that explain patterns of these resource mobilization behaviors. The current literature offers little direction toward understanding how entrepreneurs creatively accumulate and employ resources at hand innovatively to survive resource limitations.
Context
Academic research in bootstrapping and bricolage currently focuses on fast-growing small businesses across multiple industries. Scholars have also conducted various studies focusing on women entrepreneurs and how their resource mobilization behaviors differ from their male counterparts (Brush et al. 2006; Carter et al. 2003). However, most of the existing research focuses on studying ventures in developed countries in Europe or North America (Baker et al. 2003; Brush et al. 2006; Carter et al. 2003; Carter and van Auken 2005; Ebben 2009; Ebben and Johnson 2006; Senyard et al. 2014). Both the literature consist of multi-disciplinary works on different aspects of resource management like strategic motives, stages of the business, and uncertainty and institutional voids in the environment (Brush et al. 2006; Ebben and Johnson 2006; Mair and Marti 2009; Winborg 2009). There has also been a considerable focus on the human and social capital aspect along with their impact on how the business eventually performs (Ebben 2009; Grichnik et al. 2014; Jayawarna et al. 2011; Stinchfield et al. 2013; Vanacker et al. 2011). Recent literature also studies the impact of the COVID-19 pandemic on entrepreneurial resource management behaviors, given the need to build resilience in challenging crisis driven markets (Brown et al. 2020; Kuckertz et al. 2020; Ma and Yang 2022). Overall, contextually fragmented research for developed countries is available, making it difficult to generalize and draw conclusions.
Characteristics
Studies have identified various antecedents like environmental factors such as environmental hostility and munificence, uncertainty, resource (knowledge and financial) constraints, and weak institutions, business factors like traditional financial capital access, firm size, and entrepreneurial factors like strategic orientations of founders, frugality, relational capital, creativity and passion, social capital and human capital, team heterogeneity (An et al. 2018; Grichnik et al. 2014; Grichnik and Singh 2010; Iqbal et al. 2021; Miao et al. 2017; Michaelis et al. 2020; Rahman et al. 2020; Rutherford et al. 2017; Stenholm and Renko 2016; Su et al. 2020). Further, academicians have also studied the impact of these resource mobilization behaviors on firm development, growth, performance, innovation capabilities, and sustainable competitive advantages (Brush et al. 2006; Ebben 2009; Ernst et al. 2015; Fisher 2012; Harrison et al. 2004; Jones and Jayawarna 2010; Salunke et al. 2013; Senyard et al. 2014; Vanacker et al. 2011). Despite the literature on the antecedents and outcomes of bricolage and bootstrapping, the results often conflict. While some studies suggest a positive impact of bootstrapping and bricolage on firm performance, others highlight that they can result in inferior quality business offerings and unconventional solutions that may not be suitable for the customers in the long run (Baker and Nelson 2005; Jones and Jayawarna 2010; Rutherford et al. 2012). Therefore, there is a need for more empirical research on the antecedents and outcomes of these resourcefulness behaviors.
Methods
Prior works in entrepreneurial bootstrapping and bricolage consist of a mix of empirical and inductive case-based papers. The case study-based research uses an inductive grounded theory method suitable for understanding activities in real-time both within and outside the venture to understand how these resource management activities unfold in entrepreneurial ventures (Baker et al. 2003; Baker and Nelson 2005; Ferneley and Bell 2006; Halme et al. 2012; Harrison et al. 2004; Jonsson and Lindbergh 2013; Mair and Marti 2009). Moreover, multiple studies use empirical data and quantitative statistical analysis like structural equation modeling, regression analysis, cluster analysis, and factor analysis (Brush et al. 2006; Carter et al. 2003; Ebben and Johnson 2006; Jones and Jayawarna 2010; Neeley and van Auken 2010; Senyard et al. 2014; Winborg 2009; Winborg and Landström 2001). Further, many highly cited works in this literature consist of conceptual papers (Baker 2007; Boxenbaum and Rouleau 2011; Duymedjian and Rüling 2010; Phillips and Tracey 2007; Stinchfield et al. 2013; Welter et al. 2016). Future research can also incorporate some mixed methods based studies which are currently very limited (Desa 2012; Kuckertz et al. 2020).
Discussion
Despite the similarities in their resourcefulness nature, the concepts of entrepreneurial bootstrapping and bricolage are yet to be integrated systematically into literature. Both behaviors act as potential key dynamic capabilities that help firms access, enhance and integrate limited resources that might be deemed obsolete or worthless. While bootstrapping primarily deals with how entrepreneurs access financial, material, and knowledge resources for their businesses, bricolage involves creatively applying these resources to enhance their limited resource base. Bootstrapping helps bolster the resource base, which is further made more flexible and innovative by creative reinvention of resources using bricolage. While bootstrapping deals with using informal ways to access, borrow or share resources creatively, bricolage involves deploying these resources in novel ways to extract value from them that others might ignore, making bricolage a strategic action triggered when entrepreneurs decide to strategically bootstrap their ventures. Refer to Sect. 2 for a better understanding of the differences and similarities between the two concepts.
The bricolage literature focusing more on resource mobilization in resource-limited environments with institutional voids might benefit from bootstrapping techniques. Meanwhile, research trends in bootstrapping focus on resource-led innovations, strategic use of slack resources, and strategic application of resource acquisition activities. Bricolage activities complement these research trends by involving a creative reinvention of slack resources to strategically innovate a firm's resource management. Both entrepreneurial bootstrapping and bricolage focus on the critical role of resources for operations and survival of the business in a resource-scarce environment by emphasizing taking action by managing with the existing resources without any external help to strategically respond to the uncertain environment. The unconventional “do-it-yourself" approaches and the critical role of the entrepreneurs’ social capital to access resources and test products emphasize the central role of the entrepreneur (human capital, social capital, and behavior patterns) in both behaviors. The resource led innovation and creative reinvention is critical to discovering and exploiting business opportunities.
The joint analysis of these two studies using bibliometric techniques demonstrates that the two behaviors have evolved drastically and that the latest research trends suggest a possible convergence of the two. The bibliographic coupling analysis helped understand the research trends to better understand the evolution of publications and probable joint paths hinting at a convergence between the two. Further, the intellectual structure (co-citation) analysis enables a better understanding of the origins and relationship between the two pieces of literature. Studying the two together facilitates discerning their embedded patterns and interrelationships while incorporating the entire published literature in this area (Duran-Sanchez et al. 2019; Vasconcelos Scazziota et al. 2020). Therefore, a combined effort from bricolage and bootstrapping can help build a more resilient venture and be in a better position to tackle any such resource disruptions in the case of a crisis or otherwise. This study is an effort in this direction as it suggests a probable integration between the two approaches that might help entrepreneurship scholars understand entrepreneurial resource behaviors better.
Bibliographic coupling analysis shows that both behaviors show pre-dominance of research on their impact on firm performance (venture performance and growth: BC2). The results of co-citation analysis strengthen this by presenting firm performance (entrepreneurial orientation and firm performance: CC3) as a crucial part of the intellectual structure of the two pieces of literature. Bootstrappers depend on their social networks to access resources, given the uncertainties that entrepreneurs operate in. Therefore, research on using entrepreneurial social capital (both weak and strong ties) has been a prominent research trend that can potentially benefit from bricolage. Using network bricolage (Baker et al. 2003) could be transformational in this aspect as it involves working with people in a firm's network. Bricolage will help work around the resource limitations to find new solutions to entrepreneurial problems using their creativity and networking skills.
A systematic co-citation analysis highlighted the intellectual structure of the approaches. The factor analysis suggests that, though some of these works deal directly with one of the two areas, they are also present in the other literature. Thus, a clear overlap between the two pieces of literature is present. All the factors identified in the co-citation analysis involved works referenced in both the literature. Factor CC1 pertaining to entrepreneurial resourcefulness details classical and seminal works in both approaches, along with the works that helped develop the understanding of these behaviors. Although factor CC2 deals with resourcing social entrepreneurship ventures that generally use bricolage for resource management, co-citation analysis results highlight the presence of bootstrapping, improvisation, and tinkering literature in this factor. Moreover, firm performance and entrepreneurial orientation (Factor CC3) discusses how an organization and its resource management reflect its founders and team members. The bootstrapping and bricolage literature highlights the resource-based theory by considering entrepreneurial cognition and the ability to access, accumulate and combine resources as a critical component of a firm’s resource management and performance. Finally, as noted above, research around the use of social networks (Factor CC4) for accessing and creatively deciding to use resources has proved to be an important intellectual structure aspect.
This bibliometric analysis helped to identify the differences and similarities in the knowledge base and intellectual structure (Grégoire and Cherchem 2020; Zupic and Čater 2015). The first contribution of this study is to consider these two approaches in synergy to understand them better together. These two behaviors complement each other as bootstrapping helps arrange for the finances, material resources, knowledge, and skills that are then creatively applied by the bricoleurs. Therefore, studying the two approaches together is interesting while considering their unique aspects (accumulation vs. deployment). This study will help direct scholarly attention toward including these behaviors in existing frameworks. With the increase in the number of publications, growing theoretical bases, and diverse methodologies, the current literature, which lacks integration of the two related concepts, can benefit from this bibliometric study as it helps overcome the idiosyncrasies that might arise in reviewing a large number of works, given the capabilities and biases of a researcher subjectively reviewing them alone. The results of this study depend on the systematic techniques of science mapping that helped analyze the combined literature in ways impossible through existing qualitative reviews.
Another contribution of this study is, thus, to direct literature toward understanding how bootstrapping can be considered a precursor or first step to bricolage behaviors. The two behaviors together can assist new ventures in accumulating and arranging resources in ingenious ways to build resilient businesses and provide a complete picture of the resource management processes of new entrepreneurial ventures. Academicians can also work towards the development of a common measurement scale to facilitate combined empirical research in this direction. Therefore, this study adds a new perspective for entrepreneurship researchers to explore the combined or sequential application of bootstrapping and bricolage. Finally, the TCCM analysis also helps propose some probable research gaps directed toward identifying theories that can undergo empirical tests using mixed-methods research to explain patterns of these resource mobilization behaviors. Therefore, scholars in this area should consider a joint analysis of these two approaches to develop a common measurement and test it using the qualitative grounded theory approach and empirically in quantitative studies.
Conclusion, limitations, and future work
In this study, the aim was to systematically analyze the literature on bootstrapping and bricolage to help understand the existing literature more methodically by enabling us to assimilate the contributions of scholars to get a comprehensive picture of the behaviors (Grégoire and Cherchem 2020; Zupic and Čater 2015). As suggested by Gregóire and Cherchem (2020), this bibliometric analysis further helped understand the rational structures of both the areas and the complementarities and differences between them (Donthu et al. 2021; Habib and Afzal, 2019; Rey-Martí et al. 2016). A co-citation analysis of both behaviors combined helped us understand the intellectual structures of the literature. A research trend analysis supported this as a result of the bibliographic coupling. Further, the combined TCCM analysis highlighted the theoretical developments, contexts, characteristics, and methodologies used in the two pieces of literature and suggested existing gaps in the literature.
Since the concepts of bootstrapping and bricolage have mostly developed in isolation and there exists a missing integration between the two, this study can be used as a starting point to further research on how these two behaviors can potentially converge and contribute to each other. The two entrepreneurial behaviors of bootstrapping and bricolage demonstrate convergence and mutually contribute to each other's literature. Overall, this study contributes to enhancing our understanding of the dynamics of resource management by entrepreneurs who operate in uncertain environments using flexible and informal approaches. This systematic empirical approach can also help to analyze other conceptually similar approaches in entrepreneurship. These concepts can then be applied together to contribute to other entrepreneurship and strategic management fields.
Future studies should consider the following research agendas to add to the research in this area (i) Fitting these two related behaviors in one existing framework to guide entrepreneurs employing these informal methods to sail through the challenges of the post COVID-19 pandemic; (ii) Developing a standard measurement scale for the two closely related behaviors; (iii) Generalization and validation of the measurement instrument through both qualitative and empirical research; (iv) Extending the proposed integration by delving deeper into the theoretical integration of the two streams of literature; (v) Exploring the shared antecedents of these two behaviors and devising a hierarchical framework to understand their interrelationships; (vi) Studying and adding to the literature of different combinations of action theories in entrepreneurship like bricolage and improvisation, bootstrapping and jugaad, bricolage and effectuation to develop a more generalizable and broader model or framework of entrepreneurial resource management behaviors; (vii) Exploring similar experimental research designs to further study theoretical integration and thematic similarities of various entrepreneurial behaviors whose resemblance is not explained by existing models in the entrepreneurship literature; (viii) Exploring the unique mechanisms through which social entrepreneurs create resource opportunities by employing both bootstrapping and bricolage (ix) Understanding the unique position of social enterprises in building and leveraging the credibility and creating economic and social value while bootstrapping for acquiring and mobilizing business resources.
Although the sample of documents for this study comes from one of the most comprehensive databases, some published works were outside the scope of the database. Future studies can consider a similar analysis by considering works from other databases. The sample of this study had more works in the bricolage literature as it has a higher number of research publications. This might be responsible for a stronger influence of bricolage works on the overall analysis. Moreover, publications with more referenced documents might influence the analysis. Even though bibliometric studies present a structured and systematic way of analyzing literature, they can not replace the in-depth analysis derived from a complete content analysis based study. Future studies analyzing bootstrapping and bricolage together can focus more on in-depth qualitative content-based analysis. As understood from the TCCM analysis, further studies should also consider mixed-method research in these areas for newer contexts like developing nations.
Scholars studying entrepreneurial resource behaviors can also use a similar analysis to explore convergence and differences between other approaches like improvisation, effectuation, and jugaad. Further, apart from researching these behaviors theoretically, scholars should also shift their focus on empirical qualitative and quantitative studies to understand how these approaches unfold in the case of resource-deficient ventures struggling their way through volatile pandemic hit markets. Apart from helping to understand resourcefulness behaviors in synergy, this study can further spur deliberation and research over other concepts which add value to the field of entrepreneurial research. Finally, this study should can act as a base for directing research toward exploring new ways in which entrepreneurs use these two behaviors together by developing a common measure for the two and testing it empirically in different entrepreneurial contexts.
Appendix 1: Top journals for the sample selected
S. No. Sources (Journals) No. of. articles
1 Entrepreneurship and Regional Development 16
2 Journal of Business Venturing 9
3 International Journal of Entrepreneurial Behaviour and Research 7
4 Venture Capital 7
5 International Journal of Entrepreneurship and Innovation 6
6 Small Business Economics 6
7 Entrepreneurship: Theory and Practice 4
8 Journal of Business Research 4
9 Journal of Developmental Entrepreneurship 4
10 Strategic Entrepreneurship Journal 4
11 Technological Forecasting and Social Change 4
12 International Business Review 3
13 International Entrepreneurship and Management Journal 3
14 International Journal of Entrepreneurship and Innovation Management 3
15 International Journal of Innovation Management 3
16 Journal of Business Venturing Insights 3
17 Journal of Entrepreneurship In Emerging Economies 3
18 Journal of Product Innovation Management 3
19 Management Decision 3
20 Technology Analysis and Strategic Management 3
Funding
No funding was received to assist with the preparation of this manuscript.
Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose. The authors report that there are no competing interests to declare.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
==== Refs
References
Al Issa H-E The impact of improvisation and financial bootstrapping strategies on business performance EuroMed J Bus 2020 16 2 171 194 10.1108/EMJB-03-2020-0022
Alvarez SA Barney JB Discovery and creation: alternative theories of entrepreneurial action Revis Organ Contex 2007 10.15603/1982-8756/roc.v3n6p123-152
Alvarez SA Busenitz LW The entrepreneurship of resource-based theory J Manag 2001 10.1016/S0149-2063(01)00122-2
An W Zhang J You C Guo Z Entrepreneur’s creativity and firm-level innovation performance: bricolage as a mediator Technol Anal Strateg Manag 2018 30 7 838 851 10.1080/09537325.2017.1383979
An W Rüling C-C Zheng X Zhang J Configurations of effectuation, causation, and bricolage: implications for firm growth paths Small Bus Econ 2020 54 3 843 864 10.1007/s11187-019-00155-8
Archer GR Baker T Mauer R Towards an alternative theory of entrepreneurial success: integrating bricolage, effectuation and improvisation Front Entrep Res 2009 29 6 1 23
Argyris Ch Schön DA Organizational learning: a theory of action perspective Reis 1997 10.2307/40183951
Audretsch DB The dynamic role of small firms: evidence from the U.S. Small Bus Econ 2002 18 1 13 40 10.1023/A:1015105222884
Baker T Resources in play: bricolage in the toy store(y) J Bus Ventur 2007 22 5 694 711 10.1016/j.jbusvent.2006.10.008
Baker T Nelson RE Creating something from nothing: Resource construction through entrepreneurial bricolage Adm Sci Q 2005 50 3 329 366 10.2189/asqu.2005.50.3.329
Baker T Miner AS Eesley DT Improvising firms: Bricolage, account giving and improvisational competencies in the founding process Res Policy 2003 32 2 SPEC 255 276 10.1016/S0048-7333(02)00099-9
Baker HK Kumar S Pandey N Thirty years of small business economics: a bibliometric overview Small Bus Econ 2021 56 1 487 517 10.1007/S11187-020-00342-Y/TABLES/16
Barney J Firm resources and sustained competitive advantage J Manag 1991 10.1177/014920639101700108
Beckett RC Entrepreneurial bricolage: developing recipes to support innovation Int J Innov Manag 2016 10.1142/S1363919616400107
Bellavitis C Filatotchev I Kamuriwo DS Vanacker T Entrepreneurial finance: new frontiers of research and practice: Editorial for the special issue embracing entrepreneurial funding innovations Ventur Cap 2017 19 1–2 1 16 10.1080/13691066.2016.1259733
Berends H Jelinek M Reymen I Stultiëns R Product innovation processes in small firms: combining entrepreneurial effectuation and managerial causation J Prod Innov Manag 2014 10.1111/jpim.12117
Berger AN Udell FG The economics of small business finance: The roles of private equity and debt markets in the financial growth cycle J Bank Financ 1998 10.1016/S0378-4266(98)00038-7
Bhide A Bootstrap finance: the art of start-ups Harv Bus Rev 1992 70 6 109 117 10122688
Bizri RM Refugee-entrepreneurship: a social capital perspective Entrep Reg Dev 2017 29 9–10 847 868 10.1080/08985626.2017.1364787
Block JH Fisch C Hirschmann M The determinants of bootstrap financing in crises: evidence from entrepreneurial ventures in the COVID-19 pandemic Small Bus Econ 2021 10.1007/s11187-020-00445-6
Borgatti SP Everett MG Freeman LC Ucinet for windows: software for social network analysis 2002 Harvard Analytic Technologies
Boxenbaum E Rouleau L New knowledge products as bricolage: Metaphors and scripts in organizational theory Acad Manag Rev 2011 36 2 272 296 10.5465/amr.2009.0213
Brown R Rocha A Cowling M Financing entrepreneurship in times of crisis: exploring the impact of COVID-19 on the market for entrepreneurial finance in the United Kingdom Int Small Bus j: Res Entrep 2020 38 5 380 390 10.1177/0266242620937464
Brush CG Carter NM Gatewood EJ Greene PG Hart MM The use of bootstrapping by women entrepreneurs in positioning for growth Ventur Cap 2006 8 1 15 31 10.1080/13691060500433975
Carter RB van Auken H Bootstrap financing and owners’ perceptions of their business constraints and opportunities Entrep Reg Dev 2005 17 2 129 144 10.1080/08985620500067548
Carter NM Brush CG Greene PG Gatewood E Hart MM Women entrepreneurs who break through to equity financing: the influence of human, social and financial capital Ventur Cap 2003 5 1 1 28 10.1080/1369106032000082586
Cassar G The financing of business start-ups J Bus Ventur 2004 10.1016/S0883-9026(03)00029-6
Cheung CWM Kwong C Manzoor H Rashid MU Bhattarai C Kim Y-A The co-creation of social ventures through bricolage, for the displaced, by the displaced Int J Entrep Behav Res 2019 25 5 1093 1127 10.1108/IJEBR-03-2018-0140
Churchill NC Thorne JR Alternative financing for entrepreneurial ventures Entrep Theory Pract 1989 10.1177/104225878901300302
Cohen J A coefficient of agreement for nominal scales Educ Psychol Meas 1960 10.1177/001316446002000104
Cooper AC Gimeno-Gascon FJ Woo CY Initial human and financial capital as predictors of new venture performance J Bus Ventur 1994 10.1016/0883-9026(94)90013-2
Dabić M Maley J Dana LP Novak I Pellegrini MM Caputo A Pathways of SME internationalization: a bibliometric and systematic review Small Bus Econ 2020 55 3 705 725 10.1007/S11187-019-00181-6/FIGURES/5
Denyer D and Tranfield D (2009) Producing a systematic review. In: The SAGE handbook of organizational research methods
Desa G Resource mobilization in international social entrepreneurship: bricolage as a mechanism of institutional transformation Entrep: Theory Pract 2012 36 4 727 751 10.1111/j.1540-6520.2010.00430.x
Desa G Basu S Optimization or bricolage? Overcoming resource constraints in global social entrepreneurship Strateg Entrep J 2013 7 1 26 49 10.1002/sej.1150
Desa G Koch JL Scaling social impact: building sustainable social ventures at the base-of-the-pyramid J Soc Entrep 2014 10.1080/19420676.2013.871325
di Stefano G Peteraf M Verona G Dynamic capabilities deconstructed z : a bibliographic investigation into the origins, development, and future directions of the research domain Ind Corp Chang 2010 19 4 1187 1204 10.1093/icc/dtq027
DiMaggio PJ Powell WW The iron cage revisited: institutional isomorphism and collective rationality in organizational fields Am Sociol Rev 1983 10.2307/2095101
Donthu N Kumar S Mukherjee D Pandey N Lim WM How to conduct a bibliometric analysis: an overview and guidelines J Bus Res 2021 10.1016/j.jbusres.2021.04.070
Duran-Sanchez A Alvarez García J del Río-Rama M de La C Ratten V Trends and changes in the International journal of entrepreneurial behaviour & research: a bibliometric review Int J Entrep Behav Res 2019 10.1108/IJEBR-04-2019-0249
Duymedjian R Rüling C-C Towards a foundation of bricolage in organization and management theory Organ Stud 2010 31 2 133 151 10.1177/0170840609347051
Ebben JJ Bootstrapping and the financial condition of small firms Int J Entrep Behav Res 2009 15 4 346 363 10.1108/13552550910967930
Ebben J Johnson A Bootstrapping in small firms: an empirical analysis of change over time J Bus Ventur 2006 21 6 851 865 10.1016/j.jbusvent.2005.06.007
Ernst H Kahle HN Dubiel A Prabhu J Subramaniam M The antecedents and consequences of affordable value innovations for emerging markets J Prod Innov Manag 2015 32 1 65 79 10.1111/jpim.12171
Evers N O’Gorman C Improvised internationalization in new ventures: the role of prior knowledge and networks Entrep Reg Dev 2011 23 7–8 549 574 10.1080/08985621003690299
Ferneley E Bell F Using bricolage to integrate business and information technology innovation in SMEs Technovation 2006 26 2 232 241 10.1016/j.technovation.2005.03.005
Fisher G Effectuation, causation, and bricolage: A behavioral comparison of emerging theories in entrepreneurship research Entrep: Theory Pract 2012 36 5 1019 1051 10.1111/j.1540-6520.2012.00537.x
Franquesa J Brandyberry A Organizational slack and information technology innovation adoption in SMEs Int J E-Bus Res 2009 5 1 25 48 10.4018/jebr.2009010102
Freeman J Carroll GR Hannan MT The liability of newness: age dependence in organizational death rates Am Sociol Rev 1983 10.2307/2094928
Granovetter M Economic action and social structure: the problem of embeddedness Am J Sociol 1985 10.1086/228311
Grégoire DA Cherchem N A structured literature review and suggestions for future effectuation research Small Bus Econ 2020 10.1007/s11187-019-00158-5
Grichnik D Singh L Resource bootstrapping of nascent entrepreneurs: conscious entrepreneurial decision or forced reaction? Front Entrep Res 2010 30 12 3
Grichnik D Brinckmann J Singh L Manigart S Beyond environmental scarcity: human and social capital as driving forces of bootstrapping activities J Bus Ventur 2014 29 2 310 326 10.1016/j.jbusvent.2013.02.006
Habib R Afzal MT Sections-based bibliographic coupling for research paper recommendation Scientometrics 2019 10.1007/s11192-019-03053-8
Hair JF Black WC Babin BJ Anderson RE Multivariate data analysis Vectors 2010 10.1016/j.ijpharm.2011.02.019
Halme M Lindeman S Linna P Innovation for inclusive business: intrapreneurial bricolage in multinational corporations J Manage Stud 2012 49 4 743 784 10.1111/j.1467-6486.2012.01045.x
Hambrick DC Mason PA Upper echelons: the organization as a reflection of its top managers Acad Manag Rev 1984 10.5465/amr.1984.4277628
Harrison RT Mason CM Girling P Financial bootstrapping and venture development the software industry Entrep Reg Dev 2004 16 4 307 333 10.1080/0898562042000263276
Hjørland B Citation analysis: a social and dynamic approach to knowledge organization Inf Process Manag 2013 10.1016/j.ipm.2013.07.001
Hoang H Antoncic B Network-based research in entrepreneurship a critical review J Bus Ventur 2003 10.1016/S0883-9026(02)00081-2
Honigmann JJ Levi-Strauss C The savage mind Soc Forces 1967 10.2307/2575938
Hood WW Wilson CS The literature of bibliometrics, scientometrics, and informetrics Scientometrics 2001 10.1023/A:1017919924342
Horváth K Network-based bootstrapping and performance in SMEs: the moderating role of entrepreneurship training Int J Bus Environ 2019 10 3 227 246 10.1504/IJBE.2019.097972
Hota PK Mitra S Qureshi I Adopting bricolage to overcome resource constraints: the case of social enterprises in rural India Manag Organ Rev 2019 15 2 371 402 10.1017/mor.2019.19
Hsiao RL, Ou SH, Chen HH (2014) Recreating something from everything: new product innovation through social construction of frugal resources. In: ICMIT 2014: 2014 IEEE international conference on management of innovation and technology, pp 248–255. 10.1109/ICMIT.2014.6942433
Iqbal Q Ahmad NH Halim HA Insights on entrepreneurial bricolage and frugal innovation for sustainable performance Bus Strat Dev 2021 4 3 237 245 10.1002/bsd2.147
Jacob F Evolution and tinkering Science 1977 10.1126/science.860134
Jayawarna D Jones O Macpherson A New business creation and regional development: enhancing resource acquisition in areas of social deprivation Entrep Reg Dev 2011 23 9–10 735 761 10.1080/08985626.2010.520337
Jayawarna D Jones O Marlow S The influence of gender upon social networks and bootstrapping behaviours Scand J Manag 2015 31 3 316 329 10.1016/j.scaman.2015.06.002
Jones O Jayawarna D Resourcing new businesses: Social networks, bootstrapping and firm performance Ventur Cap 2010 12 2 127 152 10.1080/13691061003658886
Jonsson S Lindbergh J The development of social capital and financing of entrepreneurial firms: from financial bootstrapping to bank funding Entrep Theory Pract 2013 37 4 661 686 10.1111/j.1540-6520.2011.00485.x
Kaiser HF The application of electronic computers to factor analysis Educ Psychol Meas 1960 10.1177/001316446002000116
Kessler MM Bibliographic coupling between scientific papers Am Doc 1963 10.1002/asi.5090140103
Keupp MM Gassmann O The past and the future of international entrepreneurship: a review and suggestions for developing the field J Manag 2009 10.1177/0149206308330558
Korsgaard S Müller S Welter F It’s right nearby: how entrepreneurs use spatial bricolage to overcome resource constraints Entrep Reg Dev 2021 33 1–2 147 173 10.1080/08985626.2020.1855479
Kuckertz A Brändle L Gaudig A Hinderer S Morales Reyes CA Prochotta A Steinbrink KM Berger ESC Startups in times of crisis: a rapid response to the COVID-19 pandemic J Bus Ventur Insights 2020 10.1016/j.jbvi.2020.e00169
Kuhn TS (1962) The structure of scientific revolutions. Chicago (University of Chicago Press) 1962. In Original edition
Landis JR Koch GG The measurement of observer agreement for categorical data Biometrics 1977 10.2307/2529310
Lin TY Cheng YY Exploring the knowledge network of strategic alliance research: a co-citation analysis Int J Electron Bus Manag 2010 8 2 152 160
Linna P Bricolage as a means of innovating in a resource-scarce environment: a study of innovator-entrepreneurs at the bop J Dev Entrep 2013 10.1142/S1084946713500155
Loasby BJ Leibenstein H Beyond economic man. A new foundation for microeconomics Econ J 1976 10.2307/2231480
Löfqvist L Product innovation in small companies: managing resource scarcity through financial bootstrapping Int J Innov Manag 2017 10.1142/S1363919617500207
Lumpkin GT Dess GG Clarifying the entrepreneurial orientation construct and linking it to performance Acad Manag Rev 1996 10.5465/AMR.1996.9602161568
Ma T Yang S Impact of entrepreneurial orientation on the performance of new ventures in China: the roles of entrepreneurial bricolage and opportunity recognition Chin Manag Stud 2022 16 4 924 941 10.1108/CMS-11-2020-0478/FULL/PDF
Mair J Marti I Entrepreneurship in and around institutional voids: a case study from Bangladesh J Bus Ventur 2009 24 5 419 435 10.1016/j.jbusvent.2008.04.006
Malmström M Typologies of bootstrap financing behavior in small ventures Ventur Cap 2014 16 1 27 50 10.1080/13691066.2013.863064
March JG Exploration and exploitation in organizational learning Org Sci 1991 10.1287/orsc.2.1.71
McCain KW Mapping authors in intellectual space: a technical overview J Am Soc Inf Sci 1990 10.1002/(SICI)1097-4571(199009)41:6<433::AID-ASI11>3.0.CO;2-Q
Merigó JM Gil-Lafuente AM Yager RR An overview of fuzzy research with bibliometric indicators Appl Soft Comput J 2015 10.1016/j.asoc.2014.10.035
Meyer JW Rowan B Institutionalized organizations: formal structure as myth and ceremony Am J Sociol 1977 10.1086/226550
Miao C Rutherford MW Pollack JM An exploratory meta-analysis of the nomological network of bootstrapping in SMEs J Bus Ventur Insights 2017 8 1 8 10.1016/j.jbvi.2017.04.002
Michaelis TL Carr JC Scheaf DJ Pollack JM The frugal entrepreneur: a self-regulatory perspective of resourceful entrepreneurial behavior J Bus Ventur 2020 10.1016/j.jbusvent.2019.105969
Miner AS Bassoff P Moorman C Organizational improvisation and learning: a field study Adm Sci Q 2001 10.2307/2667089
Moghaddam K Aidov A DuVal C Azarpanah S High-growth entrepreneurial firm funding: a qualitative study of native-born and immigrant entrepreneurs Ventur Cap 2017 19 1–2 75 94 10.1080/13691066.2016.1256295
Myers SC Majluf NS Corporate financing and investment decisions when firms have information that investors do not have J Financ Econ 1984 10.1016/0304-405X(84)90023-0
Nath R Jackson WM Productivity of management information systems researchers: does Lotka’s law apply? Inf Process Manag 1991 10.1016/0306-4573(91)90049-R
Neeley L van Auken H The relationship between owner characteristics and use of bootstrap financing methods J Small Bus Entrep 2009 10.1080/08276331.2009.10593462
Neeley L van Auken H Differences between female and male entrepreneurs’ use of bootstrap financing J Dev Entrep 2010 15 1 19 34 10.1142/S1084946710001439
Nelson R Lima E Effectuations, social bricolage and causation in the response to a natural disaster Small Bus Econ 2020 54 3 721 750 10.1007/S11187-019-00150-Z/TABLES/10
Nicholls-Nixon CL Castilla JAD Garcia JS Pesquera MR Latin america management research: review, synthesis, and extension J Manag 2011 10.1177/0149206311403151
Paul J Rosado-Serrano A Gradual Internationalization versus born-global/international new venture models: a review and research agenda Int Market Rev 2019 10.1108/IMR-10-2018-0280
Penrose E The theory of the growth of the firm Long Range Plan 1996 10.1016/s0024-6301(96)90295-2
Perry JT Chandler GN Markova G Entrepreneurial effectuation: a review and suggestions for future research Entrep Theory Pract 2012 10.1111/j.1540-6520.2010.00435.x
Persson O, Danell R, Schneider JW (2009) How to use Bibexcel for various types of bibliometric analysis. Celebrating scholarly communication studies: a festschrift for olle persson at his 60th birthday
Peterson RA A meta-analysis of variance accounted for and factor loadings in exploratory factor analysis Market Lett 2000 10.1023/A:1008191211004
Pfeffer J Salancik G External control of organizations-resource dependence perspective Org Beh 2: Essent Theor Process Struct 2015 1 1 10
Phillips N Tracey P Opportunity recognition, entrepreneurial capabilities and bricolage: connecting institutional theory and entrepreneurship in strategic organization Strateg Organ 2007 5 3 313 320 10.1177/1476127007079956
Pilkington A Chai KH Research themes, concepts and relationships: a study of International journal of service industry management (1990–2005) Int J Serv Ind Manag 2008 10.1108/09564230810855725
Pilkington A Liston-Heyes C Is production and operations management a discipline? A citation/co-citation study Int J Oper Prod Manag 1999 1 1 10 10.1108/01443579910244188
Politis D Winborg J Dahlstrand ÅL Exploring the resource logic of student entrepreneurs Int Small Bus J 2012 30 6 659 683 10.1177/0266242610383445
Rahman SA Taghizadeh SK Alam MMD Khan GM The functionality of entrepreneurial passion and entrepreneurial bricolage on micro-entrepreneur’s wellbeing J Small Bus Strateg 2020 30 3 47 64
Read S Song M Smit W A meta-analytic review of effectuation and venture performance J Bus Ventur 2009 10.1016/j.jbusvent.2008.02.005
Rey-Martí A Ribeiro-Soriano D Palacios-Marqués D A bibliometric analysis of social entrepreneurship J Bus Res 2016 10.1016/j.jbusres.2015.10.033
Rossetto DE Bernardes RC Borini FM Gattaz CC Structure and evolution of innovation research in the last 60 years: review and future trends in the field of business through the citations and co-citations analysis Scientometrics 2018 10.1007/s11192-018-2709-7
Rutherford MW Coombes SMT Mazzei MJ The impact of bootstrapping on new venture performance and survival: a longitudinal analysis Front Entrep Res 2012 32 12 4
Rutherford MW Pollack JM Mazzei MJ Sanchez-Ruiz P Bootstrapping: reviewing the literature, clarifying the construct, and charting a new path forward Group Org Manag 2017 42 5 657 706 10.1177/1059601117730574
Rutherford M Phillips D Arteaga-Fonseca J The bootstrapping-bricolage interface De Gruyter Handb Entrep Financ 2022 10.1515/9783110726312-004/HTML
Salunke S Weerawardena J McColl-Kennedy JR Competing through service innovation: the role of bricolage and entrepreneurship in project-oriented firms J Bus Res 2013 66 8 1085 1097 10.1016/j.jbusres.2012.03.005
Sarasvathy SD Causation and effectuation: toward a theoretical shift from economic inevitability to entrepreneurial contingency Acad Manag Rev 2001 10.5465/AMR.2001.4378020
Schulte-Holthaus S Entrepreneurship in the creative industries: a literature review and research agenda FGF Stud Small Bus Entrep 2018 10.1007/978-3-319-65506-2_7
Seelos C Mair J Social entrepreneurship: creating new business models to serve the poor Bus Horiz 2005 10.1016/j.bushor.2004.11.006
Senyard J Baker T Davidsson P Entrepreneurial bricolage: towards systematic empirical testing Front Entrep Res 2009 29 5 5
Senyard J Baker T Steffens P Davidsson P Bricolage as a path to innovativeness for resource-constrained new firms J Prod Innov Manag 2014 31 2 211 230 10.1111/jpim.12091
Shane S Prior knowledge and the discovery of entrepreneurial opportunities Org Sci 2000 10.1287/orsc.11.4.448.14602
Shane S Venkataraman S The promise of entrepreneurship as a field of research Acad Manag Rev 2000 10.5465/AMR.2000.2791611
Short JC Moss TW Lumpkin GT Research in social entrepreneurship: past contributions and future opportunities Strat Entrep J 2009 10.1002/sej.69
Singh S Dhir S Structured review using TCCM and bibliometric analysis of international cause-related marketing, social marketing, and innovation of the firm Int Rev Publ Nonprofit Market 2019 10.1007/s12208-019-00233-3
Sirmon DG Hitt MA Ireland RD Managing firm resources in dynamic environments to create value: looking inside the black box Acad Manag Rev 2007 10.5465/AMR.2007.23466005
Stenholm P Renko M Passionate bricoleurs and new venture survival J Bus Ventur 2016 31 5 595 611 10.1016/j.jbusvent.2016.05.004
Stinchfield BT Nelson RE Wood MS Learning From Levi-Strauss’ legacy: art, craft, engineering, bricolage, and brokerage in entrepreneurship Entrep: Theory Pract 2013 37 4 889 921 10.1111/j.1540-6520.2012.00523.x
Su Z Yang J Wang Q The Effects of top management team heterogeneity and shared vision on entrepreneurial bricolage in new ventures: an attention-based view IEEE Trans Eng Manage 2020 10.1109/TEM.2020.2984628
Teece DJ Pisano G Shuen A Dynamic capabilities and strategic management Strat Manag J 1997 10.1002/(SICI)1097-0266(199708)18:7<509::AID-SMJ882>3.0.CO;2-Z
Terjesen S Hessels J Li D Comparative international entrepreneurship: a review and research agenda J Manag 2016 10.1177/0149206313486259
Tindiwensi CK Abaho E Munene JC Muhwezi M Nkote IN Entrepreneurial bricolage in smallholder commercial farming: a family business perspective J Fam Bus Manag 2021 11 4 423 439 10.1108/JFBM-04-2020-0036
van Auken H Differences in the usage of bootstrap financing among technology-based versus nontechnology-based firms J Small Bus Manage 2005 43 1 93 103 10.1111/j.1540-627x.2004.00127.x
Vanacker T Manigart S Meuleman M Sels L A longitudinal study on the relationship between financial bootstrapping and new venture growth Entrep Reg Dev 2011 23 9–10 681 705 10.1080/08985626.2010.502250
Vasconcelos Scazziota V Andreassi T Ribeiro Serra FA Guerrazzi L Expanding knowledge frontiers in entrepreneurship: examining bricolage and effectuation Int J Entrep Beh Res 2020 26 5 1043 1065 10.1108/IJEBR-11-2019-0638
Verbeek A Debackere K Luwel M Zimmermann E Measuring progress and evolution in science and technology—I: the multiple uses of bibliometric indicators Int J Manag Rev 2002 4 2 179 211 10.1111/1468-2370.00083
Vogel R Güttel WH The dynamic capability view in strategic management: a bibliometric review Int J Manag Rev 2013 10.1111/ijmr.12000
Waleczek P Zehren T Flatten TC Start-up financing: How founders finance their ventures’ early stage Manag Decis Econ 2018 39 5 535 549 10.1002/mde.2925
Watkins MW Exploratory factor analysis: a guide to best practice J Black Psychol 2018 10.1177/0095798418771807
Weick KE The collapse of sensemaking in organizations: the mann gulch disaster Adm Sci Q 1993 10.2307/2393339
Weigand C Schulte R The effects of managerial preferences on the financial behaviour of small firms: a demand-side perspective Int J Entrep Ventur 2020 12 5 522 546 10.1504/IJEV.2020.111534
Welter C Mauer R Wuebker RJ Bridging behavioral models and theoretical concepts: effectuation and bricolage in the opportunity creation framework Strateg Entrep J 2016 10 1 5 20 10.1002/sej.1215
Wernerfelt B A resource-based view of the firm Strat Manag J 1984 10.1002/smj.4250050207
Wierenga M Uncovering the scaling of innovations developed by grassroots entrepreneurs in low-income settings Entrep Reg Dev 2020 32 1–2 63 90 10.1080/08985626.2019.1640478
Wiklund J Shepherd D Entrepreneurial orientation and small business performance: a configurational approach J Bus Ventur 2005 20 1 71 91 10.1016/J.JBUSVENT.2004.01.001
Williamson OE Transaction-cost economics : the governance of contractual relations transactions-cost economics J Law Econ 1979 22 2 223 261 10.1086/466942
Winborg J Use of financial bootstrapping in new businesses: a question of last resort? Ventur Cap 2009 11 1 71 83 10.1080/13691060802351248
Winborg J Landström H Financial bootstrapping in small businesses: examining small business managers’ resource acquisition behaviors J Bus Ventur 2001 16 3 235 254 10.1016/S0883-9026(99)00055-5
Witt P Entrepreneurs’ networks and the success of start-ups Entrep Reg Dev 2004 10.1080/0898562042000188423
Wu L Liu H Zhang J Bricolage effects on new-product development speed and creativity: the moderating role of technological turbulence J Bus Res 2017 70 127 135 10.1016/j.jbusres.2016.08.027
Zahra SA Gedajlovic E Neubaum DO Shulman JM A typology of social entrepreneurs: motives, search processes and ethical challenges J Bus Ventur 2009 10.1016/j.jbusvent.2008.04.007
Zhao D Towards all-author co-citation analysis Inf Process Manag 2006 10.1016/j.ipm.2006.03.022
Zupic I Čater T Bibliometric methods in management and organization Org Res Methods 2015 10.1177/1094428114562629
| 0 | PMC9707410 | NO-CC CODE | 2022-12-01 23:20:23 | no | Manag Rev Q. 2022 Nov 29;:1-34 | utf-8 | null | null | null | oa_other |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.